Responsive Documents - CREW: FTC: Regarding Indoor Tanning Association: 5/15/14 - Batch Two, Part Two

Dietary Supplement Fact Sheet: Vitamin D
pharmacist, or other qualified health professional about the appropriateness of taking dietary supplements and their potential
interactions with medications.
Offi ce of Dietary Supplements
National Insti tutes of Health
Bethesda. Maryland 20892 USA
Web: ht!D:I/ods ad njh goy
E mail : ods@nih.gov
Document Last Updated: 11/13/2009 2:37 PM
http://dietary-supplements.info.nih.gov/factsheets!VitaminD_pf.asp[ l 2/l/2009 8:09: 29 PM]
OFFICT OF
DII:.IARY
SUPPLE tENTS
Dietary Supplement Fact Sheet
Table of Contents
Introduction
Reference Intakes
Sources of Vitamin D
Vitamin D Intakes and Status
Vitamin D Deficiency
Groups at Risk of Vitamin D Inadequacy
Vitamin D and Health
Health Risks from Excessive Vitamin D
Interactions with Medications
Vitamin D and Healthful Diets
References
Introduction
Vitamin D
Vitamin D is a fat-soluble vitamin that is naturally present in very few foods, added to others, and available as a dietary
supplement. It is also produced endogenously when ultraviolet rays from sunlight strike the skin and trigger vitamin D synthesis
[1,2]. Vitamin D obtained from sun exposure, food, and supplements is biologically inert and must undergo two hydroxylations in
the body for activation. The first occurs in the liver and converts vitamin D to 25-hydroxyvitamin D [25(0H)D], also known as
calcidiol. The second occurs primarily in the kidney and forms the physiologicall y active 1,25-dihydroxyvitamin D [1,25(0H}2D], also
known as calcitriol [3].
Vitamin D is essential for promoting calcium absorption in the gut and maintaining adequate serum calcium and phosphate
concentrations to enable normal mineralization of bone and prevent hypocalcemic tetany. It is also needed for bone growth and
bone remodeling by osteoblasts and osteoclasts [3,4,5] . Without sufficient vitamin D, bones can become thin, brittle, or misshapen.
Vitamin D sufficiency prevents rickets in children and osteomalacia in adults [2,6, 7] . Together with calcium, vitamin D also helps
protect older adults from osteoporosis.
Vitamin D has other roles in human health, including modulation of neuromuscular and immune function and reduction of
inflammation. Many genes encoding proteins that regulate cell proliferation, differentiation, and apoptosis are modulated in part by
vitamin D [3,5,8,9] . Many laboratory-cultured human cells have vitamin D receptors and some convert 25(0H)D to 1,25(0H}zD [10] .
It remains to be determined whether cells with vitamin D receptors in the intact human carry out this conversion.
Serum concentration of 25(0H)D is the best indicator of vitamin D status. It reflects vitamin D produced cutaneously and that
obtained from food and supplements [4] and has a fairly long circulating half-life of 15 days [11 ]. However, serum 25(0H)D levels do
not indicate the amount of vi tamin D stored in other body tissues. Circulating 1,25(0H}zD is generally not a good indicator of
vitamin D status because it has a short half -life of 15 hours and serum concentrations are closely regulated by parathyroid hormone,
calcium, and phosphate [ 11]. Levels of 1,25(0H}zD do not typically decrease until vitamin D deficiency is severe [ 5, 10].
There is considerable discussion of the serum concentrations of 25(0H)D associated with deficiency (e.g. , rickets), adequacy for
bone health, and optimal overall health (Table 1 ). A concentration of <15 nanograms per milliliter (ng/ml) (or <37 .5 nanomoles per
liter [nmoi/L]) is generally considered inadequate; concentrations >15 ng/ml (>37 . 5 nmoi/L) are recommended. Higher levels are
proposed by some (>30 ng/ml or >75 nmoi/L) as desirable for overall health and disease prevention [12], but insufficient data are
available to support them [ 13]. Serum concentrations of 25(0H)D consistently >200 ng/ml (>500 nmoi/L) are potentially toxic.
Ta bl d e 1: Serum 25-Hyc roxyvitamin D [25(0H}D] Concentrations an d I h* Heat
I ngtmL ** II nmoltL ** II
Health status
1<10-11 11<25-27.5
with vitamin D deficiency, leading to rickets in infants and children and osteomalacia in
adults [ 4, 13]
1<10-15 11<25-37. 5 !I Generally considered inadequate for bone and overall health in healthy individual s [4,13]
1=15 11 =37.5 !IGenerally considered adequate for bone and overall health in healthy individuals [4]
II II II
http://dietary-supplements.info.nih.gov/factsheets!VitaminD_pf.asp[l2/l/2009 8:09:29 PM]
I
I
I
Consistently Consistently Considered potentially toxic, leading to hypercalcemia and hyperphosphatemia, although human data
>200 >500 are limited. In an ani mal model, concentrations =400 ng/ml (=1,000 nmoi/L) demonstrated no toxicity
[11,14].
. . .
*Serum concentrations of 25(0H)D are reported m both nanograms per millil iter (ng/ml) and nanomoles per liter (nmoi/L) .
** 1 ng/ml = 2.5 nmoi/L
An additional compl ication in assessing vitamin D status is in the actual measurement of serum concentrations of 25(0H)D.
Considerable variability exists among the various assays available and among laboratories that conduct the analyses [15, 16, 17]. This
means that compared to the actual concentration of 25(0H)D in a sample of blood serum, a falsely low or falsely high value may be
obtained depending on the assay or laboratory used [18]. A standard reference material for 25(0H)D became avail abl e in july 2009
that will now permit standardization of values across laboratories [19].
Reference Intakes
Intake reference values for vi t amin D and other nutrients are provided in the Dietary Reference Intakes (ORis) developed by the
Food and Nutrition Board (FNB) at the Institute of Medicine of The National Academies (formerly National Academy of Sciences) [4].
DRI is the general term for a set of reference val ues used to plan and assess nutrient intakes of healthy people. These val ues, which
vary by age and gender [4], include:
Recommended Dietary Allowance (RDA): average daily level of intake sufficient to meet the nutrient requirements of nearly
all (97%-98%) healthy people.
Adequate Int ake (AI): establi shed when evidence is insufficient t o develop an RDA and is set at a level assumed to ensure
nutritional adequacy.
Tolerable Upper Intake Level (UL): maximum daily intake unlikely to cause adverse health effects [4].
The FNB established an AI for vitami n D that represents a dail y intake that is sufficient to maintain bone health and normal calcium
metabolism i n healthy people. Ais for vitami n D are listed in both micrograms (meg) and International Units (IUs); the biological
activity of 1 meg is equal to 40 IU (Table 2). The Ais for vitamin D are based on the assumption that the vitamin is not synthesized
by exposure to sunl ight [4].
bl d k Ta e 2: A equate Inta es (Als) or VItamin D [ 4]
I
Age IIChildrenll Men IIWomeniiPregnancyiiLactationl
meg
B1rth to 13 years (
200
IU)
ID O l II I
114-18 years
5 meg 5 meg ,,,5 meg
(200 IU) (200 IU) (200 IU)

(200 IU)
119-50 years
5 meg 5 meg

(200 IU) (200 IU) (200 IU) (200 IU)
151-70 years

10 meg 10 meg
I II I
(400 IU) (400 IU)
171+ years

15 meg 15 meg
I II I
(600 IU) (600 IU)
In 2008, the American Academy of Pediatrics (MP) issued recommended intakes for vitamin D that exceed those of FNB [20]. The
AAP recommendat ions are based on evidence from more recent cl inical trials and the history of safe use of 400 IU/day of vitamin D
in pediatric and adolescent populations. MP recommends that exclusively and partially breastfed infants receive supplements of
400 IU/day of vitamin D shortly after birth and continue to receive these supplements until they are weaned and consume =1 ,000
mllday of vitamin D-fortified formula or whole milk [20]. (All formulas sold in the Uni t ed States provide =400 IU vitamin 0
3
per
liter, and the majority of vitamin D-only and multivitamin liquid supplements provide 400 IU per serving.) Simil arly, all non-
breastfed infants i ngesting <1 ,000 mllday of vitamin D-fortified formula or milk should receive a vitamin D suppl ement of 400
IU/day. MP also recommends that older children and adolescents who do not obtain 400 IU/day through vitamin D-fortified milk
and foods should take a 400 IU vitamin D supplement dai ly [20].
The FNB established an expert committee in 2008 to review the ORis for vitamin D (and calcium). The current DR!s for this nutri ent
were established in 1997, and since t hat time substantial new research has been published to justify a reevaluation of adequate
vitamin D intakes for healthy populations. Determinations of ORis are based on i ndicators of adequacy or hazard; dose-response
curves; health outcomes; life-stage groups; and relations between intakes, biomarkers, and outcomes. For vitamin D, the FNB
committee will focus on (1) effects of circulating concentrations of 25(0H)D on health outcomes, (2) effects of vitamin D intakes on
circulating 25(0H)D and on health outcomes, and (3) levels of intake associated with adverse effects [21]. The FNB expects to issue
its report, updat ing as appropriate the ORis for vitamin D and calcium, by May 2010 [22].
Sources of Vitamin D
Food
Very few foods in nature contain vitamin D. The flesh of fish (such as salmon, tuna, and mackerel) and fish liver oils are among the
best sources [4]. Small amounts of vitamin Dare found in beef liver, cheese, and egg yolks. Vitamin Din these foods is primarily in
the form of vitamin D3 (cholecalciferol) and its metabolite 25(0H)D3 [23]. Some mushrooms provide vitamin D2 (ergocalciferol) in
variable amounts [24-26]. Mushrooms with enhanced l evels of vitamin D
2
from being exposed to ultraviolet light under controlled
conditions are also avail able.
Fortified foods provide most of the vitamin Din the American diet [4,26]. For example, almost all of the U.S. milk supply is fortified
with 100 IU/cup of vitamin D (25% of the Dai ly Value or 50% of the AI level for ages 14-50 years). In the 1930s, a mi l k fortification
program was implemented in the United States to combat rickets, then a major public health problem. This program virtually
eliminated the disorder at that time [4, 14]. Other dairy products made from milk, such as cheese and ice cream, are generally not
fortified. Ready-to-eat breakfast cereals often contain added vitamin D, as do some brands of orange juice, yogurt, and margarine.
In the United States, foods all owed to be fortified with vitamin D include cereal flours and related products, milk and products
made from milk, and calcium-fortified fruit juices and drinks [27]. Maximum levels of added vitamin Dare specified by law.
Several food sources of vitamin D are listed in Table 3.
Table 3 Selected Food Sources of Vitamin D [ 30) .
I
Food
IUs per
serving*
II Percent
DV**
lcod liver oil, 1 tablespoon 1,360 340
!salmon (sockeye), cooked, 3 ounces 794 199
Mushrooms that have been exposed to ultraviolet light to increase vitamin D, 3 ounces (not yet
400 100
commonly available)
!Mackerel, cooked, 3 ounces 388 97
!Tuna f ish, canned in water, drained, 3 ounces 154 39
IMilk, nonfat, reduced fat, and whole, vitamin D-fortified, 1 cup 115-124 29-31
Orange juice fortified with vitamin D, 1 cup (check product labels, as amount of added vitamin D
100 25
varies)
Yogurt, fortified with 20% of the DV for vitamin D, 6 ounces (more heavily fortified yogurts provide
80 20
more of the DV)
!Margarine, fortified, 1 tablespoon 60 15
!sardines, canned in oil, drained, 2 sardines 46 12
!Liver, beef, cooked, 3.5 ounces 46 12
Ready-to-eat cereal , fortified with 10% of the DV for vitamin D, 0.75-1 cup (more heavily fortified
40 10
cereals might provide more of the DV)
IEgg, 1 whole (vitamin D is found in yolk) 25 6
!cheese, Swiss, 1 ounce 6 2
*IUs = International Units.
**DV = Daily Value. DVs were developed by the U.S. Food and Drug Administration to help consumers compare t he nutrient
contents of products within the context of a total diet. The DV for vitamin D is 400 IU for adults and chi ldren age 4 and older.
Food labels, however, are not required to list vitamin D content unless a food has been fortified with this nutrient. Foods
providing 20% or more of the DV are considered to be high sources of a nutrient.
The U.S. Department of Agriculture's Nutrient Database Web site, http://www.nal.usda.gov/fnic/foodcomp/search, lists the
nutrient content of many foods and provides a list of foods containing vitamin D:
http;//www.ars.usda.gov/SP2UserFiles/Piace/123545QO/Data/SR22/nutrlist/sr22a324,pdf . A growing number of foods are bei ng
analyzed for vitamin D content. Simpler and faster methods to measure vitamin D in foods are needed, as are food standard
reference materials with certified values for vitamin D to ensure accurate measurements [31 ].
Sun exposure
Most people meet their vitamin D needs through exposure to sunlight [5,31]. Ultraviolet (UV) B radiation with a wavelength of 290-
315 nanometers penetrates uncovered skin and converts cutaneous 7-dehydrocholesterol to previtamin D
3
, which in turn becomes
vitamin D
3
[9,32,33]. Season, geographic latitude, time of day, cloud cover, smog, skin melanin content, and sunscreen are among
the factors that affect UV radiation exposure and vitamin D synthesis [33]. The UV energy above 42 degrees north latitude (a l ine
I
I
I
I
I
I
I
I
I
I
I
I
I
I
approximately between the northern border of California and Boston) is insufficient for cutaneous vitamin D synthesis from
November through February [5]; in far northern latitudes, this reduced intensity lasts for up to 6 months. In the United States,
latitudes below 34 degrees north (a line between Los Angeles and Columbia, South Carolina) allow for cutaneous production of
vitamin D throughout the year [27].
Complete cloud cover reduces UV energy by 50%; shade (including that produced by severe pollution) reduces it by 60% [34]. UVB
radiation does not penetrate glass, so exposure to sunshine indoors through a window does not produce vitamin D [35]. Sunscreens
with a sun protection factor of 8 or more appear to block vitamin D-producing UV rays, although in practice people generally do not
apply sufficient amounts, cover all sun-exposed skin, or reapply sunscreen regularly [36]. Skin likely synthesizes some vitamin D
even when it is protected by sunscreen as typically applied.
The factors that affect UV radiation exposure and research to date on the amount of sun exposure needed to maintain adequate
vitamin D levels make it difficult to provide general guidelines. It has been suggested by some vitamin D researchers, for example,
that approximately 5-30 minutes of sun exposure between 10 AM and 3 PM at least twice a week to the face, arms, legs, or back
without sunscreen usually lead to sufficient vitamin D synthesis and that the moderate use of commercial t anning beds that emit
2%-6% UVB radiation is also effective [10,33]. Individuals with limited sun exposure need to include good sources of vitamin Din
thei r diet or take a supplement.
Despite the importance of the sun to vitamin D synthesis, it is prudent to limit exposure of skin to sunlight [36] and UV radiation
from tanning beds [37]. UV radiation is a carcinogen responsible for most of the estimated 1.5 million skin cancers and the 8,000
deaths due to metastatic melanoma that occur annually in the United States [36]. Lifetime cumulative UV damage to skin is also
largely responsible for some age-associated dryness and other cosmetic changes. It is not known whether a desirable level of regular
sun exposure exists that imposes no (or minimal) risk of skin cancer over time. The American Academy of Dermatology advises that
photoprotective measures be taken, including the use of sunscreen, whenever one is exposed to the sun [38].
Dietary supplements
In supplements and fortified foods, vitamin D is available in two forms, 0
2
(ergocalciferol) and 0
3
(cholecalciferol). Vitamin 0
2
is
manufactured by the UV irradiation of ergosterol in yeast, and vitamin D3 is manufactured by the irradiation of 7-dehydrocholesterol
from lanolin and the chemical conversion of cholesterol [ 10]. The t wo forms have t raditionally been regarded as equivalent based
on their ability to cure rickets, but evidence has been offered that they are metabolized differently. Vitamin 0
3
could be more than
three times as effective as vitamin 0
2
in raising serum 25(0H)D concentrations and maintaining those levels for a longer time, and
its metabolites have superior affinity for vitamin D-binding proteins in plasma [5,39,40]. Because metabolite receptor affinity is not
a functional assessment, as the earlier results for the healing of rickets were, further research is needed on the comparative
physiological effects of both forms. Many supplements are being reformulated to contain vitamin 0
3
instead of vitamin 0
2
[40]. Both
forms (as well as vitamin D in foods and from cutaneous synthesis) effectively raise serum 25(0H)D levels [5].
Vitamin D Intakes and Status
In 1988-1994, as part of the thi rd National Health and Nutrition Examination Survey (NHANES III), the frequency of use of some
vitamin D-containing foods and supplements was examined in 1,546 non-Hispanic African American women and 1,426 non-Hispanic
white women of reproductive age (15-49 years) [41 ]. In both groups, 25(0H)D levels were higher in the fall (after a summer of sun
exposure) and when milk or fortified cereals were consumed more than three times per week. The prevalence of serum
concentrations of 25(0H)D =15 ng/ml (=37 .5 nmoi/L) was 10 times greater for the African American women (42.2%) than for the
white women (4.2%).
The 2000-2004 NHANES provides the most recent data on the vitamin D nutritional status of the U.S. population. Generally, younger
people had higher serum 25(0H)D levels t han older people, males had higher levels than females, and non- Hispanic whites had
higher levels than Mexican Americans, who in turn had higher level s than non-Hispanic blacks. Depending on the population group,
1%-9% had serum 25(0H)D levels <11 ng/ml (<27.5 nmoi/L), 8%-36% had levels <20 ng/ml (<50 nmoi/L), and t he majority (50%-78%)
had levels <30 ng/ml (<75 nmoi/L) [42].
In NHANES 2000-2004, age-adjusted mean serum 25(0H)D concentrations were 2-8 ng/ml (5-20 nmoi/L) lower compared to NHANES
III [ 43]. However, after adjustment for assay shifts, age-adjusted means in NHANES 2000-2004 remained significantly lower (by 2.0-
3.6 ng/ml (5-9 nmoi/L)) in most males, but not in most females. In a study subsample, adjustment for the confounding effects of
assay differences changed mean serum 25(0H)D concentrations by - 4 ng/ml (-10 nmoi/L), and adjustment for changes in the
factors likely related to real changes in vitamin D status (such as body mass index (BMI), milk intake, and sun protection) changed
mean serum 25(0H)D concentrations by 0.4-0.64 ng/ml (1.0-1.6 nmoi/L).
Subsequent to this report, another investigator [44] evaluated vitamin D levels measured in NHANES 2001-2004 compared to NHANES
III and reported a marked decline, leading some to suggest that the majority of children and adults in the United States (and almost
all African Americans and Mexican Americans) are vitamin D insufficient. However, this analysis exaggerates the temporal and
demographic trends in vitamin D status because it uses a higher than usual cutoff to characterize vitamin D insufficiency, does not
separate the independent effects of season and latitude in data and, most seriously, fails to compensate for a change in the
25(0H)D measurement assay used between both sets of NHANES surveys [45]. Over time, mean serum 25(0H)D concentrations in the
United States have declined, but only modestly, when compensating for the assay change [43]. The real decline (-2.0-3.6 ng/ml
(-S-9 nmoi/L)) is likely due to simultaneous increases in BMI, reduced milk intake, and greater use of sun protection in the U.S.
population.
According to NHANES data from 2005-2006, only 29% of adult men and 17% of adult women (ages 19 and older) had intakes of
vitamin D from food alone that exceeded their Ais. Overall in the U.S. population, only about one-third of individuals 1 year of age
and older had vitamin D intakes from food exceeding their respective Ais [46]. However, dietary supplements as well as foods
contribute vitamin D, so both sources must be included to obtain a true picture of total intakes. In 2005-2006, 37% of people in the
United States reported the use of a dietary supplement containing vitamin D. Total intake estimates of vitamin D from both food and
supplements are currently being tabulated by the Office of Dietary Supplements.
Vitamin D Deficiency
Nutrient deficiencies are usually the result of dietary inadequacy, impaired absorption and use, increased requirement, or increased
excretion. A vitamin D deficiency can occur when usual intake is lower than recommended levels over time, exposure to sunlight is
limited, the kidneys cannot convert vitamin D to its active form, or absorption of vitamin D from the digestive tract is inadequate.
Vitamin D-deficient diets are associated with milk allergy, lactose intolerance, and strict vegetarianism [47].
Rickets and osteomalacia are the classical vitamin D deficiency diseases. In children, vitamin D deficiency causes rickets, a disease
characterized by a failure of bone tissue to properly mineralize, resulting in soft bones and skeletal deformities [34]. Rickets was
first described in the mid-17th century by British researchers [34,48]. In the late 19th and early 20th centuries, German physicians
noted that consuming 1-3 teaspoons of cod liver oil per day could reverse rickets [48]. In the 1920s and prior to identification of
the structure of vitamin D and its metabolites, biochemist Harry Steenbock patented a process to impart antirachitic activity to
foods [27]. The process involved the addition of what turned out to be precursor forms of vitamin D followed by exposure to UV
radiation. The fortification of milk with vitamin D has made rickets a rare disease in the United States. However, rickets is still
reported periodically, particularly among African American infants and children [34,48]. A 2003 report from Memphis, for example,
described 21 cases of rickets among infants, 20 of whom were African American [48].
Prolonged exclusive breastfeeding without the AAP-recommended vitamin D supplementation is a significant cause of rickets,
particularly in dark-skinned infants breastfed by mothers who are not vitamin D replete [6]. Additional causes of rickets include
extensive use of sunscreens and placement of children in daycare programs, where they often have less outdoor activity and sun
exposure [34,48]. Rickets is also more prevalent among immigrants from Asia, Africa, and the Middle East, possibly because of
genetic differences in vitamin D metabolism and behavioral differences that lead to less sun exposure [34].
In adults, vitamin D deficiency can lead to osteomalacia, resulting i n weak muscles and bones [6, 7,11 ]. Symptoms of bone pain and
muscle weakness can indicate inadequate vitamin D levels, but such symptoms can be subtle and go undetected in the initial stages.
Groups at Risk of Vitamin D Inadequacy
Obtaining sufficient vitamin D from natural food sources alone can be difficult. For many people, consuming vitamin D-fortified
foods and being exposed to sunlight are essential for maintaining a healthy vitamin D status. In some groups, dietary supplements
might be required to meet the daily need for vitamin D.
Breastfed infants
Vitamin D requirements cannot be met by human mi lk alone [4,49], which provides only about 25 IU/L [SO]. A recent review of
reports of nutritional ricket s found that a majority of cases occurred among young, breastfed African Americans [51]. The sun is a
potential source of vi tamin D, but AAP advises keeping infants out of direct sunlight and having them wear protective clothing and
sunscreen [52]. As noted earlier, MP recommends that exclusively and partially breastfed infants be supplemented with 400 IU of
vitamin D per day [20].
Older adults
Americans aged SO and older are at increased risk of developing vitamin D insufficiency [33]. As people age, skin cannot synthesize
vitamin D as efficiently and the kidney is less able to convert vitamin D to its active hormone form [4,53]. As many as half of older
adults in the United States with hip fractures could have serum 2S(OH)D levels <12 ng/ml (<30 nmoi/L) [5] .
People with limited sun exposure
Homebound individuals, people living in northern latitudes (such as New England and Alaska), women who wear long robes and head
coverings for religious reasons, and people with occupations that prevent sun exposure are unlikely to obtain adequate vitamin D
from sunlight [54,55].
People with dark skin
Greater amounts of the pigment melanin result in darker skin and reduce the skin's ability to produce vitamin D from exposure to
sunlight. Some studies suggest that older adults, especially women, with darker skin are at high risk of developing vitamin D
insufficiency [41 ,56]. However, one group with dark skin, African Americans, general ly has lower levels of 25(0H)D yet develops
fewer osteoporotic fractures than Caucasians (see section below on osteoporosis).
People with fat malabsorption
As a fat-soluble vitamin, vitamin D requires some dietary fat in the gut for absorption. Individuals who have a reduced ability to
absorb dietary fat might require vitamin 0 supplements [57]. Fat malabsorption is associated with a variety of medical conditions
including some forms of liver disease, cystic fibrosis, and Crohn's disease [27].
People who are obese or who have undergone gastric bypass surgery
Individuals with a BMI =30 typically have a low plasma concentration of 25(0H)D [58]; this level decreases as obesity and body fat
increase [59]. Obesity does not affect skin's capacity to synthesize vitamin D, but greater amounts of subcutaneous fat sequester
more of the vitamin and alter its release into the circulation. Even with orally administered vitamin D, BMI is inversely correlated
with peak serum concentrations, probably because some vitamin 0 is sequestered in the larger pools of body fat [58]. Obese
individuals who have undergone gastric bypass surgery may become vitamin D deficient without a sufficient intake of this nutrient
from food or supplements, since part of the upper small intestine where vitamin Dis absorbed is bypassed [60,61].
Vitamin D and Health
Optimal serum concentrations of 25(0H)D for bone and general health throughout life have not been established [5, 1 0] and are
likely to vary at each stage of life, depending on the physiological measures selected. The three-fold range of cut points that have
been proposed by various experts, from 16 to 48 ng/ml (40 to 120 nmoi/L), reflect differences in the functional endpoints chosen
(e.g., serum concentrations of parathyroid hormone or bone fractures), as well as differences in the analytical methods used.
In March 2007, a group of vitamin D and nutrition researchers published a controversial and provocative editorial contending that
the desirable concentration of 25(0H)D is =30 ng/ml {=75 nmoi/L) [12]. They noted that supplemental intakes of 400 IU/day of
vitamin D increase 25(0H)D concentrations by only 2.8-4.8 ng/mL (7-12 nmoi/L) and that daily intakes of approximately 1,700 IU
are needed to raise these concentrations from 20 to 32 ng/ml (50 to 80 nmoi/L).
Osteoporosis
More than 25 million adults in the United States have or are at risk of developing osteoporosis, a disease characterized by fragile
bones that significantly increases the risk of bone fractures [62]. Osteoporosis is most often associated with inadequate calcium
intakes (generally <1 ,000-1,200 mg/day}, but insufficient vitamin D contributes to osteoporosis by reducing calcium absorption [63].
Although rickets and osteomalacia are extreme examples of the effects of vitamin D deficiency, osteoporosis is an example of a
long-term effect of calcium and vitamin 0 insufficiency [64]. Adequate storage levels of vitamin D maintain bone strength and might
help prevent osteoporosis in older adults, nonambulatory individuals who have difficulty exercising, postmenopausal women, and
individuals on chronic steroid therapy [65].
Normal bone is constantly being remodeled. During menopause, the balance between these processes changes, resulting in more
bone being resorbed than rebuilt. Hormone therapy with estrogen and progesterone might be able to delay the onset of
osteoporosis. However, some medical groups and professional societies recommend that postmenopausal women consider using
other agents to slow or stop bone resorption because of the potential adverse health effects of hormone therapy [66-68].
Most supplementation trials of t he effects of vitamin D on bone health also include calcium, so it is not possible to isolate the
effects of each nutrient. The authors of a recent evidence-based review of research concluded that supplements of both vitamin D3
(at 700-800 IU/day) and calcium (500-1 ,200 mg/day) decreased the risk of falls, fractures, and bone loss in elderly individuals aged
62-85 years [5]. The decreased risk of fractures occurred primarily in elderly women aged 85 years, on average, and living in a
nursing home. Women should consult their healthcare providers about their needs for vitamin D (and calcium) as part of an overall
plan to prevent or t reat osteoporosis.
African Americans have lower levels of 25(0H)D than Caucasians, yet they develop fewer osteoporotic fractures. This suggests that
factors other than vitamin D provi de protection [69]. African Americans have an advantage in bone density from early childhood, a
function of their more efficient calcium economy, and have a lower risk of fracture even when they have the same bone density as
Caucasians. They also have a higher prevalence of obesity, and the resulting higher estrogen levels in obese women might protect
t hem from bone loss [69]. Further reducing the risk of osteoporosis in African Americans are their lower levels of bone-turnover
markers, shorter hip-axis length, and superior renal calcium conservation. However, despite this advantage in bone density,
osteoporosis is a signifi cant health problem among African Americans as they age [69].
Cancer
Laboratory and animal evidence as well as epidemiologic data suggest that vitamin 0 status could affect cancer risk. Strong
biological and mechanistic bases indicate that vitamin D plays a role in the prevention of colon, prostate, and breast cancers.
Emerging epidemiologic data suggest that vitamin D has a protective effect against colon cancer, but the data are not as strong for
a protective effect against prostate and breast cancer, and are variable for cancers at other sites [70, 71 ]. Studies do not
consistently show a protective effect or no effect, however. One study of Finnish smokers, for example, found that subjects in t he
highest quintile of baseline vitamin D status have a three-fold higher risk of developing pancreatic cancer [72].
Vitamin D emerged as a protective factor in a prospective, cross-sectional study of 3,121 adults aged =50 years (96% men) who
underwent a colonoscopy. The study found that 10% had at least one advanced cancerous lesion. Those with the highest vitamin D
intakes {>645 IU/day) had a significantly lower risk of these lesions [73]. However, the Women's Health Initiative, in which 36,282
postmenopausal women of various races and ethnicities were randomly assigned to receive 400 IU vitamin D plus 1,000 mg calcium
daily or a placebo, found no significant differences between the groups in the incidence of colorectal cancers over 7 years [74].
More recently, a clinical trial focused on bone health i n 1,179 postmenopausal women residing in rural Nebraska found that subj ects
supplemented daily with calcium (1 ,400-1,500 mg) and vitamin 0
3
(1, 100 IU) had a significantly lower incidence of cancer over 4
years compared to women taking a placebo [64] . The small number of cancers reported (50) precludes generalizing about a
protective effect from either or both nutrients or for cancers at different sites. This caution is supported by an analysis of 16,618
participants in NHANES III, where total cancer mortality was found to be unrelated to baseline vitamin D status [76]. However,
colorectal cancer mortality was inversely related to serum 25(0H)D concentrations.
Further research is needed to determine whether vitamin D inadequacy in particular increases cancer risk, whether greater exposure
to the nutrient is protective, and whether some individuals could be at increased risk of cancer because of vitamin D exposure
[70, 77] .
Other conditions
A growing body of research suggests that vitamin D might play some rol e in the prevention and treatment of type 1 [78] and type 2
diabetes [79]. hypertension [80]. glucose i ntolerance [81]. multiple sclerosis [82], and other medical conditions [83,84]. However,
most evidence for these rol es comes from i n vitro, ani mal, and epidemiol ogical studies, not the randomized cl i ni cal trials
considered to be more definitive. Unti l such trials are conducted, the implications of the avai labl e evidence for public health and
patient care will be debated. A systematic review of health outcomes related to vitamin D and calcium i ntakes, both alone and in
combination, was published in August 2009 [85].
A recent meta-analysis found that use of vitamin D supplements was associated with a reduction in overall mortality from any cause
by a statistically signifi cant 7% [86,87]. The subj ects in these trials were pri mari ly healthy, middl e aged or elderly, and at high risk
of fractures; they took 300-2,000 IU/day of vitamin D supplements.
Health Risks from Excessive Vitamin D
Vitamin D toxicity can cause nonspecific symptoms such as nausea, vomiting, poor appetite, constipation, weakness, and weight loss
[88]. More seriously, it can also raise blood levels of calci um, causing mental status changes such as confusion and heart rhythm
abnormali ties [7]. The use of supplements of both calcium (1 ,000 mg/day) and vitamin D (400 IU/day) by postmenopausal women
was associated with a 17% increase in the risk of kidney stones over 7 years in the Women's Health Initiative [89] . Deposition of
calcium and phosphate in the kidneys and other soft tissues can also be caused by excessive vitamin D levels [47]. A serum 25(0H)D
concentration consistently >200 ng/ml (>500 nmoi/L) is considered to be pot entially toxic [11 ] . In an animal model, concentrations
=400 ng/ml (=1,000 nmoi/L) were not associated with harm [ 14].
Excessive sun exposure does not result in vitamin D toxi city because the sust ained heat on the skin is thought to photodegrade
previtamin 0
3
and vitamin 0
3
as it is formed [10,35]. High intakes of dietary vitami n Dare very unlikely to result in toxicity unless
large amounts of cod liver oil are consumed; toxicity is more likely to occur from high intakes of supplements.
Long-term int akes above the UL increase the risk of adverse health effects [4] (Table 4). Substantially larger doses administered for
a short time or periodically (e.g., 50,000 IU/week for 8 weeks) do not cause toxicity. Rather, the excess is stored and used as
needed to maintain normal serum 25(0H)D concentrations when vitamin D intakes or sun exposure are limited [ 11, 90].
T bl 4 T I bl U a e : o era e 1pper nta e eve s or 1tamm k l I (Uls) f v
. D [4]
I
Age II Childrenll Men II Women IIPregnancyiiLactation I
Birth to 12 months
25 meg
CJCJI II I
(1 ,000 IU)
11-13 years
ISO meg
. (2,000 IU)
CJCJI II I
114+ years meg meg meg
. (2,000 IU) (2,000 IU) (2,000 IU)
meg
(2,000 IU)
Several nutrition scientists recently challenged these Uls, first published in 1997 [90]. They point to newer clinical trials conducted
in healthy adults and conclude that t he data support aULas high as 10,000 IU/day. Although vitamin D supplements above
recommended levels given in clinical trials have not shown harm, most trials were not adequately designed to assess harm [5].
Evidence is not sufficient to determine the potential risks of excess vitamin D in infants, children, and women of reproductive age.
As noted earlier, the FNB is current ly reviewi ng data to determine whether updates to the ORis (incl uding the Uls) for vitamin Dare
appropriate [4].
Interactions with Medications
Vitamin D suppl ements have the potential to interact with several types of medications. A few examples are provided below.
Individuals taking these medications on a regular basis should discuss vitamin D intakes with their healthcare providers.
Steroids
Corticosteroid medications such as prednisone, often prescri bed to reduce inflammation, can reduce calcium absorption [ 91-93] and
impair vitamin D metaboli sm. These effects can further contribut e t o the loss of bone and the development of osteoporosis
associated with their long-term use [ 92, 93].
Other medications
Both the weight-loss drug orli stat (brand names Xenical and aiiPM) and the cholesterol-lowering drug cholestyramine (brand names
Questran, LoCholest, and Prevalite) can reduce the absorption of vitamin D and other fat-soluble vitamins [ 94,95]. Both
phenobarbital and phenytoin (brand name Dilantin), used to prevent and control epi l eptic seizures, increase the hepatic
metabolism of vitamin D t o inactive compounds and reduce calcium absorption [96].
Vitamin D and Healthful Diets
According to the 2005 Dietary Guidelines for Americans, "nutrient needs should be met primarily through consuming foods. Foods
provide an array of nut rient s and other compounds that may have beneficial effects on health. I n cert ain cases, fortified foods and
dietary supplements may be useful sources of one or more nutrients that otherwise might be consumed i n less than recommended
amounts. However, dietary supplements, while recommended in some cases, cannot replace a healthful diet. "
The Dietary Guidelines for Americans describes a healthy di et as one that
Emphasizes a variety of fruits, vegetables, whole grains, and fat-free or low-fat milk and milk products.
Mil k is fortified with vitamin D, as are many ready-to-eat cereals and a few brands of yogurt and orange j uice. Cheese
naturall y contains small amounts of vitamin D.
Incl udes lean meats, poultry, fish, beans, eggs, and nuts.
Fish such as salmon, tuna, and mackerel are very good sources of vitamin D. Small amounts of vitamin D are also found
in beef liver and egg yolks.
Is low in saturated fats, trans fats, cholesterol, salt (sodium}, and added sugars.
Vitamin D is added to some margarines.
Stays within your dail y calorie needs.
For more informat ion about building a healthful diet, refer to the Dietary Guidelines for Americans
(http:/ /www.health.gov/djetaryguidelines/dga2005/document/default.htm) and the U.S. Department of Agriculture's food guidance
system, My Pyramid (http:/ /www.mypyramid. gov).
References
1. Deluca HF, Zierold C. Mechanisms and functions of vitamin D. Nutr Rev 1998;56:54-10. [PubMed abstract]
2. Deluca HF. Overview of general physiologic features and functions of vitamin D. Am j Clin Nut r 2004;80:1689S-96S. [PubMed
abstract]
3. van den Berg H. Bioavailability of vitamin D. Eur j Clin Nutr 1997;51 :576-9. [PubMed abstract]
4. Institute of Medi cine, Food and Nutrition Board. Dietary Reference Intakes: Calcium, Phosphorus, Magnesium, Vitamin D, and
Fluoride. Washington, DC: National Academy Press, 1997.
5. Cranney C, Horsely T, O'Donnell S, Weil er H, Ooi D, Atkinson S, et al. Effectiveness and safety of vitamin D. Evidence
Report/Technology Assessment No. 158 prepared by the University of Ottawa Evidence-based Practice Center under Contract
No. 290-02.0021. AHRQ Publication No. 07-E013. Rockville, MD: Agency for Heal t hcare Research and Qual ity, 2007. [PubMed
abstract]
6. Goldring SR, Krane S, Avioli LV. Disorders of calcification: osteomalacia and rickets. In: DeGroot Lj, Besser M, Burger HG,
jameson j L, Loriaux DL, Marshall j C, et al., eds. Endocrinology. 3rd ed. Philadelphia: WB Saunders, 1995:1204-27.
7. Favus Mj, Christakos S. Pri mer on the Metabolic Bone Diseases and Disorders of Mineral Metaboli sm. 3rd ed. Phi ladelphia, PA:
Lippincott-Raven, 1996.
8. Holick MF. Evol ution and function of vitami n D. Recent results. Cancer Res 2003; 164:3-28. [PubMed abstract]
9. Hayes CE, Hashold FE, Spach KM, Pederson LB. The immunological functions of the vitamin D endocri ne system. Cell Mol Bioi
2003;49:277-300. [ PubMed abstract]
10. Holick MF. Vitamin D deficiency. N Englj Med 2007;357:266-81. [PubMed abstract]
11 . j ones G. Pharmacokinetics of vitamin D toxicity. Am j Clin Nutr 2008;88:582S-6S. [PubMed abstract]
12. Vieth R, Bischoff-Ferrari H, Boucher Bj, Dawson-Hughes B, Garland CF, Heaney RP, et al. The urgent need to recommend an
i ntake of vitamin D that is effective. Am j Clin Nutr 2007;85: 649-50. [PubMed abstract]
13. Scientific Advisory Committee on Nutrition. Update on Vitamin D. Position Statement by the Scientific Advisory Committee on
Nutrition. London: The Stationery Office, Limited, 2007.
14. Shepard RM, Deluca HF. Plasma concentrations of vitamin D3 and its metabolites in the rat as influenced by vitamin D3 or
245-hydroxyvitamin D3 intakes. Arch Biochem Biophys 1980;202:43-53. [PubMed abstract]
15. Carter GD. 25-hydroxyvitamin D assays: the quest for accuracy. Clio Chern 2009;55:1300-02.
16. Hollis BW. Editorial: the determination of circulating 25-hydroxyvitamin D: no easy task. j. Clio Endocrinol Metab
2004;89:3149-3151.
17. Lensmeyer GL, Wiebe DA, Binkley N, Drezner MK. HPLC method for 25-hydroxyvitamin D measurement: comparison with
contemporary assays. Clio Chern 2006;52:1120-26. [PubMed abstract]
18. Binkley N, Krueger D, Cowgill CS, Plum L, Lake E, Hansen KE, et al. Assay variation confounds the diagnosis of
hypovitaminosis D: a call for standardization. J Clio Endocrinol Metab 2004;89:3152-57. [PubMed abstract]
19. National Institute of Standards and Technology. NIST releases vitamin D standard reference material, 2009.
20. Wagner CL, Greer FR; American Academy of Pediatrics Section on Breastfeeding; American Academy of Pediatrics Committee
on Nutrition. Prevention of rickets and vitamin D deficiency in infants. children. and adolescents . Pediatrics 2008;122: 11 42-
1152. [PubMed abstract]
21. Yetley EA, Brule D, Cheney MC, Davis CD, Esslinger KA, Fischer PWF, et al. Dietary Reference Intakes for vitamin D:
justification for a review of the 1997 values. Am J Clio Nutr 2009;89:719-27. [PubMed abstract]
22. Institute of Medicine, Food and Nutrition Board. Pietary Reference Intakes for vitamin p and calcium.
23. Ovesen L, Brot (,Jakobsen J. Food contents and biological activity of 25-hydroxyvitamin D: a vitamin D metabolite to be
reckoned with? Ann Nutr Metab 2003;47:1 07-13. [PubMed abstract]
24. Mattila PH, Piironen VI, Uusi-Rauva EJ, Koivistoinen PE. Vitamin D contents in edible mushrooms. J Agric Food Chern
1994;42:2449-53.
25. Outila TA, Mattila PH, Piironen VI, Lamberg-AIIardt CJ E. Bioavailability of vitamin D from wild edible mushrooms
(Cantharellus tubaeformis) as measured with a human bioassay. Am j Clio Nutr 1999;69:95-8. [PubMed abstract]
26. Calvo MS, Whiting SJ, Barton CN. Vitamin D fortification in the United States and Canada: current status and data needs. Am
j Clio Nutr 2004;80:171 OS-6S. [PubMed abstract]
27. Holick MF. Vitamin D. In: Shils ME, Shike M, Ross AC, Caballero B, Cousins RJ, eds. Modern Nutrition in Health and Disease,
10th ed. Philadelphia: Lippincott Williams & Wilkins, 2006.
28. PenningtonjA, DouglassjS. Bowes and Church's Food Values of Portions Commonly Used. 18th ed. Philadelphia: Lippincott
Williams & Wilkins, 2004.
29. Nutrition Coordinating Center. Nutrition Data System for Research (NDS-R). Version 4.06/34. Minneapolis: University of
Minnesota, 2003.
30. U.S. Department of Agriculture, Agricultural Research Service. USDA Nutrient Database for Standard Reference. Release 22,
2009.
31. Byrdwell we, DeVriesj, Exler J , Harnly JM, Holden J M, Holick MF, et al. Analyzing vitamin Din foods and supplements:
methodologic challenges. Am J Cli o Nutr 2008;88:554S-7S. [PubMed abstract]
32. Holick MF. McCollum Award Lecture, 1994. Vitamin D: new horizons for the 21st century. Amj Clio Nutr 1994;60:619-30.
[PubMed abstract]
33. Holick MF. Vitamin D: the underappreciated D-lightful hormone that is important for skeletal and cellular health. Curr Opin
Endocrinol Diabetes 2002;9:87-98.
34. Wharton B, Bishop N. Rickets. Lancet 2003;362:1389-400. [PubMed abstract]
35. Holick MF. Photobiology of vitamin D. In: Feldman D, PikejW, Glorieux FH, eds. Vitamin D, Second Edition, Volume I.
Burlington, MA: Elsevier, 2005.
36. Wolpowitz D, Gilchrest BA. The vitamin D questions: how much do you need and how should you get it? J Am Acad Dermatol
2006;54:301 -1 7. [PubMed abstract]
37. International Agency for Research on Cancer Working Group on ultraviolet (UV) light and skin cancer. The association of use
of sunbeds with cutaneous malignant melanoma and other skin cancers: a systematic review. Intj Cancer 2006;120:1116-22.
[PubMed abstract]
38. American Academy of Dermatology. Position statement on vitamin D. November 1, 2008.
39. Armas LAG, Hollis BW, Heaney RP. Vitamin D2 is much less effective than vitamin D3 in humans. J Clio Endocrinol Metab
2004;89:5387-91. [ PubMed abstract]
40. Houghton LA, Vieth R. The case against ergocalciferol (vi tamin D2) as a vi tamin supplement. Am J Clio Nutr 2006;84:694-7.
[PubMed abstract]
41. Nesby-O' Dell S, Scanlon KS, Cogswell ME, Gillespie C, Hollis BW, Looker AC, et al. Hypovitaminosis D prevalence and
determinants among African-American and white women of reproductive age: thi rd National Health and Nutrition Examination
Survey, 1988-1994. Am j Clio Nutr 2002;76: 187-92. [PubMed abstract]
42. Yetley EA. Assessing the vitamin D status of the US population. Am J Clio Nutr 2008;88:558S-64S. [PubMed abstract]
43. Looker AC, Pfeiffer CM, Lacher DA, Schleicher RL, Picciano MF, Yetley EA. Serum 25-hydroxyvitamin D status of the US
population: 1988-1994 compared with 2000-2004. Am J Clio Nutr 2008;88:1519-27. [PubMed abstract]
44. Ginde AA, Liu MC, Camargo J r, CA. Demographic differences and trends of vitamin D insufficiency in the US population, 1988-
2004. Arch Intern Med 2009;169:626-32. [PubMed abstract]
45. National Center for Health Statistics. Analytical note for NHANES 2000-2006 and NHANES III 0988 -1994) 25-hydroxyyitamin D
analysis, 2009.
46. Moshfegh A, Goldman J, Ahuja J, Rhodes D, LaComb R. 2009. What We Eat in America, NHANES 2005-2006: Usual Nutrient
Intakes from Food and Water Compared to 1997 Dietary Reference Intakes for Vitamin D, Calcium, Phosphorus, and
Magnesium. U.S. Department of Agriculture, Agricultural Research Service. [http://www.ars.usda.gov/ba/bhnrc/fsrg]
47. Biser-Rohrbaugh A, Hadley-Miller N. Vitamin D deficiency in breast-fed toddlers.J Pediatr Orthop 2001;21:508-11. [PubMed
abstract]
48. Chesney R. Rickets: an old form for a new century. Pediatr Int 2003;45: 509-11. [PubMed abstract]
49. Picciano MF. Nutrient composition of human milk. Pediatr Clin North Am 2001 ;48:53-67. [PubMed abstract]
50. Gartner LM, Greer FR, American Academy of Pediatrics Committee on Nutrition. Prevention of rickets and vitamin D
deficiency: new guidelines for vitamin D intake. Pediatrics 2003:111 :908-10. [PubMed abstract]
51. Weisberg P, Scanlon KS, Li R, Cogswell ME. Nutritional r ickets among children in the United States: review of cases reported
between 1986 and 2003. Am J Clin Nutr 2004;80:16975-7055. [PubMed abstract]
52. American Academy of Pediatrics Committee on Environmental Health. Ultraviolet light: a hazard to chi ldren. Pediatrics
1999;104:328-33. [PubMed abstract]
53. Need AG, Morris HA, Horowitz M, Nordin C. Effects of skin thickness, age, body fat, and sunlight on serum 25-hydroxyvitamin
D. Am J Clin Nutr 1993;58:882-5. [PubMed abstract]
54. Webb AR, Kline L, Holick MF. I nfluence of season and latitude on the cutaneous synthesis of vitamin 03: Exposure to winter
sunlight in Boston and Edmonton will not promote vitamin 03 synthesis in human skin. J Clin Endocrinol Metab 1988;67:373-8.
[PubMed abstract]
55. Webb AR, Pilbeam C, Hanafin N, Holick MF. An evaluation of the relative contributions of exposure to sunlight and of diet to
the circulating concentrations of 25-hydroxyvitamin D in an elderly nursing home population in Boston. Am J Clin Nutr
1990;51 :1075-81. [PubMed abstract]
56. Harris 55, Soteriades E, Coolidge JAS, Mudgal S, Dawson-Hughes B. Vitamin D insufficiency and hyperparathyroidism in a low
income, multiracial, elderly population. J Clin Endocrinol Metab 2000;85:4125-30. [PubMed abstract]
57. Lo CW, Paris PW, Clemens TL, Nolan J, Holick MF. Vitamin D absorption in healthy subjects and in patients with intestinal
malabsorption syndromes. Am J Clin Nutr 1985;42:644-49. [PubMed abstract]
58. Wortsman J, Matsuoka LV, Chen TC, Lu Z, Holick MF. Decreased bioavailability of vitamin D in obesity. Am J Clin Nutr
59. Vilarrasa N, MaravaiiJ, Estepa A, Sanchez R, Masdevall C, Navarro MA, et al. Low 25-hydroxyvitamin D concentrations in
obese women: their clinical significance and relationship with anthropometric and body composition variables. J Endocrinol
Invest 2007;30:653-8. [PubMed abstract]
60. Malone M. Recommended nutritional supplements for bariatric surgery patients. Ann Pharmacother 2008;42:1851-8. [PubMed
abstract]
61. Compher CW, Badellino KO, Boullatajl. Vitamin D and the bariatric surgical patient: a review. Obes Surg 2008;18:220-4.
[PubMed abstract]
62. Reid IR. The roles of calcium and vitamin Din the prevention of osteoporosis. Endocrinol Metab Clin North Am 1998;27: 389-
98. [PubMed abstract]
63. Heaney RP. Long-latency deficiency di sease: insights from calcium and vitamin D. AmJ Clin Nutr 2003;78:912-9. [PubMed
abstract]
64. Parfitt AM. Osteomalacia and related disorders. In: Avioli LV, Krane SM, eds. Metabolic bone disease and clinically related
disorders. 2nd ed. Philadelphia: WB Saunders, 1990:329-96.
65. LeBoff MS, Kohlmeier L, Hurwitz S, Franklin J, Wright J, Glowackij. Occult vitamin D deficiency in postmenopausal US
women with acute hip fracture. JAMA 1999;251 :1505-11. [PubMed abstract]
66. Kirschstein R. Menopausal hormone therapy: summary of a scientific workshop. Ann Intern Med 2003;138:361-4. [PubMed
abstract]
67. American College of Obstetricians and Gynecologists. Frequently Asked Questions About Hormone Therapy. Nm
Recommendations Based on ACOG's Task Force Report on Hormone Therapy.
68. North American Menopause Society. Role of progestrogen in hormone therapy for postmenopausal women: position statement
of The North American Menopause Society. Menopause 2003;10:113-32. [PubMed abstract]
69. Aloia JF. African Americans, 25-hydroxyvitamin D, and osteoporosis: a paradox. Am J Clin Nutr 2008;88:5455-505. [PubMed
abstract]
70. Davis CD. Vitamin D and cancer: current dilemmas and future research needs. Am J Clin Nutr 2008;88:5655-95. [PubMed
abstract]
71. Davis CD, Hartmuller V, Freedman M, Hartge P, Picciano MF, Swanson CA, Milner JA. Vitamin D and cancer: current dilemmas
and future needs. Nutr Rev 2007;65:S71 -S74. [PubMed abstract]
72. Stolzenberg-Solomon RZ, Vieth R, Azad A, Pietinen P, Taylor PR, Virtamo J, et al. A prospective nested case-control study of
vitamin D status and pancreatic cancer ri sk in male smokers. Cancer Res 2006;66:1 0213-9. [PubMed abstract]
73. Lieberman DA, Prindiville S, Weiss DG, Willett W. Risk factors for advanced colonic neoplasia and hyperplastic polyps in
asymptomatic individuals. JAMA 2003;290:2959-67. [PubMed abstract]
74. Wactawski-Wende J, Kotchen J M, Anderson GL, Assaf AR, Brunner RL, O'Sullivan MJ, et al. Calcium plus vitamin D
supplementation and the risk of colorectal cancer. N Eng I J Med 2006;354:684-96. [PubMed abstract]
75. LappeJM, Travers-Gustafson D, Davies KM, Recker RR, Heaney RP. Vitamin D and calcium supplementation reduces cancer
risk: results of a randomized trial. Am J Clin Nutr 2007;85:1586-91. [PubMed abstract]
76. Freedman OM, Looker AC, Chang S-C, Graubard Bl. Prospective study of serum vitamin D and cancer mortality in the United
States. J Natl Cancer I nst 2007;99:1594-602. [PubMed abstract]
77. Davis CD, Dwyer JT. The 'sunshine vitamin': benefits beyond bone? J Nat I Cancer Inst 2007;99:1563-5. [PubMed abstract]
78. Hypponen E, Uiara E, Reunanen A, jarvelin MR, Virtanen SM. Intake of vitamin D and risk of type 1 diabetes: a birth-cohort
study. Lancet 2001 ;358: 1500-3. [PubMed abstract]
79. Pittas AG, Dawson-Hughes B, Li T, Van Dam RM, Willett we, Manson J E, et al. Vitamin D and calcium intake in relation to
type 2 diabetes in women. Diabetes Care 2006;29:650-6. [PubMed abstract]
80. KrauseR, BUhring M, HopfenmUIIer W, Holick MF, Sharma AM. Ultraviolet Band blood pressure. Lancet 1998;352:709-10.
[PubMed abstract]
81. Chiu KC, ChuA, Go VL, Saad MF. Hypovitaminosis D is associated with insulin resistance and beta cell dysfunction. Am J Clin
Nutr 2004;79:820-5. [PubMed abstract]
82. Munger KL, Levi n LI, Hollis BW, Howard NS, Asche rio A. Serum 25-hydroxyvitamin D levels and risk of multiple sclerosis. JAMA
83. Merlino LA, Curtis J, Mikuls TR, Cerhan J R, Criswell LA, Saag K. Vitamin D intake is inversely associated with rheumatoid
arthritis: results from the Iowa Women's Health Study. Arthritis Rheum 2004;50:72-7. [PubMed abstract]
84. Schleithoff SS, Zittermann A, Tenderich G, Berthold HK, Stehle P, Koerfer R. Vitamin D supplementation improves cytokine
profi les in patients with congestive heart failure: a double-blind, randomized, placebo-controlled trial. Am J Clin Nutr
85. Chung M, Balk EM, Brendel M, IpS, Lauj, Leej, et al. Vitamin D and calcium: a systematic review of health outcomes.
Evidence Report/Technology Assessment No. 183 prepared by the Tufts Evidence-based Practice Center under Contract No.
290-2007-10055-I. AHRQ Publication No. 09-E015. Rockville, MD: Agency for Healthcare Research and Quality, 2009.
86. Autier P, Gandini S. Vitamin D supplementation and total mortality: a meta-analysis of randomized control led t rials. Arch
Intern Med 2007;167:1730-7. [PubMed abstract]
87. Giovannucci E. Can vitamin D reduce total mortality? Arch Intern Med 2007;167:1709-10. [PubMed abstract]
88. Chesney RW. Vitamin D: can an upper limit be defined?J Nutr 1989;119 (12 Suppl):1825-8. [PubMed abstract]
89. jackson RD, LaCroix AZ, Gass M, Wallace RB, RobbinsJ, Lewis CE, et al. Calcium plus vitamin D supplementation and the risk
of fractures. N Engl J Med 2006;354:669-83. [PubMed abstract]
90. HathcockJN, Shao A, Vieth R, Heaney R. Risk assessment for vitamin D. AmJ Clin Nutr 2007;85:6-18. [PubMed abstract]
91. Buckley LM, Leib ES, Cartularo KS, Vacek PM, Cooper SM. Calcium and vitamin D3 supplementation prevents bone loss in the
spine secondary to low-dose corticosteroids in patients with rheumatoid arthritis. A randomized, double-blind, placebo-
controlled trial. Ann Intern Med 1996;125:961-8. [PubMed abstract]
92. Lukert BP, Raisz LG. Glucocorticoid- induced osteoporosis: pathogenesis and management. Ann Intern Med 1990;112:352-64.
[PubMed abstract]
93. de Sevaux RGL, Hoitsma AJ, Carstens FHM, Wetzels J FM. Treatment with vitamin D and calcium reduces bone loss after renal
transplantation: a randomized study. J Am Soc Nephrol 2002;13:1608-14. [PubMed abstract]
94. McDuffie J R, Calis KA, Booth SL, Uwaifo GI, Yanovski JA. Effects of orlistat on fat-soluble vitamins in obese adolescents.
Pharmacotherapy 2002;22:814-22. [PubMed abstract]
95. Compston J E, Horton LW. Oral 25-hydroxyvitamin D3 i n treatment of osteomalacia associated with ileal resection and
cholestyramine therapy. Gastroenterology 1978;74:900-2. [PubMed abstract]
96. Gough H, Goggin T, Bissessar A, Baker M, Crowley M, Callaghan N. A comparative study of the relative infl uence of different
anticonvulsant drugs, UV exposure and diet on vitamin D and calcium metabolism in outpatients with epilepsy. Q J Med
Disclaimer
Reasonable care has been taken in preparing this document and the information provided herein is believed to be accurate.
However, this i nformation is not intended to constitute an "authoritative statement" under Food and Drug Administration rules
and regulations.
About ODS
The mission of the Office of Dietary Supplements (ODS) is to strengthen knowledge and understanding of dietary supplements by
evaluating scientific information, stimulating and supporting research, disseminating research results, and educating the public to
foster an enhanced quality of life and health for the U.S. population.
General Safety Advisory
Health professionals and consumers need credible i nformat ion t o make thoughtful decisions about eating a healthful diet and
using vitamin and mineral supplements. These Fact Sheets provide responsible information about the role of vitamins and
minerals in health and disease. Each Fact Sheet in this series received extensive review by recognized experts from the academic
and research communities.
The information is not intended to be a substitute for professional medical advice. It is important to seek the advice of a
physician about any medical condition or symptom. It is also important to seek t he advice of a physician, registered dietitian,
pharmacist, or other qualified health professional about the appropriateness of taking dietary supplements and their potential
interactions with medications.
Offi ce of Dietary Supplements
National Insti tutes of Health
Bethesda. Maryland 20892 USA
Web: ht!D:I/ods ad njh goy
E mail : ods@nih.gov
Document Last Updated: 11/13/2009 2:37 PM
http://dietary-supplements.info.nih.gov/factsheets!VitaminD_pf.asp[ l 2/l/2009 8:09: 29 PM]
PREVENTING CHRONIC DISEASE
PUBLIC HEALTH RESEARCH , PRACTICE , AND POLICY
VOLUME 5: NO. 4 OCTOBER 2008
ORIGINAL RESEARCH
Enforcement of State Indoor Tanning Laws
in the United States
Joni A. Mayer, PhD, Katherine D. Hoerster, BA, Latrice C. Pichon, MPH, Debra A. Rubio, BA,
Susan I. Woodruff, PhD, Jean L. Forster, PhD, MPH
Suggested citation for this article: Mayer JA. Hoerster
KD, Pichon LC. Rubio DA, Woodruff SI, Forster JL.
Enforcement of state indoor tanning laws in the United
States. Prev Cluonic Dis 2008;5(4). http://ll\rww.cdc.gov/
pcd/issues/2008/oct/07 _ 0194. htm. Accessed [date].
PEER REVIEWED
Abstract
Introduct ion
Twenty-eight US states have passed legislation for
indoor tanning facilities. To our knowledge, whether
these state laws are actually enforced has not been
evaluated previously in all 28 states. Therefore, we
interviewed key informants in these states to assess
enforcement practices.
Methods
Two trained interviewers used a structured survey
instrument to interview 28 key informants who were
knowledgeable about enforcement practices for laws
regarding indoor tanning. Respondents provided informa-
tion specific to the most populous city in their states.
Results
Licensure for indoor tanning businesses was required
in 22 of the 28 cities. Slightly less than half of the cities
gave citations to tanning facilities that violated state law.
Approximately 32% of the cities did not inspect indoor tan-
ning facilities for compliance with state law, and another
32% conducted inspections less t han annually. Of t hose
cities that inspected at all, most conducted unannounced
inspections.
Conclusion
The relatively low rates of annual inspections and cita-
tions are of concern. We recommend that future studies
assess whether legislation, enforcement practices, or a
combination of the 2 affects the practices of indoor tanning
facilities or of consumers.
Introduction
Indoor tanning with UV radiation lamps has been
linked to melanoma (1), squamous cell carcinoma (1),
molecular damage associated with skin cancer (2), and
other acute damage to eyes and skin (3,4). Commercial
indoor tanning facilities are prevalent in the United
States (5), and "all-you-can-tan" discount pricing pack-
ages make indoor tanning inexpensive (6). The rates of
indoor tanning for teen girls in the United States are
high (7-10); in a national sample, approximately 40% of
17- to 18-year-old girls had used indoor tanning in the
past year (7).
Some US states have passed legislation regulating
indoor tanning facilities, with the intent of reducing risks
to consumers. Ongoing systematic updates on the number
and content of these laws have been provided, focusing on
youth access restrictions (11-13). A recent report quanti-
fied the stringency of state indoor tanning legislation in
the 28 states that had a state law as of2006 (14). However,
in order to assess the level at which the laws are imple-
mented and the effect of these laws on the industry and
consumers, information about enforcement practices is
needed. Consequently, we conducted telephone interviews
of key informants in states with indoor tanning legislation
to assess enforcement practices.
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the US Department of Health and Human Serv1ces,
the Public Health Serv1ce, the Centers for D1sease Control and Prevention, or the authors' affiliated mstitut1ons. Use of trade names IS for identification only
and does not imply endorsement by any of the groups named above.
www.cdc.gov/pcd/issues/2008/ocV07 _0194.htm Centers for Disease Control and Prevention 1
VOLUME 5: NO. 4
OCTOBER 2008
Methods
Settings and participants
The CITYlOO (Correlates of Indoor Tanning in Youth)
project assesses factors that may influence use of indoor
tanning by adolescents (10,14); 1 objective is to better
understand current legislation that pertains to indoor tan-
ning. In the current study, we targeted the most populous
city in each of the 28 states to evaluate how state laws are
enforced at the local level.
Our goal was to interview, by telephone, the person who
was t he most knowledgeable about enforcement prac-
tices in each city or county. From a list of contacts for each
state's legislation presented on a Web site operated by the
tanning industry (15), we telephoned these contacts (typi-
cally at the state or county health department) and asked
them to identify the best key informant. The process for
identifying each city's respondent continued until we found
a knowledgeable potential 1espondent. We then mailed
an introductory letter to each potential respondent that
explained the purpose ofthe study and its voluntary nature
and assured anonymity of t he respondent and that data
would not be linked to the city's name in any published
reports. Approximately 1 week after mailing the letter, we
attempted to contact informants until we reached them
and they completed the interview. The 2 interviewers
(K.D.H. and L.C.P.) had previous experience in conducting
telephone interviews and received training for this study.
Survey
The survey questions were based on a combination of
previous study in this area (16), expert opinion about mea-
suring enforcement activities in the tobacco control area
(a good model for indoor tanning), and select enforcement
and monitoring activities mentioned in the indoor tanning
laws (14). Initially, we developed a longer version of the
survey that asked for specific data on various activities
(eg, number of facilities inspected in the previous year,
number of complaints received). That version assumed a
high level of inspection and other enforcement activities,
assumed that enforcement agencies kept detailed records
of those activities, and requested that informants obtain
that information before the interview and provide it during
the interview. During the informant identification process,
we became aware that the level of enforcement activities
was fairly low. Therefore, to better match the depth of our
assessment to actual practices, reduce the amount of work
required of respondents, and achieve a higher response
rate, we retained only the basic items and eliminated the
more elaborate, labor-intensive items.
The following factors were assessed: number of staff allo-
cated to carry out enforcement activities in the city/county;
whether indoor tanning businesses were required to have
a license; frequency of inspections (in absence of a com-
plaint); whether inspections were announced in advance;
whether inspection included review of customer records
and, if so, whether customer's age, parental consent forms,
number and dates of tanning sessions, and duration of
sessions were examined; and whether businesses received
citations when they violated the law. We also assessed the
types of penalties for selling sessions to underage youth or
not obtaining parental permission for minors and whether
graduated penalties (more severe penalties for each suc-
cessive violation) were used. A copy of the survey is pro-
vided in the Appendix.
A draft of the survey was reviewed for clarity by 2 public
health department professionals. All survey procedures
and materials were approved by t he institutional review
borud at San Diego State University. Interviews were con-
ducted in April and May of 2007.
Statistical analysis
Descriptive statistics, including frequencies or means,
were computed for each vruiable. Additionally, we con-
ducted bivariate tests (x
2
and correlations) to assess the
associations between the stringency of the written law (14)
and reported enforcement practices. Specifically, we exam-
ined the relationship between reported inspection fiequen-
cy and overall law stringency score, youth access subscore,
enforcement subscore, and 1 individual inspection item.
For the items about penalties specific to youth access, we
computed frequencies for only the 21 cities in states with
youth access laws. We did not perform multivariate tests
because of the small sample size. All analyses were con-
ducted in SPSS 13.0 (SPSS Inc, Chicago, Illinois).
Results
Response rates and respondent characteristics
We identified 28 respondents (1 for the most populous
The opinions expressed by authors contributing to th1s journal do not necessarily reflect the opinions of the US Department of Health and Human Serv1ces,
the Public Health S e r v ~ c e the Centers for D1sease Control and Prevention, or the authors' affiliated mstitutlons. Use of trade names is for identification only
2 Centers for Disease Control and Prevention www.cdc.gov/pcd/issues/2008/ocV07 _0194.htm
VOLUME 5: NO. 4
OCTOBER 2008
city in each of the 28 targeted states) who confirmed that
they were knowledgeable about enforcement practices.
Data were obtained for all 28 cities. If respondents told
the interviewer that their states had no law (n = 5) or that
the cities engaged in no enforcement activities (n = 2), the
interviewer contacted additional informants to confirm
that laws were not enforced. The interviewer then coded
the remaining survey items to indicate nonperformance of
enforcement activities.
Enforcement resources and practices
More than three-fourths of the respondents were
employed by a state or local health agency (Table 1). The
organizations that employed the respondents also consti-
tuted the primary enforcement entity for the state indoor
tanning legislation in the designated city.
The number of full-time employees available for inspec-
tions and other enforcement activities ranged from 0 to 15,
with a mean of 3.29 (standard deviation 3.89) staff and a
median of 2. Approximately 29% of the cities had no full-
time enforcement staff (Table 2). Licensure for indoor tan-
ning businesses was required in most cities. Slightly less
than half of the cities gave citations (ie, penalties) to tan-
ning facilities that violated the state law. Approximately
32% of the cities did not inspect indoor tanning facilities
for compliance with the state law, and another 32% con-
ducted inspections less than annually. Of those cities that
inspected, most conducted unannounced inspections.
Of the 19 cities that conducted inspections, most reviewed
customer records as part of the inspection process. Of
these, most reviewed information about customers' ages,
parental consent forms, number and dates of tanning ses-
sions, and tanning session duration (Table 3).
Of the 21 cities in states that had youth access laws,
approximately half penalized these violations (Table 4).
Warnings, monetary fines, and license suspensions were
used for both kinds of youth access violations, with no
strong predominance by type of penalty. Of the cities that
penalized violations, most gave graduated penalties for
each of the youth access- related violations, in which each
additional violation results in a larger penalty.
Bivariate associations
We conducted Pearson correlational tests between
inspection frequency and the variables from an earlier
assessment of state indoor tanning laws (14). These cor-
relations (N = 28) were 0.51 (P = .006) for enforcement
subscore, 0.34 (P = .075) for minor's access stringency sub-
score, and 0.58 (P = .001) for overall law stringency score.
Reported inspection frequency was positively correlated
with the number of full-time enforcement staff reported
by the respondent (r = 0.48, P = .Oll). We then dichoto-
mized reported inspection frequency (less than annually
vs at least annually) and the individual inspection item
score from the earlier analysis of state laws (less strict vs
more strict) . These variables were significantly associated
(x
2
= 5.18, P = .023) . Of cities whose laws on inspections
were less strict (n = 21), only 23.8% conducted inspections
at least annually. Of those whose inspection requirement
was stricter (n = 7), 71.4% conducted inspections at least
annually. A license requirement in the written law was
significantly associated with actual (self-reported) license
requirement (x
2
= 5.06, P = .024). In cities in which the
state law did not mention licensure (n = 6), 50% required
licensure, whereas in cities whose law mentioned licensure
(n = 21), 90.5% required licensure.
Discussion
To our knowledge, this article is 1 of only 3 to report
actual enforcement practices related to state indoor tan-
ning laws (16, 17) and the only article to date that pro-
vides enforcement information for all 28 states. Our data
indicate that routine annual inspections, which are a pre-
requisite for other enforcement activities such as levying
penalties for violations, are not conducted in 64% of the cit-
ies. However, for those cities that conduct regular inspec-
tions, most conduct unannounced inspections, which likely
increases their effectiveness. Additionally, the annual
inspections routinely included review of client records and
encompassed information that may reflect UV radiation
exposure levels (eg, duration and frequency of sessions)
and youth access (eg, customer age and parental consent
forms) . Thus, the annual inspections appear to be of high
quality. The relationship between inspection frequency
and staffing level suggests that cities that do not conduct
annual inspections need more resources. However, we can-
not infer causality because of the study's cross-sectional
design. Results from a study of sanitarians within 1 large
metropolitan area in both Massachusetts (21 munici-
palities of Boston) and Minnesota (21 jurisdictions of the
Twin Cities) indicated the rates of routine inspections
VOLUME 5: NO. 4
OCTOBER 2008
by Massachusetts agencies were higher than those in
Minnesota (89.9% vs 28.6%) (16). Responses from state
health department staff in Texas, lllinois, and Wisconsin
showed a large amount of variability (for each state as a
whole) on both penalties to facilities for youth access viola-
tions and facility inspection/auditing practices (17).
The low rates of penalizing any violation and youth
access violations are also cause for concern. As is the case
with enforcement activities regarding tobacco and alcohol
control, businesses rue less likely to comply with age-of-
sale laws if noncompliance is not penalized (18, 19).
The strong association between various aspects of the
law and enforcement activities is encouraging. However,
in the 7 states with more stringent inspection require-
ments, the inspection requirements tended to overestimate
the actual level of inspections. Moreover, 5 respondents
incorrectly reported that their states, at the time of the
interview, had no law on indoor tanning. Therefore, even
though some states had laws, they were not being enforced
even minimally.
One limitation of this study is that the data were based
on the report of enforcement professionals, and we did
not attempt to verify them with other measures such as
interviews with tanning facility managers. Although we
assured their anonymity, respondents may have over-
reported enforcement practices. Second, we used the most
populous city in each state to represent enforcement of the
state law, so our findings may not generalize to other cit-
ies and rural areas in each state. Initially, in 18 of the 28
states, we interviewed participants about the enforcement
practices of at least 1 additional large city in that state.
However, because the key enforcement practices were
almost perfectly consistent between cities in each state, we
ultimately used the "largest city in state" approach. That
experience leads us to believe that the data for each city
generalize to other large cities in each state. Finally, we
neglected to ask about each state's licensing fees, if any;
such fees could help fund enforcement activities (20).
Strengths of our study included an optimal response
rate and our access to reliable data on the stringency of
each state's law (14) . These data facilitated compruisons
between "ideal" and "real" enforcement activities. As noted
earlier, we promised respondents that we would not link
published data to individuals or cities; this assurance of
anonymity probably improved both our response rate and
the accuracy of the data. Even though we are unable to
reveal which states were enforcing at lower levels, we will
be providing each respondent with feedback on how the
city's enforcement level compares with enforcement for all
other cities combined. For states in which enforcement is
low, the feedback may increase enforcement practices.
In tobacco control, antitobacco organizations historically
focused their efforts on passing new laws, but enforcement
of existing laws is viewed by many to have been critical to
reducing tobacco use (21-26). In the field of indoor tanning
legislation, we cannot say whether legislation, enforce-
ment practices, or a combination of the 2 has any effect on
the practices of indoor tanning facilities or of consumers.
Therefore, we recommend that future evaluations of indoor
tanning legislation measure not only the Wl"itten law but
also its implementation and enforcement; research should
also attempt to assess the relationship between legislation
strictness/enforcement level and the practices of business-
es and consumers, especially adolescent consumers.
Acknowledgments
This study was supported by the National Cancer Institute,
grants R01 CA093532, R01 CA093532S1, and K05 100051.
The following were partially responsible for study concept
and design: Dr George Belch, Dr James Sallis, Dr Donald
Slymen, Ms Elizabeth Clapp, and Ms Rebecca Garrow,
San Diego State University; Dr Todd Gilmer, University
of California Medical School, San Diego; and Dr Mrutin
Weinstock, Veterans Administration Medical Center and
Brown University. We thank Ms Lorri Freitas of the San
Diego County Health and Human Services Agency and Dr
Donald Bishop of the Minnesota Department of Health for
valuable feedback on the survey instrument and the survey
participants for their time and effort.
Author Information
Corresponding Author: Joni A. Mayer, PhD, Graduate
School of Public Health, San Diego State University, 9245
Sky Pruk Ct, Ste 220, San Diego, CA 92123. Telephone:
619-594-7916. E-mail: jmayer@mail.sdsu.edu.
Author Affiliations: Katherine D. Hoerster, J oint Doctoral
Program in Clinical Psychology, Latrice C. Pichon, Joint
Doctoral Program in Public Health, Health Behavior, San
VOLUME 5: NO. 4
OCTOBER 2008
Diego State University/University of California, Debra
A. Rubio, Susan I. Woodruff, School of Social Work, San
Diego State University, San Diego, California; J ean L.
Forster, School of Public Health, University of Minnesota,
Minneapolis, Minnesota. At the time of this 1esearch,
Susan I. Woodruff was affiliated with the Graduate School
of Public Health, San Diego State University, San Diego,
California.
References
1. International Agency for Research on Cancer, working
group on artificial ultraviolet (UV) light and skin can-
cer. The association of use of sunbeds with cutaneous
malignant melanoma and other skin cancers: a sys-
tematic review. [Published erratum in: Int J Cancer
2007;120(11):2526]. Int J Cancer 2007;120(5):1116-
22.
2. Whitmore SE, Morison WL, Potten CS, Chadwick C.
Tanning salon exposure and molecular alterations. J
Am Acad Dermatol 2001;44(5):775-80.
3. Centers for Disease Control and Prevention. Injuries
associated with ultraviolet tanning devices- Wisconsin.
MMWR Morb Mortal Wkly Rep 1989;38(19):333-5.
4. Walters BL, Kelley TM. Commercial tanning facili-
ties: a new source of eye injury. Am J Emerg Med
1987;5(5):386-9.
5. Palmer RC, Mayer JA, Woodruff SI, Eckhardt L, Sallis
JF. Indoor tanning facility density in eighty US cities.
J Community Health 2002;27(3):191-202.
6. Kwon HT, Mayer JA, Walker KK, Yu H, Lewis EC,
Belch GE. Promotion of flequent tanning sessions
by indoor tanning facilities: two studies. J Am Acad
Dermatol 2002;46(5):700-5.
7. Cokkinides VE, Weinstock MA, O'Connell MC, Thun
MJ. Use of indoor tanning sunlamps by US youth,
ages 11-18 years, and by their parent or guard-
ian caregivers: prevalence and correlates. Pediatrics
2002; 109(6): 1124-30.
8. Demko CA, Borawski EA, Debanne SM, Cooper KD,
Stange KC. Use of indoor tanning facilities by white
adolescents in the United States. Arch Pediatr Adolesc
Med 2003;157(9):854-60.
9. Geller AC, Colditz G, Oliveria S, Emmons K, Jorgensen
C, Aweh GN, et al. Use of sunscreen, sunburning rates,
and tanning bed use among more than 10,000 US chil-
dren and adolescents. Pediatrics 2002;109(6):1009-14.
10. Hoerste1 KD, Mayer JA, Woodruff SI, Malcarne V,
Roesch SC, Clapp E. The influence of pruents and
peers on adolescent indoor tanning behavior: find-
ings flom a multi-city sample. J Am Acad Dermatol
2007;57(6):990-7.
11. Dellavalle RP, Parker ER, Cersonsky N, Hester EJ,
Hemme B, Burkhardt DL, et al. Youth access laws:
in the dark at the tanning parlor? Arch Dermatol
2003; 139( 4):443-8.
12. Francis SO, Burkhardt DL, Dellavalle RP. 2005: a
banner year for new US youth access tanning restric-
tions. Arch Dermatol2005; 141(4):524-5.
13. McLaughlin JA, Francis SO, Burkhardt DL, Dellavalle
RP. Indoor UV tanning youth access laws: update
2007. Arch Dermatol 2007;143(4):529-32.
14. Woodruff SI, Pichon LC, Hoerster KD, Forster JL,
Gilmer T, Mayer JA. Measuring the stringency of states'
indoor tanning regulations: instrument development
and outcomes. JAm Acad Dermatol 2007;56(5):774-
80.
15. A guide to state radiation control offices. Phoenix (AZ):
National Tanning Training Institute. http://www.
tanningtraining .com/reginfo/sta te.h tml. Accessed
September 25, 2006.
16. Hickle A, Forster J , Lazovich D, Allwood P, Remba
N, Grossmeier J, et al. Sanitarians' work with indoor-
tanning businesses: findings from interviews in
two major metropolitan areas. J Environ Health
2005;67(8):30-6, 54.
17. Hester EJ, Heilig LF, D'Ambrosia R, Drake AL,
Schilling LM, Dellavalle RP. Compliance with youth
access regulations for indoor UV tanning. Arch
Dermatol 2005;141(8):959-62.
18. Dent CW, Grube JW, Biglan A. Community level alco-
hol availability and enforcement of possession laws as
predictors of youth drinking. Prev Med 2005;40(3):355-
62.
19. Forster JL, Wolfson M. Youth access to tobacco: poli-
cies and politics. Annu Rev Public Health 1998;19:203-
35.
20. Demierre MF. Time for national legislation of
indoor tanning to protect minors. Arch Dermatol
2003;139:520-4.
21. Forster JL, Komro KA, Wolfson M. Survey of city ordi-
nances and local enforcement regarding commercial
availability of tobacco to minors in Minnesota, United
States. Tob Control1996;5(1):46-51.
22. Jacobson PD, Wasserman J. The implementation and
enforcement of tobacco control laws: policy implica-
tions for activists and the industry. J Health Polit
VOLUME 5: NO. 4
OCTOBER 2008
Policy Law 1999;24(3):567-98.
23. Jason LA, Porkorny SB, Schoeny ME. Evaluating the
effects of enforcements and fines on youth smoking.
Critical Public Health 2003; 13:33-45.
24. Lazovich D, Forster J, Widome R, VanCoevering P.
Tobacco possession, use, and purchase laws and pen-
alties in Minnesota: enforcement, tobacco diversion
programs, and youth awareness. Nicotine Tob Res
2007;9(Suppl1):S57-64.
25. Ma GX, Shive S, Tracy M. The effects of licensing
and inspection enforcement to reduce tobacco sales
to minors in greater Philadelphia, 1994-1998. Addict
Behav 2001;26(5):677-87.
26. Woodhouse LD, Sayre JJ, Livingood WC. Tobacco
policy and the role of law enforcement in prevention:
the value of understanding context. Qual Health Res
200 1; 11(5):682-92.
Tables
Table 1. Respondent and Organization Characteristics (N =
28), CITY100 Enforcement Survey, April and May 2007
Variable n (%)
Sex
-
Women 7 (25.0)
Men 21 (75.0)
Organization
State health, environmental health, or radiologic health 15 (53.6)
agency
City or county health, environmental health, or radiologic 7 (25.0)
health agency
State cosmetology board 4 (14.3)
Other state agency 2 (7.1)
Occupation/title
Department head or director 6 (21.4)
- -- -
Supervisor or manager 6 (21.4)
Health physicist 4 (14.3)
Environmental health or industrial hygiene 3 (10.7)
-
Sanitarian 3 (10. 7)
Inspector 2 (7.1)
L Other"
-
4 (14.3)
Abbreviat ion: CITY100, Correlates of Indoor Tanning in Youth.
a These titles included investigator, board administrator, regulatory enforce
ment, and compliance officer.
PUBLIC HEALTH RESEARCH , PRACTICE, AND POLICY
VOLUME 5: NO. 4
OCTOBER 2008
Table 2. Indoor Tanning Legislation Enforcement Resources and General Practices (N = 28), CITY100 Enforcement Survey,
April and May 2007
Variable
No. of full-time enforcement staff
0
1-2
3-4
2:5
1-
Ucensure required
No
1-
Yes
Don't know
Give citation if facility violates law
No
Yes
Don't know
Inspection schedule
Never
Less than annually
Annually
1-
Twice a year
Announce inspection in advance
8
1-
Never/rarely
Sometimes
Often/always
Abbreviation: CITY100, Correlates of Indoor Tanning in Youth.
a Based on the 19 cities that ever conducted inspections.
n (%)
8 (28.6)
7 (25.0)
7 (25.0)
6 (21.4)
-
5 (17.9)
22 (78.6)
1 (3.6)
r
14 (50.0)
13 (46.4)
1 (3.6)
-
9 (32.1)
9 (32.1)
6 (21.4)
-
4 (14.3)
17 (89.5)
2 (10.5)
1
0
-
the Public Health Serv1ce, the Centers for D1sease Control and Prevention, or the authors' affiliated mstitutlons. Use of trade names is for identification only
PUBL I C HEALT H RESEARCH, PRACTICE, AND POLICY
VOLUME 5: NO. 4
OCTOBER 2008
Table 3. Indoor Tanning Facility Inspection Practices Related to Customer Records, CITY100 Enforcement Survey, April and
May 2007
Frequency
--------------------------------------------
Practice
Inspection includes customer record review"
Review includes customer's ageb
Review includes parental consent formsb
1--
Review includes number and dates of sessionsb
Review includes session duration!>
Abbreviation: CI1Y100, Correlates of Indoor Tanning in Youth.
a Based on the 19 cities t hat ever conducted inspections.
b Based on the 17 cities t hat included customer record review when inspecting.
Never/Rarely
n (%)
2 (10.5)
1 (5.9)
2 (11.8) l
1 (5.9)
2 (11.8)
Sometimes Often/Always
n (%) n (%)
2 (10.5) 1 15 (78.9) 1
0 16 (94.1)
0 15 (88.2)
-
3 (17.6) 13 (76.5)
1 (5.9) 14 (82.4)
Table 4. Penalties for Youth Access Violations in States With Indoor Tanning Access Laws, CITY100 Enforcement Survey, April
and May, 2007
Violation
-------------------------------------------------------------------
Selling Sessions to Underage Youth Not Obtaining Parental Consent for Minors
Penalty n (%)" n (%)b
Penalty type
Any type 11 (55.0) 10 (47.6)
Warning 10 (50.0) 9 (42.9)
Monetary fine 8 (40.0) 8 (38.1)
License suspension 7 (35.0) 7 (33.3)
Other 3 (15.0) 1 (4.8)
Graduated penalties given
I
No 4 (36.4) I 3 <3o.o> I
Yes
-
a Based on 20 respondents because of missing data.
b Based on 21 respondents.
c Based on those respondents whose cities gave any type of penalty for the violation.
7 (63.6) I 7 (70.0)
the Public Health Serv.ce, the Centers for D1sease Control and Prevention, or the authors' affiliated mstitutlons. Use of trade names is for identification only
VOLUME 5: NO. 4
OCTOBER 2008
Appendix: CI1Y100 Enforcement Survey
Study Introduction
This interview is part of the CITYiOO Indoor Tanning Project. CITYiOO
(Correlates of Indoor Tanning in Youth) is a project funded by the National
Cancer Institute that will help us better understand the factors that influ
ence teens to use indoor tanning. The project is based at the Graduate
School of Public Health at San Diego State University and is focusing on
over iOO cities in the US. One goal of CITYiOO is t o evaluate enforcement
activities in cities located in states with indoor tanning laws. Your state has
a law governing indoor tanning.
If you decide to participate, I will ask you questions about the enforcement
activities in [city/county]. such as inspections of indoor tanning facilities.
These are activities at the city or county level to enforce the state
law. We would like you to participate, irrespective of whether your city has
many or few enforcement activities. The interview will take only around 5
to 7 minutes. The researcher in charge of this study is Dr Joni Mayer; you
may have her phone number if you wish to write it down. Collect calls are
accepted. She will be able to answer any questions you have. Or if you
have any questions now, I can answer them for you. I want to assure you
that your responses in this interview wil l not be linked with your narne.
city, or county in any written reports or publications. All data will be kept in
locked file cabinets, and only the CITYiOO research staff will have access
to the data. Participation is voluntary, and you are free to end the interview
at any point.
Resources (R)
I'd like to first ask you about your employment, and [city'S/county's)
resources related to regulating indoor tanning businesses.
Ria. What is the agency you work for?
Ri b. What department within that agency?
Ric. Is your job at the
City level?
County level?
State level?
None of these? Describe.
R2. What is your occupation and/or j ob title?
R3. How many agencies are responsible for enforcing the state's indoor
tanning law in [city/county]?
None/no enforcement
One
Two or more
Don' t know
Refused
No state law
R4. The primary enforcement agency-is it at the
City-level? Name:
County-level? Name:
Statelevel? Name:
None of these? Name:
R5. How many full-time staff (or FTE) are allocated to carry out the inspec-
tions and/or other enforcement activities in [city/county]?
Gave a number:
Don't know
Refused
Not applicable
Licensure (L)
Li. Are indoor tanning businesses in [city) required to have a license?
No
Yes
Don't know
Refused
Not applicable
Inspections (I)
Now I'd like to ask you about inspection procedures.
li. In the absence of a complaint, how often is a tanning facility inspected
in [city/county]?
Never (go to question P1)
Less than once a year
Once a year
Twice a year
More than twice a year
Other (describe)
Don't know
Refused
Not applicable
12. How often is the inspection announced in advance to the tanning facility
t hat will be inspected?
Never/ rarely
Sometimes
Often/always
Don't know
Refused
Not applicable
13. Please tell me how often an inspection includes the review of customer
records.
Never/ rarely
Sometimes
Often/always
Don't know
Refused
Not applicable
14. (If the answer to 13 is sometimes, often, or always) When customer
records are reviewed, how often are the following looked at?
14a. Age of customers
Never/ rarely
Sometimes
The opinions expressed by authors contributing to th1s j ournal do not necessarily reflect the opinions of the US Department of Health and Human Serv1ces,
VOLUME 5: NO. 4
OCTOBER 2008
PUBL I C HEALT H RESEARCH , PRACTICE, AND POLICY
Often/always
Don't know
o Refused
o Not applicable
14b. Parental consent forms
o Never/rarely
o Sometimes
o Often/always
o Don't know
Refused
o Not applicable
14c. Number and dates of tanning sessions
o Never/rarely
o Sometimes
o Often/always
Dont know
o Refused
o Not applicable
14d. Duration of tanning sessions
o Never/ rarely
Sometimes
o Often/always
o Don't know
Refused
o Not applicable
Penalties/Fines (P)
The next questions are about penalties for violations of laws regulating
indoor tanning.
Pl. If a tanning facility in [city/county] violates a law regulating this type of
business, will the facility receive a citation?
No (skip the remaining items}
o Yes
Don't know
o Refused
Not applicable
P2. In general, what is the penalty if a facility is cited for selling tanning
sessions to underage youth?
P2a. A warning
. No
Yes
Don't know
Refused
o Not applicable
P2b. A monetary fine
o No
o Yes
o Don't know
o Refused
o Not applicable
P2c. License suspension
No
Yes
o Don't know
Refused
o Not applicable
P2d. Other (probe}
P3. If a facil ity is cited more than once for selling tanning sessions to
underage youth, does [city/county] use graduated penalties, in which
each repeat violation results in a larger penalty?
No
o Yes
Don't know
Refused
Not applicable
P4. In general, what are the penalties if a facility is cited for not obtaining
parental permission for minors?
P4a. A warning
No
. Yes
Don't know
Refused
Not applicable
No
o Yes
Don't know
o Refused
Not applicable
No
o Yes
Don't know
Refused
Not applicable
P4d. Other (probe}
P5. If a facility is cited more than once for not obtaining parental permis-
sion, does your city/county use graduated penalties, in which each
repeat violation results in a larger penalty?
No
Yes
Don't know
Refused
o Not applicable
That concludes all my questions. I really appreciate all of your time! Do
you have any questions or comments about this survey? Thanks again.
Goodbye.
FEDERAL REGISTER
Vol . 64, No . 26
Proposed Rules
DEPARTMENT OF HEALTH AND HUMAN SERVICES {HHS)
Food and Drug Administration (FDA)
21 CFR Part 1020
[ Docket No . 98N- 1 170 ]
Medi cal Devi ces ; Sunlamp Products Performance Standard; Request fo r
Commentsand Information
64 FR 6288
DATE : Tuesday, February 9, 1999
ACTION : Advance notice o f proposed rulemaki ng .
To vi ew t he next page , type . np* TRANSMIT .
To vi ew a spec i f i c page , t r ansmi t p* and the page number , e . g . p*1
[*6288]
SUMMARY : The Food and Drug Admi nistration {FDA) is announci ng i ts i ntent to
propose amendment s to t he performance standard for sunlamp products . The
agency is t a king thi s act ion to addr ess concer ns about the adequacy of the
warn i ngs on sunlamp p r oduct s , cur r ent recommended exposur e schedule t o
minimize risk to customers who choose to produce and maintain a tan, current
labeling f or replacement l amps , and current heal th warnings which do not
reflect recent advances in photobi o l ogical research . FDA i s sol ici t i ng
comments and i nf ormat i on from interested persons concerning the subject matter
of the proposed amendment s .
DATES : Wri t t en commen t s by May 10, 1999 .
ADDRESSES : Submit written comment s to the Dockets Management Branch {HFA-
305) , Food and Drug Administration, 5630 Fishers Lane , rm. 1061, Rockville , MD
20852. I ndividuals or organizat i ons wishing to receive copies o f draft
amendments or related documents d i s tributed for review during the development
of these amendments may have their names placed on a mailing list by wri t ing
t o Office of Science and Technology (HFZ-114) , Cent er for Devices and
Radi ologi cal Heal th, Food and Dr ug Admi ni s trat i on , 5600 Fi sher s Lane ,
Roc kville , MD 20857 , FAX 301-594-6775, e-mail address HWCCDRH . FDA . GOV.
FOR FURTHER INFORMATION CONTACT : W. Howard Cyr , Cent er for Devices and
Radiological Health (HFZ- 114) , Food and Drug Adminis t ration, 5600 Fishers
Lane , Rockville, MD 20857 , 301- 443- 7179 . [*6289)
SUPPLEMENTARY I NFORMATION :
I . Background
The Safe Medical Devices Act o f 1 990 (Pub . L. 101-629) , enacted on November
28 , 1990, t ransferred t he provisions of t he Radiation Control for Heal t h and
Safet y Act of 1968 (Pub . L. 90- 602) from Title III of t he Public Health
Ser vi ce Ac t t o Chapter V, subchapter C of t he Feder a l Food , Dr ug, and Cosmet i c
Act (21 U. S . C. 360hh et seq . ) . This authority provides for developing,
amendi ng, and admi ni s t ering radi ation safety performance standards for
e l ect ronic products .
Sunlamp products are class I medical devices exempt from premarke t
notification requirements (21 CFR 878 . 4635) . These products are intended to
provide u l t raviolet (UV) radiat i on to tan t he s kin . As class I devices ,
sunl amp products are subject t o general cont rols such as registration,
listing, and current good manufacturing pract ices . Sunlamp products are also
subject to the regulations for electronic product radiation control including
par t s 1000 t hrough 1010 and 1040 . 20 (21 CFR parts 1000 t hrough 1010 and 21
CFR 1040 . 20) .
The sunlamp performance standard in 1040 . 20 was ori g i nal ly publ ished in
the Federa l Register of November 9, 1979 (44 FR 65352) . On September 6, 1985
(50 FR 36548) , FDA amended 1040 . 20 and made it appl i cabl e to a l l sunlamp
p roducts manufactured on or after September 8 , 1986 . On August 21 , 1986, FDA
issued a guidance entitled " Policy on Maximum Timer Interval and Exposure
Schedule for Sunlamp Products ." The guidance explained the criteria FDA uses
to evaluate the adequacy of the exposure schedule and the recommended maximum
exposure t ime for sunlamp products . On September 2 , 1986, FDA issued another
guidance ent itled " Policy on Lamp Compatibi l i t y ." The guidance l ist ed the
criteria FDA use s to evaluat e lamp compat i b i l i ty for sunlamp products .
Before proposing any electronic product performance standards , FDA is
required to consult a statutory advisory committee , the Technical Electronic
Product Radi a t ion Safety Standards Commi t t ee ( TEPRSSC) (21 U. S . C.
360kk(f) (1) (A)) . At the September 23 and 24, 1998 , meeting of TEPRSSC, FDA
presented general concepts for amendments to the performance standard for
sunlamp products . The commit tee recommended that FDA pursue development of t he
amendments . FDA intends to present more specific proposals t o amend the
performance s t andard to TEPRSSC prior t o t he publ i cat i on of a proposed rul e
in the Federal Regis t er .
FDA is concerned that inadequate attention is being paid to the recommended
exposure schedule which should be designed to minimize risks for those who
choose to produce and maintain a tan . FDA is fur ther concerned t hat the
warnings f or sunlamp product s are not reachi ng many users of sunlamp product s
and that t he exi s t ing exposure schedul e does not t ake into account t he
var i ations i n indi vi dual human UV sensit ivity. I n order t o update the current
sunlamp product standards , FDA is considering revising 1040 . 20 .
I n addi t i on, sunl amp technology cont i nues t o change . These changes can
a f fec t bot h the intensit y and the spectral characteristi cs of the UV f rom
sunl amps . Because there is no uni f orm grading/rat i ng system, choosing a
replacement lamp can be confusing for tanning bed owners . Owners choosing
replacement lamps must consider lamp compatibility as well as compliance with
FDA ' s performance standard in order to protect users from excessive exposure
to UV .
I n addi t i on t o concerns about t he warni ngs , l abel ing, and exposure
schedule, FDA is aware of new research f indi ngs that suggest a stronger
associ a t ion between exposures t o ultravi o l et radiati on and the i ncreased
incidence of skin cancer that has been observed in the U. S . populat ion . Some
of this increase has b een linked to intense, intermi ttent exposures to solar
r a diation, but other research suggests that chronic, less intense exposures to
ultraviolet radiation also contribute to skin cancer . Research has ident ified
the fundamental chemical damage that occurs in the genet ic material of humans
and has l i nked some s kin cancers t o changes in speci f i c genes . These
scientific findi ngs have l ed many in the medical communi ty t o strongly suggest
that consume r s avoi d intense , inte rmittent exposures ( the type that could
produce sunburns ) to ult raviole t radiation, and also minimize o t her UV
e xposures as we ll .
There are o t her de l e t eri ous e f f ect s f rom human e xposure t o UV radi a t i on .
They i nclude bl i s t ering bur ns , ski n eryt he ma, phot oagi ng, and
photoallergi c/photosensitive drug int eractions . UV radiation may induce damage
in the cornea, lens , and retina of the eye , which in ext r eme cases lea d s to
permanent loss of vision . UV exposure is immunosup p ressive , and can have an
impact on the development of many d iseases .
Some research has l inked s kin cancer to e xposures t o sunl amp products , and
some research has even sugges t ed an associ a tion be t wee n t he use o f sunlamps
and mal i gnant me lanoma . Thi s associ a t i on i s not de f i ni t i ve . FDA solic i t s
comments and informat i on as to whether a warning about possible melanoma
induction should be pa rt of s unlamp l abel s . To provide user s with s uffici e nt
information for the safe use of these devices at tanning salons and for home
sunlamp product s , FDA s eeks comments and informat ion on suggested change s to
the current sunlamp labels .
Aft er c onsi de r i ng t he risks , some consumers may s t i l l choose t o t an, e i t her
by e xposure t o t he sun or by use of sunl amp product s . Those consumers who use
s unlamp product s should obtain their tan with t he least amount of risk from
s unburn and eye damage . Therefore, FDA seeks a dvice on a recomme nded exposure
schedule which would minimize the risks of a dve r se effects while still
producing and maintaining a t an .
I I . Revi s i ons Under Considerat ion
FDA believe s t hat ame ndme nt of t he current pe r f ormance s t andard i s
necessary to keep pace with changes in technology and advances in research
rel a t e d to the use of sunlamp prod uct s . The following di scussion i s inte nded
to describe the need for the revi s ion and FDA ' s p roposed approac h . Comments
received from thi s a d vance not ice of proposed rulemaking (ANPRM) will be used
to devel op any proposed amendments . Any proposed regul a t ory changes or
s t andar ds amendments wi ll be i ncluded in a f ut ure proposed rul e . FDA i s
sol i c i t i ng comme nt s on al l aspects of t his ANPRM, and specifical ly reque s t s
comments on t he following proposed amendments :
1 . FDA i s considering revising a nd updat ing the current s unlamp p roduc t
performance s t andard ( 1040 . 20) and harmoni z i ng i t wi th t he I nt ernat i onal
El ect rot echni cal Commi t t e e St andard 335-2-27 f o r UV and infrared e mi t t ing
appliances . Aft er consulting with int ernational standards organizations and
evalua t ion of t he current scientific knowledge , FDA intends to develop a
recommende d exposure schedule which will become part of the directions for use
of the sunlamp product . As part of the development process , FDA intends to
review the material on effects of UVA and UVB on skin, the effects of UV
exposure on melanoma induction, and the use of photobiological action spectra
as a basis for risk assessment in health protection and product safety
discussed at the American Society for Photobiology and European Society for
Photobiology Joint Workshop on UV and Melanoma , Snowbird, Utah , July 11
[*6290] through 15, 1998; the International Symposium and Workshop on
Measurements of Optical Radiation Hazards , at the National Institute for
Standards and Technology , Gaithersburg , MD , September 1 through 3 , 1998 ; and
(3) the Research Workshop on Risks and Benefits of Exposure to Ultraviolet
Radiation and Tanning , at the National Institutes of Health, Bethesda, MD ,
September 16 through 18, 1998 . The proceedings of these meet ings describe
current r esearch findings that show a stronger corre l a tion betwee n UV exposure
and s kin cancer , photoaging , and photoimmunological e ffect s .
2 . FDA is considering revising and updating its August 21 , 1986, guidance
on the determination of the maximum timer interval and recommended exposure
schedule for sunlamp products entitled, " Policy on Maximum Timer Interval and
Exposure Schedul e for Sunl amp Products ." FDA i s concerned that i nadequat e
a t t ent ion is bei ng pai d to current recommended exposure schedules and that
current guidance may allow higher exposures than are necessary t o produce and
maint ain a tan, and it does not incorporate the differences in individual
human sensitivity to UV exposure . FDA intends to update this guidance after
reviewing and evaluating material presented at the meetings listed previously
and other available information . FDA is further considering incorporating the
previous guidance into the sunlamp product performance standard because it
beli eves such incorporat ion woul d result i n a more comprehensi ve regul a t ory
s t andard wit h a l l rel e vant i nformat i on for compl iance in t he standard.
3 . FDA is considering adding a provision clarifying that manufacturing
includes the modification of a sunlamp product , previously certified under
1010 . 2 , by any person engaged in t he business o f manuf acturi ng, assembl i ng or
modi fying sunl amp products i f the modi f i cat i on affects any aspect o f t he
product ' s performance , information or intended function for which 1040 . 20
has an applicable requirement . This addition would clarify t hat sunlamp
products are bei ng regul a t ed l i ke o t her products regul a t ed under 1010 . 2 . FDA
is also considering requiring the manufacturer who performs such modification
to recertify and re- identify the product in accordance with the p rovisions of
1010 . 2 and 1010 . 3 . This potent i a l amendment i s intended to c l ari f y the
respons i b i l i t i es of f i rms and indi vi dual s who are in t he business of
instal l i ng ult ravi o l et l amps and new timers wi th different performance
characteristics than the original lamps and t imers in previously cert ified
products .
4 . FDA is concerned t hat the current warning label is not read by many
tanning salon patrons because it is too long and detailed . Therefore , FDA is
considering updating the warning statement required by 1040 . 20(d) (1 ) (i ) to
simplify the wording and to highlight the risk of skin cancers . In order to
update the warning statements , FDA intends to review and e va luate
epidemiological and mechanistic information on uv exposure- related skin
cancers , i ncludi ng possi b l y fatal cutaneous mal i gnant melanoma . In devel oping
i t s speci f ic proposal for this i tem, FDA wi l l be reviewi ng t he mat erial
presented a t t he meet i ngs c i ted previously and o t her avai l able i nf ormat i on .
5 . FDA is considering requiring the reproduction of the text of the warning
s t a t ement speci f ied in 1040 . 20(d) (1 1 ( i) in cat a l ogs , speci f i cat i on sheet s ,
and brochures pert a i ni ng t o sunl amp products . FDA is concerned t hat consumers
who purchase sunlamp product s through catalog mai l order or through catalogs
on electronic media may not receive information about the associated hazards
and risks until the products are delivered to their homes and unpacked .
6 . To simplify appropriate lamp replacement , FDA is considering the
development of a biological efficacy rating scale for ultraviolet lamps
intended f or use i n sunl amp products . Lamp t echnol ogy cont inues to evolve ,
a f fect ing the levels of UV exposure , t he spect ral characteri s t ics and,
t herefore, t he biologi cal e f f i cacy of ul traviolet lamp radiat i on . At present,
a label that specifies the type of lamps suitable for replacement in the
product is required on sunlamp products and in the user instructions . As new
l amps and new l amp manufacturers enter the marketplace , while other
manufacturers abandon the marketplace , it is increasingly cumbersome to keep
track of individual lamp designations which are compatible with the product
and compl i ant wi th t he standard . I n order to simpl i f y the process , especial l y
for indust ry and State regul a t ors , FDA i s considering a uni f orm grading/rat i ng
syst em.
III . Comments
Interested persons may, on or before May 10, 1999, submit to the Dockets
Management Branch (address above) writ ten comments regarding t h is ANPRM. Two
copi es of any comments are to be submi t t ed, except that indi v i dual s may submi t
one copy . Comment s are t o be i dentifi e d wi t h the docke t number found i n
bracke t s in t he heading of this document . Received comments may be seen in the
office above between 9 a . m. and 4 p . m., Monday through Friday . Thi s ANPRM is
i ssued under section 5 31 e t seq . of the Fe de r a l Food , Drug, and Cosmet i c Act
(21 U.S.C. 360hh et seq .) and under authority o f t he Commi ssioner o f Food and
Drugs .
Dated : Februa ry 2 , 1999 .
William K. Hubbard,
Associ a t e Commiss i oner for Policy Coordi nat i on .
[FR Doc . 99-3109 Fi l ed 2-8-99; 8 : 45 am]
BILLING CODE 4160- 01- F
DATE : SEPTEMBER 10, 2008
CLIENT : TANNING
LIBRARY : HEALTH
FILE : FE DREG
YOUR SEARCH REQUEST IS :
TANNING
AND ULTRAVIOLET
NUMBER OF DOCUMENTS FOUND WITH YOUR REQUEST THROUGH :
LEVEL 1. .. 645 LEVEL 2 ... 71
FEDERAL REGISTER
Vol. 66, No. 143
Rules and Regulations
DEPARTMENT OF HEALTH AND HUMAN SERVICES (HHS)
Food and Drug Administration (FDA)
21 CFR Parts 862, 864, 866, 868, 870, 872, 874, 876, 878, 880, 882, 884, 886,888, 890, and 892
[Docket No. 01N-0073]
Medical Devices; Exemption From Premarket Notification Requirements; Class !Devices; Technical Amendment
Part II
66 FR 38786
DATE: Wednesday, July 25,2001
ACTION: Final rule; technical amendment.
To view the next page, type .np* TRANSMIT.
To view a specific page, transmit p* and the page number, e.g. p* 1
[*38786]
SUMMARY: The Food and Drug Administration {FDA) is amending the language in its medical device
classification regulations for class l devices for consistency, to include in sections where it was not present, a
specific reference to the limitations on exemptions from premarket notification requirements for each generic device
classified. The specific reference language was included when some class I generic devices were first exempted
under provisions of the Food and Drug Administration Modernization Act of 1997 (FDAMA). These amendments
will provide the same reference for devices that were exempted before that time. The language is intended to
conveniently provide the reference, and make the sections clear and easy to read. The status of the devices is not
being changed.
DATES: This rule is effective July 25, 200 I.
FOR FURTHER INFORMATION CONTACT: HeatherS. Rosecrans, Center for Devices and Radiological
Health (HFZ-404), 9200 Corporate Blvd., Rockville, MD 20850,301-594-1190.
SUPPLEMENTARY INFORMATION:
I. Background
Under section 513 of the act (21 U.S. C. 360c), FDA must classify devices into one of three regulatory classes:
Class I, class II, or class III. FDA classification of a device is determined by the amount of
regulation necessary to provide a reasonable assurance of safety and effectiveness.
Under the 1976 amendments (Public Law 94-295), as amended by the SMDA (Public Law 101-629), clevicet are
to lite clauifiecl into clast I <eneral control$) if there ltlnfor111atlon thowln that the neral
control$ of the act are tufflclent to enture tafet" ancleHedlueneu; into class II (special controls), if
general controls, by themselves, are insufficient to provide reasonable assurance of safety and effectiveness, but
there is sufficient inf01mation to establish special controls to provide such assurance; and into class ill (premarket
approval), if there is insufficient infonnation to support classifying a device into class I or class II and the device is a
life-sustaining or life-supporting device, or is for a use which is of substantial importance in preventing impairment
of human health, or presents a potential unreasonable risk of illness or injury.
Most generic types of devices that were on the market before the date of the 1976 amendments (May 28, 1976)
(generally referred to as preamendments devices) have been classified by FDA under the procedures set forth in
section 513(c) and (d) of the act through the issuance of classification regulations into one of these three regulatory
classes. Devices introduced into interstate commerce for the first time on or after May 28, I 976 (generally referred to
as postamendments devices) are classified through the premarket notification process under section 510(k) of the act
(21 U.S. C. 360(k)). Section 510(k) of the act and the implementing regulations in 21 CFR part 807 require persons
who intend to market a new device to submit a premarket notification report (51 O(k)) containing information that
allows FDA to determine whether the new device is substantially equivalent within the meaning of section 513(i) of
the act to a legally marketed device that does not require premarket approval. Unless exempted from premarket
notification requirements, persons may not market a new device under section 51 O(k) of the act, unless they receive a
substantial equivalence order from FDA or an order reclassifying the device into class I or class IT, under section
513(f) of the act.
On November21, 1997, the President signed FDAMA into law (Public Law 105-115). Section 206 of FDAMA, in
part, added a new section 51 0(1) to the act. New section 51 0(1) of the act became effective
February 19, 1998. It provides that a class I device is exempt from the premarket
notification requirements under section 51 O(k) of the act, unless the device is
intended for a use that is of substantial importance in preventing impairment of
human health or it presents a potential unreasonable risk of illness or injury.
To implement this provision, FDA evaluated all class I devices to determine which
device types should become exempt under new provision 51 0(1) of the act and which
device types should remain subject to the requirements of 510(k) of the act. FDA then
amended its classification regulations, in part, by publishing in the Federal Register of February 2, 1998 (63 FR
5387), a list of certain class I devices that would become exempt from 510(k) requirements on February 19, 1998,
subject, however to the limitations found in each classification regulation section (e.g., 2/ CFR 862.9, 864.9, etc.),
63 FR 5387, February 2, 1998. The limitations language of each classification states that if a class I or IT devices is
intended for a use different from that of a legally marketed device in that generic type, or if the modified device
operates using a different fundamental scientific technology than that of a legally marketed device in that generic
type, a new 51 O(k) submission and clearance is required. The limitations language also lists specific intended uses
for in vitro diagnostics devices that would preclude an exemption from the requirements of 51 O(k). FDA issued a
proposed rule in the Federal Register of November 12, 1998 (63 FR 63222), to designate class I devices that are
exempt from the premarket notification requirements, subj ect to certain limitations, and to designate class I devices
that remain subject to premarket notification requirements under the new statutory criteria. The designations of these
devices were codified by a final rule in the Federal Register of January 14, 2000 (65 FR 2296).
As published in the January 14, 2000, Federal Register, the amendments state, in part, that the limitations in each
classification regulation apply to the premarket notification exemptions for each generic device classified in each
section. In addition to mentioning the limitations generally in each classification regulation, FDA noted in the
Federal Register of January 14, 2000, publication that, for clarity and convenience, the classification section for each
generic device newly exempted under section 51 0(1) of the act specifically states that the exemptions are subject to
limitations. The agency fm1her noted that for individual device classification sections that had been codified
previously as exempt from premarket notification requirements, it would add the same subject-to-limitations
language in the future. These amendments now add that language. For example, with this regulation, 21 CFR
862.1190 states that the copper test system "is exempt from the premarket notification procedures in subpart E of
part 807 of this chapter subject to the limitations in 862.9. " (Emphasis added.) FDA is adding this specific
reference to the limitations for [*38787] consistency, clarity, and convenience. The status of the devices is not
changing.
This document is published as a final rule with the effective date shown under the DATES section above. FDA has
already established by regulation that exemptions from premarket notification are subject to certain limitations (e.g.,
21 CFR862.9). This rule merely cross-references, for clarity and convenience, in individual classification regulations
the sections that establish these limitations. FDA, therefore, has determined that this final rule has no substantive
impact on the public. FDA, therefore, for good cause, finds under 5 U.S. C. 553(b)(3)(B) and (d)(3) that notice and
public comment are unnecessary and that this rule may take effect upon publication.
IT. Environmental Impact
The agency has determined under 21 CFR 25.30(h) that this action is of a type that does not individually or
cumulatively have a significant effect on the human environment. Therefore, neither an environmental assessment
nor an environmental impact statement is required.
III. Analysis oflmpacts
FDA has examined the impact of the rule under Executive Order 12866 and the Regulatory Flexibility Act (5
U.S. C. 601-612) (as amended by subtitleD of the Small Business Regulatory Fairness Act of 1996 (Public Law 104-
121)), and the Unfunded Mandates Reform Act of 1995 (Public Law 104-4). Executive Order 12866 directs agencies
to assess all costs and benefits of available regulatory alternatives and, when regulation is necessary, to select
regulatory approaches that maximize net benefits (including potential economic, environmental, public health and
safety, and other advantages; distributive impacts; and equity). The agency believes that this final rule is consistent
with the regulatory philosophy and principles identified in the Executive order. In addition, this rule is not a
significant regulatory action as defined by the Executive order and so is not subject to review under the Executive
order.
The Regulatory Flexibility Act requires agencies to analyze regulatory options that would minimize any significant
impact of a rule on small entities. Because this rule does not change the status quo for these devices, the agency
certifies that this final rule will not have a significant economic impact on a substantial number of small entities.
Section 202(a) of the Unfunded Mandates Reform Act of 1995 requires that agencies prepare a written statement of
anticipated costs and benefits before proposing any rule that may result in an expenditure by State, local, and tribal
governments, in the aggregate, or by the private sector, of $100 million in any one year (adjusted annually for
inflation). The Unfunded Mandates Refonn Act does not require FDA to prepare a statement of costs and benefits
for the final rule, because the final rule is not expected to result in any 1-year expenditure that would exceed $100
million.
IV. Paperwork Reduction Act of 1995
This final rule contains no collections of information. Therefore, clearance by the Office of Management and
Budget under the Paperwork Reduction Act of 1995 is not required.
List of Subjects
21 CFR Parts 862, 866, 868, 870, 872, 874, 876, 878, 880, 882, 884, 888, and 890
Medical devices.
21 CFR Part 864
Biologics, Blood, Laboratories, Medical devices, Packaging and containers.
21 CFR Part 886
* * * * *
******
282. Section 878.4635 is amended by revising paragraph (b) to read as follows:
878.4635 - Ultraviolet lamp for tanning.
* * * * *
(b) Classification. Class I (general controls). The device is exempt from the
premarket notification procedures in subpart E of part 807 of this chapter, subject to
the limitations in 878.9.
283. Section 878.4660 is amended by revising paragraph (b) to read as follows:
*******
FEDERAL REGISTER
Vol. 62, No. 111
Notices
DEPARTMENT OF HEALTH AND HUMAN SERVICES (HHS)
Centers for Disease Control (CDC)
[Program Announcement 775]
Primary Prevention Skin Cancer Strategies for Children, Parents, andCaregivers
62 FR 31604
DATE: Tuesday, June 10, 1997
To view the next page, type .np* TRANSMIT.
To view a specific page, transmit p* and the page number, e.g. p* 1
[*31604]
Introduction
The Centers for Disease Control and Prevention (CDC) announces the availability of fiscal year (FY) 1997 funds
for cooperative agreement projects for ptimary prevention of skin cancer, and to build a national primary prevention
effort that targets children (aged 0-13), parents, and caregivers. Caregivers are defined as those individuals who
spend a significant number of consecutive hours with a child or children on a daily basis, i.e., grandparents, day-care
workers, teachers, foster parents, etc. Project activities will be developed to complement previous and ongoing
efforts of the National Skin Cancer Prevention Education Program (NSCPEP) and focus on two program options.
Applicants may choose one or both of the options. The strategies or activities proposed for each option chosen must
be clearly identified and stand alone, and applications must include separate narratives and budgets for each option
selected.
Applicants not adhering to this requirement will be disqualified.
Option One: Develop and conduct a skin cancer primary prevention intervention.
Option Two: Develop partnerships, coalitions, or interest groups with the lay, professional, and scientific
community that supplement and support the primary prevention efforts of the NSCPEP.
CDC is committed to achieving the health promotion and disease prevention objectives of "Healthy People 2000,"
a national activity to reduce morbidity and mortality and to improve the quality of life. This announcement is related
to the priority area of Cancer. (For ordering a copy of "Healthy People 2000", see the section "Where To Obtain
Additional Information.")
Authority
This program is authorized under section 317(k)(2) of the Public Health Service Act, as amended (42 U.S. C.
247b(k)(2)). Applicable program regulations are found in 42 CFR part 5lb-Project Grants for Preventive Health
Services.
Smoke-Free Workplace
CDC strongly encourages all grant recipients to provide a smoke-free workplace and to promote the non-use of all
tobacco products, and Pub. L. 103-227, the Pro-Children Act of 1994, prohibits smoking in certain facilities that
receive Federal funds in which education, library, day care, health care, and early childhood development services
are provided to children.
Eligible Applicants
Eligible applicants are public and private not-for-profit organizations, governments, and their agencies. Thus,
universities, colleges, research institutions, other not-for-profit public and private organizations, State and local
governments or their bona fide agents, federally recognized Indian tribal governments, Indian tribes or Indian tribal
organizations, and small, minority-and/or women-owned not-for-profit businesses are eligible to apply.
Note: Organizations described in section 50l(c)(4) of the Internal Revenue Code of 1966 that engage in lobbying
are not eligible to receive Federal grant and cooperative agreement funds.
Availability of Funds
Approximately $800,000 is available in FY 1997 to fund approximately four awards. A minimum of one award
will be made for each of the Options. The average award will be $200,000, with awards ranging from approximately
$150,000 to $250,000. It is expected that the awards will begin on or about September 30, 1997, and will be for a 12-
month budget period within a project period of up to 3 years. Funding estimates may vary and are subject to change.
Continuation awards within the project period will be made on the basis of satisfactory progress and the
availability of funds.
Use of Funds
Restrictions on Lobbying
Applicants should be aware of resttictions on the use of HHS funds for lobbying of Federal or State legislative
bodies. Under the provisions of 31 U.S. C. 1352 (which has been in effect since December 23, 1989), recipients (and
their subtier contractors) are prohibited from using appropriated Federal funds (other than profits from a Federal
contract) for lobbying Congress or any Federal agency in connection with the award of a particular contract, grant,
cooperative agreement, or loan. This includes grants/cooperative agreements that, in whole or in part, involve
conferences for which Federal funds cannot be used directly or indirectly to encourage participants to lobby or to
instruct participants on how to lobby.
In addition, the FY 1997 HHS Appropriations Act, which became effective October I, 1996, expressly prohibits
the use of 1997 appropriated funds for indirect or "grass roots" lobbying efforts that are designed to support or defeat
legislation pending before State legislatures. This new law, Section 503 of Public Law 104-208, provides as follows:
Section 503(a) No part of any appropriation contained in this Act shall be used, other than for normal and
recognized executive-legislative relationships, for publicity or propaganda purposes, for the preparation, distribution,
or use of any kit, pamphlet, booklet, publication, radio, television, or video presentation designed to support or
defeat legislation pending before the Congress,* **except in presentation to the Congress or any State legislative
body itself.
(b) No part of any appropriation contained in this Act shall be used to pay the salary or expenses of any grant or
contract recipient, or agent acting for such recipient, related to any activity designed to influence legislation or
appropriations pending before the Congress or any State legislature.
Department of Labor, Health and Human Services, and Education, and Related Agencies Approptiations Act,
1997, as enacted by the Omnibus Consolidated Appropriations Act, 1997, Division A, Title I, section !Ol(e), Public
Law I 04-208 (September 30, 1996).
Background
Skin cancer is the most common form of cancer in the United States, which accounts for more than one million
new cases annually or roughly one third of all new cancer cases. Basal and squamous cell skin cancers are the most
common types of skin cancer and tend to have a low mortality but high morbidity that may result in disfigurement
and disability. Melanoma has a lower incidence, but a higher mortality rate among the skin cancers. The American
Cancer Society estimates that in I 997, 40,300 persons will be diagnosed with melanoma of the skin and 7,300 will
die from the disease. There will be a projected total of 9,490 deaths, 2,100 resulting from basal cell, squamous cell,
and a small proportion of more rare skin cancers. From 1973-1992, the overall percentage increase in the rate of
death of melanoma (34.1%) was the third highest of all cancers. Incidence rates are over 10 times higher among
whites than among blacks (11.7 per 100,000 v. 0.8 per 100,000 for the period 1985-1989). Mortality from cutaneous
melanoma has increased, [*31605] although less rapidly than the incidence. Survival has improved partly because of
an increase in the proportion of cases diagnosed at the localized stage.
Unprotected exposure to ultraviolet radiation, from the sun or nonsolar sources
such as tanning beds, is strongly associated with skin cancer. Melanoma appears to
have a strong association with early life sun exposure and sunburns. Because of the
apparent link between severe sunburns during childhood and increased risk of
melanoma later in life, special efforts should be made to protect children from the
sun. Basal cell cancer and melanoma appear to be occurring at earlier ages, which
implies the early initiation of activities that significantly increase sun exposure
among children.
There are some predisposing risk factors that appear to heighten the propensity for the development of skin cancer
such as the presence or family history of skin cancer; large mole count; fair or light colored complexion, hair and
eyes; and skin that readily burns from sun exposure.
Currently, it is recommended that people of all ages, and especially those with light complexions, limit sun
exposure. Parents and caregivers should limit sun exposure for infants and children. Childhood education is
considered a priority target for prevention because children receive an estimated 70-80 percent of lifetime sun
exposure before the age of 18; excessive sun exposure early in life appears to increase the risk of the subsequent
development of skin cancer later in life, and beneficial behavior patterns established during early childhood often
persist throughout life. Children are particularly at risk for sun exposure and have the greatest lifetime potential to
benefit from positive sun protection habits. Strategies should identify discrete actions children, parents, and
caregivers can take to assure adequate protection from the sun.
Since 1994, CDC has been developing and implementing the NSCPEP program. Related projects funded by CDC
include: development and evaluation of skin cancer primary prevention education strategies; media campaigns with
resultant widespread media dissemination; national skin cancer prevention education agenda-setting meetings;
development of partnerships; development of educational brochures with other agencies and organizations, and
development of guidelines for skin cancer prevention in the school and community. In the fall of 1996, CDC co-
sponsored and participated in a workshop related to basal cell and squamous cell skin cancers, spear-headed by the
National Institute of Arthritis, Musculoskeletal, and Skin Diseases, National Institutes of Health. Workshop
deliberations affirmed the need to develop strategies aimed at the protection of children from over exposure to the
sun and the recommendations reflected this. The previously mentioned activities have provided guidance and focus
to CDC's advances in skin cancer prevention. As a result, CDC will continue to focus efforts on primary prevention
strategies that support the initiation, growth, and maintenance of the NSCPEP, partnerships with national
professional organizations, agencies, institutions, and the media.
Purpose
This program will assist in developing and building upon efforts that are consistent with the NSCPEP. The primary
goal of this program is to develop, conduct, and evaluate strategies that effectively reach children, parents, and
caregivers, and are aimed at reducing skin cancer through the adoption of preventive behaviors and the institution of
sun protection measures. These measures may include environmental interventions, such as physical/structural
modifications or incentives. Such strategies could include providing physical structures and accompanying
incentives to seek shade, and requiring the use of hats, protective clothing, etc., when outside or altered times for
outdoor activities.
Program Requirements
In conducting activities to achieve the purpose of this program, the recipient will be responsible for the activities
under A. (Recipient Activities) and CDC will be responsible for the activities under B. (CDC Activities).
A. Recipient Activities
I. Seek input from persons in the targeted population, representative interest groups, and persons who can
complement activities and provide expertise such as medical, behavioral, and public health perspectives.
2. Inventory resources needed to develop, conduct, and evaluate the intervention, such as hardware, software,
skills, capabilities, and material and logistic resources, e.g. training materials, transportation, etc.
3. Develop the intervention.
4. Develop procedures and tools for collecting pre-intervention data, intervention process data, and post
intervention data.
5. Create a marketing plan. Include testing of the plan to ensure that adequate numbers of the targeted population
are informed and have the opportunity to participate.
6. Pilot test the intervention among a representative sample of the targeted population.
7. Conduct the intervention in a defined targeted population, taking into account modifications and adjustments
identified during the pilot test.
8. Analyze and evaluate the results of the intervention using appropriate qualitative or quantitative methods.
Include an assessment of the fidelity of the methodology and protocol, and a description of results with respect to
awareness, knowledge, and to the degree possible, behavioral change atttibuted to the intervention in the targeted
population.
9. Participate in conferences, workshops, and meetings convened by CDC.
1. Define and provide justification for the scope of the proposed partnerships, coalition(s), or interest group(s). The
scope can be a diverse group of interested agencies and organizations, including public health; public and private
education agencies; voluntary organizations; advocacy groups; not-for-profit and for profit organizations, etc., or a
more narrowly defined group of interested agencies and organizations that has as their constituent base the
populations for which this program is intended, for example, children and youth organizations; schools; media and
private sector partners; parks and recreation organizations; U.S. sport and athletic organizations, parent
organizations, etc. The magnitude of reach should describe the level at which the activities will occur (local, State,
regional, or national).
2. Develop the purpose, mission, objectives, and expected outcomes of the partnerships, coalition(s), or interest
group(s).
3. Develop criteria for selecting members based on #2, include length of the term and ways to optimize member
involvement and buy in.
4. Define the level of involvement and expected contributions of members. Address issues related to organizational
structure and function; composition of subcommittees and ad hoc committees; decision making processes, etc.
[*31606]
5. Identify ways to enhance process efforts, such as building infrastructure, facilitating group process and
communication, and planning and attending to meeting logistics.
6. Establish an initial agenda for action and facilitate group process to develop a purpose, short-and long-term
goals, and activities.
7. Develop a strategy to sustain partnerships, coalition(s), or interest group(s).
8. Describe plans for integrating efforts and activities into ongoing national efforts.
9. Develop a mechanism for monitoring and reporting coalition activities and accomplishments. This may include,
but is not limited to, meeting minutes, attendance logs, operational and procedural manuals, etc.
10. Participate in conferences, workshops, and meeti ngs convened by CDC.
B. CDC Activities
1. Provide scientific and programmatic technical assistance.
2. Participate with and assist recipient in identifying appropriate agencies and organizations that will enhance
project activities.
3. Collaborate with recipients to develop, implement, evaluate, and disseminate project activities designed to
improve and change the knowledge, attitude, and impact on behaviors of the targeted groups.
4. Monitor the recipient's performance of project activities, attainment of project objectives, and compliance with
other CDC requirements.
5. Provide periodic updates about skin cancer prevention public knowledge, attitudes, and practices, and scientific
data when available.
6. Assist with the design and conduct of the evaluation plan, including project outcomes and process measures, and
modifications, as deemed necessary.
7. Coordinate dissemination of recipients' experiences and results through grantee meetings, workshops, and
conferences with other CDC recipients, other NSCPEP projects, and CDC.
8. Assist recipients with dissemination of project results in the public domain, through venues such as professional
publications, presentations at conferences, etc.
Technical Reporting Requirements
Semi-annual progress reports are required and must be submitted no later than 30 days after each semi-annual
reporting period. The semi-annual progress reports must summarize the following: (I) A comparison of actual
accomplishments to the goals and objectives established for the reporting period; (2) the reasons for slippage if
established goals were not met; and (3) other pertinent infonnation, including, when appropriate, analysis and
explanation of unexpectedly high costs for perfonnance.
An annual financial status report must be submitted no later than 90 days after the end of each budget period. Final
financial and performance reports are required no later than 90 days after the end of the project period. All reports
must be submitted to the Grants Management Branch, Procurement and Grants Office, CDC.
Application Content
All applicants must develop their applications in accordance with PHS Form 5161- 1 (Revised 7/92, OMB Number
0937-0189), information contained in this program announcement, and the instructions outlined below. Applicants
are required to submit an original and two copies of the application. Pages must be clearly numbered, and a complete
index to the application and its appendixes must be included. Begin each separate section on a new page. The
original and each copy of the application must be submitted unstapled and unbound. All materials must be
typewtitten, single-spaced, with unreduced type on 8 112 by 11" paper, with at least 1" margins, headers and footers,
and ptinted on one side only.
Appendixes should be of a reasonable length; only include documents necessary to support the application, such as
Letters of Support and examples of relevant work, as requested.
Applicants should discuss technical, programmatic, and public health expertise they can offer in the development
of national skin cancer prevention efforts and in participation in national meetings and on committees and task
forces. An evaluation plan should be included with the application.
Applicants may elect to submit proposals that address one or both of the options. Each option must be treated as a
separate submission or application and the application(s) should not exceed 30 pages, excluding appendixes.
A. Executive Summary
Provide a clear, concise, one-page summary of: (J) The capabilities and experience in conducting activities related
to the Option selected. Include any activities conducted in skin cancer prevention; (2) the major objectives of the
proposed project; (3) roles and responsibilities of proposed project personnel, including collaborators; and (4) the
estimated total cost of the project, including the total funds requested.
B. Demonstrated Capabilities
Provide evidence, based on previous projects, of the ability to:
1. Describe examples of previous primary prevention intervention work, including those in skin cancer prevention
or in other health areas. Discuss organization capability, scope, magnitude of reach (local, State, regional, national),
targeted population, process and evaluation methodology, and desctiption of the outcomes and efficacy.
2. Include evidence of adequate resources to develop, conduct and evaluate interventions, such as staff expertise,
facilities, hardware, and software. Describe the capabilities available to obtain additional resources when
appropriate.
3. Include evidence of direct work with children, parents, and caregivers, and/or evidence of collaborative efforts
on projects with interest groups and organizations, representing children, parents, and caregivers, that have
conducted primary prevention interventions, including those in skin cancer prevention or in other health areas.
community that supplement and support the current efforts of the primary prevention educational activities of the
NSCPEP.
1. Describe previous experiences and provide examples of development or substantive participation and sustain
ability of previous partnerships, coalition(s), or interest group(s). Include names or types of members, scope,
magnitude of reach (local, State, regional, national), process and evaluation methodology, and a description of
outcomes and efficacy.
2. Describe the organizational role and processes employed to ensure adequate [*31607] resources to develop,
implement, evaluate, and sustain partnerships, coalition(s), or interest group(s).
3. Describe and include evidence of past or current experience and participation in partnerships, coalition(s), or
interest group(s) that have children, parents, or caregivers as constituents, and that have conducted prevention
activities.
4. Include final reports, proceedings, materials developed, or a list of accomplishments resulting from group
activities in the appendix.
C. Project Objectives
Submit overall project objectives that are specific, measurable, realistic, and time-phased. Activities during year
01 through year 03 should be related and build on previous work. This should be reflected in the overall project
objectives. The objectives and activities related to year 01 should be described in detail. Year 02 and 03 objectives
and activities should be briefly described. End-of-year and end-of-project expected outcomes should be included.
D. Operational Plan
Describe the operational plan for achieving each of the objectives established in section C. Provide a concise
description of each major activity, and how it will be carried out. Include proposed collaborative efforts. Include
relevance to the National Skin Cancer Prevention Education Program efforts. The plan must have a timeline for
completion of each major activity. The year 0 I timeline must include specific process steps and include CDC review
and approval.
Letters of support that specify the precise nature of proposed collaboration, and the products, services, capabilities,
or other activities that will be provided through the collaboration should be included in the appendix.
Specifically for the Option selected, the Operational Plan should include the following:
1. Describe and provide a rationale for the proposed intervention. Include specific process steps that will be
undertaken to accomplish the proposed project. These steps should include, but are not limited to:
(a) The extent of problem; targeted population selection and rationale; baseline data on knowledge, attitudes, and
practices; literature review; incorporation of existing primary prevention or skin cancer prevention eff01ts;
theoretical framework; goals and objectives; development of intervention and marketing plan, including testing of
the intervention, to ensure that adequate numbers of the targeted population are informed and have the opportunity to
participate, and development of data collection tools. Include the availability of resources to be used on this project,
such as skills, capabilities, materials, and facilities.
(b) Plans for the implementation of the intervention, following the pilot or pretesting of the intervention in a
sample population. Include sampling, mechanisms for modification and retesting, and conduct of the intervention in
the population.
(c) The fonnative, outcome, and process measures proposed, and the methodology used to evaluate these
measures.
(d) The expected impact on the efforts of the NSCPEP.
2. Include specific plans to collaborate with key agencies and organizations representing targeted populations,
CDC, other grantee recipients, and current NSCPEP efforts. Include letters of support (in the appendixes) from
agencies and organizations with a substantive role in the proposed activities.
3. Include a detailed timeline for all proposed activities.
4. Include evaluation methodology of the intervention by using appropriate qualitative or quantitative methods.
Include an assessment of the fidelity of the selected methodology and protocol, and a description of proposed results
with respect to awareness, knowledge, and to the degree possible, behavioral change attributed to the intervention in
the targeted population.
commuojty that supplement and support the primary prevention efforts of the NSCPEP.
1. Include the scope of partnerships, coalition(s), or interest group(s). This should include the proposed
composition (diverse versus narrow) and the proposed magnitude of reach (local, State, regional, or national).
2. Include the proposed purpose, objectives, and expected outcomes of the partnerships, coalition(s), or interest
group(s).
3. Include criteria used for selecting members, ways to use and optimize member involvement, plans to sustain
membership and proposed members or types of members. Include in the appendix, Letters of Support from persons
interested and willing to participate.
4. Include process steps used to conduct the meetings; facilitate group process; build group infrastructure;
communi cate with the group before, during, after, and between meetings; and manage and plan for meeting activity
logistics, including travel, meeting space, etc.
5. Include an initial plan for action and methods for facilitating the group to develop the purpose; short- and long-
term goals; and activities of the group.
6. Include a detailed timeline for all proposed activities.
7. Include plans to coordinate with other grantees, and other NSCPEP skin cancer prevention coalitions currently
in progress, and CDC.
E. Project Management
Describe the capabilities, function, time dedication, and qualifications required for each position. Include
collaborators, their qualifications, and reason for their selection.
Specifically for Option selected, Project Management should include the following:
Option One: Provide evidence that a well-balanced team of experts bas been assembled to assure that the
intervention selected will be designed and developed by using necessary sciences. Include behavioral scientists,
evaluation scientists, dermatologists, public health personnel, and the targeted audience in all steps of the process.
Option Two: Provide evidence that a staff person or a consultant has been retained who has expertise in group
process and facilitation, as well as substantive experience in coalition development, management, and evaluation.
Include evidence of strong management, organizational , and human relations skills.
F. Budget
Provide a detailed budget request (using Standard Form 424A "Budget Information") and line-item justification of
all proposed operating expenses consistent with the option selected and the proposed activities. Use the sample
budget included in the application kit as a guide to budget development. Include the folJowing:
1. Travel plans in year 01: Budget two trips to CDC in Atlanta, Georgia, for conferences, workshops, or a reverse
site visit. Plan to travel one or two persons, for one to three days.
2. All proposed contracts must indicate the following: (1) Name of contractor, (2) Method of selection, (3) Period
of performance, (4) Scope of work, (5) Method of accountability, and [*31608] (6) Detailed budget with a
justification for costs.
Evaluation Criteria (Total of 100 Points)
The application will be reviewed and eval uated according to the following criteria:
A. Demonstrated Capabilities (20 Points Each)
The extent to which all items in the application content are addressed for Option selected including:
1. Provides examples of previous work similar to the nature of Option selected. Includes targeted populations,
scope, magnitude of reach (local, State, regional, national), evaluation methodology, and outcomes and efficacy.
2. Provides evidence of adequate resources to develop, conduct, and evaluate activities, such as staff expertise,
working knowledge of Option selected, facilities, logistical support, and hardware and software.
3. Provides evidence of direct work with children, parents and caregivers, or evidence of collaborative efforts on
projects with interest groups and organizations representative of these that have conducted prevention activities.
B. Project Objectives (20 Points)
The appropriateness of proposed objectives that are specific, measurable, time-phased, and realistic for year 01
activities, and a brief description of proposed objectives for years 02 and 03, and the extent to which end-of-year,
and end-of-project expected outcomes are described and effect the effort of the National Skin Cancer Prevention
Education Program. Epidemiologic data should be included to support and prioritize the need for a targeted primary
prevention activity in the Option selected.
C. Operational Plan (Option One: 40 Points Total, 25 Points for the General Operational Plan and 15 Points for the
Evaluation Plan; Option Two: 35 Points Total, 25 Points for the General Operational Plan and 10 Points for the
Evaluation Plan)
I. Provides evidence of a planning process that includes data and needs assessment, literature review, activity
selection, and selection of the targeted population (Option One), partnerships, coalition(s), or interest group(s)
(Option Two).
2. Provides a cogent, logical, complete description and process steps of activities.
3. Provides goals, project objectives, and expected outcomes.
4. Provides a timeline that includes CDC review and approval at critical decision-making and work-related steps.
5. Provides evidence of resources necessary to successfully address the activities, such as skills, capabilities and
staff, logistical support, and hard and software necessary to carry out Option selected.
6. Provides a plan to market and disseminate activities.
7. Provides an Evaluation Plan that includes the methodology for monit01ing formative process, and outcome
measures. Includes a description of data collection tools; CDC collaboration, review and approval; Human Subjects,
Minorities and Women Research review and other agency review.
D. Project Management (Option One: 20 Points Each; Option Two: 25 Points)
Provides a description of the capabilities, function, and qualifications of the proposed staff, staff functions, and
other resources needed to effectively pe1i'orm requested activities in selected Option.
E. Budget (Not Weighted)
The extent to which the applicant provides a detailed budget and justification consistent with the stated objectives
and proposed project activities for Option selected included in the application content and with this program
announcement.
F. Human Subject (Not Weighted)
Whether or not exempt from the Department of Health and Human Services (DHHS) regulations, are procedures
adequate for the protection of human subjects? Recommendations on the adequacy of protections include: (1)
Protections appear adequate and there are no comments to make or concerns to raise; (2) protections appear
adequate, but there are comments regarding the protocol, (3) protections appear inadequate and the Objective
Review Group (ORG) has concerns related to human subjects; or (4) disapproval of the application is recommended
because the research risks are sufficiently serious and protection against the risks are inadequate as to make the
entire application unacceptable.
Noncompeting Continuation Application Content
In compliance with 45 CFR 74.12l(d) and 92.10(b)(4), as applicable, noncompeting continuation applications
submitted within the project period need only include:
A. A brief progress report that describes the accomplishments of the previous budget period.
B. Any new or significantly revised items or information (objectives, scope of activities, operational methods,
evaluation, etc.) not included in the Year 01 application.
C. An annual budget and justification. Existing budget items that are unchanged from the previous budget period
do not need re-justification. Simply list the items in the budget and indicate that they are continuation items.
Supporting justification should be provided where appropriate.
Executive Order 12372 Review
Applications are subject to Intergovernmental Review of Federal Programs as governed by Executive Order (E.O.)
12372. E.O. 12372 sets up a system for State and local government review of proposed Federal assistance
applications. Applicants should contact their State Single Point of Contact (SPOC) as early as possible to alert them
to the prospective applications and receive any necessary instructions on the State process. For proposed projects
serving more than one State, the applicant is advised to contact the SPOC of each affected State. A current list of
SPOCs is included in the application kit. If SPOCs have any State process recommendations on applications
submitted to CDC, they should send them to Sharron P. Orum, Grants Management Officer, Grants Management
Branch, Procurement and Grants Office, Centers for Disease Control and Prevention (CDC), 255 East Paces Ferry
Road, NE., Room 314, Mailstop E-18, Atlanta, GA 30305, no later than 60 days after the application deadline date.
The Program Announcement Number and Program Title should be referenced on the document. The granting agency
does not guarantee to "accommodate or explain" State process recommendations it receives after that date.
Indian tribes are strongly encouraged to request tribal government review of the proposed application. If tribal
governments have any tribal process recommendations on appli cations submitted to CDC, they should forward them
to Sharron P. Orum, Grants Management Officer, Grants Management Branch, Procurement and [*31609] Grants
Office, Centers for Disease Control and Prevention (CDC), 255 East Paces Ferry Road, NE., Room 314, Mailstop E-
18, Atlanta, GA 30305.
This should be done no later than 60 days after the application deadline date. The granting agency does not
guarantee to "accommodate or explain" for tribal process recommendations it receives after that date.
Public Health System Rep011ing Requirements
This program is subj ect to the Public Health System Reporting Requirements. Under these requirements, all
community-based nongovernmental applicants must prepare and submit the items identified below to the head of the
appropriate State ancllor local health agency(s) in the program area(s) that may be impacted by the proposed project
no later than the receipt date of the Federal application. The approptiate State ancllor local health agency is
determined by the applicant. The following information must be provided:
a. A copy of the face page of the application (SF 424).
b. A summary of the project that should be titled "Public Health System Impact Statement" (PHSIS), not exceed
one page, and include the following:
(1) A description of the population to be served;
(2) A summary of the services to be provided; and,
(3) A description of the coordination plans with the appropriate State ancllor local health agencies.
If the State and/or local health official should desire a copy of the entire application, it may be obtained from the
state Single Point of Contact (SPOC) or directly from the applicant.
Catalog of Federal Domestic Assistance Number
The Catalog of Federal Domestic Assistance Number is 93.283.
Other Requirements
Paperwork Reduction Act
Projects that involve the collection of information from 10 or more individuals and funded by the cooperative
agreement will be subject to review by the Office of Management and Budget (OMB) under the Paperwork
Reduction Act.
Human Subjects
If the proposed project involves research on human subjects, the applicant must comply with the Department of
Health and Human Services Regulations, 45 CFR part 46, regarding the protection of human subjects. Assurance
must be provided to demonstrate that the project will be subj ect to initial and continuing review by an appropriate
institutional review committee. The applicant will be responsible for providing assurance in accordance with the
appropriate guidelines and form provided in the application kit.
In addition to other applicable committees, Indian Health Service (IHS) institutional review committees also must
review the project if any component of IHS will be involved or will support the research. If any American Indian
community is involved, its tribal government must also approve that p01tion of the project applicable to it.
Women, Racial, and Ethnic Minorities
It is the policy of the CDC and the Agency for Toxic Substances and Disease Registry (ATSDR) to ensure that
individuals of both sexes and the various racial and ethnic groups will be included in CDC/ATSDR-supported
research projects involving human subjects, whenever feasible and appropriate. Racial and ethnic groups are those
defined in OMB Directive No. 15 and include American Indian, Alaskan Native, Asian, Pacific Islander, Black and
Hispanic. Applicants shall ensure that women, racial and ethnic minority populations are appropriately represented
in applications for research involving human subjects. Where clear and compelling rationale exist that inclusion is
inappropriate or not feasible, this situation must be explained as part of the application. In conducting review for
scientifi c merit, review groups will evaluate proposed plans for inclusion of minoriti es and both sexes as part of the
scientific assessment of scoring.
This policy does not apply to research studies when the investigator cannot control the race, ethnicity and/or sex of
subjects. Further guidance to this policy is contained in the Federal Register, Vol. 60, No. 179, pages 47947-47951,
dated Friday, September 15, 1995.
Application Submission and Deadline
The original and two copies of the application PHS Form 5161-1 (Revised 7/92, OMB Number 0937-0189), must
be submitted to Sharron P. Orum, Grants Management Officer, Grants Management Branch, Procurement and
Grants Office, Centers for Disease Control and Prevention (CDC), 255 East Paces Ferry Road, NE., Room 314,
Mailstop E-18, Atlanta, GA 30305, on or before July 29, 1997.
1. Deadline: Applications shall be considered as meeting the deadline if they are either:
(a) Received on or before the deadline date; or
(b) Sent on or before the deadline date and received in time for submission to the objective review group.
(Applicants must request a legibly dated U.S. Postal Service postmark or obtain a legibly dated receipt from a
commercial carrier or U.S. Postal Service. Private metered postmarks shall not be acceptable as proof of timely
mailing.)
2. Late Applications: Application whi ch do not meet the cri teria in l.(a) or l.(b) above are considered late
applications. Late applications will not be considered in the current competiti on and will be returned to the applicant.
Where To Obtain Additional Information
To receive additional written information, call (404) 332-4561. You will be asked to leave your name, address, and
telephone number. Please refer to Announcement 775. You will receive a complete program description, information
on application procedures and application fonns. If you have questions after reviewing the contents of all the
documents, business management technical assistance may be obtai ned from Glynnis D. Taylor, Grants Management
Specialist, Grants Management Branch, Procurement and Grants Office, Centers for Disease Control and Prevention,
255 East Paces Ferry Road, NE., Room 314, Mailstop E-18, Atlanta, GA 30305, telephone (404) 842-6593, or
Internet or CDC WONDER electronic mail at gldl cdc.gov.
Programmatic technical assistance may be obtained from Barbara A. Bewerse, M.N. , M.P.H., Division of Cancer
Prevention and Control, National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease
Control and Prevention (CDC), 4770 Buford Highway, NE., Mailstop K-57, Atlanta, GA 30341-3724, telephone
(404) 488-4347, or Internet or CDC WONDER electroni c mail at byb0cdc.gov.
Please refer to Announcement 775 when requesting information and submitting an application.
You may obtain this and other announcements from one of two sites on the actual publication date: CDC's
homepage at http://www.cdc.gov or the Government Printing Office homepage (including free on-line access to the
Federal Register at http://www.access.gpo.gov).
Potential applicants may obtain a copy of "Healthy People 2000" Full [*3161 O]Report, Stock No. 017-001-00474-
0) or "Healthy People 2000" Summary Report, Stock No. 017-00 1-00473-1) referenced in the "Introduction" through
the Superintendent of Documents, Government Printing Office, Washington, DC 20402-9325, telephone (202) 512-
1800.
Dated: June 4, 1997.
Joseph R. Carter,
Acting Associate Director for Management and Operations Centers for Disease Control and Prevention (CDC).
[FR Doc. 97-15062 Filed 6-9-97; 8:45am]
Issue Number 7- SPECIAL EDITION Summer2005
The Truth About Tanning:
What You Need to Know to Protect Your Skin
There's no such thing as a safe tan.
Don't mistake the tan you get from hours spent by the
pool or under tanning lamps for a healthy summer glow,
it's actually a sign of sun damage from UV rays and can
~ ~ cause premature aging and skin cancer. But protecting
your skin now can help prevent the side effects caused by
too much sun.
The Inescapable UV Ray
Ultraviolet (UV) radiation is all around us. The most common source is sunlight, which produces three main
types of UV rays: UVA, UVB, and UVC. While UVA and UVB rays are transmitted through the atmosphere, all
UVC and some UVB rays are absorbed by the Earth's ozone layer. Most of the UV rays that reach the Earth's surface
are composed of UVA with a small amount of UVB.
UV light is classified by wavelength. UVB rays have a sh01t wavelength that reaches the outer layer of your skin,
called the epidermis. UVA rays have a longer wavelength that can penetrate and damage the lower layer of your skin,
called the dermis. It's imp01tant to use protection when you're out in the sun
because both UVA and UVB rays can cause sunburn, premature aging, skin cancer,
and damage to the eyes and immune system.
Because UV rays are strongest fromlO a.m. until 4 p.m., it's a good idea to
check the Ultraviolet Index (UVI) before you go outside. UVI is a number from
1-11 that indicates the amount of skin-damaging UV rays reaching the eatth's sur-
face at any point in time. The daily UVI number, listed in the weather section of
most city newspapers, forecasts the amount of UV you'll be exposed to during the
sun's highest point in the sky-usually around noon. The higher the UVI number is,
the more intense the exposure. If your local newspaper doesn't list the UVI for your
area, the Envimnmental Protection Agency (EPA) offers UVI forecasts by ZIP code
at http://www.epa.gov/sunwise/uvindex. html.
When the UVI is 5 or higher, you should always protect yourself from UV
exposure with sunscreen, a brimmed hat, and sunglasses; taking extra care to
reapply sunscreen and seek shade or stay indoors. Also remember that exposure
doesn't come only from above; snow, sand, water, and even concrete reflect UV
In This Issue
Recipe for a Tan
Sunburn
Long-Term Sun Damage
Skin Cancer
Prevention
Sunscreen
In the Salon
Sunless Tanning
Tanning Pills
Learn More
FDA& YOU
rays. In addition, clouds don' t block UVB and you can still get sunburned on a
cloudy day. So it's important to wear sunscreen and protective gear in all types of
weather.
Recipe for a Tan
Page2
DID YOU KNOW?
Direct sun isn't the
only cause of sunburn.
Whether from a day on the beach or hours spent in a tanning salon, the "tan"
color your skin gets after baking under UV rays is a sign of skin damage.
You can get sun-
burned even on a
cloudy day because
UV rays can filter
through the water
droplets that make up
When it's exposed to UV rays your skin produces a pigment called melanin to
protect skin cells from damage. Melanin is the same pigment that already colors
your hair, eyes, and skin. When your skin is exposed to UV rays it produces extra
melanin and may become darker over the next few days.
clouds.
Contrary to what you may have heard, getting a tan doesn't protect your skin from further UV damage.
The extra melanin in tanned skin provides a Sun Protection Factor (SPF) of about 2 to 4; far below the
minimum recommended SPF of 15.
While it's true that sunlight can have the benefit of helping your body produce vitamin D, about 10 to 15
minutes of unprotected sun on your face and bands 2 to 3 times a week provides you with a healthy dose. Too
much sun exposure can lead to sunburn, premature aging, or skin cancer.
Sunburn
Like a tan, a sunburn is a sign of short-term sun damage. Sunburn, also called erythema, is the skin's
natural defense against overexposure to UV rays. When UV rays reach your skin they begin damaging skin
cells in the epidermis. In response, your immune system increases blood flow in the affected areas, making
the skin feel warm and look red. White blood cells, which help protect you from infection and disease, attack
and remove the damaged skin cells. The process of removing the damaged cells can cause the skin to itch and
peel. Meanwhile, the damaged skin cells are releasing chemicals that send messages to your brain. Your
brain translates these messages into a painful burning sensation to let you know you've been sunburned.
Because it can take up to 8 hours for the full effects of sunburn to kick in, you won't realize that you've
been burned right away.
A mild sunburn can be treated with cool baths, over-the-counter
hydrocortisone creams, and aspirin to ease pain and swelling,
according to the American Academy of Dermatology (AAD). A
severe sunburn, usually characterized by a large area of red,
blistered skin with a headache, fever, or chills should be treated
as a medical emergency and examined by a doctor right away.
Studies have shown a link between severe sunburn and
melanoma, the most serious form of skin cancer, so any
sunburn should been taken seriously.
Leathery, wrinkled skin and dark spots are
common earmarks of a lifelong sunbather.
Unfortunately, since these signs of sun damage
don't usually show up until many years later,
you may think you're immune to the long-term
effects of tanning.
FDA& YOU
Everyone, no matter their skin tone, is at risk
for skin damage. There are six skin categories
recognized by the FDA and the AAD. Each is
classified by sensitivity to the sun and typical
skin tone. Check out the table to see which skin
type you are.
The best way to prevent sun damage is to
practice sun safety everyday. That includes wear-
ing sunscreen, a brimmed bat, and sunglasses
every time you go outside. Remember, even if
you can't see it now, the damage done today will
catch up with you later.
Skin Cancer
According to the American Cancer Society,
"Many of the more than 1 million skin cancers
that are expected to be diagnosed in 2005 could
have been prevented by protection from the sun's
rays. "
Skin
Type
I
II
Ill
IV
v
VI
Page3
Sun History Example
Always burns easily, Red-headed, freckles,
never tans, extremely Irish/Scots/Welsh
sun sensitive skin
Always burns easily, Fair-skinned, fair-haired,
tans minimally, very sun blue or green-eyed,
sensitive skin Caucasians
Sometimes burns, tans Average skin
gradually to light brown,
sun sensitive skin
Burns minimally, always Mediterranean-type
tans to moderate brown, Caucasians
minimally sun sensitive
Rarely burns, tans well , Middle Eastern, some
sun insensitive skin Hispanics, some
African-Americans
Never burns, deeply pig- African-Americans
mented, sun insensitive
skin
Experts agree that natural and artificial sunlight, particularly the UV rays, damages the skin. UV rays
cause the obvious short-term damage seen in a sunburn or a tan, as well as the long-term damage that
accumulates with each exposure.
When you tan you greatly increase your risk of developing skin cancer. This is especially true if you
spend time tanning each year because damage to the skin accumulates over time. Unlike skin cancer,
premature aging of the skin will occur in everyone who is repeatedly exposed to the sun over a long time,
although the damage may be less apparent and take longer to show up in people with darker skin.
There are three main types of skin cancer: melanoma, basal cell carcinoma, and squamous cell carcinoma.
Melanoma is the least common but most serious because it's responsible for most of the skin cancer deaths
each year. The other two types, basal cell and squamous cell carcinomas, are often refened to as non-
melanoma skin cancer. Basal cell cancer is the most common skin cancer, followed by squamous cell carcino-
ma, which can also become a killer
A fourth type of growth, actinic or solar keratosis, is also of concern because it can progress into cancer.
It's the most common pre-malignant skin condition, occumng in more than 5 million Americans.
DID YOU KNOW?
Australia has the highest incidence
of skin cancer in the world. To make
it easy for Australians to remember
how to protect their skin,
The Cancer Council Victoria coined
the catchy slogan: Slip! Slop! Slap!
Slip! - Slip on a shirt
Slop! - Slop on SPF 15+ sunscreen
Slap! - Slap on a wide-brim hat
Researchers still aren't sure why some people develop skin cancer
and others don't, but there are some preventive measures you can
take that may reduce your chances of getting skin cancer. While sun-
screens protect against sunburn, they don't necessarily prevent cancer.
If you use sunscreens to spend more time in the sun, your skin could
still be exposed to a high dose of UV, especially the longwave rays.
So it's still a good idea to stay out of the sun at midday, and to protect
yourself with sunglasses, a wide-brim hat, and protective clothing like
a long-sleeved shirt made of thick, light-colored fabric.
Moles or freckles that change shape, color, texture, or get crusty
and bleed could be a sign of skin cancer. Early-stage melanomas
often show up as a light brown to black flat mark that is usually about
FDA& YOU Page 4
Perform A Self Skin Cancer Check
No matter how much time you spend in the sun, you should
protect yourself by checking for signs of skin cancer. Visit
http://www.fda.gov/cdrh/fdaandyou/issue03.html#7 to learn how.
one-quarter inch in size. Any suspect spot
should be checked out by your doctor as soon as
possible. When detected in its earliest stages,
skin cancer is often curable.
For more information on skin cancer, visit
the American Academy of Dermatology's Web
..._ ____________________ ___. site at http://www.aad.org.
Prevention
The best way to protect your skin from the dangerous effects of UV rays is to take simple precautions
every day. Wearing sunscreen, shielding your face and eyes with a wide-brim hat, and sunglasses with a
UV A/UVB rating of 99% or higher, and seeking shade when possible can help decrease your risk.
Clothing can also help protect you from harmful UV in the form of protection you don't need to reapply.
Fabrics can differ greatly in their ability to shield you from UV rays, and natural fibers like cotton offer little
protection when wet.
The ideal sun protective fabrics are lightweight, comfortable, and protect against exposure even when wet.
SPF clothing are available that have thick, tightly woven fabrics with special fibers and dyes to help shield
you from the sun's rays. Remember that light-color fabrics will be cooler in the summer heat.
Certain medications, such as antibiotics, can make you more sensitive to the sun and put you at greater
risk for sunburn. Ask your doctor whether you're taking a medication that could affect your sensitivity to the
sun and what you should do.
Sunscreen
Sunscreen doesn't completely protect you from harmful UV rays, but it can drastically reduce their effects
if used properly. Sunscreen is available in a variety of forms for you to choose from, including sprays, lotions,
gels and wax sticks. Most sunscreens are made of chemicals
that absorb UV radiation. Others create a banier that reflects
the UV radiation away from the skin.
When shopping for sunscreen, chose one that's labeled as
broad-spectrum because it will help protect you from both UVA
and UVB rays. Check the sunscreen label for broad-spectrum
ingredients, such as benzophenones (oxybenzone), cinnamates
(octylmethyl cinnamate and cinoxate), sulisobenzone, salicy-
lates, titanium dioxide, zinc oxide, and avobenzone (Parsol
1789).
All sunscreens are labeled with SPF numbers. The higher
the SPF number, the more protection against sunburn the sun-
screen provides. To get the most protection out of sunscreen
choose one with an SPF of at least 15.
Some sunscreens are labeled as being water-resistant.
These sunscreens stay on the skin longer even if they get wet
from pool water, ocean water, or sweat. But water-resistant
doesn't mean waterproof. Water-resistant sunscreens still need
to be reapplied, so check the label for reapplication times.
Protect Yourself with
These Sun Safety Tips:
* Avoid the sun, or seek shade, from 10 a.m.
to 4 p.m. when the sun's rays are
strongest.
* Apply an SPF 15 or higher sunscreen
* Al low 30 minutes for skin to absorb
sunscreen before going outside.
* Check the label and reapply sunscreen
according to the instructions.
* Wear a wide-brimmed hat.
* Protect eyes with sunglasses that have a
UV/UVB protection of at least 99%.
* Check with your doctor to find out if you're
taking medications that will make you
more sensitive to the sun.
FDA & YOU PageS
The effectiveness of a sunscreen is reduced if it's applied inconectly or if it's washed off, rubbed off, or
sweated off. To make sure you're getting the maximum sunscreen protection, apply an even layer of sunscreen
and reapply it according to the directions.
Sunscreen usually needs about 15-30 minutes to soak in to the skin before you go outside. Read the label
to see how long you should wait. If the label doesn't indicate how long, wait 30 minutes to be safe.
In the Salon
With the convenience offered by a tanning salon, it may be tempting to lie in a tanning bed or sit in front
of a tanning lamp. Fight the urge! Tanning beds and lights are just as dangerous as tanning at the pool or on
the beach. The UVA rays emitted by a tanning lamp or bed are often much more intense than those produced
by the sun. The aging and cancer risks associated with outdoor tanning are the same as tanning in a salon.
For these reasons, the FDA doesn't recommend the use of indoor tanning equipment--EVER
If you insist on using a tanning lamp or bed, follow these steps to reduce the dangers of UV exposure.
Be sure to wear the goggles provided, making sure they fit snugly and aren't cracked.
Start slowly and use short exposure times to build up a tan over time.
Follow manufacturer-recommended exposure times for your skin type. Check the label for exposure times.
Stick to your time limit.
After a tan is developed, tan no more often than twice a week.
The key is to take it slow. If you get the maximum exposure the
first time, you'll probably get burned. And because sunburn takes
several hours to develop you won't realize your skin is burned until
much later.
FDA has a radiation safety perfonnance standard for sunlamp
products. All sunlamp products must have a warning label, an
accurate timer, an emergency stop control, and include an exposure
schedule and protective goggles.
You should NOT use a tanning bed or lamp if:
You sunburn easily and don't tan. Skin that doesn't tan in the
sun probably won't tan with sunlamps either.
You get frequent cold sores. UV radiation may cause them to
appear more frequently.
You're taking medicines that can make you more sensitive to
UV rays. Check with your doctor or pharmacist.
Sunless Tanning
The spray-on glow offered at sunless tanning booths make them a
popular place to maintain a tan year-round. What you may not realize
is that even sunless tanning can have risks.
Sunless tanning delivers a faux glow by coating your skin with
the chemical dihydroxyacetone (DHA). DHA interacts with the dead
Sun Screen Review
* Choose a water-resistant, broad-
spectrum sunscreen with SPF 15
or higher.
* Apply an even coat of sunscreen
over all exposed skin, including
your eyelids, lips, nose, ears, neck,
hands and feet.
* Allow 15-30 minutes for the
sunscreen to be absorbed by your
skin before going outside.
* Reapply sunscreen according to
the directions on the label-usually
about once every hour.
FDA& YOU Page 6
surface cells in the epidermis to darken skin color and simulate a tan, and the result
usually last for several days. DID YOU KNOW?
You should know that while the FDA allows DHA to be "externally applied" for
skin coloring, there are restrictions on its use. DHA should not be inhaled, ingested,
or used in such a way that the eyes and eye area are exposed to it because the risks,
if any, are unknown.
Many self-tanners
don't have sunscreen
in them.
Check the label. If it
doesn't have sun-
screen, apply one with
SPF 15 or higher
before going outside
Before using a sunless tanning booth, ask the tanning salon these questions to
make sure you'll be protected:
Will my eyes and the area surrounding them be protected?
Will my nose, mouth and ears be protected?
Willi be protected from inhaling the tanning spray through my nose or mouth?
If the answer to any of these questions is "no," look for another salon. Otherwise you're putting yourself
at risk for exposure to chemicals with potentially dangerous effects.
You should also take precautions if you're applying a self-tanner at home. Most self tanners contain the
same DHA used in sunless tanning salons. Self-tanners are available in many forms, including lotions, creams
and sprays that you apply and let soak in to your skin. Follow the directions on the self-tanner label carefully
and take care not to get the self-tanner in your eyes, nose, or mouth.
Tanning Pills
You may have seen ads that promise to give you a golden glow just by swallowing a
pill. Tb.is may sound too good to be true, because it is. These, so-called, tanning pills are
unsafe and none are approved by the FDA
Some tanning pills contain the color additive canthaxanthin. When large amounts of canthaxanthin are
ingested, the substance can turn the skin a range of colors from orange to brown. They can also cause serious
health problems including liver damage; a severe itching
condition called urticaria; and an eye disorder called
canthaxanthin retinopathy, in which yellow deposits form
in the retinas.
Learn More
To learn more about UV rays, tanning, and sun safety
visit:
The Environmental Protection Agency's SunWise Web
site at http://www.epa.gov/sunwise.
The FDA's Web page on sunscreens, tanning products,
and sun safety at http://vm.cfsan.fda.gov/-dms/cos-220.html
VISIT OUR BOOTH
FDA & YOU will be exhibiting at the National
Association of Health Education Centers
(NAHEC) Conference in Houston, Texas
August 30 & 31 st
About FDA & You
FDA & You is an FDA publication to inform and encourage
health educators and students to learn about the latest FDA
medical device and health news.
The publication's contents may be freely reproduced.
Comments may be sent to the editors.
Editor: Alicia Witters
Editor: Edle Seligson
Email: FDAandyou@cdrh.fda.gov
Read us online at
hUp://www.fda.goy/cdrhlfdaandyou.html
Department of Health and Human Services
Food and Drug Administration
Center for Devices and Radiological Health, HFZ-230
Rockville, MD 20850
Special thanks to
Sharon Miller, the OCER Radiation Experts and
Tammy Wallace
for contributing to this issue.
Issue Number 7- SPECIAL EDITION Summer2005
The Truth About Tanning:
What You Need to Know to Protect Your Skin
There's no such thing as a safe tan.
Don't mistake the tan you get from hours spent by the
pool or under tanning lamps for a healthy summer glow,
it's actually a sign of sun damage from UV rays and can
~ ~ cause premature aging and skin cancer. But protecting
your skin now can help prevent the side effects caused by
too much sun.
Ultraviolet (UV) radiation is all around us. The most common source is sunlight, which produces three main
types of UV rays: UVA, UVB, and UVC. While UVA and UVB rays are transmitted through the atmosphere, all
UVC and some UVB rays are absorbed by the Earth's ozone layer. Most of the UV rays that reach the Earth's surface
are composed of UVA with a small amount of UVB.
UV light is classified by wavelength. UVB rays have a sh01t wavelength that reaches the outer layer of your skin,
called the epidermis. UVA rays have a longer wavelength that can penetrate and damage the lower layer of your skin,
called the dermis. It's imp01tant to use protection when you're out in the sun
because both UVA and UVB rays can cause sunburn, premature aging, skin cancer,
and damage to the eyes and immune system.
Because UV rays are strongest fromlO a.m. until 4 p.m., it's a good idea to
check the Ultraviolet Index (UVI) before you go outside. UVI is a number from
1-11 that indicates the amount of skin-damaging UV rays reaching the eatth's sur-
face at any point in time. The daily UVI number, listed in the weather section of
most city newspapers, forecasts the amount of UV you'll be exposed to during the
sun's highest point in the sky-usually around noon. The higher the UVI number is,
the more intense the exposure. If your local newspaper doesn't list the UVI for your
area, the Envimnmental Protection Agency (EPA) offers UVI forecasts by ZIP code
at http://www.epa.gov/sunwise/uvindex. html.
When the UVI is 5 or higher, you should always protect yourself from UV
exposure with sunscreen, a brimmed hat, and sunglasses; taking extra care to
reapply sunscreen and seek shade or stay indoors. Also remember that exposure
doesn't come only from above; snow, sand, water, and even concrete reflect UV
In This Issue
Recipe for a Tan
Sunburn
Skin Cancer
Prevention
Sunscreen
In the Salon
Sunless Tanning
Tanning Pills
Learn More
FDA& YOU
rays. In addition, clouds don' t block UVB and you can still get sunburned on a
cloudy day. So it's important to wear sunscreen and protective gear in all types of
weather.
Recipe for a Tan
Page2
DID YOU KNOW?
Direct sun isn't the
only cause of sunburn.
Whether from a day on the beach or hours spent in a tanning salon, the "tan"
color your skin gets after baking under UV rays is a sign of skin damage.
You can get sun-
burned even on a
cloudy day because
UV rays can filter
through the water
droplets that make up
When it's exposed to UV rays your skin produces a pigment called melanin to
protect skin cells from damage. Melanin is the same pigment that already colors
your hair, eyes, and skin. When your skin is exposed to UV rays it produces extra
melanin and may become darker over the next few days.
clouds.
Contrary to what you may have heard, getting a tan doesn't protect your skin from further UV damage.
The extra melanin in tanned skin provides a Sun Protection Factor (SPF) of about 2 to 4; far below the
minimum recommended SPF of 15.
While it's true that sunlight can have the benefit of helping your body produce vitamin D, about 10 to 15
minutes of unprotected sun on your face and bands 2 to 3 times a week provides you with a healthy dose. Too
much sun exposure can lead to sunburn, premature aging, or skin cancer.
Sunburn
Like a tan, a sunburn is a sign of short-term sun damage. Sunburn, also called erythema, is the skin's
natural defense against overexposure to UV rays. When UV rays reach your skin they begin damaging skin
cells in the epidermis. In response, your immune system increases blood flow in the affected areas, making
the skin feel warm and look red. White blood cells, which help protect you from infection and disease, attack
and remove the damaged skin cells. The process of removing the damaged cells can cause the skin to itch and
peel. Meanwhile, the damaged skin cells are releasing chemicals that send messages to your brain. Your
brain translates these messages into a painful burning sensation to let you know you've been sunburned.
Because it can take up to 8 hours for the full effects of sunburn to kick in, you won't realize that you've
been burned right away.
A mild sunburn can be treated with cool baths, over-the-counter
hydrocortisone creams, and aspirin to ease pain and swelling,
according to the American Academy of Dermatology (AAD). A
severe sunburn, usually characterized by a large area of red,
blistered skin with a headache, fever, or chills should be treated
as a medical emergency and examined by a doctor right away.
Studies have shown a link between severe sunburn and
melanoma, the most serious form of skin cancer, so any
sunburn should been taken seriously.
Leathery, wrinkled skin and dark spots are
common earmarks of a lifelong sunbather.
Unfortunately, since these signs of sun damage
don't usually show up until many years later,
you may think you're immune to the long-term
effects of tanning.
FDA& YOU
Everyone, no matter their skin tone, is at risk
for skin damage. There are six skin categories
recognized by the FDA and the AAD. Each is
classified by sensitivity to the sun and typical
skin tone. Check out the table to see which skin
type you are.
The best way to prevent sun damage is to
practice sun safety everyday. That includes wear-
ing sunscreen, a brimmed bat, and sunglasses
every time you go outside. Remember, even if
you can't see it now, the damage done today will
catch up with you later.
Skin Cancer
According to the American Cancer Society,
"Many of the more than 1 million skin cancers
that are expected to be diagnosed in 2005 could
have been prevented by protection from the sun's
rays. "
Skin
Type
I
II
Ill
IV
v
VI
Page3
Sun History Example
Always burns easily, Red-headed, freckles,
never tans, extremely Irish/Scots/Welsh
sun sensitive skin
Always burns easily, Fair-skinned, fair-haired,
tans minimally, very sun blue or green-eyed,
sensitive skin Caucasians
Sometimes burns, tans Average skin
gradually to light brown,
sun sensitive skin
Burns minimally, always Mediterranean-type
tans to moderate brown, Caucasians
minimally sun sensitive
Rarely burns, tans well , Middle Eastern, some
sun insensitive skin Hispanics, some
African-Americans
Never burns, deeply pig- African-Americans
mented, sun insensitive
skin
Experts agree that natural and artificial sunlight, particularly the UV rays, damages the skin. UV rays
cause the obvious short-term damage seen in a sunburn or a tan, as well as the long-term damage that
accumulates with each exposure.
When you tan you greatly increase your risk of developing skin cancer. This is especially true if you
spend time tanning each year because damage to the skin accumulates over time. Unlike skin cancer,
premature aging of the skin will occur in everyone who is repeatedly exposed to the sun over a long time,
although the damage may be less apparent and take longer to show up in people with darker skin.
There are three main types of skin cancer: melanoma, basal cell carcinoma, and squamous cell carcinoma.
Melanoma is the least common but most serious because it's responsible for most of the skin cancer deaths
each year. The other two types, basal cell and squamous cell carcinomas, are often refened to as non-
melanoma skin cancer. Basal cell cancer is the most common skin cancer, followed by squamous cell carcino-
ma, which can also become a killer
A fourth type of growth, actinic or solar keratosis, is also of concern because it can progress into cancer.
It's the most common pre-malignant skin condition, occumng in more than 5 million Americans.
DID YOU KNOW?
Australia has the highest incidence
of skin cancer in the world. To make
it easy for Australians to remember
how to protect their skin,
The Cancer Council Victoria coined
the catchy slogan: Slip! Slop! Slap!
Slip! - Slip on a shirt
Slop! - Slop on SPF 15+ sunscreen
Slap! - Slap on a wide-brim hat
Researchers still aren't sure why some people develop skin cancer
and others don't, but there are some preventive measures you can
take that may reduce your chances of getting skin cancer. While sun-
screens protect against sunburn, they don't necessarily prevent cancer.
If you use sunscreens to spend more time in the sun, your skin could
still be exposed to a high dose of UV, especially the longwave rays.
So it's still a good idea to stay out of the sun at midday, and to protect
yourself with sunglasses, a wide-brim hat, and protective clothing like
a long-sleeved shirt made of thick, light-colored fabric.
Moles or freckles that change shape, color, texture, or get crusty
and bleed could be a sign of skin cancer. Early-stage melanomas
often show up as a light brown to black flat mark that is usually about
FDA& YOU Page 4
Perform A Self Skin Cancer Check
No matter how much time you spend in the sun, you should
protect yourself by checking for signs of skin cancer. Visit
http://www.fda.gov/cdrh/fdaandyou/issue03.html#7 to learn how.
one-quarter inch in size. Any suspect spot
should be checked out by your doctor as soon as
possible. When detected in its earliest stages,
skin cancer is often curable.
For more information on skin cancer, visit
the American Academy of Dermatology's Web
..._ ____________________ ___. site at http://www.aad.org.
Prevention
The best way to protect your skin from the dangerous effects of UV rays is to take simple precautions
every day. Wearing sunscreen, shielding your face and eyes with a wide-brim hat, and sunglasses with a
UV A/UVB rating of 99% or higher, and seeking shade when possible can help decrease your risk.
Clothing can also help protect you from harmful UV in the form of protection you don't need to reapply.
Fabrics can differ greatly in their ability to shield you from UV rays, and natural fibers like cotton offer little
protection when wet.
The ideal sun protective fabrics are lightweight, comfortable, and protect against exposure even when wet.
SPF clothing are available that have thick, tightly woven fabrics with special fibers and dyes to help shield
you from the sun's rays. Remember that light-color fabrics will be cooler in the summer heat.
Certain medications, such as antibiotics, can make you more sensitive to the sun and put you at greater
risk for sunburn. Ask your doctor whether you're taking a medication that could affect your sensitivity to the
sun and what you should do.
Sunscreen
Sunscreen doesn't completely protect you from harmful UV rays, but it can drastically reduce their effects
if used properly. Sunscreen is available in a variety of forms for you to choose from, including sprays, lotions,
gels and wax sticks. Most sunscreens are made of chemicals
that absorb UV radiation. Others create a banier that reflects
the UV radiation away from the skin.
When shopping for sunscreen, chose one that's labeled as
broad-spectrum because it will help protect you from both UVA
and UVB rays. Check the sunscreen label for broad-spectrum
ingredients, such as benzophenones (oxybenzone), cinnamates
(octylmethyl cinnamate and cinoxate), sulisobenzone, salicy-
lates, titanium dioxide, zinc oxide, and avobenzone (Parsol
1789).
All sunscreens are labeled with SPF numbers. The higher
the SPF number, the more protection against sunburn the sun-
screen provides. To get the most protection out of sunscreen
choose one with an SPF of at least 15.
Some sunscreens are labeled as being water-resistant.
These sunscreens stay on the skin longer even if they get wet
from pool water, ocean water, or sweat. But water-resistant
doesn't mean waterproof. Water-resistant sunscreens still need
to be reapplied, so check the label for reapplication times.
Protect Yourself with
These Sun Safety Tips:
* Avoid the sun, or seek shade, from 10 a.m.
to 4 p.m. when the sun's rays are
strongest.
* Apply an SPF 15 or higher sunscreen
* Al low 30 minutes for skin to absorb
sunscreen before going outside.
* Check the label and reapply sunscreen
according to the instructions.
* Wear a wide-brimmed hat.
* Protect eyes with sunglasses that have a
UV/UVB protection of at least 99%.
* Check with your doctor to find out if you're
taking medications that will make you
more sensitive to the sun.
FDA & YOU PageS
The effectiveness of a sunscreen is reduced if it's applied inconectly or if it's washed off, rubbed off, or
sweated off. To make sure you're getting the maximum sunscreen protection, apply an even layer of sunscreen
and reapply it according to the directions.
Sunscreen usually needs about 15-30 minutes to soak in to the skin before you go outside. Read the label
to see how long you should wait. If the label doesn't indicate how long, wait 30 minutes to be safe.
In the Salon
With the convenience offered by a tanning salon, it may be tempting to lie in a tanning bed or sit in front
of a tanning lamp. Fight the urge! Tanning beds and lights are just as dangerous as tanning at the pool or on
the beach. The UVA rays emitted by a tanning lamp or bed are often much more intense than those produced
by the sun. The aging and cancer risks associated with outdoor tanning are the same as tanning in a salon.
For these reasons, the FDA doesn't recommend the use of indoor tanning equipment--EVER
If you insist on using a tanning lamp or bed, follow these steps to reduce the dangers of UV exposure.
Be sure to wear the goggles provided, making sure they fit snugly and aren't cracked.
Start slowly and use short exposure times to build up a tan over time.
Follow manufacturer-recommended exposure times for your skin type. Check the label for exposure times.
Stick to your time limit.
After a tan is developed, tan no more often than twice a week.
The key is to take it slow. If you get the maximum exposure the
first time, you'll probably get burned. And because sunburn takes
several hours to develop you won't realize your skin is burned until
much later.
FDA has a radiation safety perfonnance standard for sunlamp
products. All sunlamp products must have a warning label, an
accurate timer, an emergency stop control, and include an exposure
schedule and protective goggles.
You should NOT use a tanning bed or lamp if:
You sunburn easily and don't tan. Skin that doesn't tan in the
sun probably won't tan with sunlamps either.
You get frequent cold sores. UV radiation may cause them to
appear more frequently.
You're taking medicines that can make you more sensitive to
UV rays. Check with your doctor or pharmacist.
Sunless Tanning
The spray-on glow offered at sunless tanning booths make them a
popular place to maintain a tan year-round. What you may not realize
is that even sunless tanning can have risks.
Sunless tanning delivers a faux glow by coating your skin with
the chemical dihydroxyacetone (DHA). DHA interacts with the dead
Sun Screen Review
* Choose a water-resistant, broad-
spectrum sunscreen with SPF 15
or higher.
* Apply an even coat of sunscreen
over all exposed skin, including
your eyelids, lips, nose, ears, neck,
hands and feet.
* Allow 15-30 minutes for the
sunscreen to be absorbed by your
skin before going outside.
* Reapply sunscreen according to
the directions on the label-usually
about once every hour.
FDA& YOU Page 6
surface cells in the epidermis to darken skin color and simulate a tan, and the result
usually last for several days. DID YOU KNOW?
You should know that while the FDA allows DHA to be "externally applied" for
skin coloring, there are restrictions on its use. DHA should not be inhaled, ingested,
or used in such a way that the eyes and eye area are exposed to it because the risks,
if any, are unknown.
Many self-tanners
don't have sunscreen
in them.
Check the label. If it
doesn't have sun-
screen, apply one with
SPF 15 or higher
before going outside
Before using a sunless tanning booth, ask the tanning salon these questions to
make sure you'll be protected:
Will my eyes and the area surrounding them be protected?
Will my nose, mouth and ears be protected?
Willi be protected from inhaling the tanning spray through my nose or mouth?
If the answer to any of these questions is "no," look for another salon. Otherwise you're putting yourself
at risk for exposure to chemicals with potentially dangerous effects.
You should also take precautions if you're applying a self-tanner at home. Most self tanners contain the
same DHA used in sunless tanning salons. Self-tanners are available in many forms, including lotions, creams
and sprays that you apply and let soak in to your skin. Follow the directions on the self-tanner label carefully
and take care not to get the self-tanner in your eyes, nose, or mouth.
Tanning Pills
You may have seen ads that promise to give you a golden glow just by swallowing a
pill. Tb.is may sound too good to be true, because it is. These, so-called, tanning pills are
unsafe and none are approved by the FDA
Some tanning pills contain the color additive canthaxanthin. When large amounts of canthaxanthin are
ingested, the substance can turn the skin a range of colors from orange to brown. They can also cause serious
health problems including liver damage; a severe itching
condition called urticaria; and an eye disorder called
canthaxanthin retinopathy, in which yellow deposits form
in the retinas.
Learn More
To learn more about UV rays, tanning, and sun safety
visit:
The Environmental Protection Agency's SunWise Web
site at http://www.epa.gov/sunwise.
The FDA's Web page on sunscreens, tanning products,
and sun safety at http://vm.cfsan.fda.gov/-dms/cos-220.html
VISIT OUR BOOTH
FDA & YOU will be exhibiting at the National
Association of Health Education Centers
(NAHEC) Conference in Houston, Texas
August 30 & 31 st
About FDA & You
FDA & You is an FDA publication to inform and encourage
health educators and students to learn about the latest FDA
medical device and health news.
The publication's contents may be freely reproduced.
Comments may be sent to the editors.
Editor: Alicia Witters
Editor: Edle Seligson
Email: FDAandyou@cdrh.fda.gov
Read us online at
hUp://www.fda.goy/cdrhlfdaandyou.html
Center for Devices and Radiological Health, HFZ-230
Rockville, MD 20850
Special thanks to
Sharon Miller, the OCER Radiation Experts and
Tammy Wallace
for contributing to this issue.
5600 Fishers Lane (HFI-40)
Rockville, MD 20857
September 2000
(FDA) 99-1279
~ c = ~ ~ ~ ~ ~ ~ 0 ~ @
0 ~
~ ~ ~ ~ ~ ~ v ~ ~ ~
~ U.S. Food and Drug Administration
The Food and Drug
Administration, or FDA, is a
United States government agency.
FDA makes rules about many
products consumers use when
sunning or tanning.
These rules help protect the
health of consumers.
When Outside in
The Sun
Beware
Of the
Dangers
Harmful rays from the sun,
sunlamps and tanning beds may
cause:
skin cancer, which can be
deadly
eye problems
weakened ability to fight
disease
unsightly ski n spots
wtinkles and "leathery"
skin.
Take Extra Care
Be sure to follow the seven steps
to safer sunning especially if
you answer yes to any of
these questions:
( Do you have pale white skin?
-o Do you have blonde, red or light brown hair?
Were you ever treated for skin cancer?
() Has a family member ever had skin cancer?
-() Do you have an illness? If so, ask your doctor
about extra care.
()Do you take medicines? If so, ask your doctor
about extra care.
Give babies and children extra care in the sun.
Protect Yourself
With the Seven Steps
To Safer Sunning
1. Stay in the shade.
Avoid the sun from:
to
10 a.m. 4p.m.
This is when sun rays are strongest. Don' t be fooled
by cloudy skies. Harmful rays pass through clouds.
2. Use sunscreen
products on your skin.
Many suntan products have
sunscreens to protect your skin
from the sun.
Products with sunscreens have
an "SPF" number on the label.
SPF stands for Sun Protection
Factor. A higher number means
it protects longer. Buy products
with an SPF number of 15 or
more.
Buy products whose label also
says:
( "broad spectrum," meaning
it protects against the two
types of harmful sun rays
( "water resistant," meaning
it stays on your skin
longer, even if you get wet
or sweat a lot.
Follow These Tips
For Using Sunscreen Products
Put a sunscreen of at least SPF 15 on your skin
15 to 30 minutes before going outside.
Rub the sunscreen evenly on all uncovered skin.
Be sure to put it on your eyelids, lips, nose, ears,
neck, hands and feet. If you do not have much
hair, put it on the top of your head.
Do not get a sunscreen in your eyes. It can sting.
Once in a while, put on more sunscreen while
you're in the sun. Read the label to see how
often to put it on.
-( Do not use a sunscreen on babies under 6
months old.
On children older than 6 months, use a sunscreen
every time they go out.
3. Wear a hat.
A hat with a wide brim helps
shade the neck, ears, eyes, and
head.
4. Wear sunglasses.
Buy only sunglasses with a
label sayi ng the glasses block
99 to 100 percent of the sun's
rays. If there is no label, do not
buy the glasses.
5. Cover up.
Wear loose, lightweight, long-
sleeved shirts and long pants or
long skirts when in the sun.
6. A void artificial
tanning methods.
This includes sunlamps and
tanning beds, as well as tanning
pills and tanning makeup.
Tanning pills have a color
additive that turns your ski n
orange after you take them.
FDA has OK'd this color
additive for coloring foods but
not for ta1ming the skin. The
large amount of color additive
in tanning pills may be harmful.
Tanning makeup is put on the
skin to make it look tan.
Sometimes the color can be
washed off with soap and water.
Other times, it wears off after a
few days. These products are
not sunscreen lotions and will
not protect your skin from the
sun.
7. Check your skin
regularly for signs of
skin cancer.
Look for changes in the size,
shape, color or feel of
birthmarks, moles and spots.
If you find any changes or find
sores that are not healing, see
your doctor.
1. Look at the
back of your
neck and scalp
with the help of
a hand minor. 2. Look at your body-
front, back and sides-
in the minor.
3. Bend your
elbows and
look at the
undersides of
your arms.
5. Check parts that
are hard to see- like
your back- with
a hand minor.
4. Look at the
backs of your
legs and feet.
Do You Have More
Questions About
Safer Sunning?
Ask your doctor or other health-
care worker.
And ask FDA. There may be an
FDA office near you. Look for the
number in the blue pages of the
phone book.
You can also contact FDA
through its toll-free number,
1-888-INFOFDA
( 1-888-463-6332).
Or, on the World Wide Web at
www.fda.gov.
Cancer Epidemiology, l C ~
Biomarkers & Prevention
Tanning Beds, Sunlamps, and Risk of Cutaneous Malignant
Melanoma
Richard P. Gallagher, John J. Spinelli and Tim K. Lee
Cancer Epidemiol Biomarkers Prev 2005;14:562-566.
Updated version
Cited Articles
Citing articles
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cebp.aacrjournals.org/content/14/3/562
This article cites by 25 articles, 6 of which you can access for free at:
http:/ /cebp.aacrjournals.org/content/14/3/562. full . htm l#ref -list -1
This article has been cited by 15 HighWire-hosted articles. Access the articles at:
http://cebp.aacrjournals.org/content/ 14/3/562.full .html#related-urls
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at pubs@aacr.org.
To request permission to re-use all or part of this article, contact the AACR Publications
Department at permissions@aacr.org.
Downloaded from cebp.aacrjournals.org on April 9, 2013. 2005 American Association for Cancer Research.
562 Cancer Epidemiology, Biomarkers & Prevent ion
Minireview
Tanning Beds, Sunlamps, and Risk of Cutaneous
Malignant Melanoma
Richard P. Gallagher,
1
'
2
,3 John J. Spinelli,
1
'
2
and Tim K. Lee l ,
4
' Cancer Control Research Program, British Columbia Cancer Agency; 'Department of Health Care and Epidemiology and
'Division of Dermatology, Faculty of Medicine, University of British Columbia, Vancouver, Canada;
and ' School of Computing Sciences, Simon Fraser University, Burnaby, Canada
Abstract
Background: A number of studies have been conducted
evaluating the risk of cutaneous malignant melanoma after
exposure to sunlamps and/or sunbeds. The proportion of
subjects in the individual studies who have reported
exposure has, in general, been modest, and the resulting
risk estimates for melanoma have been unstable with wide
95% confidence intervals (95% Cl). The inconclusive results
seen in individual studies have resulted in confusion as to
the carcinogenicity of these devices.
Methods: We conducted a systematic review and meta-
analysis of these studies. A review of the literature from Jan
1, 1984 to April 2004 using MEDLINE identified 12 case-
control studies and 1 cohort study which quantitatively
evaluated the use of sunlamps and/or sunbeds and subse-
quent melanoma. After applying exclusion/inclusion criteria,
9 case-control and 1 cohort study provided data for the
analysis. Summary odds ratios (OR) and 95% Cis for sunlamp/
Introduction
There is good experimental and epidemiologic evidence that
UV radiation exposure (mainly from sunlight) is causally
related to all forms of human skin cancer including cutaneous
ma)jgnant melanoma (1). Melanoma incidence has most
strongly and consistently been associated with reported
"intermittent sun exposure" mostly accrued through recrea-
tional activities. A quantitative review of studies of sun
exposure and melanoma found a positive association between
intermittent exposure and risk of cutaneous malignant
melanoma in 21 (statistically significant in 16) of 23 studies
included in the analysis (2}.
Because by its nature, exposure to artificial UV radiation
through sunlamp and sunbed use is intermittent in character,
there has been consistent concern over the past 15 years that
use of such devices for recreational tanning may increase risk
of melanoma (3). In addi tion, data from surveys conducted in
Europe (4, 5) and Nort h America (6) indicate that sunbeds are
now being used by an increasing proportion of the population,
particularly young people.
A series of epidemiologic investigations have attempted to
determine the nature of this putative association. However,
analysis of sunlamp/sunbed use has been hampered by small
Recei\ed 7 / 21J/ 04; revised 10/14/04; accepted 10/27/04.
Grant support: Michael Smith Fuundation for llealth Research ln1rastructure award.
The costs of publication of this artide were defrayed in part by the payment of page charges.
This article must therefore be hereby marked advertisement in accordance with 18 U.S.C.
Section 1734 solely to indicate this fact.
Requests for reprints: Richard P. Gallagher, Cancer Control Research Program, British
Columbia Cancer Agency, 600 West 10th Avenue, Vanc<>uver, BC, Canada VSZ 4E6.
Phone: 6048776000; Fax: 604-S77 1868. Email: richardgbccanccr.bc.ca
Copyright t'> 2005 American Association for Cancer Researd1.
sunbed use and subsequent melanoma were calculated using
a random-effect model.
Results: Ten studies provided data for assessment of melano-
ma risk among subjects who reported " ever" being exposed
compared with those " never'' exposed. A positive association
was found between exposure and risk (summary OR, 1.25;
95% Cl, 1.05-1.49). Significant heterogeneity between studies
was present. Evaluation of the metrics "first exposure as a
young adult" (5 studies) and ''longest duration or highest
frequency of exposure" (6 studies) also yielded significantly
elevated risk estimates (summary OR, 1.69; 95% Cl, 1.32-2.18,
and 1.61; 95% CJ, 1.21-2.12, respectively, with no heterogeneity
in either analysis).
Conclusions: Results indicate a significantly increased risk
of cutaneous melanoma subsequent to sunbed/sunlamp
exposure. (Cancer Epidemiol Biomarkers Prev 2005;14(3):
562- 6)
numbers of exposed subjects in individual studies. In order to
evaluate the strength and consistency of association between
use of these devices and melanoma, we have conducted a
systematic review and meta-analysis.
Materials and Methods
To identify relevant studies of sunbeds/sunlamps and
melanoma, we conducted a MEDLJNE database search for
the years 1984 to 2004, using the terms "melanoma (etiology,
epidemiology) and sunbeds, sw1lamps, and solarium" as
keywords. We also selected all review articles in English on
UV radiation, tanning devices, solaria, and melanoma, and
scrutinized these to locate articles with data on sunlamp and
sunbed use missed by the MEDLINE search. We did not
attempt to locate unpublished data. We chose 1984 as the
year of start of our search as that year saw the first
publications from large-scale epidemiologic studies of mela-
noma and UV exposure with good control for phenotype
confounders. The search yielded a total of 12 case-control
studies reporting risk estimates for melanoma subsequent to
sunbed/sunlamp use (7-18). In addition, one cohort study was
found which reported on usage among Norwegian and Swedish
women (19).
Selection of Studies. All case-control studies of cutaneous
malignant melanoma that reported on use of sunlamps,
sunbeds, or both were initially considered for the analysis.
For inclusion, we required that numbers (or percentages) of
exposed cases and controls were presented in the study, and
odds ratios (ORs) with 95% confidence intervals (95% Cl) were
Cancer Epidemiol Biomarkers Prev 2005;14(3). March 2005
Downloaded from cebp.aacrjournals.org on Apri l 9, 2013. 2005 American Association for Cancer Research.
available, evaluating at least "ever" versus "never" use of
these devices. Because of these requirements, two earlier
studies were omitted; one due to missing Cis (7) and the other
due to missing data on numbers of exposed cases and controls
(8). A study by MacKie et al. (10) was also omitted because it
was not possible from the information presented to determine
accurately the proportions or numbers of cases and controls
exposed. All other studies were included in the analysis.
In order to attempt to make the analysis reflect, as far as
possible, recreational rather than medical use of sunlamps/
sunbeds, we did not include studies of psoralen and UV A
radiation therapy [e.g., that of Stern et al. (20)]. Where studies
provided separate estimates of risk for medical and nonmed-
ical use, the estimates for nonmedical use were used. Where
separate estimates were not given, the overall ORs and Cis
were used. Whenever possible, ORs adjusted for phenotype
factors (hair, skin, eye color, phototype, number of nevi, etc.)
and sun exposure were used in preference to crude or
unadjusted values. The studies of Swerdlow et al (13) and
Osterlind et al. (9) present risk estimates as crude ORs
unadjusted for phenotype factors. Swerdlo"..,S indicated the
"ever versus never exposed" value was adjusted only for age,
sex, and region of residence. In the case of the study by Walter
et al. (14), risk estimates were calculated as crude and adjusted
ORs, and because the two analyses gave "essentially the same
effect" (p.236), the authors presented the unadjusted values.
Thus, the Walter estimates can be expected to closely
approximate adjusted values.
A number of the studies presented data which allowed us to
evaluate whether initial exposure to sunbeds/sunlamps
occurring earlier in life "as a young adult" conferred a
different risk than if exposure began closer to the time of the
study. A total of five studies contributed to this analysis and,
with the exception of the study of Walter et al. (14), the
estimates used were the adjusted values.
Finally, a number of studies attempted to determine
whether a dose-response gradient with exposure was seen.
We produced summary ORs comparing melanoma risk
between subjects with the longest duration or highest
frequency of exposure and subjects never exposed. Six studies
provided data for this analysis.
Two of the authors (R.P.G. and T.K.L.) examined each of the
studies independently to determine which risk estimate was
the best indicator of "first exposure as a young adult" and
which was the best for "longest duration or highest frequency
of use," and after a discussion concerning one study, agreed on
the measures considered to be most appropriate.
Summary ORs and 95% Cis were calculated for the three
measures of sunlamp/sunbed exposure noted above using the
method of DerSimonian and Laird (21). The Q statistic was
used as a test for heterogeneity among the original study
estimates (22). Published gender-specific ORs (14) or period-
specific ORs (15) were treated as separate entries or "studies"
in the meta-analysis. Sensitivity analyses were conducted by
recalculating summary ORs after eliminating specific studies.
Results
A total of 10 published articles (with 12 ORs) were used in
assessing the relationship betv.reen ever versus never use of
stmlamp/sunbed and melanoma (Table 1). The summary OR
showed a modest elevated risk (OR, 1.25i 95% Cl, 1.05-1.49).
Positive associations were seen in 8 of 10 individual studies,
although only 4 risk estimates were statistically significant.
One of the studies (9) showed an inverse association. There
5
Personal communication (April 30, 2004).
Cancer Epi demiology, Biomarkers & Prevention 563
was significant heterogeneity (Q = 28.9; P = 0.0024) likely due
to the studies being conducted over a long period of time with
different designs. Also, one study (9) showed results markedly
different from the others. Recalculation excluding the Oster-
lind study substantially reduced but did not eliminate the
heterogeneity, and had only a slight effect on the summary OR.
Excluding both the Osterlind study and the Swerdlow study
(not adjusted for phenotype factors) again had essentially no
effect (OR, 1.24; 95% CI, 1.09-1.41). The cohort study of Veierod
et al. (19) used "never/rarely" as the index in assessing
exposure. We included this measure in the analysis as
combining these two groups as the index should give a
conservative estimate of risk among users. Excluding the
Veierod study made little difference to the summary OR (OR,
1.21; 95% CI, 1.02-1.44).
Five studies contributed data to the analysis of first
exposure as a young adult (Table 2) which showed a positive
association with subsequent melanoma (OR, 1.69; 95% CI, 1.32-
2.18). Confidence intervals for this analysis were wider than
those seen for the previous metric because the estimates
contributing to the summary ORs were based on relatively
small numbers of subjects. All five of the individual studies
showed a positive association although only two were
statistically significant. This group of estimates showed no
evidence of heterogeneity (Q = 3.81; P = 0.58). Recalculation
excluding the Swerdlow study (no control for phenotype
factors) had only a slight effect on the summary measure (OR,
1.65; 95% CI, 1.28-2.13).
Data from six studies were entered into the analysis of
longest duration or highest frequency of use (Table 3). A
higher point estimate of risk was seen in this analysis (OR,
1.61; 95% CI, 1.21-2.12) than in the ever versus never analysis,
although the Cis for the two estimates overlapped slightly. All
six individual study estimates showed a positive association
with melanoma. Again, this group of estimates showed no
evidence of heterogeneity (Q = 5.90; P = 0.55), and recalcula-
tion excluding the Swerdlow study (no control for phenotype
factors) had virtually no effect on the summary OR (OR, 1.57i
95% CI, 1.19-2.09).
Discussion
This meta-analysis is subject to a number of limitations. The
estimates of risk for melanoma subsequent to using sunlamps/
sunbeds are based on published data in a series of 10 articles
over a period of 20 years. A pooled analysis of original
observations taken in the 10 studies would have provided a
more powerful approach to summarizing data on melanoma
and sunlamp/sunbed use. However, because the studies had
different overall aims, different metrics were used to record
duration and/or frequency of sunlamp/sunbed use. With the
exception of Westerdahl et al. (17), no study collected all the
information (years of use, frequency of exposure per year, and
duration of each exposure) needed to conduct a full quanti-
tative assessment of the association. This made pooling of raw
data infeasible.
Results of the analysis are, of course, dependent on the
choice of measures selected from each study. The overall
measure, ever versus never use, is fairly clear-cut, with the
caveat that we selected wherever possible the risk estimate
and Cis noted "for tanning purposes" rather than total use.
Recalculation of summary ORs using values for total sunbed/
sunlamp usage rather than use for tanning purposes
indicated that our decision did not affect the condusions
reached. The measures used to assess first exposure as a
young adult were subject to more judgment. The case-control
studies defined a young adult in different ways, with first
exposure ages ranging from "less than 25" to "less than 39
years." For the women's prospective cohort study (19), we
564 Tanning Beds and Melanoma
Table 1. Exposure to sunlamps and/or sunbeds and cutaneous malignant melanoma: case control and cohort study results
Ever vs. never exposed
Reference Place and period Cases Controls % Controls exposed Metric
Osterlind East Denmark 1982-1985 474 926 18% Ever used sunbeds
et al. (9)
Swerdlow UK (Scotland) 1979-1984 180 197 8.3% Ever used UV lamps
et al. (13) or sunbeds
Walter Ontario Canada 1984-1986 583 608 Males 14% Ever used sunbeds/
et al. (14)
Females 21%
sunlamps
Garbe Germany 1984-1987 856 705 7% Use of stmbeds-yes
et al . (11)
Males 14%
Au tier Germany, Belgium, 420 447 Ever exposed
et al. (15) France 1991-1993 to sunlamps
Females 17"/}
for tanning
Ever exposed
to sunbeds
for tanning
Westerdahl South Sweden 1988-1990 400 640 25% Ever used sunbeds/
et al. (16)
38%11
sunlamps
Holly San Francisco USA 1981-1986 452 930 Ever use of sunlamp
et a.t. (12)
Chen Connecticut USA 1987-1989 624
et al. (18)
512 Males 16% Ever used sunlamp
Females 22"/c,
Westerdahl South Sweden 1995-1997 571 913 Males 33% Ever used stmbeds
et al. (17)
Females 57%
Veierod Sweden and Norway; Total cohort, 2% of total female Exposed 1 I mo
et al. (19) female cohort 1991-1999 106,379 women cohort exposed in anv month
at age 10-39
1
Summary OR
No. studies= 12 (10 investigations, 1 with sex-specific, and 1 with exposure-specific risk estimates)
NOTE: Abbreviation: NS, not stated.
*Odds ratio adjusted in the original study for age, sex, host factors, and in some studies, sun exposure.
tOdds ratio tmadjusted.
!Sunbed use only; no OR given for sunlamp use.
Used for tanning purposes.
II Study conducted among female subjects only.
tComparison group is never I rarely.
OR (95% CI)
0.7 t, t (0.5-1.0)
2.9 t (1.3-6.4)
Males 1.88 t (1.20-2.98)
Females 1.45t (0.99-2.13)
M + F 1.62t (1.21-2.16)
1.5* (0.9-2.4)
1.77t (1.00-3.23)
Sunbeds 0.95t (0.64-1.41)
Ails 1.16 t (0.83-1.61)
1.3* (0.9-1.8)
0.94 t (0.74-1.2)
1.13* (0.82-1.54)
1.2 (0.9-1.6)
1.55* (1.04-2.32)
1.25 (1.05-1.49)
defined first exposure as a young adult to be women who
used sunbeds for the first time at ages 10 to 19. As this cohort
was comprised of women who were recruited in 1991 to 1992
at ages 30 to 50, early exposure would have occurred
before 1980.
The longest duration or highest frequency of use analysis
combined what we considered to be the best measure of
cumulative sunbed/sunlamp exposure available within each
study. The intention of this analysis was to compare risk
estimates in subjects with maximal cumulative usage to that in
Table 2. Exposure to sunlamps and/or sunbeds and cutaneous malignant melanoma: case control and cohort studies
First exposure as young adult
Reference Place and period Cases Controls % Controls exposed
Swerdlow UK (Scotland) 1979-1984 180 197 3%
et a!. (13)
Walter Ontario Canada 1984-1986 583 608 Males 7%
et a!. (14)
Females 12%
Chen Connecticut USA 1987-1989 624 512 8%
et al. (18)
Westerdahl South Sweden 1995-1997 571 913 9%
et al. (17)
2% of total female
cohort
Veierod Sweden and Norway; Total cohort
et a!. (19) female cohort 1991-1999 106,379 women
Summary OR
No. studies = 6 (5 studies, 1 with sex-separate risk estimates)
*Odds ratio adjusted in the original studies for age, sex, host factors, and in some studies, stm exposure.
t Odds ratio tmadjusted.
Metric
Age at first
exposure <30 y
First use
<age 30
<age 25 at first use
of sunlamp
First exposure at
age !!35
Exposed
age 10-19
OR* (95% CI)
3.8 (0.9-16.5)
Males 2.13
1
(1.13-4.13)
Females 1.55 t (0.94-2.59)
M +,F 1.75t (1.18-2.59)
1.35 (0.88-2.08)
2.3* (1.2-4.2)
1.52* (0.56-4.12)
1.69 (1.32-2.18)
Cancer Epidemiology, Biomarkers & Prevention 565
Table 3. Exposure to sunlamps and/or sunbeds and cutaneous malignant melanoma: case control and cohort studies
Longest duration or highest frequency of use
Reference Place and Period Cases Controls % Controls Exposed Metric OR 95% CI)
Swerdlow et al. (13) UK (Scotland) 1979-1984 180 197 2% Duration of use > 1 y 3.4* (0.6-20.3)
Walter et al. (14) Ontari o, Canada 1984-1986 583 608 Males 4% Sunbed/sunlamp use Males 2.12
1
(0.90-5.28)
:;::1/mo for :;::12 mo
Females 2% Females 2.99
1
(1.08-9.57)
M + F 2.44
1
(1.27-4.71)
Au tier et al. (15) Gennany, Belgium, 420 447 Prior to 1980 5% ;;:: 10 h exposure for Prior to 1980 2.12 (0.84-5.37)
France 1991-1993 tanning before 1980
1980 or later 2% :;::10 h exposure for 1980 or later 0.99* (0.49-2.00)
South Sweden 1988-1990
tanning 1980 or later
1.8* (1.0-3.2) Westerdahl et al. (16) 400 640 5% >10 exposures/y to
sunbeds/sunlamps
1.1f (0.60-2.20) Chen et al. (18) Connecticut USA 1987-1989 624 512 8% 2:10 uses of sunlamp
Westerdahl et al. (17) South Sweden 1995-1997 571 913 6% >250 sunbed uses 1.5 (0.7-3.2)
Summary OR
No. studies = 8 (6 studies, 1 with sex-specific estimates, and one with period-specific estimates) 1.61 (1.21-2.12)
*Odds ratio adjusted in the original studies for age, sex, host factors, and in some studies, sun exposure.
tOdds ratio unadjusted.
subjects "ever exposed" and those "never exposed." This
would enable us to determine whether there was a suggestion
of a gradient of risk from none to maximal use. Of course those
subjects most heavily exposed will also be included in the ever
exposed category, so the OR differences seen between the
levels of exposure are not independent.
The published articles covered a time period of nearly 20
years. During this time the UV emissions of artificial tanning
devices changed in character. Stmlamps used up to the late
1970s were usually used in the home setting (except for
medical use) and emitted primarily UVB (sometimes with a
small component of UVC). Tn the early 1980s, two major
changes took place. Indoor tanning began to be done largely in
commercial salons rather than at home, and salons began to
use VV A lamps. Thus, the character of the exposure changed
and because of this there is some question whether results
from the early studies (9, 13, 14) can be legitimately combined
with those of more recent studies. We suggest that combining
the results of the studies is appropriate, as there is no
convincing human data demonstrating that UVB is more or
less strongly related to melanoma than UV A. Tn fact, there is
some evidence that melanocytes exposed in vivo to either UV A
or to UVB show similar levels of thymine dimer formation (23).
In addition, studies have shown that UV A as well as UVB
impairs antioxidant function and promotes reactive oxygen
species, events known to be involved in cutaneous carcino-
genesis (24-27). Finally, the risk estimates from later studies,
taken together, do not differ in character from those seen in the
earlier investigations.
Finally, although we have used risk estimates adjusted for
phenotype factors and, when possible, for sun exposure, it is
unlikely that the studies have achieved complete control for
these potential confounders. If individuals who use artificial
tanning devices are more likely to suntan, as many suspect,
then some of the elevated risk seen might be due to
recreational sun tanning.
Our results, however, suggest that any exposure to artificial
tanning devices modestly, but significantly, increases risk of
cutaneous melanoma; however, caution is needed as it is
possible that there is unpublished (or unanalyzed) data on
sunbed or sunlamp use in existence from previously con-
ducted etiologic studies of melanoma. The subgroup analysis
of risk among those exposed "early in adult life" suggests a
risk estimate slightly higher than that seen in the ever versus
never analysis. We take this to indicate that risk increases with
adequate lag time (>10 years) after commencement of
exposure. However, as noted above, this lag time is potentially
confounded by the change in tanning device emissions from
UVB to UVA in the early 1980s. The latter interpretation,
however, seems unlikely as tisk estimates for exposure during
the period when UVA was emitted by sunbeds (17) are similar
to those seen during earlier periods.
Summary ORs for those with the longest duration or highest
frequency of use suggest a higher risk of melanoma among
those most heavily exposed than among those ever exposed.
Confidence intervals of the two estimates overlap; however,
this does give some indication that a dose-response effect
might be found to be present if exposure data were more
effectively captured.
Tf there is a causal relationship between stmlamp I sunbed
exposure and melanoma, the public health question is asked:
how important is the risk? It is not possible with the data
available to answer this question wi th any certainty. However,
a large number of epidemiologic studies have been conducted
on the relationship betvveen solar UV exposure and melanoma,
and these have been summarized quantitatively by Nelemans
et al. (28) and Elwood and Jopson (2). Combining the highest
reported levels of "intermittent" exposure produced a sum-
mary OR for all studies of 1.71 (95% CI, 1.54-1.90) in the
Elwood study, and 1.57 (95% CI, 1.29-1.91) for population-
based studies in the Nelemans investigation. These values are
similar to that seen for longest duration or highest frequency of
use of sunlamps/sunbeds in the present study. Although
caution is required due to the possibility of confounding by
inadequate control for concurrent sun exposure, the results
suggest that artificial UV may be a significant contributor to
risk among those with substantial exposure.
In summary, although it is not possible to determine
accurately how much sunbed/sunlamp use contributes to
individual risk of cutaneous malignant melanoma, it seems
clear that any use of these devices elevates risk for cutaneous
malignant melanoma. Furthermore, risk further increases
with appropriate lag tin1e, and frequency and duration of
use seem likely to be positi vely related to the magnitude of
the risk.
References
1. International Agency for Research on Cancer. !ARC monographs on the
evaluation of carcinogenic risk to humans. Vol 55. Solar and ultraviolet
radiation. Lyon: !ARC Press; 1992.
2. Elwood JM, jopson j. Melanoma and sun exposure; an overview of
pubUshed studies. Int J Cancer 1997;73:198-203.
3. Young AR. Tanning devices-fast track to skin cancer? Pigment Cell Res
2004;17:2- 9.
566 Tanning Beds and Melanoma
4. Amir Z, Wright A, Kernohan E, Har t G. Attitudes, beliefs and behavior
regarding the use of sunbeds among healthcare workers in Bradford. Eur
} Cancer Care 2000;9:76- 9.
5. }erkegren E, Sandreiser L, Brandberg Y, Rosdahl I. Sun-related behavior and
melanoma awareness among Swedish university students. Eur } Cancer
Prev 1999;8:27 -34.
6. Geller AC, Colditz G, Oliveria S, et al. Use of sunscreen, sunburning rates,
and tanning bed use among more than 10000 US children and adolescents.
Pediatrics 2002;109:1009 - 14.
7. Gallagher RP, Elwood }M, Hill GB. Risk factors for cutaneous malignant
melanoma: the western Canada melanoma study. Recent Results Cancer Res
1986;102:38-55.
8. Holman CDJ, Armstrong BK, Heenan P}, et al. The causes of malignant
melanoma: r ~ s u l t s from the West Australian Lions melanoma rt-search
project. Recent Results Cancer Res 1986;102:18- 37.
9. Osterlind A, Tucker MA, Stone B}, Jensen OM. The Danish case-control
study of cutaneous malignant melanoma: II. Importance of UV light
exposure. lnt J Cancer 1988;42:319 - 24.
10. MacKie RM, Frcudenberger T, Aitchison TC. Personal risk-factor chart for
cutaneous melanoma. Lancet 1989;2:487 - 90.
11. Garbe C, Weiss J, Kruger S, et al. The German Melanoma Registry and
environmental risk factors implied. Recent Results Cancer Res 1993;128:
69- 89.
U . Holly EA, Aston DA, Cress R, Ahn OK, Kristiansen JJ. Cutaneous melanoma
in women I. Exposure to sunlight, abili ty to tan, and other risk factors
related to ultraviolet light. Am J Epidemiol1995;141:923-33.
13. Swerdlow A}, English }S, MacKie RM, et al. Fluorescent lights, ultraviolet
lamps, and risk of cutaneous melanoma. BM} 1988;297:647-50.
14. Walter SO, Marrett LD, From L, Ht,rtzman C, Sharmon HS, Roy P. The
association of cutaneous malignant melanoma with use of sunbeds and
sunlamps. Am J Epidemiol 1990;131:232-43.
15. Autier P, Dore JF, Lejeune F, et al. Cutaneous malignant melanoma and
exposure to sunlamps and sunbeds: an EORTC multicentre case-control
study in Belgium, France and Germany. lnt J Cancer 1994;58:809- 13.
16. Westerdahl }, Olsson H, Masback A, et al. Use of sunbeds or sunlamps and
malignant melanoma in Southern Sweden. Am J Epidemiol 1994;140:691-9.
17. Westerdahl }, lngvar C, Masback A, Jonsson N, Olsson H. Risk of
cutaneous malignant melanoma in relation to use of sunbeds: further
evidence for UV-A carcinogenicity. Br } Cancer 2000;82:1593- 9.
18. Chen YT, Dubrow R, Zheng T, Barnhill RL, Fine J, Berwick M. Sunlamp use
and risk of cutaneous malignant melanoma: a population-based, case-
control study in Connecticut USA. lnt} Epidemiol 1998;27:758-65.
19. Veierod M13, Weiderpass E, Thorn M, et al. A prospective study of
pigmentation, sun exposure, and risk of cutaneous malignant melanoma in
women. J Nat! Cancer Inst 2003;95:1530-8.
20. Stern RS, Nichols KT, Vakeva LH; for the PUVA Follow-up Study.
Malignant melanoma in patients treated for psoriasis with methoxsalen
(psoralen} a.nd ultraviolet A radiation (PUVA}. N Eng! J Med 1997;336:
1041- 5.
21. DerSin1onian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials
1986;7:177 - 88.
22. Petitti DB. Meta-analysis, decision analysis, and cost effectiveness analysis:
methods for quantitative medicine. New York: Oxford University Press;
1994.
23. Young AR, Potten CS, Nikaido 0, et al. Human melanocytes and
keratinocytes exposed to UVB or UVA in viw show comparable levels of
thymine dimers. J Invest Dermatol 1998;111:936-40.
24. Fuchs J, Huflejt, Rothfuss LM, Wilson OS, Carcamo G, Packer L. Acute
effects of near ultraviolet and visible light on the cutaneous antioxidant
system. Photochem Photobiol1989;50:739- 44.
25. Fuchs }, Huflejt ME, Rothfuss LM, Wilson OS, Carcamo G, Packer L.
Impairment of enzymatic and nonenzymatic antioxidants in skin by UVB
irradiation. J Invest Dermatol1989;93:769-73.
26. Sander CS, Hamm F, Elsner P, Thiele JJ. Oxidative stress in mal ignant
melanoma and non-melanoma skin cancer. Br J Dermatol 2003;148:
913- 22.
27. Sander CS, Chang H, Hamm F, Elsner P, Thiele JJ . Role of oxidative stress
and the antioxidant network in cutaneous carcinogenesis. Int J Dermatol
2004;43:326- 35.
28. Nelemans PJ, Rampen FHJ, Ruiter OJ, Verbeek ALM. An addition to the
controversy on sunlight exposure and melanoma risk. J Clin Epidemiol
1995;48:1331-42.
A Preliminary Investigation of the
Predictors of Tanning Dependence
Carolyn J. Heckman, PhD; Brian L. Egleston, PhD; Diane B. \Vilson, EdD, RD;
Karen S. lngersoll, PhD
Objectives: To investigate pos-
sible predictors of tanning depen-
dence including demographic vari-
ables, exposure and protective be-
haviors, and other health-related
behaviors. Methods: This study con-
sisted of an online survey of 400
students and other volunteers from
a university community. Results:
Twenty-seven percent of the
sample was classified as tanning
dependent. Tanning dependence
S
kin cancer is the most common
form of cancer in the United States,
accounting for half of all human
cancers, ' with over a million new cases
diagnosed yearly. Nmety percent of all
skin cancers are due to ultraviolet radia-
tion (UVR) .
2
Thanks to recent media at-
tention. the US public has demonstrated
increased awareness of the harmful ef-
fects of UVR.' Unfortunately, studies have
also demonstrated that people. particu-
larly young adults. continue to use mmi-
Carolyn J Heckman, Assoetate Member. Popu
latron Scrence, Fox Chase Cancer Center,
Cheltenham, PA. Brian L. Egleston, As:srstant
Member, Brostatisttcs Faetlity, Fox Cha!;e Cancer
Center, Pluladelpltra, PA. Drane B. Wrlson, Asso
crate Professor, Department of Internal Medrcrne,
Virgmra Commonwealth Umuersity Richmond. VA
Karen S. Ingersoll. Assocrate Professor, Depart
ment of Psychratnc Medretne, University of Vrr
grnra Center for Addrctron Research and Educa
tron, Charlottesvrlle, VA.
Address correspondence to Dr Heckman, Popu-
lattOII Snence, Fox Chase Cancer Center, 510
Tow11shrplrne Rd. First Floor, Cheltenham, PA
19012. Email: Carolyn.lleckmanrafccc edu
Am J Health Bebn ."" 2 008;32(5):451-464
was predicted by ethnicity and skin
type, indoor and outdoor tanning
and burning, and lower skin protec
tive behavi ors, as well as smoking
and body mass index. Conclusions:
Young adults are at risk for tanning
dependence, which can be pre-
dicted by specific demographic and
behavioral variables.
Key words: tanning, skin can-
cer prevention, addiction
Am J Healt h &hav. 2008;32(5):451-464
mal skin protection strategies yet re-
ceive large amounts of intentional and
incidental exposure to UVR, either
through sun exposure or from the use of
tanmng salons
1 7
As suggested by the
continued exposure to UVR \vithout ad-
equate protection despttt increasmg
awareness of risks, clearly there are
strong motivations for such exposure and/
or Significant barriers to engaging in pro
tecbve behaviors. Most of the recent lit-
erature suggests that the effect on ap-
pearance ts the pnmary motivation for
sunbathing and tanning booth use . .l-
11
It is
no wonder then that educational mter-
ventions targeting skin safety and can-
cer knowledge have fallen short in terms
of changing actual tanning behavior.
11 1
"
However, there may be some indtviduals
who tan excessively for reasons other
than appearance Like other health risk
behaviors. tannmg behtwtor probably has
multiple determinants.
Similarities between excessive tan-
ning dnd substance use disorders or ad-
diction have been discussed formally in
the literature recently and anecdotally
for years.
17
In the lay media, the term
451
Tanning Dependence
tanorexia has been used to describe a
preoccupation with the desire to be tan
paired wi.th excessive tanning, similar to
anorexia nervosa or an obsessive desire
to be thin paired wi.th compulsive dietary
behavior.
18
Tanorexia is illustrated by the
recent explosion in media images of tan
celebnties
18
and numbers of tanning sa-
lons'1Q as well as sunless tanning prod-
ucts on the market. However, tanorexia
may be a misnomer, as substance depen-
dence or addiction may be a better model
for understanding excessive tanmng than
an eating disorder model.
Several similarities exist between tan-
ning and substance use. They are both
prevalent in youth, are often initially
perceived as image enhancing, and are
health-risk behaviors that people partici-
pate in despite warnings.
20 25
A primary
motivation for tanning behavior is ap-
pearance enhancement. However, tan
ners report other benefits such as mood
enhancement, relaxation, and socializa
hon, also consistent with addiction in
which behaviors are reinforcing in spe-
cific, pleasurable ways.
26
-
28
For example,
Hillhouse and TurrisP
1
have found that a
subset of uhardcore" frequent tanners
have seasonal affective disorder. These
individuals may be using tanning for self-
medication purposes.
A possible mechanism for tanning de-
pendence (TD) is the release of endog-
enous opioids during UVR exposure.2b.n
2910
There is some evidence supporting this
hypothesis.
21
'
21
111 30
Feldman et aP
6
investi-
gated UVR as a reinforcer among 14 regu-
lar indoor tanners. The participants dem-
onstrated a preference for a UVR versus a
non-UVR tanning bed despite successful
blinding in terms of visual or tactile cues.
UVR exposure was associated with a more
relaxed and less tense mood than non-
UVR exposure. Kaur et al'
7
compared 8
frequent and 8 infrequent tanners. Fre-
quent tanners demonstrated a prefer-
ence for a UVR versus a non-UVR tanning
bed. This preference was reduced with
increasing doses of naltrexone, an opioid
antagonist. Moreover, as the dose in-
creased, 4 of the 8 frequent tanners expe-
nenced withdrawal-like symptoms (nau-
sea and/or jitteriness), and 2 dropped out
of the study entirely due to these symp-
toms. Infrequent tanners demonstrated
less preference for the UVR bed and re-
ported no adverse events at any dose of
naltrexone. These withdrawal symptoms,
452
dose-response relationship, and opioid an-
tagonism are reminiscent of traditional
substance use disorders. However, al-
though 2 studies found increased plasma
levels of endorphins during UVR expo-
sure,
2
!1
30
2 other studies have failed to
demonstrate this effect. J J < t
A few other studies have noted associa-
tions and similarities between tanning
and substance use. Demko et al
22
found
that among 6903 nationally representa-
tive white adolescents, those who used 2
or 3 substances were 3 times more likely
to be indoor tanners than others. Several
other studies have found relationships
between tanning and substance use such
as cigarette smoking.
24 13
Zeller et aP
5
found that 29% of a sample of 267 teens
aged 14- 17 who tanned indoors more than
once in the last year reported that they
would have difficulty quitting indoor tan-
ning. Variables associated with antici-
pated quitting difficulty were younger age
at initiation and higher frequency of use,
characteristics often found among sub-
stance-dependent individuals
In another attempt to explore a possible
relationship between excessive tanning
and addiction, Warthan et al
17
modified
the substance dependence criteria from
the American Psychiatric Association's
Diagnostic and Statistical Manual-1V-
TR34 (DSM-IV-TR) and those of the 4-item
CAGE scale,
35
traditionally used to screen
for potential problems with alcohol use.
CAGE is an acronym that refers to the 4
yes or no items regarding trying to Cut
down on drinking (or in this case tan-
ning), feeling Annoyed when told not to do
a behavior, feeling Guilty when doing the
behavior too much, and wanting to par-
ticipate in the behavior first thing in the
morning (Eye opener). Using these scales,
Warthan et a1
17
found that in a sample of
145 Texas beachgoers, 26% met the modi-
fied CAGE criteria and 53% met the modi-
fied DSM-IV-TR criteria for tanning de-
pendence. Using a chi -square test,
Warthan et al
17
found that the results
from the 2 scales "were significantly as-
sociated (P = .03, p 964)." Their study
provides some evidence for construct va-
lidity in that women were more likely to
have positive CAGE resuJts, and partici-
pants who scored positive on the DSM-lV-
TR reported tanning more often and going
to the beach more often to tan than did
others. Likewise, Poorsattar and
Hornung
28
found that 12% of their under-
graduate sample met the CAGE criteria.
Based on our review of the literature.
no other psychometric studies have been
conducted on these TO measures, but
prc\ious studies have examined the psy-
chometrics of the CAGE and DSM-IV-TR
criteria for substance disorder screening
and diagnosis. A number of studies have
demonstrated good reliability, sensitivity
(77-86%). specificity (93 94%), and valid-
ity of the CAGE to screen for alcohol disor-
ders. 3631 There are several strong nnd
widely used instruments used to Q!;Sess
DSM-IV criteria for the diagnosis of sub-
stance dependence. For example, the
Structured Clinical Interview for DSM-IV
(SCIO) and the Composite InternatiOnal
Diagnostic i nterview (CIDJ) have been
found to have excellent interrater reh
ability, good test-retest reliability, and
superior validity to standard clinical in-
There 1s much more cvt-
dence supporting the use of DSM-IV crite-
ria for dtagnosis of sub:itance dependence
than for use of the CAGE screener How-
ever, DSM TV measures are typicnUy no
ministered as structured interviews
rather than questionnaires as was done
with the TO measure in the study by
Warthan ct al.
17
The purpose of the current study was to
replicate and extend the work of Warthan
et nP
7
by further investigating possible
predictors of tanning dependence such as
demographic variables, exposure and pro-
tective behaviors, and other health -re-
lated behaviors (smoktng and exercise)
in a general young adult sample. We wen
interested in demographic associations
with tanning dependence that might be
used clinically to identify individuals at
risk of being or becoming tanning depen-
dent. We were also interested m sun
exposure and protective associations with
tanning dependence that might support
the construct validity of the tannmg de-
pendence measures. Finally, we were
mterested in the other health behaviors
potentially related to tanning dependence
that might provide insight into the type of
theoretical model most applicable to indi -
viduals who are preoccupied with tan-
ning.
METHODS
Participants
Participants were students and other
volunteers (n = 400) from a
metropolitan university community who
Am J Health Behav."" 2008;32(5):451-464
Heckman et al
were recruited during the 2006 spnng
semester. College-aged panlc!pants were
selected for this studv because teens and
young adults are known to be frequent
tanners and at risk for substance abuse
and dependence.
821
40
...
3
Participants were
recruited from the community in addi-
tion to the Psychology 10 l subject pool in
order to enhance gcncrahzability.
Meaaurea
Tanning dependence. Wnrthnn et o.l '
7
modified CAGE and DSM-!V-TR
(mCAGE and mDSM-IV-TR) were used to
measure TO. These mstruments are the
only formal scales currently available to
measure the construct. As ment1oned
previously, CAGE is an acronym that re-
fers to the 4 yes or no items regarding
unsuccessful attempts at Cutting down,
Annoyance with admonitions, Guilt about
the behavior, and using the behavior as
an Eye-opener. The modified 7-item DSM-
IV-TR critena address tolerance, with-
drawn!, and tanning despite negative con-
sequences, key criteria of substance de-
pendence. Sample items nrc "Do you think
you need to spend more and more time m
the sun to maintain your perfect tan?",
woo you continue tannmg so your tan Will
not fade?", "Does this !your belief that
tannmg cnn cause skin cancer! keep you
from :.pending time m the sun or going to
tanning beds?" Most of the mDSM-IV TR
items had yes or no responses, but 2
items asked about days per week spent
tanning m salons and the sun. Scoring
was conducted as in the study by Wanhan
et al
17
in which 2 or more positive items
on the mCAGE or 3 or more on the mDSM-
IV-TR were deemed to indicate TD. This
sconng scheme is the same that is used
for substance dependence screening with
the original measure,
Demographic variables. Panicipants
repor <'d their sex, age, year in school ,
racef ethntcity, whether they were psy-
chology students, and thctr likelihood of
burning nnd tnnning. Their skin type wns
classified according to

There
arc 6 skin types with 1'ype I being the
fairest t:lnd most likely to burn and Type VI
bemg t ne darkest and lenst likely to burn
Outdoor tanning, indoor tanning,
and s kin protection. The following item!->
scales were modified from simtlar ones in
the liternture.
154
f>.S
1
Participants were que-
ried about their level of intentional and
incid1ntnl summer sun exposure, tan-
453
Tanning Dependence
ning booth use, and chemical sunless
tanner use. A protective behaviors scale
was used to assess a range of skin protec-
tion behaviors: wearing clothing, staying
in the shade, avoiding or limiting midday
sun, and using sunscreen. The scale had
7 Likert-type items ranging from 1 a
never to 5 always use such protection,
for a possible range of scores from 7 to 35.
This scale was divided into tertiles (low,
medium, and high protection) for the lo-
gistc regression analyses.
Other health-related behaviors. ln or-
der to perform a preliminary assessment
of other potential health-related predictors
of TO, ever or current use of tobacco and
exerctsc ttems from the Center for Dis-
ease Control and Prevention's 1995 Na-
tional College Health Risk Behavior Sur-
vey were The following items were
divided mto tertiles (low, medium, high):
body mass index (via self-report height and
weight), days per week exercismg aerobi-
cally, and days per week exercising anaero-
bically (strengthening or toning muscles).
Procedures
The current study was an online survey
of volunteer participants from a university
community. Participants were recruited
Vta a university psychology student subject
pool, flyers. and advertisements Students
went to a website to choose the
experiments, and if they selected the cur-
rent study, they were connected directly to
the survey URL. The URL was also posted
on flyers and ads around the university and
surrounding area so that other commu-
nity members could access it. Psychology
1 01 students could earn research pomts
for their classes by signing up for any of a
variety of experiments. A raffle for depart-
ment store merchandise gift cards was
offered as an incenuve. This study was
approved by the university institutional
review board.
Analyaea
The purpose of the analyses was to
identify predictors of TO as measured by
Warthan et al's
17
modified substance de-
pendence scales. Three separate mul-
tiple logistic regression models were cre-
ated to predict TO: one including demo-
graphic variables, one including other
UVR exposure and protective behaviors,
and one including other health behavior
variables. We used 3 regressions to sepa-
rately investigate ( 1) the demographic
454
variables associated with tanning depen-
dence, (2) the UVR-related behavior vari-
ables that would support the construct
validity of tanning dependence, and (3)
the other health- related variables that
would indicate the type of theoretical model
(for example, eating disorder versus ad-
diction) that would be most associated
with tanning preoccupation. For ease of
presentation and to allow for nonlinear
effects, we categorized continuous van-
abies into 3-level ordinal variables repre-
senting low, moderate, and high values of
the variables. For the 2 continuous UVR-
related variables that were found to be
stati&tically associated with tanning de-
pendence when categorized, we refit the
models but included the 2 variables using
restricted cubic spliness.. with knots (cut-
points allowing changes in the underly-
ing basis curves) at the same cut-points
as used to create the categorical vari-
ables. We used figures to display the ef-
fects of the variables on tanning depen-
dence when entered into the model using
restricted cubic splines. Restricted cubic
splines are a more flexible method of
modeling nonlinear effects than the m-
clusion of categonzed continuous vari-
ables in regressions. Although restricted
cubic splines allow for better estimation
of nonlinear effects, the parameters from
the models in which they are included
are largely uninterpretable due to the
scaling of the spline basis functions. For
this reason, we present the point esti-
mates of the parameters when the vari-
ables were categorized, but display the
impact from the spline models visually to
give a more accurate representation of
the relationship of the variables with
tanmng dependence.
Our study had power to detect modest
effects; with the distribution of tanning
dependent and nondependent participants
in our study, we had 80% power to detect
an apprmamate 0.3 unit standardized ef-
fect (a standardized effect is an effect in
standard deviation units) between the 2
groups using a t-test with a 2-sided hy-
pothesis test and a 5% Type I error rate
and 90% power to detect an approximate
0.4 unit standardized effect. Cohen!>4 has
defined a small effect as one of 0.2 stan-
dardized units and a medtum one of 0.5
units. We did not power the study explic-
itly to detect effects m a multiple logistic
regression. Hence, the findings from the
regression analyses should be consid-
Heckman et a1
Table 1
Participant Characteristics by Tanning Dependence (N=400)
Dependent
(n)
O"erall Sample 26.5 ( 106)
t"cmale St>x 77 (82}
Psycbolol!) 101 711 (H3)
ht \ ear St udent 60 (64)
\\hlte
85 (90)
Finpotrlck Skin f ) Jlt.'
1 7 (7)
2 10 (II)
'
26 (28)
4 29 (31}
5 26 (28)
6
I (I l
Sunles l.ast Year 32 (34)
E\t'r lndoor
65 (68)
Body lndu
<25 70 (73)
25 to lO 25(26)
JO+-
6 (6)
M (SE)
Age
21 . 11 (53)
Protection
19.53 (.52)
Indoor hnnlng
# Ltlt!ume K3 15 (10)
Age Started
16.98 (.48)
#Last Year
24.20(3. 111)
#Per Month If Warm
6.90 ( 1.22)
t/ Per Month If Cool 8.0!! (1 27)
Hn per \\>eel.
uobathing 7.50 (.58)
Hn per \\cek lncldenlal II 33 ( 1.31)
Lifet ime Burn\ 2.29 ( 39)
Burns Ln t \'ear
L60 ( 14)
Day\ Smoked of last 30 day, 6.38 ( 95)
I# Cigarette per Oa)
2.42 (.59}
Day Aerobic\
2.43 (.20)
Day' nocrobic\ 2.02 (. 19)
ercd exploratory and hypothesis generat-
ing. Therefore, it 1s possible that the
nonsigmficant findings in the regression
analyses may not necessarily represent
true negative findings but merely effects
that we did not have the power to detect.
RESULTS
Descriptive Characteristics
Participants were 400 individuals, 79%
Am J Health Behav."" 2008;32(5):451-464
:\ot Dependt>nl
%(n)
P-ulut'
73.5 (294)
14 {217)
NS
79 (232)
NS
59 (173)
NS
59 (172)
< 0001
.0015
13 (37)
14 (J2)
15 (43)
23 C6H)
27 (79)
8 (24)
20 (59)
.0142
29 (85)
< 0001

63 (ItO)
22 (64)
15 (45)
\ 1 (Sl)
P-\alue
20.92 (.32> NS
23.47 (.31)
< 0001
35.93 (9)
.0009
17.47 (.43)
NS
5.75 (2.83)
< 0001
87 ( 1.09)
0003
2.35 (1.14)
.0010
2.44 (.35)
.0001
9.84 ( 79)
NS
I.R6 (.23)
NS
.67 (01!) <.0001
3.89 (.57) .0250
I 40 ( 35}
NS
2 06( 12)
NS
I 56 (II)
.0360
(n '" 316) of whom were Psychology 10 1
students and 21% (n 84) of whom were
other members of the university commu-
nity. Seventy-five percent (n 300) of the
sample was female, and the mean age of
participants was 21 years (SO 5.42}.
Sixty percent (n .. 240) of the sample were
first-year students, 25% (n 1 00) were
upper-class students, 8% (n 32) were
graduate or special students, and 7% (n
455
Tanning Dependence
Table 2
De mographi c Predictors of Tanning Dependence (n=397)
Variable OR
Fc!male 1.37
I rcshman Reference
Sophomore 0.97
Jun1or
0 78
Scmor/Other Grade 0.54
Afncan American Reference
Asian American I 93
\V hitc 7.60
Hispamc/Latino 5 50
Other Racc/Elhnicity 0 69
Skin rype I Reference
Skin Type II I 39
Skin Type lJ1 3.51
Skin T)pe rv 3.08
Skin Type\ 4.06
Sktn Jyp;;; VI 1.40
Psychology Student 0.60
28) were not in school. SlXty-six percent
(n - 264) of part1cipants were white, 17%
(n = 68) were African American, 11% (n =
44) were Asian American, and 6% (n = 24)
were other races/ ethnicitJes or multira-
cial/ethnic. Table 1 includes participant
characteristics by TO status.
The Cronbach alpha for the mCAGE
was .57 and for the mOSM-IV-TR was .56,
both of which are relatively low but reflect
the diversity of behaviors assessed by the
instruments and their brevity. Further
reliability testing and/ or modifications of
the measures should be conducted m
future studies. The mean of the mCAGE
scores was 0.47 (SO= 0.84) out of 4. and
the mean of the m S M ~ I V TR scores was
l 51 (SO = 1.31) out of 7. Based on the
mCAGE criteria, a small minority of the
sample ( 11%, n = 44) was identified as
tanning dependent. The most commonly
endorsed item referred to being annoyed
with people's advice against tanning ( 17%,
n = 68). mDSM-IV-TR criteria identified
93 (23% of the sample) tanning-depen-
dent individuals. The most commonly
endorsed item had to do with tanning
despite cancer awareness (45%. n = 178).
The measures successfully agreed in
456
P-\alue 95%CJ
0.28 (0.7R, 2.40)
0.94 (0.47. 2.01)
0.63 (0.28. 2 19)
0.22 (0.20, 1.44)
0.33 (0.52, 7 18)
0.00 (2.49, 23. 19)
0.06 (0.96, 31.59)
0.75 (0 07. 6.83)
0.55 (0.48. 3 99)
0.01 ( 1.35, 9. 12)
0.02 (l.21, 7.83)
0.01 (1.49, 11.04)
0.79 (O. n. 15.42>
0.28 (0.24, 1.51)
classifying 80% (n = 320) of the partici
pants: 294 (73%) as not tanning depen
dent. and 30 (7.5%) as tanning dependent.
Seventy-seven ( 19%) participants were
identified as tanning dependent by only
one of the 2 scales, either the mCAGE or
the mOSM-IV-TR. Because it is not yet
clear which scale measures TO more
accurately and due to the relatively low
hit rate, TO was defined for th1s study as
meeting criteria on either the mCAGE or
the mOSM-IV-TR. The foJlowmg analyses
compare TDs (n = 1 06, 26.5%) with non
TDs (n = 294, 73.5%) on a variety of
characteristics and behaviors.
Demographics
The following variables were entered
into the demographic logistic regression
model predicting TO: gender, ethnicity,
year in school, Fitzpatrick skin type, and
psychology student or not (Table 2). Signifi-
cant independent predictors of TO were
ethnicity and skin type. As expected, whites
had 7.60 times greater odds of being TO
compared to African Americans (95% CI =
2.49-23.19) conditional on other covariates
in the model. Thirty-four percent (n = 90) of
whites, 7% (n = 5) of African Americans,
and 16% (n = 1 1) of other races I ethnicities
Heckman et al
Table 3
UVR-Related Predictors of Tanning Dependence (n=387)
Variable OR P-vulue 95%CI
Lateume Lo'' R..Jerence
Moderate 144 0.31 (0.72. 2.91)
Ltfcume sunburns- Htgh 1. 15 0.72 (0.53. 2.47)
Burnt> Last Year I ow Reference
\ cur - \1oderatc 1.98 0.07 (0.95, 4. 14)
Burn' I\ ear lligh 2.85 0.01 ( 1.37, 5 91 )
I lour ... \\'cck lm:adcntal L O\\ Reference
Hour ... Week lnctdental E:11po.,.ure Moderate 1.27 0.54 (0.59. ?. 7'\}
lloUI'li'Wcck Incidental Exposure High 1.37 0.40 (0.66. 2.88)
Hours. Week Sunbathing Lo'' Rcterence
ll oul"\ \\ eek Moderate 2.32 0.06 (0.98, 5.50)
llour Week Sunbathing Uigh 7.54 0.00 (3.34, 17.02)
,\n} sunlel>' tanner tn pa ... t }car 1.21 0.59 (0.61 "'40)
l. i feu me mdoor tanmng exposure- one Reference
l 1 fcttme mdoor wnmng expo .. ure Moderate 059 0.33 (0.20, I 71)
Ltli:ttme mdoor wnnmg e\po,urc lligh 0.79 0.74 (0. 19. 3.151
Indoor tannmg tn last )ear- Reference
Indoor tannmg m last year Moderate 2.15 0.18 (0.70. 6 65)
Indoor tannmg m last year lltgh 1.73 0.40 (0.49. 6 13)
Indoor tanning In warm \\'Cather 2.99 0.03 ( 1.14. 7.82)
Indoor tannmg tn cool '"eathcr 1.57 0.34 (0.62. 3.951
Sun protet.ti\C beha\tor- Lo\\ Reference
protecth e beha\ lor - \loderatl'
Sun protecth l' behavior - I flgh
were classified as TD. Additionally, indi -
viduals with moderate skin types were
more likely to be TD than those with the
fairest or darkest skin types with ORs
ro.nging from 3.08 to 4.06 in the model for
Fitzpatnck skin types 3 through 5 com-
pared to the frurest skin type (skin type I) .
TD rates were 16% (n .. 7) for Type I's, 21',
(n 11) for Type IJ's, 39% (n - 28) for Type
Ill 's, 32% (n a 31) for Type IV's, (n 28)
for fype V's, and 4to (n 1) for Type VI's.
Surprisingly, gender was not a significant
independent predictor of TD status.
Exposure and Protective Behavior
The following variables were entend
into the UVR logtstic regression model
predtcting T D time spent sunbathing,
mcidental sun exposure, sunburns. in-
door tanning, sunless tanners, and pro
Am J Health Behav.n. 2008;32(5):451-464
0.27 0.00 (0.13. 0.55)
03 6 0.01 (0. 17. 0.74)
tecttve behaviors (Table 3). Significant
independent predictors of TD were hours
per week spent sunbathing in the sum-
mer, sunburns during the last year, in-
door tanning dunng warm weather. and
protective behaviors. Participants spent
a mean of 3. 77 (SD - 6.34) hours per week
tanntng in the summer. As expected,
those with the highest levels of summer
sunbathing had 7.54 (95% cr .. 3.34-17.02)
greater odds of bemg TD than did those
with the lowest levels conditiOnal on other
covariates in the model. Participants re-
ported a mean of 0.92 (SD 1.46) sunburn
during the last year. Those with the high-
est number of sunburns had 2.85 (95% Cl
"" 1.37 5.91) greater odds of being TD in
the model compared with those with the
fewegt sunburns. The mean score on the
protective behaviors scale was 22.41 out
457
Tanning Dependence
Figure 1
Odds Ratio of Being Tanning Dependent Relative to Baseline
Note.
0
...,
0
N
0
Point Estimate
Pointwise 95% Cl
Null Effect
Data Polrds
, .... ,,
I '
I '
I ',
II .. ____ .,.. ......
I
,
I
' I
'
'
I
'
I
'
I
I
I
I
I
I
I
I f
,' /
, ' '
," ,,"
I
vu:.: :.::;;::,:;::;;.::.: .,. .-..... ,_,,,,,, ........... .
2 5 10 20
HoltS per Week Slllbathing
50 10 15
I
I
I
I
I
I
'
'
'
I
I
' I
I
I
I
I
,
,
,
..... , .. _,,
""- .... ,.,,
20 25 30 35
Protecllon Score
Baselint- Is no time spent sunbathing and the lowest sun protection score, from Lhe model in which
tbe 2 variables were rncluded as restricted cubic splines. The effects relative to baseline are
significant where the confidence intervals do not cross I. The distribution of the dala Is
depicted by the bars on the bottom of the chart; ror ease of display, 1he bars were slightly j ittered
because of overlap.
of 35 (SO 5.52), suggesting moderate
levels of skin protection. As expected,
those who protected themselves from sun
exposure were less likely to be TD in the
model with ORs of 0.27 (95% CI = 0.13-
0.56) and 0.36 (95% CI .. 0.17-0.74) among
the moderate and high protectors, re-
spectively.
Figure 1 displays the impact of hours
per week sunbathing and sun protective
score on the odds of being classified as
tanning dependent conditional on the
other variables in the model. The odds of
being tanning dependent, compared to
those who do not spend any time sunbath-
mg increases as the hours per week
spent sunbathing, increases except at
very high levels in which little tanning
dependence was found. Because of sparse
data, the 95% confidence interval at the
highest levels becomes very wide, how-
ever, reflecting high uncertainty in the
458
point estimates at the highest level. The
relationship of sun protective behavior
with the odds of being tanning dependent
is more consistent as protective behavior
increases. Relative to those who have the
least protective behavior, those with more
sun protective behavior are less likely to
be tanning dependent.
Thirty-eight percent (n = 152) of the
sample had tanned indoors, and only 23%
(n = 92) of the sample had used sunless
tanners during the last year. The mean
age of tanning booth initiation was 17.26
years (SD = 3.91). The mean number of
lifetime tanning booth uses was 56.74
(SO = 87.99) . It is important to note that
the mean age of participants was 21 , so
the typical participant had used a tanning
booth 14 times per year during the past 4
years since age 1 7. Those who tanned
indoors during warm weather had 3.39
times the odds of being TD compared to
Heckman et al
Table 4
Other Health Behavioral Predictors of Tanning Dependence (n=393,
\'ar lnble OR
Ne\ crregularsmokcr Rcten . .-nce
m past I 14
Current smoker 1.81
J\naeroll1c exercise Low Reference
Anacmbic e"tercsc Modcnuc: 1.25
Anacroo1c 1-1 1gh 150
1\crohu: cxerc1sc Low Reference
Aerob1e cxcrc1se Moderate 1.83
Aerob1c exerctse -lligh 1.26
lln\i to Normal We1gltt. BMI <25 Rclerence
Overnctght. BMI > 25 and BMI<.. 'O I 03
Obe e, 8\fi>=JO 0.34
those who did not (95% CI 1.338.64),
conditional on other covariates m the
model. Surprisingly. use of chemical
sunless tanners and overall rates of in-
door tanning did not predict TO.
Other Health-Related Behaviors
The following vunables were entered
into the olher health behaviors logtstic
regression model predictjng TD; smokmg
status, body mass mdex (BMI) group, aero-
bic exercising. and anaerobic exerctsmg
(Table 4). Twenty-five percent (n 100) of
the sample reported having been a regu-
lar smoker at some point in time, and
23% (n 93) reported smoking during the
past month. Those who were current
smokers had greater odds of being TO
compared to those who had never smoked
(OR I 81, 95% CI 1.10 2 98). The mean
body mass index (BMI) of the sample was
24.63 (SO = 5.23), which is in the healthy
range. ln lhe model, those who were obese
(BM1> .. 30) were less likely to be TO than
those who were underweight or normal
weight (BMI<25), OR 0 34 (95% Cl 0.14-
0.85). Sixty four percent (n"'258) of the
sample was underweight or normal weight
(BM1<25), 23% (n 93) overweight
(BMI>25 and BMI<30), and 13% (n 51)
obese (BMI>=30). The mean number of
days per week spent doing aerobic exer-
cise was 2.16 (SO 2 06). and the mean
number of days spent doing strengthen-
ing exercises was 1.68 (SO "" 1.94).
Am J Health Behav."'
p.,aJue
0.!10 (O.P 3 01)
11.02 ( 1.10, 2.98)
0.4X (0.!\7, 2.35)
0 25 (0 l.021
0.07 (O.<JS, 3 . .54)
0.54 (0.60, 2.63)
() 42.
(0SY,I .78)
11.02 (0. J 4, 0.85)
Sensitivity analyses were also con-
ducted in order to assess whether chang-
ing the cut-points of the CAGE and DSM-
TV-TR scales used to define TO would
change our inferences. Tightening and
loosenang the criteria for TO did not change
the general inferences compared to the
original criteria, although some variables
did become less statistically significant.
The one major exception was the rela-
tionship of indoor tanning in warm
weather w1th tanning dependence. With
stricter critcna for TO, the point estt-
mate of indoor tanning in warm weather
was protective, whereas it was a risk
factor w1th a looser defimtion. This rela-
tionship should be examined more closely
in future studies.
DISCUSSION
These data demonstrated several pre-
dictors of tanning dependence, but some
measurement issues still need to be ad-
dressed. TO was predicted by ethnicity
and skm type, lack of skin protective
behaviors, and outdoor ond indoor tan-
ning behoviors among young adults, ns
well as smoking and body mass index.
These findings may offer new avenues for
research as well as skin protection and
skin cancer prevention interventions.
However, the internal reliability of the
measures was relatively low, and future
research should include more in-depth
psychometric development and analyses
459
Tanning Dependence
such as test-retest reliabilitY or factor
analysis. -
The current study found shghtly less
than half the rates ofTD as m the Warthan
ct nll7 study, and both studies obtained
higher TO rates on the mDSM IV-TR than
the mCAGE. Using the same sconng sys-
tem, 11% of the current participants scored
tanning dependent on the mCAGE (ver-
sus 26% in Warthan et all7), and 23%
scored dependent on the mDSM-IV-TR
fversus 53% m Warthan et al
11
). The pro-
portion meeting the CAGE criterion is
very stmilar to the 12% found by Poorsattar
and Hornung
28
in their sample of Seattle
undergraduates. These rates are also
similar to undergraduates' meettng CAGE
critena for alcohol (18%) and tobacco use
( 16%) ~ ~ The higher rates observed by
Warthan et al
17
make sense given that
their sample was recruited directly from
a beach as opposed to sampling from a
general population. TO rates would also
be likely to vary based on age, with older
individuals being less at risk due to lower
rates of tanning in general. ss
56
It was
interesting that TO did not vary by gender
as in the Warthan et aJI study. Women
were overrepresented tn the current
sample, so it may be that men who were
not particularly interested in tanning did
not choose to partictpate in this study
In the current study, TO was related to
racefethnicity and Fitzpatrick
44
skin type.
As expected, whites were more likely
than nonwhites to be tanning dependent,
but nonwhites were also represented in
the TO group. It is interesting to note that
a few African Americans screened posi-
tive for TO. These tended to be lighter-
skinned indtviduals, but this trend was
not stgnlficnnt (probably due to the small
sample size of this group). Therefore, non-
whites should not be ignored in skin
protection efforts.
Regarding Fitzpatrici<44 skin type, Type
I and Type VI participants were least at
risk for TO. Dark individuals (ie, Type VI)
may not need to make an effort to tan or
do not perceive tanning as culturally ap-
propriate, and very fair individuals may
not be able to tan or refrain from tanning
because they are aware of their high risk
for burning and other skin damage. The
TO rates among Types V (26%) and VI (4%)
were somewhat surprising, but it is im-
portant to note that Fitzpatrick skin type
was self reported. Therefore, some indi-
viduals may have rated their tanned skin
460
rather than their natural skin color or
minimized their risk for burning. This is
suggested by the lifetime use of tanning
booths by 57% of Types V and VI
As expected, TO was assoctated with
sun tanning, tanning booth use sun-
bums, and lower likelihood of protecting
the skin. These constructs are some-
what overlappmg, but the findings still
support the construct validity of the mea-
sures. A lack of association would rn1se
serious valtdity concerns. The amount of
intentional sun tanning was relauvely
low; however, the amount of incidental
exposure was high and widespread. Inter-
estingly, sun tanning appeared to be more
closely related to TO than indoor tanning,
which was only related during warm
weather. It was not known a priori whether
chemical sunless tarmer use would be
associated with TO or not. os sunless
tanner use could be viewed as a protec-
tive behavior or as a UVR tanning en-
hancer. Sunless tanning was associated
with TO in univariate analysis but was
not as strongly related to TO as the other
variables included in the multivariable
model (Tables I and 3). However, sunless
tanners could be recommended as an
alternative to UVR tanning, particularly
for TDs. An overwhelming majority of
studies have demonstrated UVR to be
harmful to the skin, but very few have
found sunless tanners to be dangerous.S7
The finding that almost 40% of the
sample had used tarming booths IS alarm-
ing, particularly when considering that
the mean age of the partictpants was 21,
the mean age of initiation was l 7, and the
mean number of lifetime uses was 57.
Tanning booth mitiation at age I 7 ts prob-
ably notable given that it likely corre
sponds with the high school prom ~ c a s o n
or spring break trips. An ideal prevention
opportunity may thus occur at a high school
just before prom time or spring break. A
previous study of 1275 adolescents found
an average age of initiation of 15 years;
however, that was a randomized house-
hold survey of teens from Boston and Min-
neapolisSt. Paul. 2 northern locnles with
cold climates.
25
Participants in the current
study were more likely to use tanning
booths during cold than warm weather.
That the risk for tanning booth use in-
creases during cold weather could be re-
Lated to greater difficulty sun tanning at
that time of year preparing for winter
vacations in warm locales. self-treatment
for seasonal affective disorder,2
1
and so on.
However in the current studv. TO was
more closely related to indoor tanning dur-
sng warm weather. Thig suggests that TOs
may be utilizing all possible means to tan
regardless of the time of year. These is-
sues should be studied further and inter-
ventions tmlorcd accordingly.
It was expected that TO would be asso-
ciated with other health-related vanables
such as smoking, weight, and exercise.
Because Demko et aJ2
1
found that sub-
stance use was related to indoor tanning,
it was hypothesized that tobacco use would
be related to TO. Smokers may be more
likely to be TO due to a similar addictive
process or because both of these behav-
iors are often viewed as image enhanc-
ing We expected low- to overage-weight
indl\nduals to be at greater risk for TO
than overweight or obese persons It is
likely that obese individuals are more
concerned about their weight than hav
ing tanned skin and may not feel comfort
able exposmg their bodies in the manner
necessary to receive high levels of UVR
We also expected that exercise would be
related to TO, but the direction of the
relationship was uncertnm. Demko et
a122 found that dieters wore more likely
to use tanning booths but that exercising
were less likely to do so. It would
make sense that dieters or exercisers
would tan more often because they are
concerned about their appearance or, al
ternatively, that dieters/exercisers may
tan less because they are concerned about
their health. '
2
:ZH&.s<J However. we found no
independent relationship between TO and
exercise. Overall, these results provide
evidence for conceptualizing tanning prc-
occupatlOn as an addiction but do not rule
out potential relationships to eating dis-
orders or other psychiatnc problems such
as obsessive-compulsive disorder.
There are st..veral ways in which the
mDSM-IV-TR scale could be made more
consistent w1th the original American
Psychiatric Association substance depen
dence A withdrawal syndrome
18 really not discussed in the modified
scale An item such as "Do you experience
Jlttcnness or nausea when you don't tan
for a while after a period of regular tan-
ning?" could be added One of the sub
stance dependence criteria is spending a
great deal of time in actlVitJes related to
the substance. It appears that Warthan ct
al'
0
arbitrarily decided to usc one or more
Am J Health Behav .TM 2008;32(5):451464
Heckman et al
days per week spent tanning to defme "a
great deal of time: but an uppropriate
amount remains an empirical question.
Finally, the item referring to missing
work due to sunburns seems to fit better
conceptually with the item referring to
m1sssng activities due to tanning, rather
than the items with which it is currently
grouped that all refer to the amount of
time spent tanning. Moreover, the cut
scores used for substances mav need to
be modified for tanning. For example, 26%
of individuals with skin Type V were found
to be tanning dependent in the current
sample. Because such dark-skinned in-
dividuals are typically not found to have
high tanning rates, TO may have been
overestimated. Moreover, the mDSM-IV-
TR and mCAGE agreed on 80% of the
classtfications, but the mOSM-IV-TR was
much more inclusive. Further psycho-
metric development could raise agree-
ment between the scales. Likewise, be
cause the items were onginally devel-
oped to assess substance use, more in-
depth testing of the items, such as cogni-
tiVe mterviewing, might help strengthen
their validity for use with tanners.
The popularity of conceptualizing ex-
cessive nonsubstance behaviors {food,
sex, gambhng, etc) as addictions has in-
creased in recent vears ' f Previous stud
ies have provided some support for con-
ceptualizing excessive tanning as an ad-
diction. Excessive tanning has been mea
sured using standard addiction mea
sures,
10
tanning has been found to be
associated with substance use,n tanners
find tanning physiologically reinforcing,
26
tanners have been concerned about quit-
ting tannmg
25
and have also demonstrated
physiologic withdrawal symptoms.
17
There
are certainly some similantics among
excessive tanning and traditional addic-
tions, suggesting possible biopsychosocial
links. It may be that many tanners are
first exposed incidentally to the sun dur-
ing childhood activities, then tan inten
tionally for primarily psychosocial rea-
sons such as appearance, relaxntion, or
socialtzation; and then a small subgroup
experiences a strong physiologic reaction
to tanning, which in some cases may be
related to an underlying addictive or mood
disorder. It ts important to note that not
all tanning behavior or even frequent
behavior should be seen as indicative of
TO. People tan for a variety of reasons,
and only a subset would meet criteria for
461
Tanning Dependence
TO including having tanning-related prob-
lems such as tolerance, withdrawal, and
other negative consequences. In addi-
tion, similarities could be drawn between
excessive tanning and obsessive-com-
pulsive disorders,
6
l eating disorders,
18
and
body dysmorphic disorder, S9 each of which
also has biopsychosocial components.
Which conceptualization is most useful
has not yet been determined.
Limitations of the current study in-
clude that it used a convenience sample,
is cross-sectional, and based on self-re-
port data. Tt is possible that the sample of
primarily Virginia psychology students is
not generalizable to some other popula-
tions. Due to the single administration of
the survey, it was not possible to monitor
changes in patterns of UVR exposure over
time. Additionally, all data were self-re-
port and did not measure actual time
spent tanning or skin damage levels.
However, because the survey was con-
ducted online, responses may have been
more truthful than an in-person inter-
Vlew in which demand characteristics
are greater. As the literature expands in
this area, future research should further
refine and validate measures of TO. For
example, objective measures of WR ex-
posure could be used (eg, spectrophotom-
etry). Other possible directions for future
research would be to further explore TO
prevalence in various populations. The
relationship between TO and other addic-
tive, eating, obsessive-compulsive disor-
ders should be investigated. For example,
TO could be compared with drugs other
than tobacco use. Researchers should
further explore tanning motivations and
tailor interventions accordingly. For ex-
ample, intervention studies might at-
tempt to intervene prior to age 17 and
focus on school-related events such as
prom and spring break in order to prevent
future TD. Ideally, interventions would
reduce the importance placed on appear-
ance in general and, specifically, the
unportance placed on tanning as an ap-
pearance enhancer.
Acknowledgments
The authors would like to thank Rick
Gibbons, Heike Mahler, John Roberts,
and Sharon Manne for their consultation
during this project.
REFERENCES
l.L.im HW, Gilchrest BA, Cooper KD, et al.
462
Sunlight, tanrung booths, and VItamin D. JAm
Acad Dennatol. 2005;52(5):868-876
2.ACS. What You Need to Know About Skin
Cancer. American Cancer Society 2006.
3. Robinson JK. Rigel OS, Amonette RA Trends
in sun exposure knowledge, atmudes, and
behaviors: 1986 to 1996. J Am Acad Dennatol.
1997;37(2 Pt 1): 179- 186.
4 Beasley TM, KnteJ BS. Factors that influence
health risk behaviors among tanning salon
patrons. Eva/ Health Prof 1997 ;20(4 ):371-388.
5.Ciarke VA, Williams T. Arthey S. Skin type
and optimistic bias in relation to the sun
protection and suntanning behaviors of young
adults. J Behav Med.. 1997;20(2):207-222.
6.Hoegb HJ, Davis 80, Manthe AF. Sun avoid-
ance practices among non-Hispanic white
Californians. Health Educ Behav.
1999;26(3):360-368.
7.Tunisi R, Hillhouse J, Gebert C. Examination
of cognitive variables relevant to sunbathing.
J Behau Med. 1998;21(3):299 313.
8.Hillhouse JJ, Stair AW. 3rd, Adler CM Pre-
dictors of sunbathing and sunscreen use in
college undergraduates. J Behav Med.
1996; 19(6):543-561
9.Jones JL, Leary MR. Effects of appearance
based admonitions against sun exposure on
tanning mtentlons tn young adults. Health
Psychol. 1994; 13(1):86-90.
IO.Leary MR, Jones JL. The" social psychology of
tanning and sunscreen use: self- presenta-
tional motives as a predictor of health nsk.
Journal of Applied Social Psychology.
1993;23: 1390-1406
11 Miller AG, Ashton WA, McHoskey JW, et al.
What price attractiveness? Stereotype and
risk factors m suntanning behavior. Journal of
Applaed Social Psychology. 1990;20: 1272 1300.
12.Borland R, Hill D, Noy S Being sunsmart:
Changes in community awareness and re-
ported behaviour foUowing a primary preven
tion program for skin cancer control Behau
Change. 1990;7:126-135.
13.Bullcr DB, Buller MK, Beach B, et al. Sunny
days, healthy ways: evaluation of a skin
cancer prevention curriculum for elementary
school-aged children. J Am Acad Dermatol.
1996:35{6):911-922.
14.Robinson JK. Behavior rnodificatJon obtained
by sun protection education coupled with
removal of a skin cancer. Arch Dermatol.
1990; 126(4):477-481.
lS.Rossi JS, Blais l..M, Weinstock MA. The
Rhode Island Sun Smart Projecl: skin cancer
prevention reaches the beaches Am J Pu.bltc
Health. 1994;84(4):672-674
16.Wemstock MA, Rosst JS. The Rhode Island
Sun Smart Project: a scientific approach to
skan cancer preven lion Clin Dermatol.
1998; 16(4);411-413.
17.Warthan MM, Uchida T, Wagner RF, Jr UV
light tanning as a type of substance-related
disorder. Arch Dermacol. 2005; 141(8):963-966.
18.BBC N. Young utanorex.ics" risking cancer
(online), Available Ill: http://news.bbc.co.uk/
go/ pr I fr / /1/hi/programmes/ real _stor-y 1
3737125/stm. Accessed Apnl 28, 2006.
19.Shapiro JJ .. Teena lead, adults follow m
sk1ppmg sunscreen (onhne), Ava1lable at:
www.cancer.org. Acctssed March 4, 2003.
20.Boldeman C, Jansson B, Oat H, et at. Sunbed
use among Swed1sh adolescents m the 1990s
a deehne w1th an unchanged relationship to
health nsk behaviors. &and J Public Health.
2003;31 (3):233-237.
21.H11lhouse J , Stapleton J, Tumsi R. Associa
t1on of frequent mdoor UV UJnning w1th
seasonal affective disorder. Arch Demtatol.
2005:141(11): 1465.
22.Demko CA, Borawsk1 EA, Oeba.nne SM, et al.
Use of mdoor tannmg facilities by wh1te
adolescents in the United Stotes. Arch Pediatr
Ado/esc Med 2003; I 57(9) 854 860
23.0'Riordon Dr.. F1eld AE, Geller AC. et al.
Frequent tnnning bed use, weight concerns,
and other health risk bchav1ors ln adolescent
females (Unated States) . Cancer Causes Control.
2006; 17(5) :679-686.
24. Boldeman C, Jansson B. Nilsson B, et al.
Sunbed use m Swed1sh urban adolescents
related to behaviornl chamctcnstlcs. Preu Med
1997;26(1) : 114119.
25.Zeller S, Lazov1ch 0, Forster J, et al. Do
adolescent mdoor tanners exhibit depen-
dency? J Am Arod Dcrmatol 2006;54(4):589
596.
26.Feldman SR, L1guon A, Kucenlc M, et al.
Ultravtolet exposure IS a rcmforcang stimulus
10 frequent mdoor umners. JAm Aood DermatoL
2004;5 I( 1) :4551
27.Kaur M, L1guon A, Lang W, et ul. lnductJOn of
wtthdrawalhke symptoms m a small random
tzed, controlled tnal of op101d blockade an
frequent tannen. J Am Acad Dermatol.
2006;54(4) :709-711.
28.Poorsaunr SP, Hornung RL. UV hght abuse
and h1gh rtsk tannmg behavior among under-
graduate college students. JAm Arod Dcrmatol.
2007;56(3):375 379.
PC, Carr DB, F'1sher JE, et al.
Plasma beta endorphin and betn- hpopro-
tein response to ult r nv1olet radiution. Lan
cet . 1983;2(8342): 166.
30.Belon PE. UVA exposure and p1tunory secre
rion. Vartations of human hpotropm concen
trations (beta LPH) after UVA exposue.
Photochem Photob1ol. 1985;42(3):327329.
3l.Gamblchler T, Bader A, Vojvodic M. et al .
Plasma levels of opi01d peptidcs after sunbed
exposures. Br J Dermatol 2002; 147(6): 1207-
1211.
32.Wmtzcn M, 0St!Jn OM, Poldennan MC, et al.
Total body exposure to ultraVlolet radiation
does not influence plasma levels of Immu-
noreactive beta-endorphin m m11n.
Photodcrmatol Photo1mntunol Pltotomed.
200 I ; 17(6) 256260
33.La.zovlch 0, Forster J , Sorensen G, et al .
Charoctenshcs associated wtlh use or mten-
Am J Health Behav.'"' 2008;32(5):451 464
Heckman et a1
uon to use indoor tanning among adolescents.
Arch Ped1atr Ado/esc Med 2004; 158(9). 918
924.
34.APA. Dtagnostlc and Stausticnl Manual of
Mental Disorders, Founh Ed1tJon, Text Revt
s1on Washington, DC: Amencan Psych1atnc
2000.
35.Ma\field D, McLeod G, Hall P. The CAGE
que"stionnatre: validation of a new alcoholism
screemng tn!>lrument. Am J Psychtatry.
1974; 131(10) : 1121- 1123.
36. Liskow B, Campbell J, Nickel E.J, et al
Val1d1ty of the CAGE quesuonnalre m scretn
ing for al cohol dependence m u walk-m
(tnage) chnac. J Srud Alcohol 1995;56(3):277-
281 .
37.Malet L, Schwan R, Doltssiron D, et nl
Valid1ty of the CAGE questionnaire tn hospl
tal. Eur Psychratry 2005;20(7):484489.
38.Bisson J, Nudeau L, Demers A The Vl\hdity of
the CAGE scale to scnen for heovy drinking
and drinking problems m a general popula
tion survey. AddKCron. 1999;94(5) :715722.
39 SCIDWebpage. Rehabllty nne! Vahdit.> of the
SCID l (online) . Avatlable et: http:/ /
cpmcnet.columbta.edti/ depl / e1d. Acceued
September 29. 2006.
40.Gtbbons FX. Gerrard M Pred1cung voung
adults' health nsk behavtor J P.:rs Soc PsyclwL
1995,69(3) :505 517.
41.Gtbbons FX, Gerrard M. A soc1al rcucuon
model of adolescent health risk In Suls J ,
Wallston K, (Eds). Socnll Foun-
dations of Health. Oxford, U. K .. Blackwell
2003.
42.Hlllhouse J, Tunis1 R, Hoh\'lSkt F, et al . An
exammatton of psychological vanables rei
evant to nrtlfic1al tannmg tendencres Journal
of Health Psychology 1999;4.507-516
43.Hillhouse JJ, Tumsr R, Kastner M. Modehng
tanmng salon behav1oral tendencies usmg
appearance motivation, selfmonuoring and
the theory of planned behrwi01 . 1-Jealch Educ
Res. 2000; 15(4):405 414.
44 Fitzpatrick TB. The vahd1tV nnd practicnhty
of sun-reactive skin types I through VI. Arch
Derma toT. 1988:1 24(6):869 871.
45.Gibbons FX, Gc."rrnrd M Health imnges and
the1r effects on heulth bchnv1or. In Gibbons
BBFX, (Ed) . Health, Copmg, und Well bemg:
Perspectives from Soc1al Companson Theory.
Mahwah, NJ: l.owrence Erlbaum Assoctatcs
1997:63-94
46.Gibbons FX. Gerrard M, Lane DJ, et nl. Usmg
UV photography to reduce use of tanning
booths. a test of cogmttvc medlatton Heulth
Psychology 2005,24(4)358 363
47.Maddock RC, Fulton RL. Mot1vauon. Emo-
uons, and Leadersh1p: The S1lent S1de or
Management . Westport, CT: Quorum Books/
Greenwood Pubhshmg Group, Inc. 1998.
48,Wemstock MA, Ross1 JS, Reddmg CA, et Ill .
Sun protection behaviors and U\ges of change
for the pnmary prevention of skm cancers
among beachgoers m southeastern Ne\1. En-
461
Tanning Dependence
gland. Ann Behm Mcd. 2000:22(4):286293.
49.Mahler HI, Fitzpatnck B, Parker P, et aJ. The
relauve effects of a health-based versus an
appearance-based intervention des1gned to
ancrease sunscreen use. Am J Health Promot.
1997:11(6):426-429.
SO.Mahler HI, Kulik JA, Gibbons FX, et al.
Effects of appearance-based anterventions on
sun protection mtenuons and self-reponed
behaviors. Health Psycho/ 2003:22(2): 199-
209.
Sl.Mahler HI, Kulik JA. Harrell J, et al. Effects
of UV photographs, photoaging mformation,
and use of sunless tannmg louon on sun
protect1on behav1ors. Arch Dermatol.
2005; 141 (3):373-380.
52.CDC. Youth riak behnv1or surveillance: Na-
llonnl College Health Risk Behnv10r Survey-
Umted States, 1995. MMWR CDC SUrve1U SUmm.
1997;46(6): 156.
53.llarrell FE. Regression Modeling
New York: Springer 200 I.
54 .Cohen J. A power pnmer. Psychol Bull.
1992;112(1):155-159.
SS.Boldeman C, Branstrom R. Dal H, et al.
Tannmg habits and sunburn in a Swedish
populatiOn age 1350 years. Eur J Cancer.
200 I :37( 18):244 1-2448.
464
56.Melia J, Bulman A. Sunburn and tanning in
a British populnuon. J Public Health Med.
1995: 17(2):223-229.
57.Petersen AD, Wu1f HC, Gmadeck1 R, et al.
Dihydroxyacetone, the ocuvc browmng ingre-
dient in sunless tannmg louons, mduces
DNA damage, cell-cycle block and apoptosis
in cultured HaCaT keratmocytes Murat Res.
2004;560(2): 173-186.
58.Greene K, Brmn LS. Messages lnfluencmg
college women's tannmg bed use: stahshcnl
versus narrat1ve <"vidence format and a self
asse!lsment to mcreaKe pcrcewcd susceptibil-
ity. J Health Commun. 2003;8(5):443-461.
59.Philhps KA, Conroy M, Oufrc!lne RG, et al.
Tanning m body dysmorph1c dtsorder.
Psychiatr Q 2006;77(2):129-138.
60.Buckley PF. Brown ES. Prevulence and con
sequences of dual diagnosis. J Clm Psyclltatry.
2006;67(7) cO I
61 Marun PR Petrv NM. Are non-substance
related addtct1ons really addictions? Am J
Addrct. 2005; 14( 1): 1-7.
62.Lean MR. Snltzmnn JL, Georgeson JC. Ap-
pearance motivattOn, obsessive-compulsive
tendenctes and cxcess1ve suntanning in a
communuy sample. J 1/ealth Psycllal.
1997;2(4):493499.
Copyright of American Journal of Health Behavior is the property of PNG Publications and its
content may not be copied or emailed to multiple sites or posted to a listserv without the copyright
holders express written permission. However, users may print, download, or email articles for
individual use.
REPORTS
I
UV light abuse and high-risk tanning behavior
among undergraduate college students
Solmaz P. Poorsattar, BS,a,b and Robin L. Hornung, MD, MPHb,c
San Francisco, California, and Seattle, Washington
Background: The failure of skin cancer prevention efforts to alter tanning behaviors may be a result of the
addictive nature of UV light.
Objective: This study attempts to determine the prevalence of UV light substance-related disorder (SRD).
Methods: A survey was administered to undergraduate college students. The cut down, annoyed, guilty,
eye-opener questionnaire was used to determine existence of SRD.
Results: Of 385 respondents, 12o/o scored posi tively on the cut down, annoyed, guilty, eye-opener
indicating SRD. Women, indoor tanners, students with tanning family and friends, and frequent ta nners
were significantly more likely than their peers to score positively.
Limitations: The small size is a li mitation of this study as results may not be generalizable to larger
populations. There also may be self-report bias.
Conclusimzs: A significant proportion of college students clemonsmtte evidence of SRD with respect to UV
light. (] Am Acacl Dermatol 2007;56:375-9.)
I
n the United States, the incidence of malignant
melanoma increases each year. l.
2
Research ex-
ploring the cause of this epidemic consistently
links excessive UV light exposure early in life to skin
cancer later in life
3
-
7
and finds young skin to be
part icularly sensitive to UV radiation.s-
11
Therefore,
decreasing UV light exposure during childhood and
adolescence is a central objective of skin cancer
primary prevention effo1ts.
Most prevention programs focus on educati onal
interventions meant to increase parents' and chil-
dren's knowledge of the dangers of UV light over-
exposure. Unfortunately, studies have shown that
From the University of California, San Francisco, School of
Medicine; Children' s Hospital and Regional Medical Center,
Seattleb; and Division of Dermatology, Department of Pediatrics,
University of Washington School of Medicine.c
Funding sources: None.
Conflicts of interest: None identified.
Accepted for publ ication August 26, 2006.
Reprint requests: Robin L. Hornung, MD, MPH, Division of
Dermatology, M2-7, Children's Hospital and Regional Medical
Center, 4800 Sandpoint Way NE, Seattle, WA 98105. E-mail:
robin.hornung@seattlechildrens.org.
Published online January 31, 2007.
0190-9622/ $32.00
2007 by t he American Academy of Dermatology, Inc.
doi:l 0.1 016/j.jaad.2006.08.064
increased knowledge, especially among adolescents
and young adults, often fails to alter tanning behav-
ior and attitudes.
12
-
19
It has been suggested anecdotall y for years that
one reason tanning is so popular is that UV light is
addictive.
20
-
2 2
Many adolescents who regularly tan
indoors self-report difficulty in quitt ing tanning
23
Experiments have also shown that UV light is
a reinforcing stimulus that causes endorphin re-
lease.
24
-
26
A recent study among beachgoers in Texas
confirmed that many sunbathers met criteria for
having substance-related disorder (SRD) witb respect
to UV light.
27
The potential addictive nature of UV
light tanning might explain why educati onal preven-
tion efforts have been largely unsuccessful.
To better understand the UV light tanning behav-
iors of adolescents and young adults and their
reasons for persistently tanning, this study examines
a population of college-age tanners, their tanning
attitudes and behaviors, and prevalence of SRD with
regard to UV light.
METHODS
After permission was obtained from our human
subj ect.<; review committee, students attending a
convenience sample of undergraduate classes in all
colleges of the University of Washington in Seattle
during December 2005 and j anuary 2006 were asked
375
376 Poorsattm and Hornung
Table I. Characteristics, self-reported tanning
practices, and cut down, annoyed, guilty,
eye-opener questionnaire responses of all
survey respondents and those scoring
positively on the questionnaire
Student characteristics
Female*
Skin phot ot ype
I
II
Il l
IV
v
Age, y
17 and 18
19and20
21 24
25-30
Family history of skin
cancer*
Yes
No
Unknown
Personal history of
skin cancer
Tanning pract ices
History of blistering
sunburn*
Outdoor tanners*
Indoor t anners*
Fami ly members tan*
Friends tan*
History of t anning
bed burn*
All Respondents
survey
respo ndents
(n = 375)
65%
19%
32%
28%
16%
4%
22%
49%
22%
7%
20%
65%
1 5o/o
Oo/o
48%
70%
33%
41%
83%
1 5o/o
All uv
scoring
pos iti\'e.ly on the
CAGE ( n = 46)
87%
11%
33%
35%
20%
2%
11 o/o
63%
22%
4%
35%
50%
1 5o/o
Oo/o
63%
1 OOo/o
76%
59%
1 OOo/o
57%
light Respondents scoring
tanners positively on th e
Modified CAGE questions (n 256) CAGE (n 46)
Have you ever felt you 21 o/o 87%
ought to cut down
on your tanni ng?
Have peopl e annoyed 9o/o 41%
you by criticizing
your tanni ng?
Have you ever felt bad 20% 87%
or guilty about your
tanning?
Have you ever thought 13% 28%
about tanni ng first
thing in the morni ng?
CAGE, Cut down, annoyed, guilty, eye-opener.
*P < .05 when comparing all survey respondents with t hose
scoring positively on t he CAGE.
J AM ACAD 0 ERMATOL
MARCH 2007
to complete a 1-page anonymous multiple-choice
questionnaire. Information coll ected incl uded de-
mographic data, personal tanning practices, and
tanning practices of family and friends.
Four questions in the survey also comprised
a modified cut down, annoyed, guilty, eye-opener
(CAGE) questionnaire tool, which was used to
determine whether patticipants showed symptoms
of SRD with regard to UV light (see Table I for
modifi ed CAGE questions). The CAGE questionnaire
is a thoroughly tested tool most often used to identify
SRD with regard to alcohol.
28
Other researchers have
suggested that screening for other substances can
effectively be incorporated into the CAGE format
by simply including references to them in the ques-
tions.
29
The CAGE questionnaire is reported to be
60o/o to 90o/o sensitive for detecting Sills when two or
more questions are positive and 40o/o to 60o/o specific
for excluding substance abuse.
30
Descriptive frequencies were used to describe
survey responses. The percentages reported for each
item are based on the valid number of responses to
the item. Students who reported experiencing two or
more of the CAGE criteria were considered as having
a positive screen result, denoting suggested SRD with
respect to UV light. lf students experienced none or
only one of the 4 criteria, they were assigned a
negative CAGE result. When comparing discrete var-
iables, Fisher's exact tests were performed to deter-
rnine statistical significance. The data were analyzed
using software (STATA 7.0, StataCorp LP, College
Station, Tex).
RESULTS
In all, 385 students participated (response rate =
90o/o). Of the 385 participants, 10 were older than 30
years and their responses were excluded from this
analysis. Only students who reported ever purposely
tanning were asked to complete the CAGE question-
naire. Demographics, relevant history, self-reported
tanning practices, and CAGE questionnaire re-
sponses are reported in Table I.
In this study, 76% of female students reponed
purposely tanning their skin versus 59o/o of male
students (P = .001). Of the female students, 42o/o
reported using indoor tanning devices versus 17% of
the male students (P = .000). Of students surveyed,
9% purposefully tanned their skin 20 or more times
per month.
Overall, 12% of the total sample, 18o/o of the self-
reporting suntanners, and 28o/o of indoor tanners
scored positively on the CAGE indicating SRD with
regard to UV light. Of the students who reported
purposely tanning their skin, 22o/o of female suntan-
ners had positive CAGE results, compared with 8o/o
) AM ACAD DRMATOL
VOLUME 56, NUMBER 3
Table II. Self-reported motivations for UV light tanning
Poorsattar and Hornung 377
AU lN light Respondents scoring Indoor tanning Initial motivation for
tanners positively on the device users indoor tanning
(n 240)
To look better 7S%
To relax 41%
To gain a protective base tan 34%
For special events 2S%
To feel healthy 22%
Social activity with friends 18%
Peer pressure 3%
Special price promotion 3%
Medical reasons 2%
Other (eg, work) 7%
CAGE, Cut down, annoyed, guilty, eye-opener.
of male suntanners (P = .007). Indoor tanning device
users were much more likely to be flagged as
potentially having a UV light disorder than nonusers
(28% vs 12%, P = .000). Students with friends and
family members who used indoor tanning macb.ines
were significantly more likely than students without
such fri ends and family to test positive (P = .000 and
P = .016, respectively). The likelihood of scoring
positively on the CAGE increased as reported fre-
quency of tanning increased.
The reasons respondents gave for tanning are
reported in Table II.
First indoor tanning experience
Of respondents, 66% repott ing ever using an
indoor tanning device were also current users
(have used one within the last year). Of students
reporting ever using an indoor tanning device, 49%
had received a burn from indoor tanning device use,
and 79% of these previously burned students still
currentiy tanned indoors. The mean age of first using
a tanning device was 16.8 years, and the range was
7 to 23 years. Of indoor tanners, 87 (71 %) started
before the age of 18 years and 12 (10%) at the age
of 14 years or younger.
In all , 59% reported going to an indoor tanning
salon for the first time with friends, 18% went alone,
14% went with their parents, and 11% with their
siblings. Reasons respondents gave for using an indoor
tanning device their fi rst time are given in Table II.
DISCUSSION
In all, 18% of undergraduate students who admit-
ted to purposely tanning their skin scored positively
on the CAGE questionnaire, indicating probable
existence of SRD with respect to UV light. This
number is significant and comparable with the 18%
of drinking college students who scored positively
on the CAGE questionnaire with respect to alcohol in
CAGE ( n 43) (o 112) (n 124)
91% 82% 60%
47% 44% 14%
44% 41% 19%
44% 46% 40%
30% 29% 8%
21% 20% 11 o/o
9% 4% 10%
9% 8% 10%
S% 4% 0%
Oo/o 7% 2%
a Midwest study
31
and the 16% of college students
who reported smoking cigarettes daily in a 2002
National Inst.it.utes of Health study.
32
Many of our respondents also exhibited high-risk
tanning behavior and had experienced a bad or
blistering burn from the sun and indoor tanning
devices. Of students sutveyed, 9% purposefully tanned
their skin 20 or more times per month. The tendency
to sunburn, lifetime number of severe sunburns, and
cumulative UV light exposure are all positively asso-
ciated with risk of developing future skin c.ancer.
33
In
addition, our study shows a strong positive correlation
between high-risk tanning behavior and SRD with
respect to UV light. Significantly more students scoring
positively on the CAGE questionnaire bad experi-
enced a bad or blistering sunburn (63%) and many
more of these students who were also indoor tanners
had received a burn from an indoor tanning device
(57%). Of students with SRD, 16% pL11]10Sefully tanned
their skin 20 or more times per month.
Students with a known family history of skin
cancer are also at much greater risk of developing
cancer
4
and surprisingly are significantly more likely
to tan their skin than those without a family history
of skin cancer. Of students with a positive family
history, 77% purposely tanned their skin outdoors
and 45o/o of them used indoor tanning devices. This
finding is consistent with a previous study that found
Midwestern college students with a positive family
history of skin cancer were 1.5 ti mes more likely to
use tanning lamps than those wi thout
1 9
Personal
experi ence apparentl y fails to alter tanning behavior.
The association found in previous studies between
frequent indoor tanners and tobacco and other
substance abuse also suggests that individuals who
tan frequently may be susceptible to multiple addic-
tive disorders.
34
-
36
Almost half of tanning students rep01t tanning
to relax, which is a strong motivating factor that has
378 Poorsattm and Hornung
been noted by numerous previous studies.
13
19
37
38
As expected of an addictive practice, established
users in our study reported tanning to relax and to
feel healthy much more often than first-time users
(44% vs 14%). Furthermore, it has been shown that
adolescents who agree that tanning improves their
mood are more likely to self-report difficulty in
quitting tanning.
23
As previously discussed, education alone will
probably not stop high-risk tanning behavior.
Numerous studies examining other health behavior
changes, particularly those of addictive behaviors,
have shown that successful self-changers use differ-
ent processes at each parti cular stage of change.
39
40
Future interventions should take into consideration
the behavioral stage and moti vational readiness of
each individual to change problem tanning behavior.
In addition, if, in fact, tanning is addictive, the
argument to prohibit the use of indoor tanning
devices by minors, as recommended by the World
Health Organization,
41
is strengthened.
There are several limitations to our study. The
sample size was small as was the group identified
as positive for SRD. The behaviors of undergraduate
college students at this university may not be able
to be generalized over all undergraduate students or
the general population. Students self-reported their
tanning behavior and medical history, which is a
potential subjective limitation. Some student respon-
dents may have answered the CAGE items with less
accuracy because of social desirability concerns.
CONCLUSION
The prevalence of SRD with respect to UV light
among college tanners helps explain why it has
been particularly difficult to modify high-risk tan-
ning behavior in adolescents and young adults. The
existence of this SRD should strengthen arguments to
increase regulations on the indoor tanning industry
and further justify following in the lead of public
health efforts that have successfull y decreased use
of other addictive carcinogens.
We would like to thank Courtney McNamara and Dylan
Mart for their help with data collection, and Do Peterson,
MS, for reviewing our survey tool and study design.
REFERENCES
1. Thompson JF, Scolyer RA, Kefford RF. Cutaneous melanoma.
lancet 2005;365:687-701.
2. Beddingfield FC Ill. The melanoma epidemic: res ipsa loquitur.
Oncologist 2003;8:459-65.
3. Green A. Sun exposure and the risk of melanoma. Australas
J Dermat ol 1984;25:99-1 02.
4. Holman CDJ, Armst rong BK. Pigmentary traits, ethnic origin,
benign nevi, and family history as risk factors for cutaneous
melanoma. J Natl Cancer lnst 1984;72:257-66.
J AM ACAD 0 ERMATOL
MARCH 2007
5. Spencer JM, Amonette R. Tanni ng beds and skin cancer:
artificial sun + old sol = real risk. Clin Dermatol1998;16:487-501 .
6. Holman CDJ, Armstrong BK. Relationship of cutaneous malig-
nant melanoma to individual sunlight exposure habits. J Natl
Cancer lnst 1986;76:403-14.
7. Zanetti R, Franceschi S, Rosso S, Colonna 5, Bidoli E. Cut aneous
melanoma and sunburns in childhood in a southern European
population. Eur J Cancer 1992;28A:1172-6.
8. Stern RS, Weinstein MC, Baker SG. Risk reduct ion for non-
melanoma skin cancer with childhood sunscreen use. Arch
Dermat ol 1986;122:537-45.
9. Council on Scientific Affairs. Harmful effects of ult raviolet
radiation. JAMA 1989;262:380-4.
10. Walter SO, Marrett LD, Shannon HS, Roy P. The associat ion of
cutaneous malignant melanoma with the use of sun beds and
sunlamps. Am J Epidemiol1990;131:232-43.
11. Levine JA, Sorace M, Spencer J, Siegel OM. The indoor UV
tanning indust ry: a review of skin cancer risk, health benefit
claims, and regulat ion. J Am Acad Dermatol 2005;53:
1038-44.
12. Monfrecola G, Fabbrocini G, Posteraro G, Pini D. What do
young people t hink about the dangers of sunbathi ng, skin
cancer and sunbeds? A quest ionnaire survey among Italians.
Photodermatol Photoimmunol Photomed 2000;16:15-8.
13. Mawn VB, Fleischer AB Jr. A survey of attitudes, beliefs, and
behavior regarding tanning bed use, sunbathing, and sun-
screen use. J Am Acad Dermat ol 1993;29:959-62.
14. Arthey S, Clarke V. Suntanning and sun protection: a review
of the psychological literature. Soc Sci Med 1995;40:265-74.
15. Robinson JK, Rademaker AW, Sylvester JA, Cook B. Summer
sun exposure: knowledge, attitudes, and behaviors of midwest
adolescents. Prev Med 1997;26:364-74.
16. Gell er AC, Colditz G, Oliveria S, Emmons K, Jorgensen C, Aweh
GN, et al. Use of sunscreen, sunburning rates, and tanning bed
use among more than 10,000 US children and adolescents.
Pediatrics 2002;1 09:1009-14.
17. Robinson JK. Rigel OS, Amonett e RA. Trends in sun exposure
knowledge, attitudes, and behaviors: 1986 to 1996. J Am Acad
Dermat ol 1996;347:179-86.
18. Benjes LS, Brooks DR, Zhang Z, Livstone L, Sayers L, Powers C,
et al. Changing patterns of sun protection between t he first
and second summers for very young children. Arch Dermatol
2004;140:925-30.
19. Knight JM, Kirincich AN, Farmer ER, Hood AF. Awareness of the
risks of t anning lamps does not influence behavior among
college students. Arch Dermat ol 2002;138:1 311 -5.
20. Netburn D. Young, carefree, and hooked on sunlamps. New
York Times May 26, 2002;9.
21. Chedru MF, Dumont C, Jourdain F, Postel-vinay N, Dab W.
Cutaneous melanoma prevention: modify behaviors [in
French]. Rev Prat 2004;54:1165-6.
22. WCCO-TV. Hooked on tanning? Available from: URL:http://
www.wcco.com/specialreports/local_story _052181257.html.
Accessed July 3, 2006.
23. Zeller S, Lazovich D, Forster J, Widome R. Do adolescent
indoor tanners exhibit dependency? J Am Acad Dermatol
2006;54:589-96.
24. Feldman SR. Liguori A, Kucenic M, Rapp SR, Fleischer AB Jr,
Lang W, et al. Ultraviolet exposure is a reinforcing st imul us in
f requent indoor tanners. J Am Acad Dermatol 2004;51 :45-51.
25. Levins PC, Carr DB, Fi scher JE, Momtaz K. Parrish JA. Plasma
beta-endorphin and beta-lipot ropin response t o ultraviolet
radiation. Lancet 1983;2:166.
26. Belon PE. UVA exposure and pituitary secretion: variat ions of
human lipotropin concentrations (j3LPH) after UVA exposure.
Photochem Photobiol 1985;42:327-9.
) AM ACAD DRMATOL
VOLUME 56, NUMBER 3
27. Warthan MM, Uchida T, Wagner RF Jr. UV light tanning as a type
of substance-related disorder. Arch Dermatol2005;141:963-6.
28. Ewing JA. Detecting alcoholism: the CAGE questionnaire.
JAMA 1984;252:1905-7.
29. Brown RL Identification and office management of alcohol
and drug disorders. In: Fleming MF, Barry KL, editors. Addictive
disorders. StLouis: Mosby; 1992. p. 28.
30. Mersy OJ. Recognition of alcohol and substance abuse. Am
Fam Physician 2003;67:1 529-32.
31. Boyd U, McCabe SE, d' Arcy H. A modified version of the CAGE
as an indicator of alcohol abuse and its consequences among
undergraduate drinkers. Subst Abuse 2003;24:221-32.
32. Johnston LD, O'Malley PM, Bachman JG. Monitoring the future
national survey results on drug use, 1975-2002. Volume II:
college students and adults ages 19-40 (National Institutes
of Health publication No. 03-5376). Bethesda (MD): National
Institute on Drug Abuse; 2003.
33. Abdulla FR, Feldman SR, Williford PM, Krawchuk D, Kaur M.
Tanning and skin cancer. Pediatr Dermatol 2005;22:501-12.
34. Demko CA, Borawski EA, Debanne SM, Cooper KD, Stange
KC. Use of indoor tanning facilities by white adolescents
Poorsattar and Hornung 379
in the United States. Arch Pediatr Adolesc Med 2003;157:
854-60.
35. Lazovich D, Forster J, Sorensen G, Emmons K, Stryker J,
Demierre MF, et al. Characteristics associated with use or
intention to use indoor tanning among adolescents. Arch
Pediatr Adolesc Med 2004;1 58:918-24.
36. Boldeman C, Jansson B, Nilsson B, Ullen H. Sunbed use in
Swedish urban adolescents related to behavioral characteris-
tics. Prev Med 1997;26:114-9.
37. Diffey BL. Use of UV-A sunbeds for cosmetic tanning. Br J
Dermatol 1986;1 15:67-76.
38. Dougherty MA, McDermott RJ, Hawkins MJ. A profile of
commercial tanning salons. Health Values 1988;12:21-9.
39. Prochaska JO, Velicer WF, Rossi J5, Goldstein MG, Marcus BH,
Rakowski W, et al. Stages of change and decisional balance for
12 problem behaviors. Health Psycho! 1994;13:39-46.
40. Prochaska JO, DiClemente CC, Norcross JC. In search of
how people change: applications to addictive behaviors. Am
Psycho! 1992;47:1102-14.
41. World Health Organization. Artificial tanning sunbeds-risks
and guidance. Geneva: World Health Organization; 2003.
4th International Workshop for the Study of Itch
San Francisco, California
September 9-11, 2007
Venue: Hilton San Francisco Financial Distti ct, 750 Kearny Street, San Francisco, California 94108
Phone: 415-433-6600; Fax: 415-765-7891; Web site:
http:/ / v.rwwl . hjl ton.com/ en_ US/ hi/hoteVSFO FDHF-Hilton-San-Francisco-Financial-District-
California/ index.do.
Meeting organizers and contact: Prof. Earl Carstens, M. Carstens, Uni versity of California, Davis,
Section of Neurobiology, Physiology, & Behavior, 1 Shields Ave, Davis, CA 95616. Phone: 530-752-
7767; Fax: 530-752-5582; E-mail: eecarstcns@ucclavis.eclu.
Official meeting Web site: http://itch2007sanfrancisco.ucdavis.edu/.
Calcium/Vitamin D Supplementation and
Cardiovascular Events
Judith Hsia, MD; Gerardo Heiss, MD, PhD; Hong Ren, MS; Matthew Allison, MD, MPH;
Nancy C. Dolan, MD; Philip Greenland, MD; Susan R. Heckbert, MD, PhD;
Karen C. Johnson, MD, MPH; JoAnn E. Manson, MD, DrPH; Stephen Sidney, MD, MPH;
Maurizio Trevisan, PhD; for the Women's Health Initiative Investigators
Background-Individuals with vascular or valvular calcification are at increased risk for coronary events, but the
relationship between calcium consumption and cardiovascular events is uncertain. We evaluated the tisk of coronary and
cerebrovascular events in the Women's Health Initiative randomized tiial of calcium plus vitamin D supplementation.
Methods and Results-We randomized 36 282 postmenopausal women 50 to 79 years of age at 40 clinical sites to calcium
carbonate 500 mg with vitamin D 200 IU twice daily or to placebo. Cardiovascular disease was a prespecified secondary
efficacy outcome. During 7 years of follow-up, myocardial infarction or coronary heart disease death was confirmed for
499 women assigned to calcium/vitamin D and 475 women assigned to placebo (hazard ratio, 1.04; 95% confidence
interval, 0.92 to 1.18). Stroke was confirmed among 362 women assigned to calcium/vitamin D and 377 assigned to
placebo (hazard ratio, 0.95; 95% confidence interval, 0.82 to 1.10). In subgroup analyses, women with higher total
calcium intake (diet plus supplements) at baseline were not at higher risk for coronary events (P= 0.91 for interaction)
or stroke (P= O. l4 for interaction) if assigned to active calcium/vitamin D.
Conclusiom-Calciumlvitamin D supplementation neither increased nor decreased coronary or cerebrovascular risk in
generally healthy postmenopausal women over a 7-year use period. (Circulation. 2007;115:846-854.)
Key Words: calcium cerebrovascular disorders coronary disease stroke women
V
ascular calcification and valvular calcification predict ath-
erosclerotic risk
1
2
and are prevalent in cluonic diseases
such as diabetes,
3
systemic lupus erythematosus,
4
and chronic
kidney disease,
5
in which the risk of coronary events is high.
Arterial calcification and valvular calcification are organized,
regulated processes similar in many respects to bone fonnation
and remodeling.
6
-9 Because of this relationship, bisphospho-
nates have been proposed as antiatherosclerotic agents,
10
and
patients with coronruy calcification commonly ask if they should
reduce their calcium consumption.
11
The literature on this topic
is scant and contlicting
12
-
16
; even less is known about the
relationship between vitamin D and coronruy risk.
17
Editorial p 827
Clinical Perspective p 854
We randomized 36 282 postmenopausal women to calcium
plus vitamin D or to placebo in a fracture trial and report here
Received November 2, 2006; accepted December 14, 2006.
the impact of 7 years of supplementation on cardiovascular
outcomes, including time trends and analyses of subgroups.
Methods
Details of the study design have been published previously,
18
as have
the baseline characteristics.
1
9-
2 1
Eligible postmenopausal women 50
to 79 years of age joined the Women's Health Initiative hormone
therapy and/or dietary modification trials between 1993 and 1998.
One year later, they were invited to join the double-blinded calcium
plus vitamin D trial; 91 % joined the calcium/vitamin D trial at their
first annual visit and 9% during the following year. Women provided
written informed consent in a fom1 approved by local institutional
review boards and were randomly assigned to calcium carbonate 500
mg with 25-hydroxy vitamin D
3
200 IU twice daily or matching
placebo (GiaxoSmithKiine, Research Triangle P<uk, NC). Concur-
rent calcium supplementation was pennitted, as was vitamin D. up to
400 IU daily.
From the Department of Medicine, George Washington University, Washington, DC (J.H.); Department of Epidemiology, University of North Carolina School
of Public Health, Chapel Hill (G.H.); Fred Hutchinson Cancer Research Center, Seanle, Wash (H.R.); Department of Family and Preventive Medicine. University
of California at San Diego, San Diego (M.A.); Departments of Medicine (N.C.D., P.G.) and Preventive Medicine (P.G.), NOithwestern University, Chicago, Ill;
Depmtment of Epidemiology, University of Washington, Seattle (S.R.H.); Depattment of Preventive Medicine, University of Tennessee Health Science Center,
Memphis (K.C.J.); Division of Preventive Medicine. Brigham and Women's Hospital and Harvard Medical School, Boston, Mass (J.E.M.); Kaiser Penuanente,
Oakland, Calif (S.S.); and University at Buffalo School of Public Health and Health Professions, Buffalo, NY (M.T.).
Clinical trial registration information- URL: http://www.clinicaltrials.gov. Unique identifier: NCT00000611.
The online-only Data Supplement, consisting of a list of investigators, is available with this article at http://circ.ahajournals.orglcgi/content/full/
ClRCULA TIONAHA.l06.673491/DC1.
Guest Editor for this article was Robert H. Eckel , MD.
Correspondence to Judith Hsia, MD, 2150 Pennsylvania Ave, NW No. 4- 414, Washington, DC 20037. E-mailjhsia@mfa.gwu.edu
2007 American Heart Association, Inc.
Circulation is available at http://www.circulationaha.org DOl: 10.1161/CJ.RCULA TIONAHA.106.673491
846
Hsia et al
Clinical Outcomes
Weight, blood pressure, and waist circumference were recorded
annually. Blood samples were collected at baseline, ie, at the time of
enrollment into the hormone therapy and/or dietary modification
trials, from all participants; in a random 6% sample, blood samples
also were collected I and 3 years later.
22
Participants reported emergency room visits, overnight hospital
stays, and outpatient coronary revascularization procedures semian-
nually. Medical records for all overnight hospitalizations and outpa-
tient coronary revascularization procedures were scruti nized for
potential outcomes of interest . Centrally trained physician-
adjudicators classified outcomes on the basis of medical record
review. Myocardial infarction was categorized through the use of an
algorithm that included symptoms, ECG findings, and cardiac
enzymes.23 Confirmed angina required hospitalization for angina
with confirmatory stress test or obstructive coronary disease by
angiography.
14
Stroke required rapid onset of a persistent neurolog-
ical deficit not due to trauma, tumor, infection, or other cause
25
;
strokes were coded as "other" if procedure related or if the adjudi-
catm: could not classify the event as hemorrhagic or ischemic. All
deaths were centrally adjudicated; other outcomes were adjudicated
on the basis of hospital record review by centrally trained, local
adjudicators blinded to treatment assigrunent. Composite outcomes
were defined during development of the analytical plan.
Statistical Methods
Statistical methods have been described.
20
21
In brief, hazard ratios
with 95% confidence intervals (Cls) were calculated from Cox
proportional-hazards models stratified by age, prevalem cardiovas-
cular disease at baseline. and randomization status in the hormone
and dietary modification trials. Subgroup analyses were planned
a priori. Subgroup analyses were stratified by age. prevalent cardio-
vascular disease at baseline, and randomization status in the hormone
and dietary modification trials. Consistency of treatment effects
among subgroups was assessed by formal tests of interaction; tests
for linear trend were used when appropriate. Nineteen subgroups
were evaluated for coronary heart disease (CHD) and for stroke;
subgroup results should be interpreted with caution because 1
significant finding would be expected by chance for each outcome
based on a 0.05 nominal level of statistical significance. All reported
probability values are 2 sided. Analyses were carried out by the
coordinating center statistics unit using the SAS System for Win-
dows, version 9 (SAS Institute. Cary, NC).
The authors had full access to and take responsibility for the
integrity of the data. All authors have read and agree to the
manuscript as written.
Results
Between 1995 and 2000, 36 282 women were randomized at
40 clinical sites; when the trial closed in April 2005, the mean
duration of follow-up was 7 .0:!: L4 years. Baseline character-
istics were balanced between treatment groups (Table I)
except for hypertension (P=0.03). At baseline, mean calcium
intake (diet plus supplements. exclusive of study medication)
was 1148:!:654 mg/d in the active treatment group and
1154:!: 658 mg/d in the placebo group, close to the recom-
mended intake of 1200 mg dai ly.
26
Vitamin D consumption
was 365:!:265 IU/d in the active treatment group and
368:!:266 IU/d in the placebo group. Sixty percent of study
participants took at least 80% of their study medication
through year 6.
Intermediate Biomarkers and Risk Factors for CUD
Although blood samples were collected on the entire cohort,
bioassays were performed in only a 6% random sample. At
baseline, total cholesterol was 5.64 mmoi/L, low-density
lipoprotein cholesterol was 3.28 mmoi/L, high-density li -
Calcium/Vitamin D and Cardiovascular Risk 847
poprotein cholesterol was 1.54 mmoi/L, triglyccrides were
1.81 mmoi!L, glucose was 5.49 mmoi!L, and insulin was II A
~ L l U / m L Differences between mean percent. change in the
intervention group and mean percent change in the control
group are shown from baseline to year 2 after randomization
(Figure 1). Percent change from baseline differed signifi-
cantly between treatment groups for low-density lipoprotein
cholesterol (P=0.02), waist circumference and weight
(P=0.03 for both), systolic blood pressure (P=O.Ol), and
diastolic blood pressure (P<O.OI).
Clinical Cardiovascular Outcomes
Myocardial infarction or CHD death was confirmed in 499
women assigned to active calcium/vitamin D and 475 as-
signed to placebo (hazard ratio, 1.04; 95% CI, 0.92 to 1.18).
Stroke was confirmed in 362 women assigned to calcium/
vitan1in D and 377 assigned to placebo (hazard ratio, 0.95;
95% CI, 0.82 to 1.10; Figure 2). Among women taking at
least 80% of study medication, the hazard ratio for myocar-
dial infarction/CHD death was 1.05 (95% CJ, 0.88 to 1.25)
and for stroke was 0.97 (95% cr. 0.79 to 1.20) (data not
shown). Risks of coronary revascularization, confirmed an-
gina, hospitalized heart failure, transient ischemic attack, and
composite outcomes also were similar in the 2 treatment
groups (Table 2).
Temporal Trends
Hazard ratios with nominal 95% Cis for myocardial infarc-
tion/CHD death at 1-year intervals of follow-up were as
follows: year l, 1.13 (95% CI, 0.79 to 1.61); year 2, 1.10
(95% CI, 0.75 to 1.61); year 3, 1.00 (95% Cl , 0.69 to 1.46);
year 4, 0.92 (95% Cl, 0.66 to 1.28); year 5, 1.00 (95% CI,
0.72 to 1.39), year 6, 1.11 (95% CI, 0.81 to 1.51 ), and year
?.7, 1.07 (95% Cl, 0.80 to 1.42). The z score for trend, based
on Cox proportional-hazards modeling with time-dependent
treatment effects, was 0.22 (P= 0.82), indicating no signifi-
cant trend in tisk over time.
Hazard ratios with 95% Cis for stroke were as follows:
year 1, 1.09 (95% CI, 0.69 to 1.72); year 2, 0.62 (95% CI,
0.41 to 0.94 ); year 3, 1.22 (95% Cl, 0.82 to I .82); year 4, 1.07
(95% CI, 0.70 to 1.65); year 5, 1.01 (95% Cl, 0.69 to 1.47),
year 6, 0.71(95% CI, 0.50 to 1.01), and year ?.7, 1.11 (95%
cr. 0.81 to 1.52). The z score was 0.55 (P=0.58).
Trends by Age
Cumulative hazard ratios for myocardial infarction/CHD
death and for stroke were evaluated by age decade. For
women 50 to 59, 60 to 69, and 70 to 79 years of age at
baseline, hazard ratios with 95% Cis for CHD were 0.94
(95% Cl, 0.70 to 1.27), 1.08 (95% C1, 0.90 to 1.30), and 1.05
(95% CI, 0.85 to 1.30), respectively (P= 0.53 for interaction).
Hazard ratios with 95% Cis for stroke were 0.90 (95% CI,
0.62 to 1.32), 0.97 (95% Cl, 0.78 to 1.20), and 0.96 (95% Cl,
0.76 to 1.20), respectively (P=0.72 for interaction).
Additional Subgroup Analyses
We evaluated several demographic and clinical characteristics to
determine whether other subgroups of women were at lower or
higher risk for myocardial infarction/CHD death with calcium!
848 Circulation February 20, 2007
TABLE 1. Baseline Characteristics by Treatment Group Assignment
CalciumNitamin D Placebo
(N=18 176) (N=18106)
p
Age, y 62.4:!: 7.0 62.4:!:6.9 0.97
Body mass index, kg/m
2
29.1 :!: 5.9 29.0 5.9 0.24
Waist circumference, em 88.9:!:13.7 88.8 13.7 0.46
Systolic blood pressure, mm Hg 12717 128 17 0.48
Diastolic blood pressure, mm Hg 76:!:9 76:!:9 0.56
Total calcium intake (supplements, diet, and medications), mg/d 1148654 1154658 0.40
Total vitamin D intake (supplements and diet), IU/d 365 265 368 266 0.36
Vitamin D intake (supplements), IU/d 190235 192235 0.46
Vitamin D intake (diet), IU/d 175117 176117 0.47
Elhnicity 0.45
White 15 047 (82.8) 15 106 (83.4)
Black 1682 (9.3) 1635 (9.0)
Hispanic 789 (4.3) 718 (4.0)
American Indian/Alaskan native 77 (0.4) 72 (0.4)
Asian/Pacific islander 369 (2.0) 353 (1.9)
Unknown 212(1.2) 222 (1.2)
Hypertension 0.03
None 10 915 (66.7) 10 858 (66.5)
Untreated 1260 (7.7) 1384 (8.5)
Treated 4187 (25.6) 4092 (25.1)
Diabetes mellitus 1055 (5.8) 1036 (5.7) 080
High cholesterol requiring pills 2008 (12.3) 1964 (12.1) 0.49
Cigarette smoking 0.31
Current 1405 (7.8) 1356 (7.6)
Past 7255 (40.3) 7133 (39.8)
Never 9325 (51.8) 9428 (52.6)
CHD at baseline 487 (2.7) 456 (2.5) 0.34
Cardiovascular disease at baseline 881 (4.8) 860 (4.7) 0.66
Stalin use 1178 (6.5) 1149 (6.3) 0.60
Aspirin use 3552 (19.5) 3517 (19.4) 0.78
NSAID use 5983 (32.9) 5891 (32.5) 0.44
Use of postmenopausal hormone therapy
Assigned to active therapy in hormone trials 4039 (22.2) 4078 (22.5) 0.49
Current use, including exposure in hormone trials 9358 (51 .5) 9484 (52.4) 0.09
NSAID indicates nonsteroidal antiinflammatory drug, including aspirin; CHD at baseline, self-reported myocardial
infarction or coronary revascularization at baseline; cardiovascular disease at baseline, self-reported myocardial infarction,
coronary revascularization, stroke, or transient cerebral ischemia at baseline; aspirin use, >80 mg taken at least twice
weekly; and stalin, 3' -hydroxy-3-methyglutaryl coenzyme A reductase inhibitor. Values are meanSD or n (%).
vitamin D (Figure 3) or for stroke (Figure 4). The hazard ratios
for CHD (P=0.91 for interaction) and stroke (P=O.J4 for
interaction) did not differ by total calcium intake (dietary plus
supplemental) at baseline. Similarly, hazard ratios did not differ
by vitamin D intake at baseline (P=0.45 for interaction for CHD
and P =O.l2 for stroke). Hazard ratios also did not differ by
ethnicity, although numbers of events were small among His-
panic, American Indian, and Asian women (P=0.54 for inter-
action for CHD and P=0.63 for stroke).
CHD risk with active calcium/vitamin D was inversely related
to body mass index (P =0.04 for interaction); ie, women with
higher body mass index were at lower CHD 1isk with active
calcium/vitamin D supplementation, whereas those with lower
body mass index were at higher CHD risk. Stroke risk with
active calcium/vitamin D was lower among women with high
cholesterol and those taking statins at baseline (P=0.04 for
interaction for both). Stroke risk with active calcium/vitamin D
was inversely related to the number of CHD risk factors; ie,
women with fewer risk factors were at higher stroke risk with
calcium/vitamin D supplementation (?=0.02 for interaction).
Discussion
Calcium/vitamin D supplementation neither increased nor
decreased the risk for CI-ID or stroke in generally healthy
Hsia et al Calcium/Vitamin D and Cardiovascular Risk 849
Total Cholesterol
Triglyceride
HDLC
LDLC
Insulin
Glucose
HOMA
WHR
Waist
Weight
Systolic BP
Diastolic BP
-20
-6.1
-6.7
-15 -10
~
-0.9 ~
: ~
t.4
_.( 0.4
-5 0 5
Figure 1. Differences in mean percent
change from baseline to year 2 between
women assigned to active calcium/vita-
min D and those assigned to placebo for
several intermediate outcomes. Horizon-
tal lines represent 95% Cis. Physical
measures were performed on the entire
cohort; laboratory measures, in a random
6% subsample. Treatment group differ-
ences were significant for low-density
lipoprotein cholesterol (LDL-C; P= 0.02),
waist circumference and weight {both
P=0.03), and systolic (P=0.01) and dia-
stolic (P< 0.01) blood pressures. HDL-C
indicates high-density lipoprotein choles-
terol; HOMA, homeostasis model
assessment; WHR, waist-to-hip ratio;
and BP, blood pressure.
Mean Percent Change lrom Baseline in the Intervention Group
postmenopausal women throughout the 7-year duration of
this randomized trial. Neither total calcium intake (dietary
plus supplemental) nor total vitamin D intake at baseline
affected cardiovascular risk with calcium/vitamin D
supplementation.
Possible explanations of this null finding include the
following: ( 1) Background calcium use impaired our ability
to identify a treatment effect; (2) the dose of vitamin D was
inadequate; (3) poor adherence to study medication blunted
any treatment effect: (4) concurrent postmenopausal hormone
therapy interfered with treatment effects; (5) the trial was
designed to evaluate the effects of calcium/vitamin D supple-
mentation on fracture, not cardiovascular disease; or (6)
calcium and vitamin D do not, in fact, affect cardiovascul ar
risk.
0.06
0.06
0.04
I
I
0.03
0.02
0,01
0.00
:.:::.
HR 1.04
95l(.CI 10.92,1.18)
0 .. ..
0 .. ..
,.,,
-
,,..
*'"'
.,.,.,
-
..
3
.. ..
,.
..
"
,.
""'
,,... ,,..
, ....
" ..
-
----
IUI!ber"de.oetlts
....,_,.""'
--- -..
--Cored
. ~
/.?
.-
"
~ ~
_....,;
6
.. .. ..
77 ..
"
- - -
- - -
A limitation of the trial was that women were allowed to
continue their own calcium supplements because it would
have been unethical to prohibit concunent calcium use in a
long-term, placebo-controlled trial. Baseline calcium con-
sumption (diet plus supplements) was balanced between
treatment groups, and no significant interaction between
dietary or total calcium consumption at baseline and random-
ized treatment assignment was observed for either CHD or
stroke.
Baseline vitamin D consumption (diet plus supplements)
and regional solar irradiance
21
also were balanced between
treatment groups. Parathyroid hormone levels are maximally
suppressed at 25-hydroxy vitamin D blood levels > 75
nmol/L (30 ng/mL).
27
In our trial , despite consuming 365 IU
vitamin D daily (supplements plus dietary vitamin D) at
--- -..
--Control
Q.04
i
"
t
J
Q03
OD2
.
,.
.. ..
.,
..
"'
..
"'
-
. .. 57 .. .. .. 72 ..
,.
..,.
...
"""
,,.. ,,...
-
"""
-
..,.
..... ..... ......
""" '""
,,.,
.....
""
-
----
_ ........
N:lmln ... Airl<
Figure 2. Kaplan-Meier estimates of cumulative hazard rates for CHD (myocardial infarction or coronary death; left) and for stroke
(right). HR indicates hazard ratio.
TABLE 2. Cardiovascular Events by Treatment Group Assignment
CalciumNitamin 0 Placebo
(N=18176), (N=18106), Hazard Ratio
n (Annualized %) n (Annualized %) (95% Cl)
p
Myocardial infarction or CHD death 499 (0.39) 475 (0.37) 1.04 (0.92-1.18) 0.50
Myocardial infarction 411 (0.32) 390 (0.31) 1.05 (0.91-1.20) 0.52
CHD death 130 (0.10) 128(0.10) 1.01 (0.7!H.29) 0.92
CABG or PCI 674 (0.53) 607 (0.48) 1.09 (0.98-1.22) 0.12
Myocardial infarction/CHD death/CABG/PCI 920 (0.72) 841 (0.66) 1.08 (0.99--1.19) 0. 10
Confirmed angina 404 (0.32) 377 (0.30) 1.08 (0.94-1.24) 0.30
Hospitalized heart failure 394 (0.31) 407 (0.32) 0.95 (0.83-1.10) 0.50
Stroke 362 (0.28) 377 (0.30) 0.95 (0.82- 1.10) 0.51
lschem ic stroke 225 (0.18) 228 (0.18) 0.98 (0.82-1.18) 0.84
Hemorrhagic stroke 58 (0.05) 68 (0.05) 0.84 (0.59--1.19) 0.33
Other stroke 63 (0.05) 57 (0.04) 1.11 (0.77-1.59) 0.58
Transient ischemic attack 213 (0.17) 182 (0.14) 1.16 (0.95-1.42) 0.13
Stroke/transient ischemic attack 563 (0.44) 547 (0.43) 1.02 (0.91-1.15) 0.75
CABG indicates coronary artery bypass grafting; PCI, percutaneous coronary intervention. Numbers of events do not
add up to the totals for categories because some women had > 1 event.
baseline, only 13% of women with incident fractures
(n= 1589) and 15% of matched controls had serum levels
>75 nmol!L, consistent with the current view that 800 to
I 000 JU daily may be needed to achieve optimal serum
vitamin D levels.
2
R Women assigned to active calcium!
vitamin D supplementation would have been taking almost
800 IU daily, which may still have been insufficient. None-
theless, women with higher vitamin D consumption at base-
line were not at higher or lower risk for CHD or stroke if
assigned to active calcium/vitamin D. Low vitamin D levels
have been associated with acute stroke
29
; because we mea-
sured serum vitamin D levels in fracture cases and controls,
not stroke cases, we are not able to confirm this association.
At the end of the trial (mean follow-up, 7 years), 76% of
participants were taking some study pills, and 59% were
taking 2:80% of their study medication. Calcium/vitamin D
supplementation did not alter CHD or stroke risk in sensitiv-
ity analyses, in which women were censored when they
became nonadherent, reducing the likelihood that adherence
affected study results. Use of postmenopausal hormone ther-
apy, which increases the risk of stroke
2
5 and CHD,30 was
balanced in the treatment groups. Neither CHD nor stroke
risk differed with calcium/vitamin D supplementation among
women assigned to active hormone therapy in the randomized
hormone trials, making it unlikely that postmenopausal hor-
mone use affected study results.
Another limitation of our analysis is that this trial was
designed to evaluate the effect of intervention on fracture, not
cardiovascular disease. Jn fact, the number of myocardial
infarctions/CHD deaths (n=974) and strokes (n=739) was
greater than the number of hip fractures (n=374), so a
reasonable treatment effect should have been readily detect-
able. Overall, the most likely explanation for our findings is
that calciumlvitamin D supplementation did not modulate
CHD or stroke risk.
Calcium and vitamin D had a mixture of favorable and
unfavorable effects on intermediate outcomes. Systolic pres-
sw-e rose l.l :t 12.4% among calcium/vitamin D recipients
during the 2 years after randomization and 0.7:!: 12.4%
among placebo recipients (P=O.OI for the treatment group
difference in percent change from baseline to year 2).
Diastolic pressure fell 0.2:!: 12.4% in the active treatment
group and 0.6:!: 12.4% in the placebo group (?=0.007).
These findings contrast with the National Health and Nutri -
tion Examination Survey, in which dietary calcium consump-
tion was inversely associated with an age-related increase in
systolic blood pressure.
3 1
Because of adjustments in concur-
rent medication dosage, we cannot be sure that the treatment
group differences in our trial are due solely to calcium/
vitamin D supplementation. Furthermore, in later years of the
trial, the changes in blood pressure from baseline no longer
differed between treatment groups.
Weight increased in both treatment groups during the 2
years after randomization (1.4:!: 10.5% versus 1.7:!: 12.0%),
as did waist circumference (1.5:!:7.6% versus 1.8:!:8.4%), but
these increases were smaller among women assigned to
active calciumlvitamin D (?=0.03 for both). The relationship
between weight and calcium/vitamin D consumption in other
reports has been inconsistent ,
32
33
but women taking calcium
supplements gained less weight than nonusers in a recent,
large, 10-year epidemiological study. 3
4
In both treatment groups, low-density lipoprotein choles-
terol rose in the 6% of participants with measured biomark-
ers, but the increase was smaller among women assigned to
active calcium/vitamin D (0.2:!:20.9% versus 2.6:!:20.7%;
? =0.02). Tn a small 12-week trial, calcium/vitamin D sup-
plementation had no effect on low-density lipoprotein cho-
lesteroP5 The modest difference seen in our trial may reflect
changes in weight over the 2 years or adjustments in
concomitant medications and cannot be definitively attributed
to calciumlvitamin D supplementation.
We found several subgroups of women with lower hazard
ratios for CI-ID or stroke with calcium/vitamin D supplemen-
tation. Women with higher body mass index appeared to be at
Hsia et al
P Vatuefor
Subgroup CaD Placebo Interaction
No. ol cases or CHD
(annuaized percentage)
0.0
Age ' yrs
50-59 84 (0.17)
87 (0.18) ]
6069 239 (0.42) 220 (0.39) 0.53
7().79 t76 (0.82) 168 (0.79)
Body mass index, kglm
2
<25 107 (0.32)
95 (0.28) ]
25 - <30 186 (0.41) 158 (0.35) 0.04
~ 3 0
205 (0.43) 220 (0.47)
Waist circumference, em
<85 t57 (0.29)
135 (0.25) ]
85 - <97 170 (0.43) 160 (0.41) 0.11
~ 9 7 172 (0.51) 178 (0.53)
Diabetes
No diabetes 409 (0.34)
394 (0.33) ]
0.84
All diabetes 89 (1.24) 81 (1.17)
Smoking
Yes 60 (0.62) 54 (0.57) ]
0.81
No 432 (0.37) 413 (0.35)
Hypertension
Yes 279 (0.59)
289 (0.61) ]
0.16
No 180 (0.26) 150 (0.22)
High cholesterol requiring pins
Yes 111 (0.83)
108 (0.83) ]
0.74
No 321 (0.33) 307 (0.31)
CHD risk factors
None 106 (0.19)
88 (0.16) ]
12 298 (0.55) 292 (0.55) 0.14
<:3
18 (1.33) 26 {1.86)
CHD at baseline
Yes 66 (2.05)
67 (2.23) ]
0.52
No 433 (0.35) 408 (0.33)
CVD at baseline
Yes 90 (1.53)
85 (1.49) ]
0.87
No 409 (0.33) 390 (0.32)
Slatin use al baseline
Yes 72 (0.91) 66 (0.87) ]
0.95
No 427 (0.35) 409 (0.34)
Aspirin use (<: 80 mQiday) at baseline
Yes 145 {0.59)
129 (0.53) ]
0.48
No 354 (0.34) 346 (0.34)
Dietary calcium Intake, mg
<800 300 (0.43)
286 (0.41) ]
800- <1200 108 (0.32) 115 (0.34) 0.52
2: 1200 73 (0.34) 63 (0.29)
Total calolum Intake, mg
<800 210 (0.49)
181 (0.43) ]
800 -<1200 128 (0.38) 125 (0.38) 0.91
2: 1200
143 (0.29) 158 (0.32)
To1alvitamin 0 intake, IU
<200 195 (0.40)
"' ... , l
200 <400 78 (0.33) 89 (0.37)
0.45
400 <600 119 (0.41) 103 (0.34)
~ 6 0 0 89 (0.38) 80 (0.35)
Alcohol use
Non d drink per week 354 (0.44)
332 (0.42) J
1 - < 7 drinks per week 92 (0.28) 95 (0.28) 0.84
7 + drinks per week 43 (0.32) 42 (0.32)
HRT arm
No1enrolled 241 (0.33)
238 (0.33) ]
ActiVe 129 (0.46) 116 (0.41) 0.69
Placebo 129 (0.46) 121 (0.44)
OM arm
Not enrolled 180 (0.47)
163 (0.43) ]
Intervention 131 (0.39) 128 (0.37) 0.65
Comparison 188 (0.34) 184 (0.34)
Calcium/Vitamin D and Cardiovascular Risk
Hazard Ratio
0.5
(-- 95% Ct - --l
1.0 1.5
.
0.94 '
!
1.08
'
1.05
.
.
'
1.16
.
0.91 '
1.18
'
'
'
'
1.17
i
1.02
0.96
'
.
.
.
1.04
1.05
'
'
1.04
1.04
'
I
0.98
I
' 1.17
.
.
0.98
.
'
1.04
I
I
'
1.19
.
1.01
0.68
I
I
I
0.89 '
1.06
I
I
I
1.04
1.04
'
~ 0 0
1.04
I
' .
1.13
.
'
1.02
I
I
1
1.03
0.93
I
' 1.21
' I
I
1.12
'
0.93
.
1.01
' I
0.97
I
I
0.91
I
I
1.19
'
1.07

I
I
.
1.05
0.95
I
1.07
.
I
0.99
I
I
I
1.13
;
I
1.08
I
I
1.1 1
' 1.05
0.97
.
851
2.0
Figure 3. Risk of CHD (myocardial infarction or CHD death) by treatment group assignment in various subgroups. Hazard ratios with nominal 95%
Cis {horizontal bars) are adjusted for age and prevalent CHD at baseline. The red dotted vertical line represents the hazard ratio for CHD in the over-
all cohort. Probability values are for the interaction between the subgroup variable and treatment assignment. CHD includes nonfatal myocardial
infarction and coronary death. Hypertension was defined as treated hypertension or a measured blood pressure of 2:140190 mm Hg. Risk factors
for CHD included current cigarette smoking, hypertension, self-reported diabetes, and high cholesterol. The presence of CHD at baseline was
defined as self-reported myocardial infarction or coronary revascularization. The presence of cardiovascular disease (CVD) at baseline was defined
as self-reported myocardial infarction, coronary revascularization, stroke, or transient cerebral ischemia. Because of missing data on some variables,
the numbers of cases do not always add up to the total number of cases in the treatment group. Statin indicates 3'-hydroxy-3-methylglutaryl coen-
zyme A reductase inhibitor, HRT, hormone replacement trial; OM, dietary modification trial; and CaD, calcium plus vitamin D.
Subgroup cao Placebo
No. ol cases of Stroke
(annualized percentage)
Age, yrs
5(}-59 51 (0.10)
56 (0.11) J
6(}-69 164 (029) 170 (0.30)
7(}-79 147 (0.69) 151 (0.71)
Body mass index, kg/m
2
<25 96 (028)
95 (028) ]
25- <30 129 (028) 140 (0.31)
;,.30 136 (028) 142 (0.30)
Waist circumference, em
<85 130 (0.24)
142 (0.26) ]
85. <97 l17 (029) 130 (0.33)
~ 9 7 115 (0.34) 102 (0.30)
Diabetes
No diabetes 318 (0.26) 324 (027) J
All diabetes 44 (0.61) 53 (0.77)
Smoking
Yes 36 (0.37)
33 (0.35) ]
No 322 (0.27) 339 (029)
Hypertenslon
Yes 215 (0.45)
242 (0.51) ]
No 124 (0,18) 103 (0.15)
High cholesterol requiring pins
Yes 45 (0.33) 65 (0.50) ]
No 273 (028) 262 (0.27)
CHD risk factOfs
None 85 (0.15)
75 (0.14) ]
1-2 220 (0.41) 238 (0.45)
"3
7 (0.52) 8 (0.57)
CHD at baseline
Yes 19 (0.59)
30 (1.00) ]
No 343 (0.27) 347 (028)
CVD at baseline
Yes 41 (0.70) 58 (1.01) ]
No 321 (0.26) 319 (0.26)
Statin use at baseHne
Yes 21 (0.27)
36 (0.47) ]
No 341 (028) 341 (028)
Aspirin use ("' 80 mg/day) at baseHne
Yes 99 (0.40) Ill (0.46) ]
No 263 (0.25) 266 (0.26)
Dietary calcium intake, mg
<800 202 (029)
205 (029) ]
800. <1 200 93 (0.27) 91 (0.27)
~ 1200 57 (027) 70 (0.32)
Total calcium Intake. mg
<800 139 (0.32)
127 (0.30) ]
800- <1 200 92 (028) 87 (0.27)
" 1200
121 (0.25) 152 (0.31)
Total vitamin o intake. IU
<200 139 (028)
130 (0.27) l
200 - <400 76 (0.32) 76 (0.31)
400 - <600 82 (028) 86 (029)
"600
55 {0.23) 74 (0.32)
Alcohol use
Non - <I drink per week 243 (0.30)
253 (0.32) J
I - < 7 drinks per week 76 (0.23) 84 (0.25)
7 +drinks per week 39 (029) 37 (028)
HRTarm
194 (0.27) ]
Active 97 (0.34) 110 (0.39)
Placebo 83 {0.30) 73 (027)
DMarm
120 (0.32) J
Intervention 90 (0.26) 94 (027)
Comparison 146 (0.26) 163 (0.30)
P Value tor
Interaction
0.72
0.98
0.60
0.26
0.64
0.06
004
0.02
0.08
0.11
0.04
0.51
0.58
0.14
0.12
085
0.44
0.44
0.0
(-95%CI-) Hazard Ratio
0.5 1.0 1.5
0.90
0.97
.
0.96
0.92
1.04
0.93
0.92
0.86
1.12
0 9 8 ~
-
0.78
:
0.95---+
1.01
-
0.88
1.19
0.69
'
t
1.04
1.14
0.90
~
0.76
' .
'
0.61
0.99-t -
0.71
0.99:'
,....._
0.54
!
--l.
.---1.00
.
0.89
0.99
:
0.97
0.98
~
0.85
~
1.07
. .04
0.80
l 1.02
0.98
:
1.04
0.74
0.96
0.90
0.92
0.90
1.14
0.99
1.05
0.86
2.0 2.5
1.12
Fi gure 4. Risk of stroke by treatment group assignment in various subgroups. Hazard ratios with nominal 95% Cis (horizontal bars) are
adjusted for age and prevalent cerebrovascular disease at baseline. Abbreviations as in Figure 3.
lower risk for CHD with calcium/vitamin D supplementation
(?=0.04 for interaction), but in view of the number of
subgroups examined, this finding may be due to chance.
We also found that women with more coronary risk factors
were at lower risk for stroke with calcium/vitamin D supple-
mentation (?=0.02 for interaction) but think this may be due
to chance as well, particularly because the number of women
with ;;:::3 risk factors was very small. On the other hand,
women with self-reported hypercholesterolemia and those
who used statins were at lower risk for stroke if assigned to
active calcium/vitamin D (?= 0.04 for interaction for both).
The clinical link between high cholesterol and stalin use and
the proposed, albeit controversial, effect of stalin use on bone
enhance the plausibility of this interaction.
Statins increase new bone fonnation in vitro and enhance
trabecular bone formation in rodents
36
through blockade of
Hsia et al
the mevalonate pathway.
37
In women, as opposed to rodents,
statins had no effect on markers of bone turnover,
38
bone
density, fracture risk,39 or progression of coronary calcifica-
tion40; thus, the relationship between statin use and fracture
risk remains controversial. Overall , the relationship between
consumption of calcium/vitamin D supplements and statin
use remains quite unclear with regard to cardiovascular
disease risk. Another placebo-controlled trial the size and
duration of the Women's Health Initiative is unlikely to be
undertaken, although it is possible that trials of bisphospho-
nates or other agents may shed some light. Populations in
those trials are not at particularly high risk for cardiovascular
disease, however, so the number of atherosclerotic events
may prove inadequate.
Calcium and vitamin D supplementation did not increase
the risk for myocardial infarction, CHD death, stroke, coro-
nary revascularization, hospitalized angina, heart failtue, or
transient ischemic attack. Thus. women taking these supple-
ments need not fear adverse cardiovascular consequences
while protecting their bone health.
Source of Funding
The Women's Health Initiative program is funded by the National
Heart, Lung, and Blood Institute, United States Department of
Heallh and Human Services.
Disclosures
Dr Hsia received a research g rant from GlaxoSmithKline. The other
authors report no conflicts.
References
I. Pletcher MJ. Tice JA. Pignone M. Using the coronary artery calcium
score to predict coronary heart disease events: a systematic review and
meta-analysis. Arch Intern Med. 2004; I 64: 1285-1292.
2. Allison MA, Cheung P, Criqui MH, Langer Rd, Wright CM . .Mitral and
aortic annular calcification are highly associated with systemic calcified
atherosclerosis. Circulation. 2006; 113:861-866.
3. Reaven PD, Sacks J, for the Investigators for the V ADT. Coronary artery
and abdominal aortic calcification are associated wi th cardiovascular
disease iu type 2 di abetes. Diabetologia. 2005:48:379- 385.
4. Molad Y. Levin-Iaina N, Yaturi M, Sulkes J, Sagie A. Heart valve
calcification in young patients with systemic lupus erythematosus: a
window to premature atherosclerotic vascular morbidity and a risk factor
for all-cause mortality. Atlwrosclerosis. 2006; 185:406-4 I 2.
5. Jono S. Shioi A. Ikari Y. Nishizawa Y. Vascular calcificat ion in chronic
kidney disease . ./ Bone Miner Metab. 2006;24:176-181.
6. Anand DV, Lahiri A, Lim E. Hopkins D, Corder R. The relationship
between plasma osteoprotegerin levels and coronary artery calcification
in uncomplicated type 2 diabetic subjects. J Am Coli Cardiol. 2006:47:
1850-1857.
7. Doherty T.M, Asotra K. Fitapatrick LA, Qiao JH, Wilkin OJ, Detrano RC,
Dunstan CR. Shah PK, Rajavashisth TB. Calcification in atherosclerosis:
bone biology and chronic inflammation at the arterial crossroads. Proc
Nat/ Acad Sci US A. 2003: 100: 11201- 11206.
8. Yattikuti R. Towler DA. Osteogenic regulation of vascular calcification:
an early perspective. Am J Pil ysiol E11docrinol Metab. 2004;286:
E686-E696.
9. Caira FC, Stock SR, Gleason TG, .McGee EC, Huang J, Bonow RO.
Spelsberg TC, McCarthy PM, Rahimtoola SH, Rajamannan NM. Human
degenerative valve disease is associated with up-regulation of low-density
lipoprotein receptor-related protein 5 receptor-mediated bone fom1ation.
./Am Coli Cardiol. 2006:47: 1707-1712
10. Bevilacqua M. Dominguez U, Rosini S. Barbagallo M. Bispbosphonates
and atherosclerosis: why? Lupus. 2005:14:773- 779.
I I. Davis W. Ask the doctor: calcium intake and vascular calcification. Life
Extension Magazine. August 2005. Available at: www.lef.org/magazinc/
mag2005/aug2005_atd_O l.htm. Accessed May 28, 2006.
Calcium/Vitamin D and Cardiovascular Risk 853
12. Umesawa M, lso H, Date C, Yamamoto A, Toyoshima H, Watanabe Y.
Kikuchi S. Koizumi A, Kondo T, lnaba Y, Tanabe N, Tamakoshi A.
Dietary intake of calcium in relation to mortality from cardiovascular
disease. Stroke. 2006;37:20-26.
13. Van der Vijver LP. van der Waal .MA. Weterings KG. Dekker JM,
Schouten EG. Kok FJ. Calcium intake and 28-ycar cardiovascular and
coronary heart disease mortality in Dutch civil servants. /111 J Epitlemiol.
1992;21: 36-39.
14. Bostick RM. Kushi LH, Wu Y. Meyer KA, Sellers TA, Folsom AR.
Relation of calcium. vitamin D, and diary food intake to ischemic heart
disease mortality among postmenopausal women. Am J Epidemiol. 1999:
149:151- 161.
15. Nerbrand C. Svardsudd K, Ek J. Tibblin G. Cardiovascular mortality and
morbidity in seven counties in Sweden in relation to water hardness and
geological settings: the project: myocardial infarction in mid-Sweden.
Eur Heart J. 1992; 13:721-727.
16. Rosenlund M, Berglind N, Hallqvist J. Bellander T, Bluhm G. Daily
intake of magnesium and calcium from drinking water in relation to
myocardial infarction. Epidemiology. 2005;16:570- 576.
17. Kununerow FA. Nutrition imbalance and angiotoxins as dietary risk
factors in coronary heart disease. Am J Clin Nutr. 1979:32:58-83.
18. Women's Health Initiative Study Group. Design of the Women's Health
Initiative clinical trial and observational stttdy. Control Clin Trials. 1998;
19:61-109.
19. Jackson RD. LaCroix AZ, Cauley JA. McGowan J. The Women's Health
Initiative calcium-vitamin D trial: overview and basel ine characteristics
of participants. Ann Epidemiol. 2003;13:S98-Sl06.
20. Wactawski-Wende J, Kotchen JM, Anderson GL, Assaf AR, Brunner RL,
O'Sullivan MJ. Margolis KL. Ockene JK. Phill ips L, Pottem L. Prentice
RL, Robbins J, Rohan TE, Sarto GE, Shanna S, Stefanick ML, Van Hom
L, Wallace RB, Whitlock E, Bassford T, Beresford SA, Black HR. Bonds
DE. Brzyski RG, Caan B, Chlebowski RT, Cochrane B, Garland C. Gass
.M. Hays J. Heiss G. Hendrix SL, Howard BY, Hsia J. Hubbell FA.
Jackson RD, Johnson KC, Judd H. Kooperbcrg CL, Kuller LH. LaCroix
AZ, Lane OS, Langer RD. Lasser NL, Lewis CE. Limacher MC. Manson
JE, for the Women's Health Initiative Investigators. Calcium plus vitamin
D supplementation and the risk of colorectal cancer. N Eng/ J Med.
2006;354:684-696.
21. Jackson RD, LaCroix AZ, Gass M, Wallace RB, Robbins J, Lewis CE,
Bassford T, Beresford SA, Black HR. Blanchette P, Bond.< DE, Brunner
RL, Brzyski RG. Caan B, Cauley JA. Chlebowski RT. Cununings SR,
Granek I. Hays J . Heiss G. Hendrix SL. Howard BY. Hsia J. Hubbell FA.
Johnson KC. Judd H, Kotchen JM, Kuller LH. Langer RD. Lasser NL,
Limacher MC, Ludlam S, Manson JE, .Margolis KL, .McGowan J, Ockene
JK, O' Sullivan .MJ, Phill ips L, Prentice RL, Satto GE, Stefanick .ML, Van
Hom L, Wactawski-Wende J, Whitlock E, Anderson GL, Assaf AR,
Barad D. for the Women's Health Initiative Investigators. Calcium plus
vitamin D supplemeutation and the risk of fractures. N Errgl J Med.
2006;354:669- 683.
22. Anderson GL, .Manson JE, Wallace R, Lund B, Hall D. Davis S.
Shumaker S, Wang CY, Stein E, Prentice RL. Implementation of the
Women's Health Initiative study design. Ann Epidemiol. 2003:13:
S5-SJ7.
23. Curb JD, McTiernan A, Heckbert SR. Kooperberg C, Stanford J, Nevitt
M. Johnson KC. Proulx-Burns L. Pa<tore L, Criqui M. Daugherty S, for
the WHI Morbidity and .Mortality Committee. Outcomes ascertainment
and adjudication methods in the Women's Health Initiative. Ann Epi-
demiol. 2003; 13:SI22-SI28.
24. Hsia J, Aragaki A, Bloch M, LaCroix AZ, Wallace R. Predictors of
angina vs myocardial infarction from the Women's Health Initiative.
Am J Cardio/. 2004;93:673-678.
25. Wassertheii-Smoller W, Hendrix SL, Limacher M. Heiss G. Kooperberg
C, Baird A. Kotchen T, Curb JD, Black H, Rossouw JE. Aragaki A,
Safford M, Stein E. Laowattana S, Mysiw WJ, for tbe WHI Investigators.
Effect of estrogen plus progestin on stroke in postmenopausal women: the
Women's Health Initiative: a randomi zed trial. JAMA. 2003;289:
2673-2584.
26. Office of Dietary Supplements, National Institutes of Health. Dietary
supplement fact sheet: calcium. Available at: http://dietary-supplemen-
ts.info.nih.gov/factsheets/calcium.asp. Accessed June I. 2006.
27. Vieth R. What is the optimal vitamin D status for health? Prog Biophys
Mol Bioi. 2006:92:26- 32.
28. Reginster JY. Zcgels B, Lejeune E. Micheletti MC, Kvsaz A, Seidel L.
Sarlet N. Influence of daily regimen calcium and vitamin D supplemen-
tation on parathyroid hormone secretion. Calcif Tissue 1111. 2002;70:
78- 82.
29. Poole KES, Loveridge N. Barker PJ, Halsall DJ, Rose C, Reeve J.
Warburton EA. Reduced vitamin D in acute stroke. Stroke. 2006;37:
243-245.
30. Manson JE. Hsia J, Johnson KC. Rossouw JE. Assaf AR, Lasser NL.
Trevisan M. Black HR. Heckbert SR. Detrano R. Strickland OL, Wong
ND, Crouse JR, Stein E, Cushman M, for the Women's Health Initiative
Investigators. Estrogen plus progestin and risk of coronary heart disease.
N Eng/ J Med. 2003:349:523- 534.
31. Haijar IM. Grim CE, Kotchen TA. Dietary calci um lowers the age-related
rise in blood pressure in the United States: the NHANES ffi survey . .I C/in
Hypertens. 2003;5: 122-126.
32. Shapscs SA. Hcshka S. Heymsfield SB. Effect of calcium supplemen-
tation on weight and fat loss in women. J Clin Endocriuol Metab.
2004:89:632-637.
33. Barr SI. Increased dairy product or calcium intake: is body weight or
composition affected in humans? J Nutr. 2003: 133:245S- 248S.
34. Gonzalez AJ. White E. Krista! A, Littman AJ. Calcium intake and 10-year
weight change in middle-aged adults . .I Am Diet Assoc. 2006; 106:
1066-1073.
35. Gannagc-Yared MH. Azoury M. Mansour I, Baddoura R, Halaby G,
Naaman R. Effects of a short-term calcium and vitamin D treatment on
serum cytoki nes. bone markers. insulin and li pid concentrations in
heallhy postmenopausal women. J E1ulocrinol Invest. 2003;26:748- 753.
36 . .Mundy G, Garrett R. Harris S. Chan J. Chen D. Rossini G. Boyce B, Zhao
M. Gutierrez G. Stimulation of bone formation in vitro and in rodents by
HMG CoA reductase inhibitors. Science. 1999;286:1946-1949.
37. Fisher JE. Rogers MJ. Halasy JM. Luckman SP, Hughes DE, Masarachia
PJ. Wesolowski G, Russell RG. Rodan GA. Reszka AA. Alendronate
mechanism of action: geranylgeraniol , an intermediate in the mevalonate
pathway, prevents inhibition of osteoclast formation, bone resorption, and
kinase activation in vitro. Proc Nat! Acad Sci US A. 1999:96:133- 138.
38. Hsia J. Morse M. Levin G. Effect of sinwastatin on bone markers in
osteopenic women: a placebo-controlled, dose-ranging uial. BMC Mus
culoskeletal Disorders. 2002:3:7.
39. LaCroix AZ. Cauley JA. Peltingcr M. Hsia J. Bauer DC, McGowan J.
Chen Z, Lewis CE. McNeeley SG, Passaro MD. Jackson RD. Statin use,
clinical fracture, and bone density in postmenopausal women: results
from the Women's Health Initiative Observational Study. Ann lntem
Med. 2003;139:97- 104.
40. Raggi P, Davidson M. Callister TQ, Welty FK. BacbmaruJ GA, Hecht H,
Rumberger JA. Aggressive versus moderate lipid-lowering therapy in
hypercholesterolemic postmenopausal women: Beyond Endorsed Lipid
Lowering With EBT Scannjng (BELLES). Circulation. 2005; 112:
563- 571.
CLINICAL PERSPECTIVE
Use of calcium and vitamin D supplements is widespread, particularly among older women. In observational studies,
calcium has been associated wilh lower blood pressure and weight loss, which might be expected to lower Jisk for coronary
heart disease and stroke. On the other hand, individuals with coronary artery calcification are at higher tisk for coronary
events, raising concern among patients Lhat calcium supplementation may be deleterious. In the Women's Health Initiative
placebo-controlled trial, calcium 1000 mg plus vitamin D 400 IU daily neither increased nor decreased the risk of coronary
heart disease or stroke during a 7 -year follow up of 36 282 postmenopausal women. Thus, women taking these supplements
need not fear adverse cardiovascul ar consequences whi le protecting their bone health.
Int. J. Cancer: 120, 1116-1122 (2006)
2006 Wiley-Liss. Inc.
The association of use of sunbeds with cutaneous malignant melanoma
and other skin cancers: A systematic review
The International Agency for Research on Cancer Working Group on artificial ultraviolet (UV) light and skin cancer
Exposure to solar ultraviolet (UV) radiation is a known cause of
skin cancer. Sunbed use represents an increasingly frequent
source of artificial UV exposure in light.-skinned populations. To
assess the available evidence of the association between sunbed
use and cutaneous malignant melanoma (melanoma) and other
skin cancers, a systematic r eview of the literature till March 2006
on epidemiological and biological studies on sunbed use was per-
formed in Pubmed, lSI Web of Scienc.e, Em base, Pascal , Cochrane
library, Lilacs and Medcarib. Search for keywords in the title and
in the abstract was done systematicall y and supplemented by man-
ual searches. Only case-control, cohort or cross-sectional studies
were selected. Data were abstracted by means of a standardized
data-collection protocol. Based on 19 informative studies, ever-use
of sunbeds was positively associated with melanoma (summary
relative risk, 1.15; 95% CI, 1.00-1.31), although there was no con-
sistent evidence of a dose-response relationship. First exposure to
sunbeds before 35 years of age significantly increased the risk of
melanoma, based on 7 informati ve studies (summary relative risk,
1.75; 95% CI, 1.35-2.26). The summary relative risk of 3 studies
of squamous cell carcinoma showed an increased risk. For basal
cell carcinoma, the studies did not support an association. The evi-
dence does not support a protective effect of the use of sunbeds
against damage to the skin from subsequent sun exposure. Young
adults should be discouraged from using indoor tanning equip-
ment and restricted access to sunbeds by minors should be
strongly considered.
2006 Wiley-Liss, Inc.
Key words: ;utificial UV; sunbeds; melanoma; skin cancer; meta-
analysis
Sun exposure is the main envi ronmental cause of skin cancer,
and ultraviolet (UV) radiation is the solar wavelength involved in
skin cancer, including the malignant cutaneous melanoma.
1
People
may also be exposed to UV radiation through many artificial sour-
ces at home and in the workplace, with some individuals receiving
high doses. Sources of artificial UV radiation include various lamps
used in medicine, industry, business and research, as well as for
domestic and cosmetic purposes. Sunbeds and sunlamps used for
tanning purposes are the main source of deliberate exposure to arti -
ficial UV radiation.t Although the contexts of sun exposure and
indoor tanning differ, both deliver UV radiation, and their health
effects would therefore be expected to be similar.
UV radiation wavelengths range between I 00 and 400 nm and
are broadly categorized into UVA (> 315-400 nm). UVB (>280-
315 run) and UVC (100-280 nm). Modern indoor tanning equip-
ment mainly emits in the UVA range, but a fraction (i.e . < 5%) of
this spectrum is in the UVB range.
Before 1990, UVB was usually considered the only ca.cino-
genic part of the solar spectrum, but since then UV A as well has
been suspected of having carcinogenic potential. In 1992, the
International Agency for Research on Cancer (IARC) classified
UVB and UV A radiation, as well as "use of sunlamps and sun-
beds," as " probably carcinogenic to humans" (Group 2A of the
IARC classification of carcinogenic agents). ' More recently, the
I Oth Report on Carcinogens published by the National Toxicology
Program in the USA classified UV A radiation as a "known to be a
human carcinogen. "
2
Biological mechanisms by which chronic
sun exposure causes squamous cell cancer (SCC) of the skin have
become better known and chronic exposure to high UVB doses is
now considered as the main environmental cause of that ski n
cancer.
3
Biological mechanisms implicated in basal cell carci-
~ u i
Publ ication of the International Union Against Cancer
noma (BCC) start to be better known. In contrast, we still have
poor knowledge of the UV wavelength and the dose delivery pat-
tern at skin level implicated in the genesis of melanoma and of
BCC.
4
Indoor tanning is widely practiced in most developed countries,
particularly in Northern Europe and the USA, and is 1.\aining popu-
larity even in sunny countries such as Australia.
6
The likely
impact of this fashion on skin cancer incidence is of substantial
concern, mainly for cutaneous malignant melanoma (hereafter
melanoma), a cancer of poor prognosis when diagnosed at an
advanced stage.
This paper summarizes a systematic review of epidemiological
and experimental studies on use of indoor tanning equipment and
skin cancer developed by a Working Group convened by IARC.
UV spectra from sunlight and indoor UV tanning appliances
During a sunny day on the Mediterranean coast, the solar UV
spectrum at noon contains 4-5% UVB and 95-96% UV A. When
UV output of a typical indoor tanning appliance is calculated in
terms of biological activity, as estimated by the erythema-effec-
tive inadiance, the emission of many tanning appliances is equiva-
lent to or exceeds the emission of the midday sun in southern
Europe.
7
8
The UV intensity of powerful tanning appliances may
be 10-15 times higher than that of the midday sun.
8
leading to
UV A doses per unit of time received by the skin during a typical
tanning session that are well above those experienced during ordi-
nary daily activities or even during sunbathing. As a result, the an-
nual UV A doses received by frequent indoor tanners may be 1.2-
4.7 times those received from the sun, in addition to those received
from the sun.
9
This widespread repeated exposure to high doses of
UV A constitutes a new phenomenon for human beings.
Members of the Working Group: Adele Green (Chair), Queensland
Institute of Medical Research, PO Royal Brisbane Hospital, Brisbane, Qld,
Australia; Philippe Autier. Unit of Epidemiology. Prevention and Screen-
ing. Jules Bordet Institute, Brussels, Belgium (current address is: Interna-
tional Agency for Research on Cancer, Lyon, France); Mathieu Boniol and
Peter Boyle, International Agency for Research on Cancer, Lyon. France;
Jean-Francais DortS, INSERM U590, Centre Leon Berard, Lyon, France;
Sara Gandini, Division of Biostatistics and Epidemiology, European Insti -
tute of Oncology, Milan, Italy; Juli a Newwn-Bishop, Cancer Research UK
Genetic Epidemiology Division. St James's University Hospital , Leeds,
UK; Beatrice Secretan,* International Agency for Research on Cancer,
Lyon, France; Stephen J Walter, Cl inical Epidemiology and Biostatistics,
McMaster University, Hamilton, Ont., Canada; Martin A. Weinstock, Der-
matoepidemiology Unit, VA Medical Center, Providence, and Depan-
ments of Dermatology and Community Health, Brown University, Provi-
dence, Rhode Island; and Johan Westerdahl. Department of Surgery. Lund
University Hospital, Lund, Sweden.
tThe device used for tanning may be referred to as sunbed, sunlamp, ar-
tificial UV, artificial light or tanning bed. among other terms. Also, anum-
ber of terms are used to define a place where indoor tarming may occur: so-
larium, tanning salon, tanning parlor, tanning booth, indoor tanning salon,
indoor tanning facil ity. In addition, indoor tanning may also occur in non-
commercial premises. For the purpose of this report, the term indoor Ian-
ning equipment has been used throughout.
*Correspondence to: international Agency for Research on Cancer,
150 Cours Albert Thomas, F-69372 Lyon Cedex 08, France.
Fax: + 33-4-72738319 or +33-4-72738351. E-ma.il: secretan@iarc.fr
Received 8June 2006; Accepted after revision 27 September 2006
DOl 10. 1002/ijc.22453
Published online 27 November 2006 in Wiley lnterScience (www.interscience.
wiley.com).
SUNBED USE AND RISK OF MELANOMA AND SKIN CANCERS 1117
In the 1990s, regulations in some countries (e.g., France, Swe-
den) limited to 1.5% the maximum percentage of UVB in the UV
output of tanning appliances. However, in practice, the UV output
and spectral characteristics (i.e., amounts of UV A, UVB, visible
light and infrared radiation) of tanning appliances vary consider-
ably. The proportion of UVB in UV energy output could vary
from 0.5 to 4%,
10
11
and may attain an emission spectrum similar
to the sun spectrum in the UVB range.
8
These differences are due
to sunbed design (e.g., the numbers and type of fluorescent tubes,
the presence of high pressure UV lamps. the materials composing
filters, the distance from canopy to the skin), sunbed power and
tube ageing.
Biological effects of exposure to artificial UV radiation
relevant to carcinogenesis
A large body of experimental and epidemiological data strongly
indicates that the spectrum of l!,V radiation reaching the Earth's
surface causes skin cancer ...
12
.ao UVB is a complete carcinogen
that is absorbed by DNA and can damage DNA directly.
13
Evidence of the mutagenic properties of UV A in humans has
been found in several st udies.
12
-
14
UV A radiation does cause
UVB-like cyclobutane pyrimidine dimers and 6-4 photoproducts,
albeit with a much lower efficacy than does UVB radiation. Most
of the DNA damage induced by UV A is indirect, through the
absorption of UV A photons by other cellular structures (chromo-
phores), with formation of reactive oxygen species that can trans-
fer UV A energy to DNA via mutagenic oxidative intermediates.
15
Skin of human volunteers exposed to UV A lamps used in tan-
ning appliances show DNA damage, p53 mutations induced by
oxidative damage and alterations of the p53 protein similar to
those observed after sun exposure or after exposure of experimen-
tal animals.
1
6-
18
UV A penetrates deeper into human skin than does UVB.
Because UV A represents the largest proportion of the UV spec-
trum of tanning appliances and of solar radiation reaching the
Earth' s surface, far more UVA than UVB reaches the basal layers
of the epidermis where melanocytes and early keratinocytic cells
are located.
Both UV A and UVB radiation can affect the immune resrconse
that may be involved in the promotion of melanoma,'
5
19

0
but
the 2 types of radiation seem to act differentty_21.
22
UVB induces
immunosuppression at both the local and systemic levels, while
UV A does not induce systemic immune suppression
2 3
To date, evidence obtained from experimental studies on the
involvement of high UVB doses in the causation of SCC is con-
sistent with observations in htUnans. In contrast, experimental
studies give conflicting results regarding the roles of UVB and
UV A in the induction of melanoma in humans. The same uncer-
tainties hold true for BCC, a type of tumor that shares some epide-
miological characteristics of melanoma.
Experiments carried out in animals cannot reproduce the com-
plex interplay in individuals between highly variable natural sus-
ceptibilities to UV radiation, sun exposure behaviors and exposure
to various sources of UV radiation. During indoor tanning, such
interrelationships may be critical, as users are more inclined than
the average population to engage in outdoor tanning activities,
24
and indoor tanning sessions often precede or follow active sun ex-
posure or outdoor tanning.
Effects of artificial UV on human skin
Skin redness or burning are reported by 18-55% of users of
indoor tanning equipment in Europe and North America.
25
Although UVB is far more potent than UV A in causing sunburn,
high fluxes of UV A are capable of inducing skin redness in indi-
viduals sensitive to sunl ight or with only moderate tanning ability.
In individuals who tan easily, exposure to tanning appliances
will lead first to the oxidation of melanin already present in super-
ficial keratinocytic layers of the skin, known as immediate pig-
ment darkening?
6
A more permanent tan is acquired with accu-
mulation of exposure, depending on tanning ability and on the
amount of UVB present in the UV spectrum of the lamps.
Immediate pigment darkening has no photoprotective effect
against UV-induced skin redness or sunburn.
27
Moreover a OVA-
induced permanent tan provides little photoprotection
28
29
and the
skin thickening caused by UV A affords only very little photopro-
tection.30 Studies in humans show that a prevacation tan induced
artificially offers virtually no protection against sun-induced DNA
damage.:n-
33
E1posure to artificial UV for tanning pwposes
Few people had used indoor tanning equipment before 1980 but
by the end of the 1990s more than 60% of women and 50% of
men aged 18- 50 years in Northern Europe reported having ever
used indoor tanning equipment.
34
indeed, prevalence of indoor
tanning is increasing so rapidly in many countries that current esti-
mates may be outdated rapidly. The most frequent motivations for
indoor tanning are the acquisition of a so-called safe tan and prep-
aration of the skin before sun exposure.2
5
Use of indoor tanning equipment is more prevalent among
women and among both men and women younger than 35 years.
Earliest studies in Sweden and in the USA tended to find indoor
tanning to be more prevalent among adolescents with fair skin
types who are more prone to sunburn.
35
-
37
More recent studies in
the USA found either the opposite
38
-4 or no association.
4 1
Few studies have assessed the compliance of indoor tanning facil-
ity operators or consumers with recommendations and regulations.
Overall, information provided by tanning salon operators on health
risks and on duration and frequency of exposure is often incomplete,
and there is a lack of identification of highly sun-sensitive subjects or
of subjects taking photosensitizing medications.6.4
2
-44
About 17-35% sunbed users reported that they did not wear eye
protection.
10
.4
1
43
In some surveys, 16% of sunbed users may have
had more than I 00 sessions per year,
10
and most users tend to
exceed the recommended exposure times.
41
44
45
Since 1989, a total of 16 studies (18 reports) have examined
prevalence of indoor tanning among children and adolescents
aged 8- 19 years in Australia, Europe and the USA
46
47
All studies
showed a frequent use by adolescents and children, sometimes at a
very young age. According to the most recent studies, 30% of ado-
lescents in Sweden and 24% of adolescents in the USA aged 13-
19 years reported ever-use of indoor tanning equipment and 8 and
12% respectively were frequent users (10 times per year or more).
In a recent survey in the United Kingdom, while 7% of children
aged 8- 11 years reported exposure to a sunbed Ln the East 6
months, as many as 48% expressed a desire to use a sunbed.
8
Epidemiological studies on indoor tanning and skin cancer
As existing animal models of human melanoma are inconsis-
tent , evidence of an association between indoor tanning and skin
cancer must be sought predominantly from epidemiological stud-
ies. Few studies have addressed this topic specifically, but some
studies included I or more secondary questions about indoor tan-
ning. We systematically analyzed the results from the relevant
studies and compiled them in a metaanalysis.
Methods
The methodology used for the literature search is summarized
in Table I. The minimal common information about exposure to
indoor tanning appliances for all studies was "ever exposed. " For
those studies wherein "ever exposed to indoor tanning appliances
versus never" was not strictly assessed
4950
we used the informa-
tion closest to this category.
Most estimates included all subjects and combined sexes in the
analysis. Some studies presented results separately for women and
men, with no combined data, in which case both estimates were
1118 !ARC WORKING GROUP
included. Since the studies used different age categories for classi-
fying age at first exposure, we considered as " young exposure"
those exposures that started before 35 years of age.
Every measure of association adj usted for the maximum num-
ber of confounding variables, and corresponding confidence inter-
TABLE I- METHOD USED FOR THE LITERATURE SEARCH
The literature to March 2006 was searched using the following
databases: Pubmed, IST Web of Science (Science Citation
Index Expanded), Embase, Pascal, Cochrane library, Lilacs
and Medcarib. The following keywords and their
corresponding French translation were used for search in the
PASCAL database: skin cancer, squamous cell carcinoma,
SCC. basal cell carcinoma, BCC and melanoma for diseases.
To define exposure, the following keywords were used:
sunbed, sunlamp, artificial UV, artiflciallight, solaria,
solarium. indoor tanning, tanning bed, tanning parlour.
tanning salon and tanning booth.
Search for keywords in the title and in the abstract was done
systematically. Manual search was done of references cited in
the selected articles, and in selected reviews or books on
melanoma and skin cancer. All participants of the working
group were asked to report any additional published or
submitted study. No language restriction was applied.
Primary inclusion criteria were developed for the selection of
relevant articles, which were case-control, cohort or cross-
sectional studies published as an original article. Ecological
srudies. case reports, reviews and editorials were not
considered eligi ble.
The selected articles were reviewed, and data were abstracted by
means of a standardized data-collection protocol. When
another article on the same study was published
simultaneously, additional relevant or missing information
was retrieved from the companion paper.
val (CI), was transformed into logarithms of rela5ive risk (log RR)
and the corresponding variance was calculat.ed.'
1
Where no esti-
mates were reported, the crude estimates were calculated from
tabular data, using asymptotic Mantei-Haenszel methods to evalu-
ate the 95% CI of the log odds ratio.
The homogeneity of the effects across studies was assessed using
the large sample test based on the x.
2
-test. The summary relative risk
was estimated using random effects models even when heterogeneity
was found to be not statistically significant, in order to be conserva-
tive. Publication bias was investigated by funnel plot regression.
52
Studies on melanoma
We identified 23 studies on use of indoor tanning equipment
and melanoma (Table II) .
34
.4
9
50
5
3-
73
All studies used the case-
control design, except for I cohort studyso A case-control study
was considered population-based when cases were derived from a
population-based cancer registry and controls were selected from
the general population. Of these 23 studies, 4 studies were
excluded from the metaanalysis because they did not include esti-
mates of the relative risk for cutaneous melanoma associated with
exposure to tanning appliances.
53
5557
62
Studies used for the metaanalysis included a total of 7,355
cases. The first study was published in 1981 and the last in 2005.
Fifteen studies were carried out in European countries, 4 of which
in Scandinavian countries, and 2 were in the United States, I in
Canada and l in Australia.
Studies on basal cell and squamous cell carcinomas
Nine case-control studies have examined the association
between indoor tanning and either BCC or SCC of the skin.
7
4-8
2
All studies reported a risk estimate except one,
74
which was there-
fore excluded. A further 3 studies that did not distinguish between
TABLE n - CHARACTERISTICS OF THE STUDIES CONSIDERED FOR THE METAANALYSIS ON MELANOMA
Reference Coumry
Number
Relative risk
2
Cases Comrols
Cohort study _
Veier0d et al. (2003)'
0
Norway, Sweden 187 106.379
1
1.55 (1.04-2.32)
Population-based case-control
studies
Adam eta/. (1981)
54
UK 169 207 2.93 (l.l6-j.40)
Gallagher eta/. (l98<]t
5
Canada 595 595
Holman eta/. (1986)' Australia 511 511 1.1 (0.6-1.8)
Osterlind et al. (198ID
59
Denmark 474 926 0.73 (0.53-1.01)
Zanetti eta/. (1988) Italy 208 416 0.9 (0.4- 2p)
Beitner eta/. (1990)
62
Sweden 523 505
Walter et a/. (1990)
63
Canada 583 608
4
Westerdahl et al. (1994)
70
Sweden 400 640 1.3 (0.9-1.8)
Holly era/. USA 452 930 0.94 (0.74-1.2)
Chen era/. (1998)
69
USA 624 512 1.13 (0.82-1.54)
Walter eta/. (1999)
64
Canada 583 608 !.54 ( 1.1 6-2.05)
Westerdahl era/. (2000)
73
Sweden 571 913 1.2 (0.9-1.6)
Other case-control studies
Klepp and Magnus

Norway 78 131
Holly er a/. ( 1987)
5
USA 121 139
Swerdlow eta/.

UK 180 120 2.94 (1.41- 6.17)
MacKie et a/. ( 1989)
6
UK 280 !80 1.3 (0.2-7.9) for men;
Dunn-Lane et al. (1993)
65
1.2 (0.5-3.0) for women
UK 100 100 1.16 (0.54-2.47)
Garbe eta/. (1993)
66
Germany 280 280 1.5 (0.9-2.4)
Autier eta/. (

Belgium, France, and Germany 420 447 0.97 (0.71-1.32)
Naldi eta/. (2000)
1
Italy 542 538 0.78 (0.45- 1.37)
Kaskel eta/. (2001)
49
Germany 271 271 1.00 (0.6-1.8)
Bataille eta/. (2004)
72
UK 41 3 416 1. 19 (0.84-1.68)
Bataille eta/. (2005)
34
Belgium, France, the Netherlands, 597 622 0.90 (0.71-1.14)
Sweden, UK
ALM, acrallentiginous melanoma; HC, histologically confirmed; LMM, lentigo maligna melanoma; M, melanoma; MM, malignant melanoma;
NM, nodular melanoma; SSM, superficial spreading melanoma.
1
Cohort size.-
2
Values in parentheses are 95% CI .-
3
Because no estimate of risk was reported in these studies, we did not include them in the
metaanalysis.-
4
The study by Walter et al. (1990)
63
was reanalyzed in the 1999 publication. We used the relative risk adjusted for potential con-
founders presented in the 1999 publication.
Studies
Adam et e.!.,
Holnwl et e.!., I
OsterliM et e.!, 1988
Swerdlow et e.!, 1988
Zanetti et e.!., 1988
MeeK ill et e.!., I (Men)
MIIC.Kie et al, 1989 (Women)
Dunn-!..an et e.l.,l993
Geroe et e.!., 1993
A ulier et e.!., 1994
WesteldAhl et e.!., 1994
Holly et e.!., 199 5
Chen et e.!.,
Welter et e.!., 1999
N eldi et e.!., 2000
WesteldAhl et e.!., 2000
Keske! et e.!, llDI
Veierod et e.!., 2003
Bataille et e.!., 2004
&taille et e.!., 2005
Summary relative risk
-
-
-
--
I
-
-

-
p
1.15 (1.00, 1.31)
I I I I I I
0.5 1.0 1.5 2.0 2.5 3.0 5.0 7.0
FIGURE 1 - Relative risk for cu-
taneous melanoma associated with
ever use of indoor tanning equip-
ment: estimates of 19 studies and
summary estimate (relative risks
were presented separately for men
and women in the study by
MacKie et a/
61
) . Relative risk
TABLE Ill - METAANALYSIS OF EPIDEMIOLOGICAL STUDIES ON INDOOR TANNING AND RISK
FOR J'vlELANOMA, SQUAMOUS CELL CARCINOMA AND BASAL CELL CARCINOMA
Exposure
Number of
1
Heterogeneity
2
sludics (pvalue)
Melanoma
Ever use of indoor tanning equipment 19 1.15 (1.00-1.31) 0.013
First exposure in youth 7 1.75 ( 1.35-2.26) 0.55
Exposure distant in time 5 1.49 (0.93- 2.38) 0.018
Exposure recent in time 5 1.10 (0.76-1.60) 0.81
Squamous cell carcinoma
Ever use of indoor tanning equipment 3 2.25 (1.08-4.70) 0.10
Basal cell carcinoma
Ever use of indoor tanning equipment 4 1.03 (0.56-1. 90) 0.06
1
Values in parentheses are 95% CI.-
2
x
2
-test: the degrees of freedom are given by the number of risk
estimates included minus I.
these 2 major types of skin cancer
75
-
77
were also excluded from
review, leaving 5 studies for consideration.
Relative risk for melanoma
Thirteen of 19 studies presented positive estimates for "ever"
versus " never" exposed to inc_loor tanning equipment, but only 4
were statistically

(Fig. 1). Seven of these stud-
ies reported only crude relative risks, and I adjusted for age and
sex only. Results of the metaanalysis are shown in Table III. The
summary estimate indicated a significant positive association
between "ever" versus "never" indoor tanning and melanoma
(RR, 1.15; Cl, 1.00-1.31) and the x
2
-test for heterogeneity was
statistically significant.
To decrease the influence of possible biases, estimates were cal -
culated including only the cohort and the 9 population-based
case-control studies. The sununary relative risk was very similar
apart from having wider Cis (RR, 1.17; CI, 0.96-1.42).ln an anal-
ysis restricted to the 8 studies that adjusted for confounders related
to sun exposure and sun sensitivity,
0
60
6
1.
64
69
-
7
1.
73
the summary
relative risk remained simi lar to that obtained from all 19 studies,
but the CI widened (RR. 1.19; CI, 0.33-4.30).
Seven studies presented estimates relevant for the evaluation of
" first exposure in youth" versus " never" (Fig. 2). All relative
risks were adjusted for confounders related to sun exposure or sun
sensitivity, except in the study by Walter e t a/
64
A sini ficant
75% increase in risk was detected (Table liT) and the X -test for
heterogeneity was nonsignificant.
Five studies investigated time since exposure and reported esti-
mates that a.llowed comparisons between recent and more distant
exposure.
34
58
63
67
69
Metaanalytic estimates were greater for
exposures more distant in time when compared to those for more
recent exposures (Table III).
There was some indication for a dose-effect relationship in 2
studies,
67
70
but not in the other two.
69
73
But metrics used for
assessing duration were all different and therefore did not permit
met.aanalytic synthesis. Only 4 studies explored the role of natural
sensitivity to sunlight on risk associated with indoor tanning, and
overall, they found no consistent resuJt.
34
64
72.
73
Type of indoor tam1ing equipment
No epidemiological study has been able to explore in a rigorous
way amounts of UV A and UVB received by indoor tanning users.
The study by Chen et a/.
69
obtained information concerning the
type of sunbed or sunlamp used (e.g. , desktop models, floor mod-
els, beds or walk-in booths). This information was obtained by
showing to subjects pictures of various types of sunlamps and sun-
Studies
Swerdlow et. al, 1988
Westerdahl et al., I 994
Chen et al. , 1998
Walter et a1 , 1999
Westerdahl et al., 2000
Veierod et al, 2003
Bataille et al., 2005
FIGURE 2- Relative risk for cu-
taneous melanoma associated with
first use of indoor tanning equip-
ment at age <35 years: estimates of
7 studies and summary estimate.
beds. The study found a nonsignificant elevated risk of malignant
melanoma associated with the use of desktop sunlamps and heavy-
weight floor-model sunbeds and a statistically significant tripled
risk associated with use of more than 2 types of sunlamps, com-
pared with no use of sunbeds. The study by Bataille et a/.
34
reported no impact of the type of device used on melanoma risk.
The relative risks of melanoma associated with ever-use of
sunbed/sunlamp reported in the studies did not vary with year of
publication or first year of study period, and funnel plot regression
gave no indication of publication bias (ever-use of sunbed/sun-
lamps, p = 0.80; first exposure in youth, p = 0.10). This observa-
tion suggests that the apparent increased risk for ever use and for
age at first use were unlikely to be explained by the earlier types
of indoor tanning appliance used.
Before 1980, exposure to artificial UV radiation was more
likely to take place at home with devices that emitted greater
amounts of UVB radiation, whereas exposure in the 1980s
increasingly occurred in commercial salons using equipment that
emitted mainly UVA. The Norway- Swedish prospective study
provided evidence that the increased melanoma risk associated
with exposure to t n n i n ~ appliances was not due to the type of UV
lamps used before 1983.
3
Relative risk for squamous cell carcinoma
and basal cell carcinoma
The metaanalysis was based on the 5 studies
78
-8
2
reporting type-
specific risk estimates (Table III). Metaanalytic estimates suggested
a significant effect of exposure to indoor talUling appliances for
SCC, but not for BCC. Funnel plot regression gave no indication of
publication bias (p = 0.26 and 0.77 for SCC and BCC, respec-
tively).
The study by Karagas et al.x
1
gave the most detailed results,
and the trends were consistent with the results reported for mela-
noma. Results were adjusted for sun sensitivity but not for sun ex-
posure, since adjustment for sun exposure did not change the risk
estimates. Depending on age at first use, the risks for BCC and
SCC were found to increase by 10% (OR, 1.1; Cl, 0.9-1.5) and
20% (OR, 1.2; CI, 0.9-1.6) respectively for each decade younger
the person was at first use of indoor tanning equipment.
-
-

-
-

-
<>
1.75 (1.35-2.26)
I I I I I I I
0.5 1.0 1.5 2.0 2.5 3.0 5.0 7.0
Relative risk
Discussion
Investigation of the assoc1atwn between indoor tanning and
skin cancers poses challenging problems, as indoor tanning has
been in widespread use only recently. Based on our knowledge
about the relationship between sun exposure and ri sk for mela-
noma, it could be stated that associations after long latency peri-
ods, such as would be expected for melanoma and BCC, may not
be detectable yet. Also, since the fashion of indoor tanning has
been increasing steadily, the failure to distinguish between distant
and recent exposures in most epidemiological studies may mask
an actual increase in risk with exposure early in life.
Our systematic review of published studies mainly from Europe
and North America of the association of use of indoor tanning
equipment with skin cancers revealed an association of age at first
use of less than 35 years with melanoma risk. These studies consis-
tently indicated a moderate strength of association, with a summary
relative risk of 1.75 (1.35-2.26). This result suggests a greater vul-
nerability of younger people to the carcinogenic impact of indoor
tanning. Also, it is in agreement with the knowledge that age at ex-
posure may influence the relative risk for skin cancer associated
with UV exposure, and that exposure to sunlight in childhood is an
important contributing factor for melanoma risk in adults.
84
85
The association with ever-use of such equipment, or use more
than '15- 20 years prior to diagnosis of melanoma. was weak, and
evidence regarding a dose- response relationship was scant. The
evidence is limited by concerns over characterization of exposure
and recall of exposure by individuals, potential confounding by
sun exposure or other variables and the low power to dete.ct asso-
ciations that become evident only following a prolonged lag pe-
riod after exposure. Our results are similar to a previous metaanal-
ysis,86 but our systematic review is more exhaustive and included
more studies.
In Scandinavian countries use of indoor tanning equipment has
been popular since the late 1970s and the prevalence of use in
those countries is the highest in the world. In the Norwegian-
Swedish prospective study the highest risk for melanoma was
found in women who used indoor tanning equipment at least once
per month when they were 20- 29 years old. These results support
the hypothesis that a certain lag period is needed before the impact
of exposure to tanning appliances on melanoma incidence
becomes apparent. It also underlines the greater vulnerability of
younger subjects to harmful effects of indoor tanning.
The positive association between use of indoor tanning equip-
ment and melanoma risk reported here is consistent with the
knowledge that melanoma is caused primarily by exposure to solar
radiation. The limited evidence for a positive association between
indoor tanning and sec is consistent with its known dependence
on dose of UV radiation to the skin. Thus the biological plausibil -
ity of a causal association between indoor tanning and risk for
melanoma and sec is strong.
On balance, the evidence pertaining to the strength, consistency,
dose- response and temporal sequence of the association of the use
of indoor tanni ng equipment with melanoma risk, and of the coher-
ence and biologic plausibility of the association, leads us to conclude
that there is convincing evidence to support a causal relationship,
particularly with exposwe before the age of 35 years. This evidence
is strongly suggestive and further studies could clarify our under-
standing of tllis association and allow more definitive conclusions.
We are cognizant of the importance of this issue for the health
of light-skinned populations. The strength of the existing evidence
suggests that policy makers should strongly consider enacting
measures such as restricting minors and discouraging young adults
from using indoor tanning equipment, in order to protect the gen-
eral population from additional risk for melanoma and squamous
cell skin cancer.
References
l. IARC. Solar and ultraviolet radiation. Lyon: International Agency for
Research on Cancer, 1992. !ARC monographs on the evaluation of
carcinogenic risks to humans , vol. 55.
2. National Toxicology Program. lOth Report on Carcinogens: substan-
ces profiles. Research Triangle Park, US Department of Health and
Human Services, Public Health Service, 2002.
3. de Gruijl FR, van Kranen HJ, Mullenders LH. UV-induced DNA
damage, repair, mutations and oncogenic pathways in skin cancer.
1 Photochem Photobiol B 2001 ;63:19- 27.
4. de Gruijl FR. Photocarcinogenesis: UVA vs. UVB radiati on. Skin
Pharmacol Appl Skin Physiol 2002; 15:316- 20.
5. Paul CL, Girgis A, Tzelepis F, Walsh RA. Solaria use by minors in
Australia: is there a cause for concern? Aust. N Z J Public Health 2004;
28:90.
6. Paul CL, Stacy F, Girgis A. Brozek I, Baird H, Hughes J. Solaria com-
pliance in an unregulated environment: the Australian experience. Eur
1 Cancer 2005:41 :1178- 84.
7. Wester U. Boldemann C, Jansson B, Ullen H. Population UV-dose
and skin area-do sunbeds rival the sun? Health Phys 1999;77:436-
40.
8. Gerber B, Mathys P, Moser M, Bressoud D, Braun-Fahrlander C.
Ul traviolet emission spectra of sunbeds. Photochem Photobiol 2002;
76:664-8.
9. Miller SA, Hamilton SL, Wester UG, Cyr WH. An analysis of UV A
emissions from sunlamps and the potential importance for melanoma.
Photochem Photobiol 1998:68:63-70.
10. MacGintley J, Manin CJ, MacKie RM. Sunbeds in currem use in
Scotland: a survey of their output and pattems of use. Br J Dermatol
1998:139:428-38.
II. Wrigllt AL, Hart GC, Kernohan E, Twentyman G. Survey of the vari-
ation in ultraviolet outputs from ultraviolet A sunbeds in Bradford.
Photodermatol Photoimmunol Photomed 1996; 12: 12- 16.
12. Pfeifer GP, You YH, Besaratinia A. Mutations induced by ultraviolet
light. Mutat Res 2005:571:19-31.
13. Griffiths HR, Mistry P, Herbe rt KE. Lunec J. Molecular and cellular
effects of ultraviolet light-induced genotoxicity. Crit Rev Clin Lab
Sci 1998;35: 189- 237.
14. Robel'! C, Muel B, Benoit A, Dubertret L, Sarasin A, Stary A. Cell
survival and shullle vector mutagenesis induced by UV A and UVB
radiation in a human cell line. J Invest Demtatol 1996; 106:721-8.
15. Halliday G. Inflammation, gene mutation and photoimmunosuppres-
sion in response to UVR-induced oxidative damage contributes to
photocarcinogenesis. Mut Res 2005;571: I 07-20.
16. Woollons A, Clingen PH, Price ML, Arlett CF, Green MH. Induction
of mutagenic DNA damage in human fibroblasts after exposure to ar-
tificial tanning lamps. Br J Dermatol1997;137:687-92.
17. Whitmore SE, Morison WL, Potten CS, Chadwick C. Tanning salon
exposure and molecular alterations. J Am Acad Dermatol 2001:44:
775-80.
18. Persson AE, Edstrom DW, Backvall H, Lundeberg J, Ponten F, Ros
AM, Williams C. The mutagenic effect of ultraviolet-A I on human
skin demonstrated by sequencing the p53 gene in single kerat inocytes.
Photodermatol Photoimmunol Photomed 2002; 18:287- 93.
19. Kripke ML. Antigenicity of murine skin tumors induced by UV light.
J Nat ! Cancer Inst 1974:53: 1333-{).
20. Singh RK, Gutman M, Reich R, Bar-Eli M. Ultraviolet B irradiation
promotes tumorigenic and metastatic properties in primary cutaneous
melanoma via induction of interleukin-8. Cancer Res 1995;55:3669-
74.
21. Phan TA, Halliday GM. Barnetson RS, Damian DL. Spectral and
dose dependence of ultraviolet radiation-induced immunosuppression.
Front Biosci 2006: I I 9 ~ 11.
22. Aboutalebi S, Strickland FM. lmmune protection, natural products,
and skin cancer: is there anything new under the sun? J Drugs Derrna-
tol2006;5:512- 7.
23. Ullrich SE. Mechanisms underlying UY-induced immune suppres-
sion. Mutat Res 2005:571:185-205.
24. Autier P, Joarlette M. Lejeune F. Lienard D, AndreJ, Acbten G. Cuta-
neous malignant melanoma and exposure to sunlamps and sunbeds: a
descriptive study in Belgium. Melanoma Res 1991; I :69-74.
25. Autier P. Perspectives in melanoma prevention: the case of sunbcds.
Eur J Cancer 2004;40:2367-76.
26. Pedeux R, AI-Irani N, Maneau C, Pellicier F, Bra11che R, Ozturk M,
Franchi J, Dore JF. Thymidine dinucleotides induceS phase ce ll cycle
arrest in addition to increased melanogenesis in human melanocytes.
1lnvest Dermatoll998; 111:472-7.
27. Black G, Matzinger E. Gange RW. Lack of photoprotection against
UVB-induced erythema by immediate pigmentation induced by 382
nm radiation. J Invest Dermatoll985:85:448- 9.
28. Gange RW, Blackett AD, Matzinger EA, Sutherland BM, Kochevar
IE. Comparative protection efficiency of UVA- and UVB-induced
tans against e rythema and formation of endonuclease-sensitive sites
in DNA by UVB in human skin. J Invest Dermatol l 985;85: 362-4.
29. Rivers JK, Norris PG, Murphy GM, Chu AC. Midgley G, Morris J,
Morris RW, Young AR, Hawk JL UV A sunbeds: tanning, photopro-
tection, acute adverse effects and immunological changes. Br J Der-
matol 1989; 120:767-77.
30. Sheehan JM. Potten CS, Young AR. Ta1ming in human skin types ll
and m offers modest photoprotection against erythema. Photochem
Photobiol 1998;68:588- 92.
31 . Hemminki K, Bykov V J, Marcusson JA. Re: Sunscreen use and dura-
tion of sun exposure: a double-blind, randomised trial. 1 Nat! Cancer
lost 1999;91 :2046-7.
32. Bykov VJ, Marcusson JA, Hemminki K. Protective effects of tanning
on cutaneous DNA damage in s itu melanoma. Dermatology 200 I ;
202:22- 6.
33. Ruegemer J. Schuetz B, Hermann K, Hein R, Ring J, Abeck D. UV-
induced skin changes due to regular use of commercial sunbeds. Pho-
todermatol Photoimmunol Photomed 2002; 18:223- 7.
34. Bataille V, Boniol M, De Vries E, Severi G, Brandbcrg Y, Sasieni P,
Cuzick J, Eggermont A, Ringborg U. Grivegnee AR, Coebergh JW.
Chignol MC, et al. A multicentre epidemiological study on sunbed
use and cutaneous melanoma in Europe. Eur J Cancer 2005;41:2141- 9.
35. Me1melstein RJ, Riesenberg LA. Changing knowledge and attitudes
about skin cancer risk factors in adolescents. Health Psycho! 1992:
11:371- 6.
36. Boldeman C, Beitner H, Jansson B, Nilsson B, Ullen H. Sunbed use
in relation to phenotype, erythema, sunscreen use and skin diseases. A
questionnaire survey among Swedish adolescents. Br J Dermatol
1996; 135:712-6.
37. Robinson JK, Rademaker AW, Sylvester JA, Cook B. Summer sun
exposure: knowledge, attitudes and behaviors of Midwest adolescents.
Prev Med 1997;26:364-72.
38. Cokkinides VE, Weinstock M, O' Connel l MC, Thun MJ. Use of
indoor tanning sunlamps by US youth, ages 11- 18 years, and by their
parent or guardian caregivers: prevalence and correlates. Pediatrics
2002; 109: 1124-30.
39. Demko CA. Borawski EA, Debanne SM. Cooper KD, Stange KC.
Use of indoor tanning faci lities by white adolescents in the United
States. Arch Pediatr Adolesc Med 2003;157:854-60.
40. Geller AC, Colditz GA. Oliveria S, Emmons K. Jorgensen C, Aweh
GN, Frazier AL. Use of sunscreen, sunburning rates, and tanning bed
use among more than 10,000 US children and adolescents. Pediatrics
2002; 109: 1009-14.
41. Oliphant JA, Forster JL, McBride CM. The use of commercial tanning
fac ilities by suburban Minnesota adolescents. Am J Public Health
1994;84:476- 8.
42. Culley CA. Mayer JA, Eckhardt L, Busic A.l , Eichenfield LF, Sall is
JF, Quintana PJ, Woodmff SI. Compliance with federal and state
legislation by indoor tanning facilities in San Diego. J Am Acad Der-
matol 200 I ;44:53-60.
43. Rhainds M, De Guire L, Claveau J. A population-based survey on the
use of artificial tanning devices in the province of Quebec, Canada.
J Am Acad Dcm1atol 1999;40:572-6.
44. Kwon HT, Mayer JA, Walker KK. YuH. Lewis EC. Belch GE. Pro-
motion of frequent tanning sessions by indoor tanning facilities: two
studies. JAm Acad Dermatol 2002;46:700-5.
45. Hornung RL, Magee KH, Lee WJ, Hansen LA. Hsieh YC. Tanning
facility use: are we exceeding Food and Drug Administration limits?
JAm Acad Dennatol2003;49:655-61.
46. Lazovich D, Forster JL, Sorensen G, Emmons K, Stryker JE, Demi-
erre MF, Hickle A, Remba N. Characteristics associated with use or
intention to use indoor tanning among adolesceots. Arch Pediatr Ado-
lese Mcd 2004;158:918- 24.
47. Lazovich D, Forster J. lndoor tanning by adolescents: prevaleoce.
practices and policies. Eur J Cancer 2005;41:20-7.
48. Hamlet N, Kennedy K. Reconnaissance study of sunbed use by pri-
mary school children in Lanarkshire. J Public Health 2004;26:31-3.
49. Kaskel P, SanderS, Kron M, Kind P, Peter RU, Krahn G. Outdoor
activities in childhood: a protective factor for cutaneous melanoma?
Results of a case-control study in 271 matched pairs. Br J Dermatol
2001; 145:602-9.
50. Veieri'Sd MB, Weiderpass E, Thorn M, Hansson J, Lund E, Armstrong
B, Adami HO. A prospective study of pigmentation, sun exposure,
and risk of cutaneous malignant melanoma in women. J Nat! Cancer
Inst 2003;95: I 530-8.
51. Greenland S. Quantitative methods in the review of epidemiologic lit-
erature. Epidemiol Rev 1987;9:1-30.
52. Macaskill P, Walter SD, Irwig L. A comparison of methods to detect
publication bias in meta-analysis. Stat Med 2001 ;20:641-54.
53. Klepp 0 , Magnus K. Some environmental and bodily characteristics
of melanoma patients. A case-control study. lnt J Cancer 1979;23:
482-6.
54. Adam SA, Sheaves JK, Wright NH, Mosser G, Harris RW, Vessey
l\tiP. A case-control study of the possible association between oral
contraceptives and malignant melanoma. Br J Cancer 1981 ;44:45-
50.
55. Gallagher RP. Elwood JM, Hill GB. Risk factors for cutaneous malig-
nant melanoma: the Western Canada Melanoma Study. Recent
Results Cancer Res 1986;102:38- 55.
56. Holman CD, A1111st rong BK, Heenan PJ, Blackwell JB, Cumming FJ,
Engl ish DR, Holl and S, Kelsall OR, Matz LR. Rouse IL, Singh A,
Ten Seldam REJ et al. The causes of malignant melanoma: results
from the West Austral ian Lions Melanoma Research Project. Recent
Results Cancer Res 1986;102:18- 37.
57. Holly EA, Kelly JW, Shpall SN. Chiu SH. Number of melanocytic
nevi as a major risk factor for mal ignant melanoma. J Am Acad Der-
matol 1987; 17:459-68.
58. Swerdlow AJ, English JS, MacKie RM, O' Doherty CJ, Hunter JA,
Clark J, Hole DJ. Fluorescent lights, ultraviolet lamps, and risk of cu-
taneous melanoma. Br Med J 1988;297:647-50.
59. Osterlind A. Tucker MA, Stone BJ, Jensen OM. The Danish case-con-
trol study of cutaneous malignant melanoma. II. Importance of UV-
light exposure. lnt J Cancer 1988;42:319-24.
60. Zanetti R. Rosso S, Faggiano F. Roffino R, Colorma S, Martina G. A
case-control study on cutaneous mal ignant melanoma in the province
of Torino, Italy. Rev Epidemiol Sante Publique 1988;36:309- 17 [in
French].
61. MacKie RM, Freudenberger T, Aitchison TC. Personal risk-factor
chart for cutaneous melanoma. Lancet 1989;2:487- 90.
62. Beitner H. Norell SE, Ringborg U. Wennersten G, Mattson B. Malig-
nant melanoma: aetiological importance of individual pigmentation
and sun exposure. Br J Dermatoll990;122:43- 51.
63. Walter SD, Marrett LD, From L, Hertzman C, Shannon HS, Roy P.
The association of cutaneous malignant melanoma with the use of
sunbeds and sunlamps. Am J Epidemioll990;131:232-43.
64. Walter SD, King WD, Marrett LD. Association of cutaneous malig-
nant melanoma with intermittent exposure to ultraviolet radiation: re-
sults of a case-control study in Ontario, Canada. lnt J Epidemiol 1999;
28:418- 27.
65. DUim-Lane J. Herity B, Moriarty MJ, Conroy R. A case control study
of malignant melanoma. lr Med J 1993;86:57-59. Erratum in: lr Med
J 1993;86: 135.
66. Garbe C, Weiss J, Kruger S, Garbe E, Buttner P, Bertz J, Hoffmeister
H, Guggenmoos-Holzmann I, Jung EO, Orfanos CE. The German
melanoma regist ry and environmental risk factors implied. Recent
Results Cancer Res 1993; 128:69- 89.
67. Autier P. Dore JF, Lejeune P, Koelmel KF, Geffe ler 0 , Hille P, Cesar-
ini JP, Lienard D, Liabeuf A, Joarlette M, ChemaJy P, Hakim K, ct al.,
for the EORTC Melanoma Cooperative Group. Cutaneous malignant
melanoma and exposure to sunlamps or sunbeds: for the EORTC mul-
ticenter case-control study in Belgium, France and Germany. lnt J
Cancer 1994;58:809-13.
68. Holly EA, Aston DA, Cress RD, Ahn DK, Kristiansen JJ. Cutaneous mel-
anoma in women. T. Exposure to sunlight, ability to tan, and other risk
factors related to ultraviolet light. Am J Epidemiol 1995; 141 : 923- 33.
69. Chen YT. Dubrow R, Zheng T. Barnhill RL, Fine J, Berwick M. Sun-
lamp use and the risk of cutaneous malignant melanoma: a popula-
tion-based case-control study in Connecticut, USA. Int J Epidemiol
1998;27:758-65.
70. Westerdabl J, Olsson H. Masback A, Jngvar C. Jonsson N, Brandt L,
Jonsson PE, Moller T. Use of sunbeds or sunlamps and malignant
melanoma in southem Sweden. Am J Epidemiol 1994; 140:691- 9.
71. Naldi L, Gall us S, Imberti GL, Cainelli T, Negri E, LaVecchia C, on
behalf of the Italian Group for Epidemiological Research in Derma-
tology. Sunlamps and sunbeds and the risk of cutaneous melanoma.
Eur J Cancer Prev 2000;9: 133-4.
72. Bataille V, Winnett A, Sasieni P, Newton-Bishop JA, Cuzick J. ExpO-
sure to the sun and sunbeds and the risk of cutaneous melanoma in the
UK: a case-control study. EUI J Cancer 2004;40:429-35.
73. Westerdahl J, Ingvar C, Masback A, Jonsson N, Olsson H. Risk of cu-
taneous mali gnant melanoma in relation to use of sunbeds: fu rther
evidence for UV-A carcinogenicity. Br J Cancer 2000;82: 1593- 9.
74. Boyd AS, Shyr Y, King LE Jr. Basal cell carcinoma in young women:
an evaluation of the association of tanning bed use and smoking.
JAm Acad Dermatol
75. O'Loughlin G, Moriarty MJ, Herity B, DaJy L. A re-appraisal of risk
factors for skin carcinoma in Ireland. A case-control study. Ir J Med
Sci 1985; 154:61-5.
76. Heri ty B. O'Loughlin G, Moriarty MJ. Conroy R. Risk factors for
non-melanoma skin cancer. J lr Med 1989;82: 151-2.
77. Hogan DJ, ToT. Wilson ER, Miller AB, Robson D, Holfeld K, Lane
P. A study of acne treatments as risk factors for skin cancer of the
head and neck. Br J Dermatol l991;125:343-8.
78. Aubry F. MacGibbon B. Risk factors of squamous cell carcinoma of
the skin. A case-control study in the Montreal region. Cancer 1985;
55:907-11.
79. Bajdik CD, Gallagher RP, Astrakianakis G, Hill GB, Fincham S,
McLean Dl. Non-solar ultraviolet radiation and the risk of basal and
squamous cell skin cancer. Br J Cancer 1996;73: 1612-4.
80. Corona R, Dogliotti E, D'Errico M, Sera F. Iavarone I, Baliva G,
Chinni LM, Gobello T, Mazzanti C, Puddu P, Pasquini P. Risk factors
for basal cell carcinoma in a Mediterranean population: role of recrea-
tional sun exposure early in li fe. Arch Dermatol 2001; 137: 1162-8.
81. Karagas MR, Stannard VA, Mott LA, Slattery MJ, Spencer SK, Wein-
stock MA. Use of tanning devices and risk of basal cell and squamous
cell skin cancers. J Nat! Cancer Inst 2002;94:224-6.
82. Walther U, Kron M, Sander S, Sebastian G, Sander R, Peter RU,
Meurer M, Krdllll G, Kaskel P. Risk and protective factors for spo-
radic basal cell carcinoma: results of a two-eentre case-control study
in southern Germany. Clinical actinic elastosis may be a protective
factor. Br J Dennatol2004;151:170-8.
83. Veier0d MB, Weiderpasss E, Lund E, Armstrong BK, Adami HO.
Re: A retrospective study of pigmentation, sun exposure, and risk of
cutaneous malignant melanoma in women. J Nat! Cancer lnst
2004;96:337-S.
84. Autier P, Dare JF, for EPIMEL and EORTC Melanoma Cooperative
Group. Influence of sun exposures during childhood and during adult-
hood on melanoma risk. Int .I Cancer 1998;77:533- 7.
85. Whiteman DC, Whiteman CA. Green A. Childhood sun exposure as a
risk factOr for melanoma: a systematic review of epidemiologic stud-
ies. Cancer Causes Control 200 l ; 12:69-82.
86. Gallagher RP, Spinell i JJ, Lee TK. Tanning beds, sunlamps, and risk
of cutaneous mal ignant melanoma. Cancer Epidemiol Biomarkers
Prev 2005; 14:562-6.
News I
Special Report: Policy
A review of human carcinogens-Part D: radiation
In June 2009, 20 scientists from nine
countries met at the International
Agency for Research on Cancer (IARC)
to reassess the carcinogenicity of the
types of radiation previously classified
as "carcinogenic to humans" (Group 1)
and to ident ify additional tumour sites
and mechani sms of carcinogenesis
(table and panel). These assessments
wil l be published as part D of Volume
100 ofthe I ARC Monographs.'
Alpha particles, consisting of two
protons and two neutrons, are a
densely ionising type of radiat ion
wi th low capaci ty to penet rate living
tissue (less than 0-1 mm). Beta
particles are electrons or positrons
that are less ionising, but more
penetrating (up to a few milimetres).
The health hazards resulting from
radionuclides that emit these
particles largely occur after internal
deposition. Epidemiological evidence
shows a number of radionuclides that
emit alpha or beta particles increase
cancer risks at several anatomical sites
(table). The Worki ng Group reaffirmed
the carcinogenicity of internally
deposited radionuclides that emit
alpha or beta particles (Group 1).
Radiation type
Alpha-particle and beta-particle emitters
After the Chernobyl accident. a
sharp increase in the risk of thyroid
cancer was found with exposure to
radioiodines, particularly iodine-131,
during chi ldhood and adolescence.u
This increased risk might be due to
higher milk intake per unit of body
weight among chi ldren; a higher thyroid
dose per unit of iodine-131 intake from
milk; a higher susceptibility per unit of
thyroid dose; or a combination of these.
Radon exposure occurs mainly
through contamination of indoor
air by radon released from soil and
building materials. Combined analyses
of case-control studies now estimate
that residential exposure to radon gas
is the leading cause of lung cancer after
tobacco smoke (8-15% attributable
risk in Europe and North America).-
5
X-rays and gamma-rays are
sparsely ion1smg electromagnetic
radiation that penetrate living tissue,
typical ly producing fast electrons that
deposit energy, resulting in tissue
damage. Extensive study of atomic-
bomb survivors shows increased
cancer risks at multiple anatomical
sites.
6
Current evidence adds to the
list of tumours caused by x-rays
Major study populations
and gamma-rays (table), and also
establishes that in-utero exposure
increases the risk of cancer at multiple
sites.
7
8
The Working Group reaffirmed
the carcinogenicity of x-radiation and
gamma-radiation (Group 1).
Neutrons are produced by nucl ear
reactions and are a main component
of cosmic radiation. They are highly
penetrating and interact with
the t raversed tissue, producing
protons, other charged part icles, and
gamma-radiation. Epidemiological
evidence is inadequate to assess the
carcinogenicity of neutrons, because
of co-exposures to other types of
radiation. However, the evidence
of cancer in experimental animals
is sufficient, and mechanistic data
show that neutrons t ransfer their
energy in clusters of ionising events-
resulting in similar, but more severe,
local damage than that induced by
x-rays or gamma-rays. On the basis
of this evidence, the Working Group
reaffirmed the carcinogenicity of
neutron radiation (Group 1).
Each type of ionising radiation
(panel) transfers energy in the
form of highly structured tracks of
Upcoming meetings
Sept 29-0ct 6, 2009
Lifestyle Factors
Oct 20-27, 2009
Chemical Agents and Related
Occupations
http://monographs.iarc.fr/
Tumour sites (and types) on which sufficient evidence is based
Radon-222 and decay products General population (residential exposure), underground miners Lung
Bone Radium-224 and decay products Medical patients
Radium-226, radium-228, and decay products Radium-dial painters
Thorium-232 and decay products Medical patients
Plutonium Plutonium-production workers
Phosphorus-32 Medical patients
Fission products, includi ng strontium-90 General population, following nuclear reactor accident
Radioiodines, including iodine-131 Children and adolescents, following nuclear reactor accident
X-radiation or gamma-radiation Atomic-bomb suNivors, medical patients; in-utem exposure (offspring
of pregnant medical patients and of atomic-bomb suNivors)
Solar radiation General population
UV-emitting tanning devices General population
CUachronic lymphocytic leukaemia. BCCwbasaltell carcinoma. SCC squamouscell carcinoma.
Table: Radiation exposureswithsufficientevidence in humans
www.thelancet.com/oncology Vol tO August 2009
Bone, paranasal sinus and mastoid process (radium-226 only)
Liver, extrahepatic bile ducts, gall bladder, leukaemia (excluding Cll)
Lung, liver, bone
Acute leukaemia
Solid cancers, leukaemia
Thyroid
Salivary gland, oesophagus, stomach, colon, lung, bone, ski n (BCC),
female breast, urinary bladder, brain and CNS, leukaemia (excluding CU),
thyroid, kidney (atomic-bomb SUIVi vors, medical patients); multiple sites
(in-ute<o exposure)
Skin (BCC. SCC. melanoma)
Skin (melanoma), eye (melanoma, particularly choroid and ciliary body)
751
I News
Monograph Working Group
Members
B Armstrong-Co-Chair
(Australia), E (ardis-Co-Chair
(Spain); A Green (Australia);
D Krewski. R Mitchel. N Priest
(Canada); L Tomasek (Czech
Republic); K Baverstock (finland);
J-F Dore, J Hall, L Saba tier
(France); M Sokolnikov (Russian
federation); M Hill, M Little,
M Marshall, C Muirhead,
A Riddell (UK); D Brenner [unable
to attend). R Guilmette, D Hoel.
D Richardson, R Ullrich (USA)
of interest
NPworks for, and RM is a
consultant to. Atomic Energy of
Canada ltd. CM receives funding
from the UK Ministry of Defence.
JH receives funding from
Ele<tricite de France. AG receives
funding from l'Oreal Recherche.
752
Invited Specialists
None
Panel: Types of radiation classified in
Groupl
Ionising radiation
Alpha-particle emi tters
Beta-particle emi tters
X-rays and gamma-rays
Neutron radiation
Solar radiation
Ultraviolet radiation (wavelengths
lOG-400 nm, encompassing UVA,
UVB, and UVC)
ionisation and excitation events that
can produce a variety of molecular
lesions and clustered, complex DNA
damage
9
Subsequent processing of
this damage induces many responses
(eg, cell killing, chromosomal
aberrations, mutations, genomic
instability, cell transformation, and
bystander effects) that contribute
to carcinogenesis. Based on these
mechanistic considerations, all types
of ionising radiation were classified by
the Working Group as "carcinogenic to
humans" (Group 1).
Solar radiation is the main source of
human exposure to ultraviolet (UV)
radiation, which is further subdivided
into UVA, UVB, and UVC. The
ultraviolet component that reaches the
earth's surface comprises around 95%
UVA and 5% UVB; UVC is blocked by
stratospheric ozone. Epidemiological
studies have established a causal
association between exposure to solar
radiation and all major types of skin
cancer (table). The Working Group
reaffirmed the carcinogenicity of solar
radiation (Group 1).
Exposure to solar radiation causes
a specific mutation fingerprint
(cytidine to thymidine transition),
as a result of cyclobutane pyrimidine
dimers in DNA. This pattern had long
been att ri buted to UVB.'
0
However,
this same cytidine to thymidine
transition has been detected in
the ski n of UVA-treated mice" and
in the Tp53 gene of UVA-induced
or UVB-induced skin tumours in
hairless mice.
10
In humans, this
transition has been seen in TPS3 in
premalignant solar keratosis and in
malignant skin tumours.'' Based on
t hese mechanistic data, the Working
Group classified UV radiation as
"carcinogenic to humans" (Group 1).
The use of UV-emitting tanning
devices is widespread in many
developedcountries,especiallyamong
young women. A comprehensive
meta-analysis concluded that the risk
of cutaneous melanoma is increased
by 75% when use of tanning devices
starts before 30 years of age.
13
Additionally, several case-control
studies provide consistent evidence
of a positive association between
the use of UV-emitting tanning
devices and ocular melanoma."'
5
Therefore, the Working Group raised
the classification of the use of UV-
emitting tanning devices to Group 1,
"carcinogenic to humans".
While reviewing the studies of
occupational UV exposure, the
Working Group concluded that there
is "sufficient evidence" for ocular
melanoma in welders. '
6
'
7
However,
because welders are also exposed to
other harmful agents, this association
could not be attri buted specifically
to UV radiation. A full review of the
carcinogenic hazards of welding will
be undertaken by IARC with high
pri ority.
Fatiha El Ghissassi, Robert Baan, Kurt
Straif, Yann Grosse, Beatrice
Secretan, Wronique Bouvard, Lamia
Benbrahim-Tallaa, Nee/a Guha,
Crystal Freeman, Laurent Galichet,
Vincent Cogliano, on behalf of the
WHO International Agency for
Research on Cancer Monograph
Working Group
International Agency for Research on
Cancer, Lyon, France
The IARC authors declared no conflicts of i nterest.
Grosse Y, Baan R. Straif K, et al. A review of
human carcinogens-part A: pharmaceuticals.
Lancer Oncol2009; 10: 13-14.
UN Chernobyl Forum expert group "Health"
(EGH). Health effects of the Chernobyl
accident and special health care programmes.
Geneva; 2006. http://whqlibdocwho.int/publi
cations/2006/924l594179_eng.pdf.
3 Card is E, Howe G, Ron E, et al. Cancer
consequences of the Chernobyl accident:
20years on.) Radiol Prot 2006; 26: 127-40.
4 Darby S, Hill D. Auvinen A. et al. Radon in
homes and risk of lung cancer: collaborative
analysis of individual data from 13 European
case-control studies. BM) 2005; 330: 223.
National Research Council, Committee to
Assess Health Risks from Exposure to low
Levels of Ionizing Radiation, Board on
Radiation Effects, and Research Division on
Earth and Life Studies. Health effects of
exposure to radon; BEIRVI. Washington:
National Academies Press; 1999.
6 National Research Council, Committee to
Assess Health Risks from Exposure to l ow
Earth and Life Studies. Health risks from
exposure to low levels of ionizi ng radiation:
BEIRVII, Phase 2. Washington: National
Academies Press; 2006.
7 Wakeford R, Li ttle MP. Ri sk coefficients for
childhood cancer after int rauteri ne i rradiation;
a review. fnt) Radial Biol 2003; 79: 293- 309.
8 Preston Dl . Cullings H, Suyama A, et al. Solid
cancer incidence i n atomic bomb survivors
exposed i n utero or as young children.
) Natl Cancer fnst 2008; 100: 428- 36.
9 Good head DT. Initial events i n the cellular
effects of ionizing radiations: clustered
damage i n DNA. fnt) Radiat Biol1994;
65:7-17.
10 Rungcr TM, Kappes UP. Mechanisms of
mutation formation with long-wave
ultraviolet light (UVA). Photodermatol
Photoimmunol Photomed 2008; 24: 2-10.
11 lkehata H, Kawai K, Komuraj, etal. UVA1
genotoxicity is mediated not by oxidative
damage but by cyclobutane pyrimidine dimers
i n normal mouse ski n.) Invest Dermotol2008;
128: 2289-96.
12 Agar NS, Hall iday GM, Barnetson RS,
Ananthaswamy HN, Wheeler M, jones AM.
The basal layer in human squamous tumors
harbors more UVA than UVB fingerpri nt
mutations: a role for UVA i n human skin
carci nogenesis. Proc Natl Acod Sci USA 2004;
101: 4954-59
13 !ARC Working Group. The association of use of
sunbeds with cutaneous malignant melanoma
and other skin cancers: a systematic review.
lnt) Cancer 2006; 120: 1116-22.
14 Seddon JM, Gragoudas ES. Glynn Rj , Egan KM,
Albert DM, Blitzer PH. Host factors, UV
radiation, and risk of uveal melanoma: a
casecontrol study. Arch Ophthalmol1990;
108: 1274-80.
15 Vajdic CM, Kri cker A, Giblin M, et al. Artificial
ultraviolet radiation and ocular melanoma in
Australia. In!) Cancer 2004; 112: 896-900.
16 Lutz )M, Cree I, Sabroe S, et al. Occupational
risks for uveal melanoma results from a case
control study in nine European countries.
CancerCausesControi200S; 16:437-47.
17 Shah CP, WeisE. Lajous M. Shields )A.
Shields CL. Intermittent and chronic ultraviolet
light exposure and uveal melanoma: a meta-
analysis. Ophthalmology 2005; 112: 1599-607.
www.t helancet.com/oncology Vol10 August 2009
News I
Special Report: Policy
A review of human carcinogens-Part D: radiation
In June 2009, 20 scientists from nine
countries met at the International
Agency for Research on Cancer (IARC)
to reassess the carcinogenicity of the
types of radiation previously classified
as "carcinogenic to humans" (Group 1)
and to ident ify additional tumour sites
and mechani sms of carcinogenesis
(table and panel). These assessments
wil l be published as part D of Volume
100 ofthe I ARC Monographs.'
Alpha particles, consisting of two
protons and two neutrons, are a
densely ionising type of radiat ion
wi th low capaci ty to penet rate living
tissue (less than 0-1 mm). Beta
particles are electrons or positrons
that are less ionising, but more
penetrating (up to a few milimetres).
The health hazards resulting from
radionuclides that emit these
particles largely occur after internal
deposition. Epidemiological evidence
shows a number of radionuclides that
emit alpha or beta particles increase
cancer risks at several anatomical sites
(table). The Worki ng Group reaffirmed
the carcinogenicity of internally
deposited radionuclides that emit
alpha or beta particles (Group 1).
Radiation type
Alpha-particle and beta-particle emitters
After the Chernobyl accident. a
sharp increase in the risk of thyroid
cancer was found with exposure to
radioiodines, particularly iodine-131,
during chi ldhood and adolescence.u
This increased risk might be due to
higher milk intake per unit of body
weight among chi ldren; a higher thyroid
dose per unit of iodine-131 intake from
milk; a higher susceptibility per unit of
thyroid dose; or a combination of these.
Radon exposure occurs mainly
through contamination of indoor
air by radon released from soil and
building materials. Combined analyses
of case-control studies now estimate
that residential exposure to radon gas
is the leading cause of lung cancer after
tobacco smoke (8-15% attributable
risk in Europe and North America).-
5
X-rays and gamma-rays are
sparsely ion1smg electromagnetic
radiation that penetrate living tissue,
typical ly producing fast electrons that
deposit energy, resulting in tissue
damage. Extensive study of atomic-
bomb survivors shows increased
cancer risks at multiple anatomical
sites.
6
Current evidence adds to the
list of tumours caused by x-rays
Major study populations
and gamma-rays (table), and also
establishes that in-utero exposure
increases the risk of cancer at multiple
sites.
7
8
The Working Group reaffirmed
the carcinogenicity of x-radiation and
gamma-radiation (Group 1).
Neutrons are produced by nucl ear
reactions and are a main component
of cosmic radiation. They are highly
penetrating and interact with
the t raversed tissue, producing
protons, other charged part icles, and
gamma-radiation. Epidemiological
evidence is inadequate to assess the
carcinogenicity of neutrons, because
of co-exposures to other types of
radiation. However, the evidence
of cancer in experimental animals
is sufficient, and mechanistic data
show that neutrons t ransfer their
energy in clusters of ionising events-
resulting in similar, but more severe,
local damage than that induced by
x-rays or gamma-rays. On the basis
of this evidence, the Working Group
reaffirmed the carcinogenicity of
neutron radiation (Group 1).
Each type of ionising radiation
(panel) transfers energy in the
form of highly structured tracks of
Upcoming meetings
Sept 29-0ct 6, 2009
Lifestyle Factors
Oct 20-27, 2009
Chemical Agents and Related
Occupations
http://monographs.iarc.fr/
Tumour sites (and types) on which sufficient evidence is based
Radon-222 and decay products General population (residential exposure), underground miners Lung
Bone Radium-224 and decay products Medical patients
Radium-226, radium-228, and decay products Radium-dial painters
Thorium-232 and decay products Medical patients
Plutonium Plutonium-production workers
Phosphorus-32 Medical patients
Fission products, includi ng strontium-90 General population, following nuclear reactor accident
Radioiodines, including iodine-131 Children and adolescents, following nuclear reactor accident
X-radiation or gamma-radiation Atomic-bomb suNivors, medical patients; in-utem exposure (offspring
of pregnant medical patients and of atomic-bomb suNivors)
Solar radiation General population
UV-emitting tanning devices General population
CUachronic lymphocytic leukaemia. BCCwbasaltell carcinoma. SCC squamouscell carcinoma.
Table: Radiation exposureswithsufficientevidence in humans
www.thelancet.com/oncology Vol tO August 2009
Bone, paranasal sinus and mastoid process (radium-226 only)
Liver, extrahepatic bile ducts, gall bladder, leukaemia (excluding Cll)
Lung, liver, bone
Acute leukaemia
Solid cancers, leukaemia
Thyroid
Salivary gland, oesophagus, stomach, colon, lung, bone, ski n (BCC),
female breast, urinary bladder, brain and CNS, leukaemia (excluding CU),
thyroid, kidney (atomic-bomb SUIVi vors, medical patients); multiple sites
(in-ute<o exposure)
Skin (BCC. SCC. melanoma)
Skin (melanoma), eye (melanoma, particularly choroid and ciliary body)
751
I News
Monograph Working Group
Members
B Armstrong-Co-Chair
(Australia), E (ardis-Co-Chair
(Spain); A Green (Australia);
D Krewski. R Mitchel. N Priest
(Canada); L Tomasek (Czech
Republic); K Baverstock (finland);
J-F Dore, J Hall, L Saba tier
(France); M Sokolnikov (Russian
federation); M Hill, M Little,
M Marshall, C Muirhead,
A Riddell (UK); D Brenner [unable
to attend). R Guilmette, D Hoel.
D Richardson, R Ullrich (USA)
of interest
NPworks for, and RM is a
consultant to. Atomic Energy of
Canada ltd. CM receives funding
from the UK Ministry of Defence.
JH receives funding from
Ele<tricite de France. AG receives
funding from l'Oreal Recherche.
752
Invited Specialists
None
Panel: Types of radiation classified in
Groupl
Ionising radiation
Alpha-particle emi tters
Beta-particle emi tters
X-rays and gamma-rays
Neutron radiation
Solar radiation
Ultraviolet radiation (wavelengths
lOG-400 nm, encompassing UVA,
UVB, and UVC)
ionisation and excitation events that
can produce a variety of molecular
lesions and clustered, complex DNA
damage
9
Subsequent processing of
this damage induces many responses
(eg, cell killing, chromosomal
aberrations, mutations, genomic
instability, cell transformation, and
bystander effects) that contribute
to carcinogenesis. Based on these
mechanistic considerations, all types
of ionising radiation were classified by
the Working Group as "carcinogenic to
humans" (Group 1).
Solar radiation is the main source of
human exposure to ultraviolet (UV)
radiation, which is further subdivided
into UVA, UVB, and UVC. The
ultraviolet component that reaches the
earth's surface comprises around 95%
UVA and 5% UVB; UVC is blocked by
stratospheric ozone. Epidemiological
studies have established a causal
association between exposure to solar
radiation and all major types of skin
cancer (table). The Working Group
reaffirmed the carcinogenicity of solar
radiation (Group 1).
Exposure to solar radiation causes
a specific mutation fingerprint
(cytidine to thymidine transition),
as a result of cyclobutane pyrimidine
dimers in DNA. This pattern had long
been att ri buted to UVB.'
0
However,
this same cytidine to thymidine
transition has been detected in
the ski n of UVA-treated mice" and
in the Tp53 gene of UVA-induced
or UVB-induced skin tumours in
hairless mice.
10
In humans, this
transition has been seen in TPS3 in
premalignant solar keratosis and in
malignant skin tumours.'' Based on
t hese mechanistic data, the Working
Group classified UV radiation as
"carcinogenic to humans" (Group 1).
The use of UV-emitting tanning
devices is widespread in many
developedcountries,especiallyamong
young women. A comprehensive
meta-analysis concluded that the risk
of cutaneous melanoma is increased
by 75% when use of tanning devices
starts before 30 years of age.
13
Additionally, several case-control
studies provide consistent evidence
of a positive association between
the use of UV-emitting tanning
devices and ocular melanoma."'
5
Therefore, the Working Group raised
the classification of the use of UV-
emitting tanning devices to Group 1,
"carcinogenic to humans".
While reviewing the studies of
occupational UV exposure, the
Working Group concluded that there
is "sufficient evidence" for ocular
melanoma in welders. '
6
'
7
However,
because welders are also exposed to
other harmful agents, this association
could not be attri buted specifically
to UV radiation. A full review of the
carcinogenic hazards of welding will
be undertaken by IARC with high
pri ority.
Fatiha El Ghissassi, Robert Baan, Kurt
Straif, Yann Grosse, Beatrice
Secretan, Wronique Bouvard, Lamia
Benbrahim-Tallaa, Nee/a Guha,
Crystal Freeman, Laurent Galichet,
Vincent Cogliano, on behalf of the
WHO International Agency for
Research on Cancer Monograph
Working Group
International Agency for Research on
Cancer, Lyon, France
The IARC authors declared no conflicts of i nterest.
Grosse Y, Baan R. Straif K, et al. A review of
human carcinogens-part A: pharmaceuticals.
Lancer Oncol2009; 10: 13-14.
UN Chernobyl Forum expert group "Health"
(EGH). Health effects of the Chernobyl
accident and special health care programmes.
Geneva; 2006. http://whqlibdocwho.int/publi
cations/2006/924l594179_eng.pdf.
3 Card is E, Howe G, Ron E, et al. Cancer
consequences of the Chernobyl accident:
20years on.) Radiol Prot 2006; 26: 127-40.
4 Darby S, Hill D. Auvinen A. et al. Radon in
homes and risk of lung cancer: collaborative
analysis of individual data from 13 European
case-control studies. BM) 2005; 330: 223.
National Research Council, Committee to
Assess Health Risks from Exposure to low
Earth and Life Studies. Health effects of
exposure to radon; BEIRVI. Washington:
National Academies Press; 1999.
6 National Research Council, Committee to
Assess Health Risks from Exposure to l ow
Earth and Life Studies. Health risks from
exposure to low levels of ionizi ng radiation:
BEIRVII, Phase 2. Washington: National
Academies Press; 2006.
7 Wakeford R, Li ttle MP. Ri sk coefficients for
childhood cancer after int rauteri ne i rradiation;
a review. fnt) Radial Biol 2003; 79: 293- 309.
8 Preston Dl . Cullings H, Suyama A, et al. Solid
cancer incidence i n atomic bomb survivors
exposed i n utero or as young children.
) Natl Cancer fnst 2008; 100: 428- 36.
9 Good head DT. Initial events i n the cellular
effects of ionizing radiations: clustered
damage i n DNA. fnt) Radiat Biol1994;
65:7-17.
10 Rungcr TM, Kappes UP. Mechanisms of
mutation formation with long-wave
ultraviolet light (UVA). Photodermatol
Photoimmunol Photomed 2008; 24: 2-10.
11 lkehata H, Kawai K, Komuraj, etal. UVA1
genotoxicity is mediated not by oxidative
damage but by cyclobutane pyrimidine dimers
i n normal mouse ski n.) Invest Dermotol2008;
128: 2289-96.
12 Agar NS, Hall iday GM, Barnetson RS,
Ananthaswamy HN, Wheeler M, jones AM.
The basal layer in human squamous tumors
harbors more UVA than UVB fingerpri nt
mutations: a role for UVA i n human skin
carci nogenesis. Proc Natl Acod Sci USA 2004;
101: 4954-59
13 !ARC Working Group. The association of use of
sunbeds with cutaneous malignant melanoma
and other skin cancers: a systematic review.
lnt) Cancer 2006; 120: 1116-22.
14 Seddon JM, Gragoudas ES. Glynn Rj , Egan KM,
Albert DM, Blitzer PH. Host factors, UV
radiation, and risk of uveal melanoma: a
casecontrol study. Arch Ophthalmol1990;
108: 1274-80.
15 Vajdic CM, Kri cker A, Giblin M, et al. Artificial
ultraviolet radiation and ocular melanoma in
Australia. In!) Cancer 2004; 112: 896-900.
16 Lutz )M, Cree I, Sabroe S, et al. Occupational
risks for uveal melanoma results from a case
control study in nine European countries.
CancerCausesControi200S; 16:437-47.
17 Shah CP, WeisE. Lajous M. Shields )A.
Shields CL. Intermittent and chronic ultraviolet
light exposure and uveal melanoma: a meta-
analysis. Ophthalmology 2005; 112: 1599-607.
www.t helancet.com/oncology Vol10 August 2009
Press Release N 171 Page 1 of 3
International Agency for Research on Cancer
PRESS RELEASE
N171
WHO 29 Novembre 2006
Sunbed use in youth unequivocally associated with skin cancer
In October 2004, the French Mini stry of Healt h contacted Dr Peter Boyle, the Director of the
rai sing a particular concern about the increase in incidence of melanoma ir
Fr ance and in the world. Melanoma is the most deadly form of skin cancer. For reference, while
Europe registers 10- 15 cases per 100,000 population and per year, Australia has an incidence
of 50 cases/100,000 population ever y year. Melanoma is on the increase everywhere in the
world and the number of cases doubl es every 12-15 years in the most affected areas.
In 1992, an IARC Working Gr oup made a thorough review of the available scientific data anc
concluded that solar radiation is carcinogenic to humans (Group 1)(2). Solar radiat ion
cutaneous melanoma and non-mel anoma skin cancer. However, there is stil l
surrounding the role of exposure to artificial sources of UV radiation, i.e. from the use ol
sunbeds or sunlamps in t anning parlours, in t he aet iology of skin cancer.
IARC Review of the literature
To respond to these concerns, IARC convened an international Working GroupC
3
) that assessee
the availabl e evidence relating to health effects, both positive and detrimental, of exposure tc
arti f icial UV r adiation through t he use of indoor tanning facilities, in particular whether thei1
use increases t he ri sk for skin cancer.
Assessment
Epidemiologic studies do not provide consistent evidence that use of indoor tanning faci lities ir
general is associated with the development of melanoma or non- melanoma skin cancer<
4
).
There are various technical explanations for this conclusion. Firstly, knowledge of levels of U'v
exposure during indoor t anning is ver y imprecise. Furthermore, early studies published had lo'll
power to detect long-term associations with artif icial UV exposure that become evident
following a prolonged lag period. "Consideri ng simultaneously all the avai lable data", Dr Boyle
said, "made it possible for us to reach a number of clear conclusions."
Conclusions
1. Clear increase in melanoma risk associated with use of sun beds in teens and twenties
The data showed a prominent and consist ent increase in risk for melanoma in people who firs1
used sunbeds in t heir t wenti es or t een year s: a 75% increase in risk of melanoma
calculated for such users of artificial tanning appliances, whi le this increase in the genera
population, although not statistically significant, is still not negligible.
2. Increase in risk of squamous cell cancer (SCC) of the skin associated with use of sunbeds ir
teens
Li mited data suggest that the ri sk of squamous cell carcinoma is similarly increased after f i rs1
use as a teenager.
3. Immune system affected
http://www .iarc.fr/ENG/Press_Releases/pr 171 a.html 4/17/2008
Press Release N 171 Page 2 of3
Data also suggest detrimental effects from use of sunbeds on the skin's immune response and
possibly on the eyes (ocular melanoma).
4. No positive health effects
Artificial tanning confers little if any protection against solar damage to the skin, nor does use
of indoor tanning facil ities grant protection against vitamin D deficiency. Data also suggest
detrimental effects from use of indoor tanning facilities on the skin's immune response and
possibly on the eyes (ocular melanoma).
Public Health message
Dr Boyle concluded that "while !ARC's mandate is one of scientific expertise and assessment of
epidemiologic risk, in view of the strength and seriousness of the findings, effective action to
restrict access to artificial tanning facilities (solariums, tanning salons, tanning parlours) to
minors and young adults should be strongly considered."
Notes:
(1) The International Agency for Research on Cancer (IARC) is part of the World Health
Organization. Its mission is to coordinate and conduct research on the causes of human
cancer, the mechanisms of carcinogenesis, and to develop scientific strategies for cancer
control. The mandate of the world cancer research agency is to coordinate international
research to take advantage of synergies and disseminate scientific information through
publications, meetings, courses, and fellowships.
(2) UV radiations A, B and C were all three independently rated as probably carcinogenic to
humans by the IARC Working Group in 1992. In addition, they concluded that the "use of
sunlamps and sunbeds entails exposures that are probably carcinogenic to humans (Group
2A)" . While natural UV radiation does indeed stimulate the production of indispensable vitamin
D, exposure to natural sun light should be moderate, and in any case artificial UV radiation
exposure, as provided by sunbed exposure, should be avoided at all cost.
(3) The members of the Working Group were: Dr Philippe Autier
1
, Dr Mathieu Boniol
1
, Dr Peter
Boyle
1
, Dr Jean-Francais Dore
2
, Dr Sara Gandini
3
, Dr Adele Green (Chair)
4
, Dr Julia Newton-
Bishop5, Dr Martin A. Weinstock
6
, Dr Johan Westerdah1
7
, Dr Beatrice Secretan (Coordinator)
1
,
Dr Stephen Walter
8
. Their conclusions are being published online by the I nternatlonaLJo_urnal
of Cancer
1
IARC -
2
INSERM U590, Centre Leon Berard, Lyon, France -
3
European Institute of Oncology,
Milan, Italy -
4
Queensland Institute of Medical Research, Australia -
5
Cancer Research UK -
6
Brown University Medical School - USA -
7
Lund University Hospital - Sweden [unable to
attend] -
8
McMaster University - Ontario, Canada
( 4) There are three main types of cancers of the skin: Basal cell carcinoma (BCC) is the
most common form of cancer and the least aggressive. Basal cells are cells that line the
deepest layer of the epidermis. A tumor of this layer is known as basal cell carcinoma. The sun
is responsible for over 90 percent of all skin cancers, including BCC, and chronic overexposure
to sunlight is the cause for most cases of basal cell carcinoma. BCCs occur most frequently on
the face, ears, neck, scalp, shoulders, and back. Squamous cell cancer (SCC) is a malignant
tumor. It is more aggressive than basal cell cancer, and is more likely than basal cell cancer to
metastasize. Melanoma is the most dangerous type of skin cancer. It involves the cells that
produce pigment (melanin), which is responsible for skin and hair color. The development of
melanoma is related to sun exposure, particularly to sunburns during childhood, and is most
common among people with fair skin, blue or green eyes, and red or blond hair.
Press Release N 171
World Health Organization
Page 3 of3
Organisation mondiale de Ia Sante
Centre international de Recherche sur le Cancer
150, cours Albert-Thomas 69372 lyon Cedex 08 (France)
Telephone: 33 472 738 485 Facsimile: 33 472 738 311 http:/ /www.iarc.fr
----------ICNIRP Statement ----------
HEALTH ISSUES OF ULTRA VIOLET TANNING APPLIANCES
USED FOR COSMETIC PURPOSES
-----------ICNIRP Statement-----------
HEALTH ISSUES OF ULTRAVIOLET TANNING APPLIANCES
USED FOR COSMETIC PURPOSES
The International Commission on Non-Ionizing Radiation Protection*
BACKGROUND
Consumer demand for a cosmetic tan is the eco-
nomic basis of the "suntanning industry," which devel-
ops and distributes equipment for commercial suntanning
and markets suntanning services to consumers.
Suntanning is caused by ultraviolet radiation
(UVR).t Exposure from sunbeds and other tanning ap-
pliances has the same potential risks as exposure to the
UVR in solar radiation. The term "sunbed" is frequently
used to describe all tanning appliances consisting of
either a single UVR-emitting lamp (emitting UV A and/or
UVB radiation) as in some facial tanners or a number of
such lamps incorporated into a bed, canopy, panel, or any
combination thereof.
Potential adverse health effects of exposure to UVR
are well documented and reasonably well quantified
(WHO 1994). The purpose of this document is to
summarize the potential adverse effects of exposure to
ultraviolet radiation from tanning appliances and to
provide recommendations to minimize the risks of such
effects. Recommendations in this document apply o ly to
the use of sunbeds for cosmetic purposes.
TANNING APPLIANCES: TYPES AND
EMISSION CHARACTERISTICS
There are two distinctly different categories of
ultraviolet appliances used in tanning applications, each
with different UVR emission characteristics and different
requirements for filtering to eliminate undesirable wave-
lengths:
* ICNIRP c/o BfS, Institut fur Strahlenhygiene, Ingolstaedter
Landstr. L 85764 Oberscbleissheim, Gennany.
For correspondence or reprints contact: R. Matthes at the above
address, or email at r.matthes@icnirp.org.
t The International Commission on Illumination (CIE) defines
UVR as optical radiation between I 00 and 400 nm, and this spectral
region is divided into three photobiological spectral regions: UVC
(100-280 nm), UVB (280-315 nm) and UVA (315-400 nm).
(Manuscript received 1 May 2002; accepted 23 July 2002)
0017-9078/03/0
Copyright 2003 Health Physics Society
119
Low-pressure fluorescent tubes, emitting mostly UV A
or mostly UVB, with broad-band or narrowband emis-
sion; and
Filtered high pressure and high intensity discharge
lamps, emitting virtually only UV A or a mixture of
UVA and UVB.
Tanning appliances are currently the subject of an
international standard established by the International
Electrotechnical Commission (IEC 1995), which has
regulatory status in some countries. Four types of appli-
ances are recognized in this standard and defined by this
standard as follows:
UV type 1 appliances are those that emit UV radiation
such that the biological effect is caused by radiation
having wavelengths longer than 320 run and charac-
terized by a relatively high irradiance (;?; 0.15 W m-
2
)'
in the range 320 nm to 400 nm. The emission at
wavelengths less than 320 run is limited to 0.5 mW
-2
m,
having wavelengths both shorter and longer than 320
run and characteiized by a relatively high irradiance
(;?; 0.15 W m-
2
;) in the range 320 run to 400 nm. The
it-radiance at wavelengths less than 320 run is in the
range 0.5-150 mwm-
2
;
having wavelengths both shorter and longer than 320
nm and characterized by a limited it-radiance ( :5 0.15
W m
2
) in each UV radiation band; and
such that the biological effect is mainly caused by
radiation having wavelengths shorter than 320 nm (at
an irradiance greater than 0.15 W m
2
, and in the
wavelength range 320-400 nm, the it-radiance is
limited to 0.15 W m
2
).
*All it-radiances are erythemally weighted. In their international
standard, the International Electrotechnical Commission (IEC), has
used the wave length ranges from 320-400 nm and <320 nm to
characterize the UV radiation.
120 Health Physics
The emission characteristics and the health risks
associated with the use of each type of appliance are
different (Gies et al. 1986). Type 4 appliances, associated
with high levels of emission of UVB (280-3 15 nm), are
intended to be used following medical advice and should
not be used for tanning purposes, mainly because of the
publicized association between UVB and skin cancer.
Most of the appliances that are in use today are UY A-
emitting (315-400 nm) appliances, UV types I, 2, and 3.
The term "sun beds" in this document refers to UV-
emitting appliances of UY type 1, 2, and 3, although UV
type 4 appliances are still marketed to commercial
suntanning establishments.
During the last decade, increasing evidence of long
term UV A-induced risks for the skin and the eye has led
the sun-tanning industry to increase the UVB content in
the emission spectrum of tanning lamps in order to more
closely simulate natural sun exposure. This change has
also permitted shorter tanning exposures. It is known that
the ratio of UV A to U VB in the solar spectrum changes
during the day and undergoes large variations according
to season and latitude. It is also important to recognize
that there is no firm scientific evidence to indicate that
tanning with either UVA-dominated or a UVB-
dominated sources poses less risk; and, likewise, the use
of a simulated solar spectrum is not necessarily "safer"
than other artificial sources.
EFFECTS ON SKIN
Tanning
When skin .is exposed to UVR, two distinct tanning
reactions ensue:
Immediate pigment darkening (IPD). IPD begins
immediately on exposure to UVR and is caused by the
darkening of the pigment melanin that is already present
in the skin; it is n01mally seen only in people who have
at least a moderate constitutive tan. Such pigmentation
begins to fade within a few minutes after cessation of
exposure. Radiation between 320 and 400 nm is regarded
as being most effective for IPD (Irwin et al. 1993).
January 2003, Volume 84. Number I
Delayed tanning (neo-melanogenesis). The visible
pigmentation takes at least 3 d to develop and results
from exposure to U VA but is more effectively produced
by UVB (Parrish et al. 1982; Gange et al. 1985).
UVR-induced neo-melanogenesis is strongly dependent
on cellular responses arising from UVR-induced DNA
damage in cellular nuclei (Eller et al. 1996; Gilchrest et
al. 1996). Delayed tanning is more persistent than IPD
and results from an increase in the number, size, and
pigmentation of melanin granules (Bech-Thomsen et al.
1994). Exposure to UVB results also in an increase in the
thickness and scattering properties of the epidermis
(outer layer of the skin) (Bech-Thomsen and Wulf 1995).
Due, at least in part, to these processes, the tan obtained
from a pure UV A appliance, while perhaps cosmetically
acceptable, is not as effecti ve in protecting against
further exposure to solar UYR as the equivalent pigmen-
tation induced by exposure to solar radiation.
The degree to which an individual successfully tans
as the result of exposure on a sunbed depends critically
on his/her skin phototype as judged by his/her ability to
tan and the susceptibility to sunburn as a result of
exposure to solar UVR. Different skin phototypes (clas-
sified as phototypes I through VI) are presented in Table
l. In principle, the reaction of a person to UVR with
respect to tanning or sunburning is similar whether the
exposure is on a sunbed or to solar radiation.
Among users of artificial suntanning devices, persons
of skin phototypes I and II, who do not tan well and/or who
sunbum easily, are likely to be disappointed with the
cosmetic results of using a sunbed. It has also been
recognized that many users experience minor adverse cuta-
neous effects such as mild erythema, itching, and skin
dryness (Diffey 1986; Rivers et al. 1989; Diffey et al. 1990).
Individual users of tanning appliances and attendants at
tanning salons may incorrectly evaluate individual skin
sensitivity to UVR and underestimate the user's sensitivity.
Sunburn
Minor sunburn is a skin reddening (actinic ery-
thema) that appears up to 12 h after UVR exposure.
Table 1. Classification of skin phototypes based 011 their susceptibility to sunburn in sunlight and their ability to tan
(Fitzpatrick et al. 1995).
Skin
phototype Sun sensitivity Sunburn susceptibility" Tanning ability Classes of individuals
I Very sensitive Always sunburn ( <2 SED) No tan Melano-compromizedb
IT Moderately sensitive High (2-3 SED) Light tan Melano-compromizedb
rn Moderately insensiti ve Moderate (3-5 SED) Medi.um tan Melano-competent
IV Moderately resistant Low (5-7 SED) Dark tan Melano-competent
v Resistant Very low (7- 10 SED) Natural brown skin Melano-protected
VI Very resistant Extremely low (> I 0 SED) Natural black skin Melano-protected
1 Tbe ranges of SEDs in parentheses are only indicative.
b Melano-compromized individuals have a greater risk of developing skin cancers than melano-competent individuals.
Health issues of ultraviolet tanning appliances e ICNTRP 121
Erythema gradually fades after a few days, replaced by
some tanning in individuals with tanning capability.
Severe sunburn is painful and results in inflammation,
blistering, and peeling of the skin. Aside from photoim-
munological effects (discussed later), systemic effects of
severe sunbum are unknown except for transient fever.
Sunburn severity depends Ciitically on skin phototype
(Table I) and UV dose. For fair-skinned people (skin
phototypes I and II, melano-compromised), the relative
effectiveness of UVR for tanning and for erythema is
approximately the same over the entire range of UVB
and UV A wavelengths (PaiTish et al. 1982). For people
who tan well and who rarely sunburn (skin phototypes ill
and IV, melano-competent), the tanning efficacy of UVA
is higher than its erythemal efficacy (Gange et a!. 1985).
For a given sunbed type, there is a range of radiant
exposures which can be expressed in terms of Standard
Erythemal Dose (SED) unit related to the Minimal
Erythemal Doses (MEDs), that will produce noticeable
effects in people of different skin pbototypes.
Skin cancer
The most serious long-term effects attributed to
UVR exposure of the skin are the skin cancers (!ARC
1992). Squamous and basal cell carcinomas are common,
rarely fatal forms of skin cancer, which are often referred
to collectively as non-melanoma skin cancers (NMSCs).
Experimental studies clearly indicate that UV A and UVB
can cause squamous cell carcinomas (SCCs) in mice. The
same effects are likely in humans (Sterenborg 1987; van
Weelden et al. 1988). The erythemal potential of a UVR
source is gaining acceptance as a reasonable approxima-
tion for a quantitative measure of its carcinogenic poten-
tial (Cole et al. 1985). A series of publications (Berget
al. 1993; de Gruijl et al. 1993; de Gruijl and van der Leun
1994; de Gruijl and Forbes 1995) has been a source for
the proposal of an action spectrum for photocarcinogen-
esis by CIE TC 6-32 (CIE 2000). It will be used to
evaluate the maximal number of authorized yearly ses-
sions as the basis of a recommendation in the next
revision of the international standard, IEC 335-2-27.
Efforts to estimate the increased 1isk of a series of
tanning sessions have been made; however, these have
A minimum erythemal dose (MED) is defined as the UVR
exposure that will produce a "just noticeable" erythema on previously
unexposed skin of an individual. If the exposure is spectrally weighted
by the CIE erythemal action spectrum, the l\IIED corresponds to an
effective radiant exposwe expressed in Standard Erythemal Dose (SED)
units, depending on individual skin phototype (Table I). Erythemally
effective it-radiances (W m
2
eff.), expressed in SEDs per hour, can be
calculated for any source of UVR using spectral k-radiance data for tbe
source, at the poi.nt of interest, and the relative spectral weighti11g factors
for erythema, promulgated by the CIE (McKinlay and Di ffey 1987;
CIE/ISO 1999). One SED (Diffey et al. 1997) is a CIE/ISO official unit,
and effective radiant exposure of 100 J m
1
.
been limited to non-melanoma skin cancer, and such
estimates necessarily require a number of assumptions
such as the dose-response function, the use of ecological
rather than individual-based data to estimate the relation
between UV exposure and risk, the extent of natural
exposure, and neglecting its intermittency. For example,
estimates of the tisk of incidence of NMSC due to the use
of UV A sunbeds suggest a doubling of risk for no more
than 20 sessions per year over 30 y in Northern Europe
population (Diffey 1987). Despite these health risks, if
individuals insist on acquiring a tan, there are conclu-
sions that can be drawn from the scientific data to
minimize risk such as presented in the recommendation.
Based upon a modeling of human skin-cancer risk
(Diffey 1987), 10 sessions of 30 min per year will
increase by 5% the risk of skin cancer compared with
non-users of solaria. A "safe" level of solarium use does
not exist.
Cutaneous malignant melanoma (MM), while much
less hequent than NMSC, is much more serious and
accounts for the majotity of deaths from skin cancer.
There are some mammalian data (Ley et al. 1989; Ley
1997) for MM that indicate a strong UVB melanoma
etiology, but these data are not entirely consistent with
data from a fish melanoma model that indicate a strong
implication of UV A in addition to UVB in the induction
of fish melanoma (Setlow 1993). The animal models
demonstrate an increased impact of neonatal exposure on
tumor induction including malignancy (Robinson et al.
2000; Noonan et al. 2001). The evidence for the indict-
ment of sunlight as a causal agent is limited to epidemi-
ological data. The data indicate that intermittent exposure
to high levels of solar UVR, particularly at an early age,
may be a contributing causal factor (for recent review see
Armstrong and Kricker 1995; Gilchrest eta!. 1999). The
individual risk of MM is higher in people who have a
large number of nevi (moles) and who sunburn readily
and tan poorly with exposure to solar UVR. Some data
suggesting an association between the use of sunbeds and
an increased risk of MM have been published, but it is
unclear as to the relative importance of UVB, UVA, and
other factors (especially, general sun-seeking behavior)
in causing this association (Swerdlow et al. 1988; Walter
et al. 1990; Au tier et al. 199 l ; Higgins and du Vivier
1992; Westerdahl et al. 1994; Autier et al. 1994; Spencer
and Amonette 1995; Stem et al. 1997; Miller et al. 1998;
Swerdlow and Weinstock 1998; Westerdahl et al. 2000).
Taken collectively, both experimental and epidemi-
ological data on skin cancer all indicate that cumulative
exposure increases the risk for skin cancers. This in-
cludes childhood exposures. Therefore, the added expo-
sure from UV tanning appliances is likely to add to the
detrimental consequences of natural solar exposure.
122 Health Physics
Premature skin aging
There is considerable evidence that cumulative
UVR (UV A and UVB) exposure results in premature
skin aging characterized by a dry, coarse, leathery, and
wrinkled appearance. It bas been clearly demonstrated
that UV A causes skin damage in mice (Kligman et al.
1987; Bissett et al. 1989). Similar effects might be
expected in humans as a result of excessive use of
sun beds.
Daily exposures to suberythemogenic purely UVA
within the spectral region 320-400 nm for 8 d or
exposure to longer UV A wavelengths between 340-400
nm for 2 mo result in cumulative morphological skin
alterations, which are indicative of tissue injury (Lavker
et al. 1995; Lowe et al. 1995; Seite et al. 1998). In a
5-year longitudinal study of women who used or did not
use tanning salons (Pierard 1998), serious modifications
of skin elasticity and extensibility were found in the
tanning salon user group. In that group, the severity of
skin disorders was inversely correlated with their natural
pigment capacities. It has been concluded from the study
that the unremitting use of sunbeds induces a functional
decline of the dermis resembling premature aging.
Table 2. Agents producing photosensitivity.
Agents Incidence
January 2003, Volume 84, Number 1
Other skin effects
People who have excessively used UV A sunbeds have
exhibited increased skin fragility and blistering (Parr- et al.
1988; Murphy et al. 1989) and atypical melanocytic lesions
(Jones et al. 1987; Williams et al. 1988; Roth et al. 1989;
Salisbury et al. 1989; Kadunce et al. 1990).
Photodermatosis and photosensitivity
Polymorphic light eruption (PLE) is a common
photodennatosis readily produced in some people by
exposure to UV sunbed radiation (Rivers et al. 1989).
Other photoaggravated dermatoses such as systemic
lupus erythematosus are also exacerbated by the use of
UV sunbeds (Stern and Docken 1986). Certain medicines
and chemical and topical products (Table 2) such as
perfumes and lotions may cause skin photosensitization
in sunbed users.
PHOTOIMMUNOLOGICAL EFFECTS
UVR exposure causes localized skin and systemic
modifications due to photoimmuno1ogical reactions
(Noonan and De Fabo 1990), also a particular concern
with UV A sunbed radiation (Hersey et al. 1988; Rivers et
Type of reaction
Effective
wavelength
range
Agents producing photosensitivity at local administration
Sulphonamides and rela1ed chemicals n.a.a phototoxic and
(sunscreens. blancophores) photoallcrgic
Disinfeclw11s (salicylanilide compounds n.a. phototoxic and
in soaps and deodorants) photoattergic
Phenothiazines (creams, dyes and n.a. pbot.otoxic and
insecticides) photoallergic
Dyes n.a. phototoxic
hypetpigmcntation
Coal tar and derivalives (phenolic n.a. phototoxic
compounds)
Essential oils (perfumes and colognes) n.a. phototoxic
hyperpigmentation
Furocoumarines compounds (psoratens) n.a. phot.otoxic
hyperpigmentation
Cadmium sulphide (tanoos) o.a. phototoxic
Agents producing pbot.oseositivity after oral or parenteral administration
Amiodarone
Thiazide diuretics
Chlorpromazine and related
phmo1hiazines
Nalidixic acid
Non sleroid anli injlomnut/Ot)' drugs
Protriptyline
Psora/ens
Sulfon.amides (bacteriostatic and
antidiabetic)
Tetracyclines (antibiotics)
n.a. - not available.
High
Medium
Medium
High
Low
High
High
Low
Mediun1
pbot.otoxic
photoallergic
phot.otoxic and
photoallergic
phototoxic
phototoxic
photoattcrgic
phototoxic
phot.otoxic
pbotoallergic
phototoxic
290-320 nm
290-400 nm
320-visible
Visible
340-430 nm
290-380 mn
290-400 nm
380-445 nm
300-400 mn
300-400 nm
320-400 nm
320-360 nm
310-340 om
290-320 nm
320-380 om
315-400 nm
350-420 nm
Health issues of ultraviolet tanning appliances e ICNIRP 123
al. 1989). The clinical improvement of atopic dermatitis
with sub-erythemal UV A exposure indicates that UVR is
able to modify substantially normal and pathological
immunological reactions. UVB radiation can promote the
development of skin cancer, perhaps by suppressing the
immune system allowing the tumor to escape immune
surveillance (Duthie et al. 1999). The action spectra in
the UVB for mixed lymphocyte reaction and mixed
epidern1al cell lymphocyte reaction were found to be
similar to the induction of thymine dimers upon DNA
irradiation (Hurks et al. 1995).
There is also evidence that exposure to UVR can
activate and accelerate the growth of human viruses
(Otani and Moti 1987; Perna et al. 1987), including
human immunodeficiency virus (HIV) (Zmudzka and
Beer 1990), and have effects on infectious disease
(Halliday and Norval 1997). At present, the significance
of these observations with respect to human health is
unclear. The effects of UV A exposure on the immune
system are even more unce1tain (Schwarz 1998; Vermeer
et al. 1998).
OCULAR EFFECTS
The cornea
The principal adverse effect of the absorption of
UVR (UVC and UVB) by the cornea is termed photo-
keratitis ("welder' s f1ash" or "snow-blindness"), and
damage is generally limited to the epithelial (front
surface) cells of the cornea (Sliney and Wolbarsht 1980).
After a 6- to 12-h latent period that depends inversely on
the sevetity of the exposure, there is severe corneal pain,
photophobia, lacrimation, and eyelid spasm. These
symptoms are terribly distressing (with incapacitation),
but typically resolve in 24 h. There is some evidence of
possible long-term effects of UVR absorption by the
cornea (Taylor et al. 1989) and evidence of endothelial
thinning (Pitts et al. 1987).
The lens
Transmission of UV A to the lens is much greater
than that of UVB. Animal data indicate that threshold
lenticular damage is limited mainly to UVR exposure in
the 295 to 325 nm wavelength band. Experimental
spectral efficiency for acute cataracts in laboratory ani-
mals has been measured only in this spectral region.
Although UV A sunbeds produce limited 295-325 nm
radiation, it should not be infeiTed that UV A is safe with
respect to lens exposure. Crystalline lens aging is char-
acterized in part by loss of elasticity and browning
(brunescence), both of which may be caused pa1tly by
UV A. Certain medicines may act as UV A photosensitiz-
ers of the crystalline lens.
The retina
At least for the chronic effects of exposure to solar
radiation and to radiation from conventional light
sources, the most important retinal damage mechanism is
photochemical injury from short-wavelength light
(Sliney and Wolbarsht 1980). The gradual brunescence
of the lens as it ages results in its decreased transmission
of blue-light and of UVR thereby affording increased
protection to the retina. Young children and people who
have had a lens surgically removed (aphakes) are at a
higher tisk of retinal damage from UVR and blue-light.
Until the last decade, many implanted artificial lens did
not effectively absorb UV A. Macular injuries to two
tanning booth users were mentioned in one study but not
confirmed (Walters and Kelley 1987).
The crystalline lens and cornea serve in considerable
measure to protect the retina from most UV tanning
booth radiation, even without protective goggles. As
discussed previously, the crystalline lens blocks UVR
below 400 nm and the comea blocks UVR below 300
nm, but trace amounts of UV -B radiation between 300
and 315 nm may reach the retina (Boettner and Wolter
1962; Sliney 1986).
Protective eyewear
The use of protective eyewear (goggles) will prevent
exposure of the eyes to harmful levels of UVR and
blue-light. This represents a very important issue for
sunbed exposures since an individual is normally pro-
tected from most of the overhead solar UVR by geomet-
rical shading by the brow ridge and upper lids. The
exposure from sunbeds is geometrically greatly different.
Furthermore, oblique rays can be focused into the nasal
equatotial region of the lens (Coroneo et al. 1991). UVR
can only reach the ctitically important germinative re-
gion of the lens by the focusing of oblique rays. Eyewear
that does not incorporate side protection is not suitable to
protect the eye from lateral UVR.
CONCLUSION AND RECOMMENDATIONS
A review of scientific evidence shows that solar
UVR is a cause of squamous cell cancer, basal cell
cancer, and cutaneous melanoma as well as causing
accelerated skin aging and other adverse health effects.
Because of this strong evidence on the adverse health
effects of UVR, even though there is not conclusive
direct evidence that sun bed exposure causes skin cancer,
it is ICNIRP's view that any use of suntanning appli-
ances is likely to raise the risk of cancer. This tisk is
particularly high for people having skin phototypes I and
II and for children.
124 Health Physics
/CNIRP, therefore, recommends against the use of
W-emitting appliances for tanning or other non-medical
purposes. The following groups are at particularly high
risk of incurring adverse health effects from UVR, and
therefore should be particularly counseled against the
use of tanning appliances:
People who have skin phototypes I or II;
Children (i.e., less than 18 y of age);
People who have large numbers of nevi (moles);
Persons who tend to freckle;
Individuals who have a history of frequent childhood
sunburn;
People who have premalignant or malignant skin
lesions;
People who have sun-damaged skin;
Those who are wearing cosmetics. These may enhance
their sensitivity to UV exposure; and
Persons taking medications. In this case they should
seek advice from their physician to determine if the
medication will make them UV -sensitive.
Although contrary to the above recommendation, if
persons decide to use suntanning appliances, steps should
be taken to minimize the risk. The recommendations
provided in Appendix A have been made by several
workshops, meetings, and publications which have
pointed to the identification of optical radiation hazards
associated with the development and popularity of the
"suntanning industry" (Council on Scientific Affairs
1989; ICNIRP/CIE Measurements of Optical Radiation
Hazards 1998; Miller et al. 1998; Moseley et al. 1998;
USNIH 1998; Greinert et al. 2001). ICNIRP generally
supports these recommendations as a likely means to
reduce the risk of skin cancers from the use of UV
tanning appliances and particularly to minimize the risks
for high-risk groups. Suggestions for fmther reading can
be found in Appendix B.
Acknowledgments -The support received by ICNIRP from the Interna-
tional Radiation Protection Association, the World Health Organization.
the International Labor Organization, the European Commission, and the
German Government is gratefully acknowledged.
During the preparation of this statement, the composition of the Jmema-
tional Commission on Non-Ionizing Radiation Protection was as follows:
A.F. McKinlay, Chairman (UK), Vice-chairman umil 2000
J.H. Bemhardr. Vice-chairmtm (Germany), Clwimum lmfil 2000
A. Ahlbom (Sweden)
U. Bergqvisr (Sweden) (j200J)
J. P. Cesarini (France)
M. Grwulolfo (/lilly), until 2000
F. R. de Gruijl (The Netherlands)
M. Hietanen (Finland)
R. Owen (USA)
D.H. Sliney (USA!
January 2003, Volume 84, Number I
J.A.J Stolwijk (USA), unril 2000
A. Swerdlow (UK). since 2000
L Szabo (Hungary), until 2000
M. Taki (Japan)
T.S. Tenforde (USA)
P. Vecchia (Italy). since 2000
B. Veyret (France), since 2000
R. Matthes. Scientific Secretary (Germany)
M.H. Repacholi. Chairman Emeritus (Swirzerlantl)
During the preparation of rhis docume/11, the composition of the ICNIRP
Standing Committee IV and task group was: D.H. Sliney (USA), Chairman
1-P. Cesarini (France) F. R. de Gruijl (The Netherlands) B.L. Diffey (U.K.)
M. Hietal!en (Finland) M.A. Mainster (USA) T. Okuno (Japan) B.E. Swck
(USA) A. Swerdlow (UK)
REFERENCES
Armstrong BK, Kricker A. Skin cancer. Dermatologic Clinics
13:583-594; 1995.
Autier P, Joarlette M, Lejeune F, Lienard T, AndreJ, Achten
G. Cutaneous malignant melanoma and exposure to sun-
lamps and sunbeds: descriptive study in Belgium. Mela-
noma Res 1:69-74; 1991.
Autier P, Dore JF, Lejeune F, Koelmel KF, Geffeler 0, Hille P,
Cesarini JP, Lienard D, Liabeuf A, Joarlette M, Chemaly P,
Hakim C, Koehl A, Kleeberg UR. (for the EORTC Mela-
noma Cooperative Group). Cutaneous malignant melanoma
and exposure to sunlamps or sunbeds: an EORTC multi-
center case-control study iJl Belgium, France, and Germany.
Int J Cancer 58:809-813; 1994.
Bech-Thomsen N, Ravnborg L, Wulf HC. A quantitative study
of the melanogenic effect of multiple suberythemal doses of
different ultraviolet radiation sources. Photodermatol Pho-
toimmunol Pbotomed 10:57- 64; 1994.
Bech-Thomsen N, Wulf HC. Photoprotection due to pigmen-
tation and epidennal thickness after repeated exposure to
ultraviolet light and psoralen plus ultraviolet A therapy.
Photodermatol Photoimmunol Photomed II :2 13-2 18;
1995.
Berg RJW, de Gruijl FR, van der Leun JC. Interaction between
ultraviolet A and ultraviolet B radiations in skin cancer
induction in hairless mice. Cancer Res 53:42 12-4217;
1993.
Bisset DL, Hannon DP, Orr TV. Wavelength dependence of
histological, physical and visible changes in chronically
UV -irradiated hairless mouse skin. Pbotochem Photobiol
50:763-769; 1989.
Boettner EA, Wolter JR. Transmission of the ocular media.
Invest Ophthalmol Vis Sci 1:776-783; 1962.
CIFJISO. Erythemal reference action spectrum and standard
erythemal dose. Geneva: ISO; ISO 17 166; 1999Nienna:
CIE; CIE S007/E; 1998.
CIE. Action spectrum for photocarcinogenesis. Vienna: CIE;
Report CIE 138-2000; 2000.
Cole CA, Forbes PD, Davies RE. An action spectrum for UV
carcinogenesis. Photochem Photobiol 43:275-284; 1985.
Coroneo MT, Mueller-Stolzenbury NW, Ho A. Peripheral light
focusing by the anterior eye and the ophthalmohelioses.
Ophthalmic Surgery 22:705-7ll; 1991.
Council on Scientific Affairs. Harmful effects of ultraviolet
Jight radiation. JAMA 262:380-384; 1989.
de Gruijl FR, Sterenborg HJCM, Forbes PD, Davies RE, Cole
C, Kelfken G, van Weelden H, Slaper H, van der Leun JC.
Health issues of ultraviolet tanning appliances e ICNIRP 125
Wavelength dependence of skin cancer induction by ultra-
violet irradiation of albino hairless mice. Cancer Res 53:53-
60; 1993.
de Gruijl FR, van der Leun JC. Estimate of the wavelength
dependency of ultraviolet carcinogenesis in humans and its
relevance to the risk assessment of a stratospheric ozone
depletion. Health Phys 67:314-325; 1994.
de Gruijl FR, Forbes PD. UV-induced skin cancer in hairless
mouse model. BioEssays 17:651-660; 1995.
Diffey BL. Use of UV A sun beds for cosmetic tanning. Br J
Dermatol115:67-76; 1986.
Dif'fey BL. Analysis of the risk of skin cancer from sunlight
and solaria in subjects living in Northern Europe. Photoder-
matol 4: 118-126; 1987.
Dif'fey BL, Farr PM, Ferguson J, Gibbs NK, de Gruijl FR,
Hawk JLM, Johnson BE, Lowe G, MacKie RM, Magnus
IA, McKinlay AF, Moseley H, Murphy GM, Nonis PG,
Young AR. Tanning with ultraviolet-A sunbeds. Br Med J
301:773-774; 1990.
Diffey BL, Jansen CT, Urbach F, Wulf HC. The standard
erythema dose: a new photo biological concept. Photoder-
matol Photoimmunol Photomed 13:64-66; 1997.
Duthie MS, Kimber I, Norval M. The effects of ultraviolet
radiation on the human immune system. Br J Derm
140:995-1009; 1999.
Eller MS, Ostrom K, Gilchrest BA. DNA damage enhances
melanogenesis. Proc Ntl Acad Sci USA 93:1087-1092;
1996.
Farr PM, Marks JM, Diffey BL, lnce P. Skin fragility and
blistering due to the use of sunbeds. Br Med J 296: 1708 -
1709; 1988.
Fitzpattick TB, Cesarini JP, Young A, Kollias N, Pathak MA.
Reported in Fitzpatrick TB, Bolognia JL. Human melanin
pigmentation: Role in pathogenesis of cutaneous melanoma.
ln: Zeise L, Chedekel MR, Fitzpatrick B, eds. Melanin: Its
Role in Human Photoprotection. Overland Park, KS: Val-
denmar Publishing Co; 1995: 177-182.
Gange RW, Park YK, Auletta M, Kagetsu N, Blackett AD,
Parrish JA, Action spectra for cutaneous responses to
ultraviolet radiation. In: Gange RW, Urbach F, eds. The
biological effects of UV A radiation. New York: Praeger;
1985:57-67.
Gies HP, Roy CR, Elliot G. Suntanning: Spectral irradiance
and hazard evaluation of ultraviolet sources. Health Phys
50:691-703; 1986.
Gilchrest BA, Park HY, Eller MS, Yaar M. Mechanisms of
ultraviolet light-induced melanogenesis. Photochem Photo-
bioi 63:1-10; 1996.
Gilchrest BA, Eller MS, Geller AC, Yaar M. The pathogenesis
of melanoma induced by ultraviolet radiation. N Engl J Med
340: 1341-1348; 1999.
Greinert R, McKinlay A, Breitbart EW. The European society
of skin cancer prevention-EUROSKIN: towards the pro-
motion and harmonization of skin cancer prevention in
Europe. Recouunendations Eur J Cancer Prev 10: 157 -162;
2001.
Halliday K.E, Norval M. The effect of UV on infectious
disease. Med Microbiol 8: 179-188; 1997.
Hersey P, MacDonald M, Henderson C, Schibeci S,
D'Alessandro G, Pryor M, Wilkinson FJ. Suppression of
natural killer cell activity in human by radiation from
solarium lamps depleted of UVB. J Invest Detmatol
90:305-3 10; 1988.
Higgins EM, du Vivier A WP. Possible induction of malignant
melanoma by sunbed use. Clin Exp Dermatol 17:357-359;
1992.
Hurks HMH, Out-luiting C, Vermeer PJ, Claas FHJ, Mommaas
AM. The action spectra for UV-induced suppression of
MLR and MECLR show that immunosuppression is medi-
ated by DNA damage. Photochem Photobiol 62:449-453;
1995.
Jones SK, Moseley H, MacKie RM. UV A-induced melano-
cytic lesions. Br J Dermatolll7:111-115; 1987.
!ARC. Solar and ultraviolet radiation. Monographs on the
evaluation of carcinogenic risk to humans. Vol. 55. Lyon:
International Agency for Research on Cancer; 1992.
ICNIRP/CIE. Measurements of Optical Radiation Hazards. A
reference book based on presentations by health and safety
experts on optical radiation hazards. Gaithersburg MD:
ICNIRP/CIE 6; 1998.
International Electtotechn.ical Commission. Safety of house-
hold and similar electrical appliances. Part 2: Particular
requirements for appliances for skin exposure to ultraviolet
and infrared radiation. Geneva: International Electrotecbni-
cal Commission; IEC 335-2-27; 1995.
Irwin C, Barnes A, Veres D, Kaidbey K. An UV radiation
action spectrum for immediate pigment darkening. Photo-
chem Photobiol 57:504-507; 1993.
Kadunce DP, Piepkom MW, Zone JJ. Persistent melanocytic
lesions associated with cosmetic tanning bed use: "sunbed
lentigines." JAm Acad Dermatol23:l029-1031; 1990.
Kligman LH, Kaidbey KH, Hitchens VM, Miller SA. Long
wavelength (> 340 nm) UV A-induced damage in hairless
mice is dose dependent. ln: Passchier W, Bosnjakovic
BFM, eds. Human exposure to ultraviolet radiatian: Risks
and regulations. An1sterdam: Elsevier Applied Science Pub-
lishers; 1987: 77- 8 I.
Lavker RM, Veres DA, Irwin CJ, Kaidbey KH. Quantitative
assessment of cumulative damage from repetitive exposures
to suberythemogenic doses of UV A in human skin. Photo-
chem Photobiol 62:348-352; 1995.
Ley RD. Ultraviolet radiation A-induced precursors of cuta-
neous melanoma in Monodelphis domestica. Cancer Res
57:3682-3684; 1997.
Ley RD, Applegate LA, Padilla RS, Stuart TD. Ultraviolet
radiation-induced malignant melanoma in Monodelphis do-
mestica. Photochem Photobiol 50: 1-5; 1989.
Lowe NJ, Meyers DP, Wieder JM, Luttman D, Borget T,
Lehman MD, Johnson A W, Scott IR. Low doses of repet-
itive ultraviolet A induce morphologic changes in human
skin. J Invest Dermatol L05:739-743; 1995.
McKinlay AF, Diffey BL. A reference action spectrum for
ultraviolet induced erythema in human skin. CIEJ 6: 17-22;
1987.
Miller SA, Hamilton SL, Wester UG, Cyr WH. An analysis of
UV A emissions from sunlamps and the potential impor-
tance for melanoma. Photochem Photobiol 68:63-70; 1998.
Moseley H, Davidson M, Ferguson J. A hazard assessment of
artificial tanning units. Photodermatol Photoimmunol Pho-
tomed 14:79-87; 1998.
Murphy GM, Wright J, Nicholls DSH, McKee PH, Messanger
AG, Hawk JLM, Levene GM. Sunbed-induced pseudopor-
phyria. Br J Dermatol 120:555-562; 1989.
Noonan FP, DeFabo EC. Ultrashort-B dose-response curves for
local and systemic immunosuppression are identical. Pho-
tochem Photobiol52:801-810; 1990.
Noonan FP, Recio JA, Takayama H, Duray P, Anver MR, Rush
WL, De Fabo EC, Merlino G. Neonatal sunburn and
melanoma in mice. Nature 413:271-272; 2001.
126 Health Physics
Otani T, Mori R. The effects of UV irradiation of the skin on
herpes simplex virus infection; Alteration in immune func-
tion mediated by epidermal cells and in the course of
infection. Arch Virol 96:1-15; 1987.
Parrish JA, Jaenicke KF, Anderson RR. Erythema and mela-
nogenesis action spectra of normal human skin. Photochem
Photobiol 36: 187-191; 1982.
Perna JJ, Mannix ML, Ronney JF, Notkins AL, Straus SE.
Reactivation of latent herpes simplex virus infection by
ultraviolet light: A human model. J Am Acad Detmatol
17:473-478; 1987.
Pierard GE. Ageing in the sun parlor. Int J Cos Sci 20:25 1-259;
1998.
Pitts DG, Bergmanson JPG, Chu LWF, Waxler M, Hitchins
VM. Ultrastructural analysis of corneal exposure to UV
radiation. Acta Ophthalmol (Copenh) 65:263-273; 1987.
Rivers JK, Norris PG, Murphy GM, Chu AC, Midgley G,
Monis J, Morris RW, Young AR, Hawk JLM. UVA
sunbeds: Tanning, photoprotection, acute adverse effects
and immunological changes. Br J Dennatol 120:767-777;
1989.
Robinson ES, Hill RH Jr, Kripke ML, Setlow RB. The
Monodelphis melanoma model: initial report on large ultra-
violet A exposures of suckling young. Photochem Photobiol
7 1:743-746; 2000.
Roth DE, HodgeS, Callen JP. Possible ultraviolet A-induced
lentigines: a side effect of chronic tanning salon usage.
J Am Acad Dermatol 20:950-954; 1989.
Salisbury JR, Williams H, du Vivier A WP. TalUling-bed
lentigines: Ultrastructural and histopathologic features.
JAm Acad Dermatol21:689-693; 1989.
Schwarz T. Effects of UV A light on the immune system. A
settled issue? In: Rougier A, Schaefer H, eds. Protection of
the skin against ultraviolet radiations. Paris: John Libbey
Eurotext; 1998: 93-95.
Seite S, Moyal D, RichardS, de Rigal J, Leveque JL, Hourseau
C, Fourtanier A. Effects of repeated suberythemal doses of
UV A in human skin. In: Rougier A, Schaefer H, eds.
Protection of the skin against ultraviolet radiations. Paris:
John Libbey Eurotext; 1998:47-58.
Setlow R, Grist E, Thompson K, Woodhead AD. Wavelengths
effective in induction of malignant melanoma. Proc Natl
Acad Sci USA 90:6666 - 6670; 1993.
Sliney DH, Wolbarsht ML. Safety with lasers and other optical
sources, a comprehensive handbook. New York: Plenum
Press; 1980.
Sliney DH. Defining biologic exposures to light. In: Cronly-
Dillon J, Rosen ES, Marshall J, eds. Hazards of light. New
York: Pergamon Press; 1986.
Spencer JM, Amonette RA. indoor tanning: Risks, benefits and
future trends. JAm Acad Dennatol33:288-298; 1995.
Sterenborg HJCM. Investigations on the action spectrum of
tumor genesis by ultraviolet radiation. Utrecht: State Uni-
versity of Utrecht; 1987. Dissertation.
Stern RS, Docken W. An exacerbation of SLE after visiting a
tanning salon. JAMA 255:3120; 1986.
Stern RS, Nichols KT, Vakeva LH. For the PUV A follow-up
study. N Engl J Med 336:1041-1045; 1997.
Swerdlow AJ, English JSC, MacKie RM, O' Doherty CJ,
Hunter JAA, Clark J, Mole DJ. Fluorescent lights, ultravi-
olet lamps, and risk of cutaneous melanoma. Br Med J
297:647-650; 1988.
Swerdlow AJ, Weinstock MA. Do tanning lamps cause mela-
noma? An epidemiologic assessment. J Am Acad Dermatol
38:89-98; 1998.
January 2003. Volume 84. Number I
Taylor HR, West SK, Rosenthal FS, Munoz B, Newland HS,
Abbey H, Emmett SK. Effect of ultraviolet radiation on
cataract formation. N Eng! J Med 319:1429-1433; 1989.
US National Institutes of Health. Program and abstract book-
Research workshop on risks and benefits of exposure to
ultraviolet and tanning. Bethesda: National Institutes of
Health; 1998.
Van Weelden H, de Gruij1 FR, van der Putte SCJ, Toonstra J,
van der Leun JC. The carcinogenic risks of modem tanning
equipment: Is UV-A safer than UV-B? Arch Dermatol
280:300-307; 1988.
Vermeer BJ, Winstzen M, Claas FHJ, Schothorst AA, Hurks
HMH. UV-induced immunosuppression. The critical role of
wavelength. In: Rougier A, Schaefer H, eds. Protection of
the skin against ultraviolet radiations. Paris: John Libbey
Eurotext; 1998: 89 -92.
Walter SD, Marrett LD, From L, Hertzman C, Shannon HS,
Roy P. The association of cutaneous malignant melanoma
with the use of sunbeds and sunlamps. Am J Epiderniol
131:232-243; 1990.
Walters BL, Kelley TM. Commercial tanning facilities: a new
source of eye injury. Am J Emerg Med 5:386-389; 1987.
Westerdahl J, Olsson H, Masback A, Ingvar C, Jonsson N,
Brandt L, Jonsson PE, Moller T. Use of sunbeds or
sunlamps and malignant melanoma in southern Sweden.
AmJ Epidem 140:691-699; 1994.
Westerdalll J, Olsson H, Masback A, lngvar C, Jonsson N.
Risks of cutaneous malignant melanoma in relation to use of
sun beds: futher evidence for UV -A carcinogenicity. Br J
Cancer 82: 1593-1 599; 2000.
Williams HC, Salisbury J, Brett J, du Vivier A. Sunbed
lentigines. Br Mecl J 296: 1097; 1988.
World Health Organization. WHO Environmental Health Cri-
teria 160. Ultraviolet radiation. Geneva: World Health
Organization; 1994: 140, 150, 224-225, 245-246.
Zmudzka BZ, Beer JZ. Activation of human immunodeficiency
by ultraviolet radiation. Photochem Photobiol6: 1153-1 162;
1990.
APPENDIX A
If tanning devices are used, then the following
specific recommendations from past workshops should
apply:
Claims of beneficial medical effects should not be
made. Any therapeutic use of tanning devices should
be clone only in medical units;
Tanning devices should comply with the requirements
of the IEC standard (1995) and be limited to UV type
1, 2, or 3 as defined in that standard;
Approptiate health warnings should be provided to the
client prior to tanning exposure;
Appropriate UV -protective goggles should be pro-
vided and worn during tanning exposures;
Operator staff should be provided with appropriate
approved training (receive appropriate certification);
Professional operators are responsible for providing
client information and guidance on the safe use of
tanning devices;
Health issues of ultraviolet tanning appliances e lCNIRP 127
Minimize the number of sessions. For example, the
French Regulations (J. 0. Republique Francaise, An-
nexe 2, 1997) require that regular exposure, for pho-
totypes Ill and IV, melano-competent skin, should not
exceed two sessions per week with a maximum of 30
sessions per year ( erythemaliy effective exposure of
500 J m
2
per session). An occasional break from the
regularity of exposure is advisable;
Manufacturers or dealers must supply exposure sched-
ules based on the tanning device lamp characteristics;
Because the sensitivities of individuals vary greatly, it
is advisable to limit the duration of the first session to
about one-half of a regular session in order to establish
the user' s skin response. If following the first session
any adverse reaction occurs, further use of the sunbed
should be discouraged;
Products designed to enhance or accelerate tanning
should not be used;
Any modifications, such as the replacement of lamps,
filters or reflectors should not change the IEC classi-
fication of the device. Tanning devices should have an
appropriate timer;
Tanning devices in hotels or in recreational facilities
should be subject to the same controls as noted above
(as for any commercial outlet);
Because of their possible misuse, unattended or coin-
operated tanning devices should not be used;
By the nature of their use, sunlamps in the home are
not subject to the same degree of control as those used
under proper supervision in commercial outlets, so
additional safety information should be provided by
the vendor or supplier of the tanning device. In these
circumstances only IEC type 3 tanning devices should
be used.
Recognizing that different countries will have different
ways of implementing and determining compliance
with these recommendations, the tanning facilities
should comply with these recommendations, and that
compliance should be checked by the appropriate
national authority where possible.
APPENDIX B
Further readings
American Academy of Detmatology. Position State-
ment on Indoor Tanning. Dermatology World; March
1999.
EUROSKIN Towards the Promotion and Harmoniza-
tion of Skin Cancer Prevention; Session VI: WHO
Workshop-UVR Tanning Devices, 2 - 5 May 2000,
Hamburg, Germany.
Finnish Center for Radiation and Nuclear Safety:
Guidance 1992.
IRPA/INIRC Guidelines. Health issues of ultraviolet
"A" sunbeds used for cosmetic purposes. Health Phys
61:285-288; 1991.
J. 0. Republique Francaise: Decret Nos. 97-6 17, Mai
1997 relatif a la vente et a la mise a disposition du
public de certains appareils de bronzage utilisant des
rayonnements ultraviolets.
Norwegian Radiation Protection Authority: Code of
practice (modification of #2 & 5 of the Royal Resolu-
tion concerning the use of x rays, 1983); 1992.
NRPB Board statement on effects of ultraviolet radi-
ation on human health and health effects from ultravi-
olet radiation. 6, No.2; 1995.
NRPB Use of Sunbeds and cosmetic tanning. State-
ment by NRPB Advisory Group on Non-Ionising
radiation. Radiological Protection Bulletin No. 2 18,
11-15; 1999.
Senat de Belgique (May 1998): Proposition de loi
reglementant 1' exploitation des centres de bronzage,
visant a responsabiliser les et les utilisa-
teurs de banes solaires (679/1).
Swedish Radiation Protection Institute: Regulatory
code concerning sunbeds (SSI FS 1998: 2).
U.S. Food and Drug Administration. Policy on maxi-
mum timer intervals and exposure schedule for sun-
lamps. August 1986; FDA Rockville MD, USA.
Radiation Protection Dosimetry-Ultraviolet Radia-
tion Exposure, Measurement and Protection.
lay AF, Repacholi MH, eds. Proceedings of an Inter-
national Workshop St Catherine' s College, Oxford,
England, October 18 -20; 1999. Radiation Protection
Dosimetry Vol. 91; 2000.
Indoor Tanning: The Risks of Ultraviolet Rays
Consumer Updates: Cosmetics
Consumer Updates: Dietary
Supplements
iiii--1H----
Consumer Updates: Drugs
Consumer Updates: Food
Consumer Updates: Medical
Devices
Consumer Updates: Radiation-
Emitting Products
Consumer Updates: Vaccines,
Blood & Biologics
Resources for You
Sign up for Consumer
Updates by E- mail
Consumer Updates RSS
Feed
A-Z Index
Indoor Tanning: The Risks
of Ultraviolet Rays

....... ,... ............... .
.. 1 .... tl ._
.. _...,.... ....
_ _......,. ..... _
............... ... ...
......................... ,_
. _ _...._...,...._
.... _ _..._
U!IIM ..-..-.-...,..111
... .. _.,..,._.Moll
......... -........... ,., __ ......,._,,.._
......... -- ............
-'"----"'
....................
S
..uMI JoU U!UIIItf
.... ....
... ,.,.l .. dtft: ..........

Nft ...... , ._ ..
...... , .... ,w.
.............. ,. ..............
... --.... ,.__,. -- -
. ..........................
..... , ...
.... ....._....-t!:1f--
"'"',.., ... . ,,.,...., ...
_ ..... ....

.. ,.,_._._ .. _.,_ ..
....... .. _. ........ .oH-
.. ....... ... -..-.... ,...
_...ll.ilu....., .... ._.,IOIW
. .... '"'_ ... ...,..,.. ..
.........................
,,._,.,
,._.,...._ ............... ....
............ _ ...... V\4
-
.. ......... -..... .....
.. _ ..
11 ........... , .... , ... ..
............ ........ _ .. ..
_..... ........... -

..,.., ................ _
', -. .. -..... ........ .,..._._
, .......
...,_PI' ............ ..
.. '" ........... ,_
' ................ ......
http://www. f da .gov /ForConsu m ers/ConsumerU pdat es/u em 186687. htm# Tan ninginChildrenandTeens[ 12/1 /2009 8:47:49 PM]
an assoc1at10n between mdoor tannmg and two types ot sKm cancer: squamous cell carcmoma
and melanoma
an association between UV-emitting tanning devices and cancer of the eye (ocular melanoma)
both UV-A and UV-B rays causing DNA damage, which can lead to skin cancer in laboratory
ammars ano numans
the risk of melanoma of the skin increasing bv 75 percent when tanning bed use started
before age 35
!ARC's review had some limitations, says Ron Kaczmarek, M.D., M.P.H., an FDA epidemiologist
wr1 o allcflyz-etllh e review. ttmttcrttcrrrs-;nd ode posstbl e ina c co
tanning experiences, not knowing the amount of UV radiation emitted by each tanning device,
a 11d tl1 e I nobi1rty-ta-s-e-parateLh-e-efferts-oHmnvit:luais'-imluo r a 11d ootduorexpusu , ".
Nevertheless..-IARC concluded tbat there is con.vlncing evidence o.Lan_as.s.o.ciati.o.tLhe.tw.e.e.o_the
use of indoor tanning equipment and melanoma risk, and that the use of tanning beds should
be discouraged.
"It's well established tbat !Bl radiation from tbe suo causes skio caocer.:...say.s Miller. "Since
lamps used in tanning beds emit UV radiation, the use of indoor tanning devices also increases
your nsK or sKm cancer.
back to too
Other Risks
In addition to the serious risk of skin cancer tanning can cause:
Premature aging. Tanning causes the skin to lose elasticity and wrinkle prematurely. This
after .
J'
immune svstem and the skin's natural defenses leaving vou more vulnerable to diseases
including skin cancer.
Gamage
rat;! i a tioR-m-ay- deve-1-0fHI-A-
itchy red rash and other adverse effects.
Advocates of tanning devices sometimes argue that using these devices is less dangerous than
'"UA-ta.r:u:ti.r.l.g-because....t.l:le
But there is no evidence to support these claims. In fact, sunlamps may be more dangerous
than the sun because they can be used at the same high intensity every day of the year-unlike
tl1e seasorr;-and cloud cmre .
back to top
Tanning in Children and Teens
FDA is particularly concerned about children and teens being exposed to UV rays. Intermittent
exposures to intense UV radiat1on leading to sunburns, espeCially in childhood and teen years,
increase the risk of melanoma, according to NCI.
FDA believes that limiting sun exposure and using sunscreen or sunblock are particularly
Important tor children smce these measures can prevent sunburn at a young age.
NL! reports that women who use tannmg beds more than once a month are percent more
likely to develop melanoma. Teenage girls and young women make up a growing number of
tannmg oea cuswmers.
roung peopre may noCffifi11<1hey are vulnerable to sk1n cancer," says Kaczmarek. 'Tfiey1lave

States who will learn they have melanoma this year, one out of eight will die from it, according
.o....N.C.l-estimat.es ..... _m pI" n"""" i <"
the second most common cancer in women 20 to 29 years old.
Some states are considering laws to ban those under age 18 from using tanning beds. And
many states now have laws that require minors to have a parent's consent or be accompanied
by a parent to the tanning facility.
FDA's current performance standard requires that a sunlamp product's label include a
recommended exposure schedule. FDA has advised manufacturers that this schedule should
,.... than tl'tfee-s-es-s-ions-ifrHte-fffs-t-wee .
In an NCI-sponsored study published in September 2009 in the Archives of Dermatology, the
study researchers h1red and tra10ed college students to pose as 15-year-oiO,ralr sKJOnea g1ns
dents as-keti-mer-e-th an 3, 69G4afl-A-. .
0
facilities in all 50 states about their practices.
Less than 11 percent onlle faCilities followed""F[)A's recommended exposure sclieaUie or mree
on
davs the first week, and many promoted frequent tanning with "unlimited tanning" discount
price packages.
I no
http://www. fda .gov /ForConsu m ers/ConsumerU pdates/u em 186687. htm# Tan ninginChildrenandTeens[ 12/1 /2009 8:47:49 PM]
that "many parents are allowing their teens to tan and are providing written consent or
accompaniment."
"Parents should carefully consider the risks before allowing their children under 18 to tan," says
Miller.
back to top
FDA Regulation
FDA regulates radiation-emitting products, including sunlamps and products that contain them,
<:11rh ri<: trinniog beds aod bootbs aod po(table born..e_uoits Maoufactu(e(S of suolrimn<: m11c;t
comply with FDA regulations, including the performance standard for sunlamp products.
FDA requires sunlamp products to carry a warning label with specific information. Based on the
results of consumer testmg, FDA IS cons1denng amendmg the warnmg label requirements to
strengthen the warnings about skin cancer and irreversible eY-e dama!le
make the warning easier for consumers to read and understand
In a December 2008 Reoort to Congress FDA noted that FDA/NCI studies found that the UV
exposures typically provided by sunlamp products are excessive, and that comparable cosmetic
effects can be oroduced with exoosures that are onlv one-third or even one-fourth the levels
currently used. FDA is evaluating the results of this research and considering whether those
resui{S warram cnanges w its performance stam:t<rrtl for son1amp proouc{s.
J.ldCKlo lUf.l
The Riskiest Practices
FDA, NCI, the American Academy of Dermatology, and other health organizations advise limiting
exposure to natural UV radiation from the sun and avoiding artificial UV sources such as tanning
u e-cts-entirei y .
All use of tanning beds increases the risk of skin cancer. Certain practices are especially
dangerous. These include:
Failing to wear the goggles provided, which can lead to short- and long-term eye inj ury .
Starting with long exposures (close to the maximum time for the particular tanning bed),
which can lead to burning. Because sunburn takes 6 to 48 hours to develop, you may not
real ize your skin is burned until it's too late .
Failing to tallow manufacturer-recommended exposure times on the label tor your sKin type.
Tanning while using certain medications or cosmetics that may make you more sensitive to
UV rays. Talk to your doctor or pharmacist first.
back to top
Melanoma: One Woman's Story
17. She s-topped-
at age 20 when she was diagnosed with melanoma, the deadliest form of skin cancer. The
former Miss Maryland says she used tanning beds at least four times a week, and sometimes
every oay.
"Growing UJ2, until I started using tanning beds, m:t )2arents were very strict about me wearing
sunscreen," says Cicala. Although she also tanned in the summer sun during her 3 years of
ldrllrlll!;; ut:u u::.t:, '-''-d'd t:Stlmate-nhdr90-p-ercent of her was In tanning bt:u::.
ri.w:.i.ng-tl:l.ls- p.er i "ri
In the 4 years since she was diagnosed with melanoma, Cicala's surgeries have left her with
airour5 scars. etcola--gets<rheod-to-toe months, wllicl1 usually-results ;,
rP lllilll.d.LoLa_s.us.p.i.cio.us_gr.o vth
Th;" a test on a II En A-
regulated products.
uate-Po-stecf:-Norre-mbeJ -3(}, zfJrJ:T
u -a-ck-to-to-,...
I
1-
;
For More Information
1-
'
Sun Safety: Save Your Skin!
1-
I
Tanning
1-
i
Report to Congress: Labeling Information on the Relationship Between the Use of Indoor
!
Tanning Devices and Development of Skin Cancer or Other Skin Damage
1-
: What is Your Risk of Developing Skin Cancer? rlil
1-
.,1
http://www. fda .gov /ForConsu m ers/ConsumerU pdates/u em 186687. htm# Tan ninginChildrenandTeens[ 12/1 /2009 8:47:49 PM]
Page Last Updated: 11 /30/2009
, v , ,.
-
Home I About FDA I Contact Us I A to Z Subject Index I Site Map I Web Site Policies I FOIA I Accessibility I No FEAR Act
-
f--Combination Products 1 Advisory Committees 1 Science & Research 1 Regulatory Information 1 Safety 1 Emergency Preparedness 1 International Programs
News & Events 1 Training and Continuing Education 1 Inspections/Compliance 1 State & Local Officials 1 Consumers 1 Industry 1 Health Professionals
-
http://www. f da .gov /ForConsu m ers/ConsumerU pdates/u em 186687. htm# Tan ninginChildrenandTeens[ 12/1 /2009 8:47:49 PM]
DEPARTMENT OF HEALTH&. HUMAN SERVICES
AJG 2 I 1986
Public Health Service
Food and Drug Adm1n1str at1on
875 7 Georgia Avenue
Silver Spring MD 209 1 0
TO: ALL MANUFACTURERS, IMPORTERS AND POTENTIAL MANUFACTURERS OF
SUNLAMP PRODUCTS.
POLICY ON MAXIMUM TIMER INTERVAL AND EXPOSURE SCHEDULE FOR
SUNLAMP PRODUCTS.
BACKGROUND:
The amended performance standard for sunlamp products (21 CFR 1040. 20) was
published in the September 6, 1985 issue of the Federal Register and will
become effective September 8, 1986. Any sunlamp product manufactured on or
after that date must comply with the amended standard.
The ten (10) minute maximum timer interval requirement was removed
from the original performance standard since there are newer
products on the market for which ten (10) minutes is not appropriate.
The maximum timer interval now depends on the intensity and spectral
distribution of ultraviolet (UV) radiation emission of each individual
model of sunlamp product and must not exceed the maximum recommended
exposure time provided on the required product warning label. Therefore,
sunlamp product manufacturers must develop an exposure schedule and
establish the maximum recommended exposure ttme (and therefore the
maximum timer interval) based on the characteristics of their particular
products.
The intended purposes of a sunlamp product timer are to provide
for reliable control of exposures and to limit acute (and delayed) damage
from unintentionally long exposures. However, the maximum timer setting
should also allow for selection of exposure times needed to build. up and
maintain a tan. The maxJmum timer interval is in no way to be considered
as a safe limit; all ultraviolet radiation is potentially hazardous.
The standard requires the manufacturer to provide an exposure schedule
in the product warning label. The purpose of the exposure schedule
is to allow a person to gradually build-up skin pigmentation and to
maintain a tan while controlling the risk of acute in.1ury and delayed
adverse effects. Since the UV radiation dose that causes a barely
discernible pink coloration (minimal erythemal dose or KED) is not the same
for different skin types, the exposure schedule for first time users will
depend on the skin type of the user. Furthermore, doses
of UV radiation received at 24 hours intervals initially lead to lowering
of the erythema and tanning thresholds. Therefore, the exposure schedule
and maximum recommended exposure time should be constrained by the
potential for erythema as well as the quantity of radiation necessary to
achieve and maintain a tan.
POLICY:
The Center for Devices and Radiological Health (CDRH) will use the
following criteria to evaluate the adequacy of the exposure schedule and
the recommended maximum exposure time (and therefore the maximum timer
interval):
Page 2
1) The maximum recommended exposure time (and maximum timer
interval) must not exceed a value which will result in an exposure
of four (4) times the minimal erythema dose (MED) for untanned
Type II skin (always burns, then tans slightly). This is based on
the CDRH Erythema Action Spectrum fproposed action spectrum of
Commission Internationale de L'Eclairage (CIE) modified by CDRHl.
See Appendix A for the action spectrum and weighting factors and
equations needed to derive it.
The formula for determining
time,"T" in seconds is:
e
the recommended maximum exposure
Te
!ViR!
where Standard MED 156J/Mt
vi = weighting factor
2
Ri irradiance in W/M
at 296nm
2) The recommended maximum exposure time must not exceed a value
which will result in an exposure of four (4) times the minimal
melanogenic dose (MMD) for untanned Type II skin. This is
based on the melanogenic action spectrum developed by Parrish
et al (1982). See Appendix B for this action spectrum.
The formula for determining the recommended maximum exposure time,
"T " in seconds is:
m
where standard MMD 459J/M
2
at 296nm
Ji weighting factor
2
irradiance in W/M
3) The recommended exposure schedule should provide for exposures
of no more than 0.75 MED three times the first week, gradually
increasing the exposure following weeks until aximua tanning
has occurred (approximate,ly four weeks total) and then provide for
maintenance of a tan by biweekly or weekly exposures of up to
four(4) MEDs or four(4) MMDs, whichever is less.
CDRH believes that the above criteria balances the need to limit acute (and
delayed) damages from unintentionally long exposure and the need to provide
for single exposure durations adequate to achieve and maintain a tan.
Appendices

E. Gundaker, Director
Office of Compliance
Center for Devices and
Radiological Health
......
.. ..._ ,._ .. . ...

.. .... ...
Appendix A
i$'rtAFT
[DATA)GIEL Y1l..B ER.Yll(EMA, Vi
CIE AC'IlON SPC'TRUM MODlFlED BY LYTLE (CORH) NORMAUZED TO 1 AT
250-302 NUMlT IS 156 11M2
WAVELENGTH READING
250 l 306 .:W99 362 .S39122E-03
251 1 307 .2692 363 .Sl9SOI&03
252 301 :1117 364 .S00607&03
253 309 . 1S92 365 .41239S&03
254 310 .1225 366 .464144&03
25S 311 .09419 367 .447931&03
2S6 312 .714399.01 361 .431636&03
251 313 . .S571991i..ol 369 .415931643
251 l 314 .4214998-41 3'10 .40018-03
259 l 3JS .03296 371 .3162107&03
2a)
l 316 . 25:J.4991i..o 1 3n .3'n14SB-03
261 I 317 .019S 373 .3SI59Q!.43
262 I 311 .01499 374 .:J.4S5318-03
263 1 319 .01153 37S
. .332t518-03 .
264 I 3210 .11719942 376 .32013SB-03
26S I 321 .612299B4'1 377 .3091S4B-Ol
~ I 322 .005141 m .l9719184J
267 I 323 .00403CS 379 .21'7QSI B4:J
261 I 324 .31CM9!JB.42 310 .2'76fOIS03
269 I 325 .212499&02 311 .266S3643
2'10 I 326 .204771641 312 .256121643
271 I 327 . 197331842 313 .247479B43
l'n I 321 . 190161641 314 .231469S43
273 I 329 .113159B-Ol 3IS .229'719643
%74 1 330 . 176f01S.C 316 .22142AS-43
%7S I 331 1'10171842 317 .2ll363S03
%76 I 332 . 161911S.C 311 .2QSSMB4S
rn I 333 . 151014641
-
. 191UlB4J
271 I 3:J.4 . IS2263S.C
,.,
. lto!JB.CD
279 1 33S .146123B.Ql 391 .li39UB O:J
210 1 336 . 141313B.Ql m .l77267Jl..O:J
211 1 337 .1362318-02 393 . 170121S03
212 1 331 . 1312118-01
,.
.164GB O:J
213 I 4339
.126SCISS.C
,
.1516ll&.o:J
214 I 340 .001219
"'
. IS215284l
215 1 :J.41 . ll7461S.C
,.,
.14129JB.O:J
216 1 :J.42 .11Sli3S.C
,.
.14193QJ.03
217 1 343 . 10901:J.B.02
,
.136T748.QJ
211 1 :J.44 .lQ!OIB.Ol 10 . l31184J
219 1 :us .t0126lS.C
""'
1
2tO 1 :J.46 .f757Sss.O:J
291 I :J.47 .940221&0:J
191 I 341 .!10591S&O:J
29J 1
,..,
.lntJB-03
2M l 3s0 .1411998-0:J
295 l 351 .11QSI2B.O:J
2t6 1 352 .7110'19S03
297 1 3S3 .7Sl6tf4J
-
1 354 .125254&43
299 1 3S5 .6JII5QIQJ
300 1
"'
.611419B4J
301 1 3S7 ~
301 1 351 .6252JB.CD
303 .7691 3S9 .6G253B-OJ
31M .5916 360 .SI06.0:J
305 .. us 361 .5S9417JI..O:J
"r'4
...., 1.0-01
..
c:x:
0
f-4 .
u
<

t:)
z
H
f-4
. . . ::r:
- t:)

-
Appendix B
(page 1)
WEIGHTING FACTORS FOR MELANOGENESIS, Ji

450 300 350 400

. WAVELENGTH (nm)
The MMD as function of wavelength has interpolated
(using log MMD) fromt the action spectrum for melanogenesis
of type II skin et al 1982)
PARRISH MELANOGENSIS T'fPE II
W1\ VEJ..EliiGTH Ji
( rrn)
250 .378409 302
251 . 374828 303
252 . 37124R 304
253 .367714 305
254 .364225 306
255 .360783 307
256 .35734 308
257 . 353943 309
258 .350547 310
.. 259 .347196 311
260 .343891 312
261 .340632 313
262 . 337419 314
263 .334206 315
264 .331039 316
265 . 327672 317
266 .327413 318
267 .326954 319
268 .326449 320
269 .32599 321
270 .325531 322
271 .325072 323
272 .324613 324
273 .324154 325
274 .323695 326
275 .323236 327
276 .321445 328
277 .319609 329
278 . 317865 330
279 .316075 331
280 .314285

332
281 .312541 333
282 .31075 334
283 .351694 335
284 .398008 336
285 . 450427 337
286 .509732 338
287 .576885 339
288 .652851 340
289 738778 341
290 .836088 342
291 . 861518 343
292 .887498 344
293 . 91435 345
294 .94212 346
295 .970625 347
296 1 348
297 .990959 349
298 350
299 .973287 'l"il
300 .96429 352
301 .886993 353
SKIN 1982
.815892
. 750391
. 690261
. 502296
.36551
.265997
.193565
.14087
. 102497
.745893-{)1
. 054301
. 395016E-<l1
.341137-Ql
.294593E-<ll
254384-Q 1
. 219683-{)1
. 189709-01
.163821-01
. 141467-01
. 122143-Qt
. 10i481E-Q1
.911137E-Q2
. 786745E-02
.679336-{)2
.586616-{)2

.437483-{)2
. 377812-{)2
.326265F:-Q2
. 281741-{)2
.243276P.-o2
.210089-{)2
.181447E-Q2
.176123-{)2
.170982E-Q2
. 165978E-Q2
.161113-{)2
.156385-{)2
.151841-{)2
. 147388E-Q2
.143074-{)2
,l38897E-<l2
.134A12E-D2
. l30864E-Q2
.127054E-Q2
.123336-{)2
.11971E-Q2
.116:>.22E-Q2
.ll2825C:-02
. l0952E-G2
. l0fi307E-Q2
. 103232-{)2
Appendix B
(page 2)
NOPAALI7.F.:n ro 296 t-1'-f
354 .100202- 0?.
35C) .972644-{)3
35n . 944186E-Q3
357 . 916645-{)3
358 . 890022-0.3
359 .S63859E-<l3
360 .A38613E-<l3
361 .813826-{)3
362 7A9958E-Q3
363 767007-{)3
3n4 .747729-03
365 .722942-<l)
366 .666q43E-<l3
367 .615075..()3
36R .567338E-Q3
369 .523272-{)3
370 .48288-{)3
371 . 44547-03
372 .4l0953P.-03
373 .379097-{)3
374 .34972-{)3
375 .322593-{)3
376 .297577-{)3
377 .274534-03
378 .253236-f))
379 .233591-{)3
380 .215506- 03
381 .213S32P.-Q3
382 .2115SaP.-oJ
3RJ .209963E-<l3
384 .207702-03
3es .2o5821E-03
386 . 203939-03
387 .202057-{)3
388 .20022l.E-Q3
389 .189296-{)3
390 .196594-{)3
391 .194R04E-G3
392 .193014E-Q3
393 . 191224E-Q3
394 .18948E-G3
395 .187735-{)J
396 .1A6037E-Q3
397 .184339-Q]
398 .18264E-Q3
399 .1aoqa8r.-o1
400 .179336E-Q3
401 . 1776A3E-Q3
402 .176077E-Q3
403 .17447-{))
.172A64E-Q3
40S .17l257E-o 3
GuideS tar - Search Results - Report Page - INDOOR T ANNTNG AS SOCIA TTON I NC Page 1 of 2
O GuideStar.org.
Welcome
Y t) U are suhstribed 10 -- The following products --
Search Results I New Search
INDOOR TANNING ASSOCIATION INC
2025 M STREET. NW ROOM 800
WASHINGTON. DC 20036
G.ENERAL INFORMATION
Who WeAre
TO PROMOTE THE fNDOOR TANNING INDUSTRY
This organization is a
This otgani7.alit)r) is required to file an IRS fonn 990 or990-EZ.
Contributions are not deductible. as pro, ,ided by law.
NTEE Code
Socie1ie.s, Associations
EIN:
Y car Founded;
Ruling Year:

You must be a or Premnun subscriber to view this information.
Information not avaiJablc
2001
No. of Board Members: Infom1ation not available
No. of FullTime Employees: Information not :;iVai I able
No. ofPartTime Employccs:lnformation nm available
No. of VoJun1eers: Inforrnatjon not avaiJ(lble
PEOPLE
You must be a GuidcStar ScJcct or Premium subscriber to view this information.
MISSION AND PROGRAMS
Mission
TO PROMOTE THE INDOOR TANNING INDUSTRY
Programs
I. TO ADVANCE nu; BUSINESS GROWTH AND IMAGE OF TliE INDOOR TANNING INDUSTRY AND THE WELFARE OF ITS
MEMBERSHIP THROUGH THE CREATION OF A NEW LEVEL OF PROI'ESSIONALISM. VISIBLE AT ALL LEVELS, BY
COLLECTIVELY ACHIEVING PROPER GOALS AND OBJECTIVES WHICH COUW NOT BE ACHIEVED INDEPENDENTLY.

FORM 990 AND EDOCS
Fonns 990 from the IRS:
2006 Form 990 m
2005 Form 990 m
2005 Form 990 m
2004 Form 990 m
2004 Form 990 m
2JlQ:t&m.22ll m
My Tools
Subscribe Now
GuidcS111r Select
G\th1e;:Star Pren1ium

Data Services
Compc_nsatiO'__q_RcP.:Qrts
Sal:lry Sea.r<: h
O_vidc::Star for GrantnulkerS

eDocs
Power Searc.hes for t>rofessionals
Our search. sort, and retrieval options
save you time and energy. mo[
http://www. guides tar .org/pqShowGsReport .do ?partner=guidestar &npold= 1003 265 94 4/9/2008
GuideS tar - Search Results - Report Page - INDOOR T ANNTNG AS SOCIA TTON INC
Additional Documents from the
Available
AboullRS Form 990
To view this organization's IRS form 990. you must have the Adl)be Acrobat Reade,r instaJied on your system. If you don't have the Reader. you
can get it ;u the Adqb_e Web_si_te for free. Having trouble viewing PDF's'? T!)' these stem;.
FINANCIAL DATA
You must be a G._uig_eStat_Premium subscriber to vie.w tllis information.
Back to Top
http://www. guides tar .org/pqShowGsReport.do ?partner=guidestar &npold= 1003 265 94
Page 2 of2
4/9/2008
'.
:'
'I . I
,; '':f . . #' I U
. I . J
, __ L_ ---" ., . ..,._z: ...... -....-.....;:::....;; . _, __ -=-
- - ---...
)' I
# - -
; ... .. . ,.. .... .. _. , .. .,., ' -' e t , . ,_,', ,, . ,._._ {,,'-
.,
't . 1\
..,. ,. , .
,- : <; .. - . 0 . .. . " ' .. ._. . ..

l
-.
-i
4
!I

'1
.: I
;
(I
j
I
I
f
1.J
{
,
j
, I
!
I
I
. ,
I
I
......
....
1
I .
-t ..__
I'
i .... 'f
..
'. ......:. ..
..
.. --- .. -. .:-- ........ .
'
.
.. . .. :
.. ,_ .: .
... ....... ..... . . .. .,._ - .
, . ._,... ; ...
'1.,
.: ...'"'-: '',\,
r
. '

. ; ' .
. ;
,
. /IC..
i
t
j
l
'
\
I
I
l.
f
1
I
I
I'
I
I
I
. I
l
J ~
Online article and related content
current as of December 10, 2008.
Correction
Citations
Related Articles published in
the same issue
Subscribe
http://jama.com/subscribe
Permissions
Randomized Trials of Antioxidant Supplementation for
Cancer Prevention: First Bias, Now Chance Next, Cause
Peter H. Gann
JAMA. publ ished onl ine Dec 9, 2008; (doi :1 0.1 001/jama. 2008.863)
http://jama.ama-assn.org/cgi/contentlfull/2008.863v1
Contact me if this article is corrected.
Contact me when this article is cited.
Vitamins E and C in the Prevention of Prostate and Total Cancer in Men: The
Physicians' Health Study II Randomized Control led Trial
J . Michael Gaziano et al. JAMA. 2008;0(2008):2008862.
Effect of Selenium and Vitamin E on Risk of Prostate Cancer and Other Cancers:
The Selenium and Vitamin E Cancer Prevention Trial (SELECT)
Scott M. Lippman et al. JAMA. 2008;0(2008) :2008864.
Email Alerts
http://jamaarchives.com/alerts
permissions@ama-assn.org
httpJ/pubs.ama-assn.org/misc/ permissions.dtl
Repri nts/ E-prints
repri nts@ama-assn.org
Downloaded from www.jama.com at Federal Trade Commission on December 10, 2008
- EDlTORIAL
Editorials represent the opinions
of the authors and lAMA and
not those of the American Medical Association.
Randomized Trials of Antioxidant
Supplementation for Cancer Prevention
First Bias, Now Chance-Next, Cause
Peter II . Gann, MD, cD
---------------------
I
N 1996, A WAVE OF !lOPE AROSE WHEN THE NUTRJTlONAL
Prevention of Cancer trial reponed a 65% reduction in
prostate in men receiving selenium supple-
mentatiOn. Thrs came only 2 years after the A TBC
(a-Tocopherol, l3eta Carotene) Cancer Prevention Trial had
reported a 3 5% reduction in prostate cancer occurrence among
men taking vitamin E supplements.
2
Suddenly, it appeared to
make sense that this cancer could be prevented by bolstering
antioxidant defenses in middle-aged and older men. Prostate
cancer was not a prespecified end point in either trial, and
although both results were based on post hoc analysis, random-
ization had worked and it seemed unlikely that the encourag-
ing findings were due to confounding bias. Indeed, confound-
ing stood as the likely explanation for discordance between the
earlier beta carotene trials that failed to demonstrate any ben-
efit and numerous observational smdies reporting that men who
consumed beta carotene-containing foods were at lower risk
for developing cancer, particularly those of the respiratory tract.
Now, 12 years later, comes the disappointing news that 2 ma-
jor trials conceived during the wave of hope found that neither
selenium nor vitamin E supplementation, alone or in combi-
nation, produced any reductions in prostate cancer or cancer
of any type. The results of these important studies, the SELECT
(Selenium and Vitamin E Cancer Prevention Trial) by Lippman
and colleagues and the Physicians' Health Study (PHS) IJ by
Gaziano and colleagues. are reported in this issue

Despite the null findings, it is important to recognize what
these trials have accomplished. SELECT had a simple, cost-
effective design, completed accrual of more than 35 000 par-
ticipants ahead of schedule-making it the largest individu-
ally randomized cancer prevention trial ever conducted-
and maintained high rates of adherence and retention for 4
to 7 years. Given its statistical power, it is unlikely that the
study missed detecting a benefit of even a very modest size.
The PHS II, \vith more than 14 000 male physicians, was also
remarkably well conducted and was especially noteworthy for
its cost-efficiency, with most of the study having been con-
ducted through the mail. Neither study has delivered its final
word on the blinded, randomized results. Preplanned sub-
group analyses are expected, particularly those aimed at base-
See also related articles.
2009 American Medical Associalion. All rights rcsei:'Ved.
line serum levels, smokingstams, and genetic factors that might
have modifi ed response. After that, like Voyager space probes,
each study will continue to contribute a wide range of valu-
able data as observational cohorts, before they fade away.
What can clinicians, researchers, and men concerned about
pros tate cancer prevention learn from the main results of these
phase 3Lrials? Three overarchingpoi.ntsare worth consideration.
Firs t, widespread use of prostate-specific antigen (PSA)
testing has s ubstantially restricted the design and interpre-
tation of prostate cancer prevention trials, perhaps perma-
nently. Neither trial mandated regular PSA testing; thus, end
points were determined by routine clinical management. This
greatly reduced the cost of each study; however, off-study
PSA testing was extremely common among these healthy,
motivated volunteers who had good access to care. In
SELECT, 83% to 86% of men reported receiving PSA tests
each year. Because of this intense surveillance, the geomet-
ric mean PSA at diagnosis was low ( <4.81 ng/mL) and, more
importantly, the proportion of cases with nonlocalized tu-
mors was very low. Of 1758 prostate cancers detected, at
diagnosis only 10 were extracapsular, l was regional node
positive, and 9 were metastatic. Even assuming that these
represent nonoverlapping cases, that would amount to only
20 nonlocali zed cancers, or L 1%. By comparison, among
l3 740 cases of pros tate cancer reported to the CAPSURE
database from 1990 to 2007, 3.7% were at clinical stage
T3bNOMO or greater, excluding T3a cases with extracap-
sul ar extension only.' Perhaps more strikingly, there was
only 1 death from prostate cancer in SELECT despite nearly
200 000 person-years of observation. Even allowing for the
requirement of a normal PSA and rectal examination at en-
try, that is an enormous deficit in expected mortality. Ap-
plying conservative assumptions about the exact age and fol-
low-up time distributions in SELECT, approximately 75 to
100 deaths would have been expected based on the US age-
specific prostate cancer mortality rates for 2000 through
2005.
6
There also were indications of a deficit in advanced
prostate cancer in PHS II, although a much smaller one.
How can an agent be shown to prevent serious, clinically sig-
nificant prostate cancers when PSA testing may be rapidly re-
moving those cancers from the population at risk before they
progress? The effects on early stage lesions may still be observed,
Author Affiliation: Department of Pathology. University of Illinois at Chicago.
Corresponding Author: Peter H. Gann, MD, SeD, Department of Pathology. Univer-
sity of Illinois at Chicago, 840 S Wood St, Chicago, IL 60612 (pgann@uic.edu).
(Reprinted) JAMA, Published onhne December 9, 2008 E1
EDlTORlAL
but if an agent has important effects on progression, those ef-
fects may be obscured. What to do? Trials could be conducted
only in populations that shun PSA testing, or mandatory biop-
sies could be performed after some reasonable interval of ex-
posure. Mandatory biopsies have the advantage of at least gath-
ering end points more quickly and also solve problems due to
the influence of an agent on PSA levels, as seen for finasteride
and now, apparently, statins.
7
Prostate cancer detection rates
are exquisitely sensitive to any factor that affects PSA levels, bi-
opsy rates, or biopsy sensitivity independent of an anticancer
effect. ln that regard, the nonsignificant increase in diabetes (pre-
sumably type 2) in the selenium group of SELECT deserves fur-
ther scrutiny because of the recognition that obesity has a small
suppressive effect on PSA and may have depressed prostate can-
cer ascertainment.
8
The increase in prostate cancer in the vi-
tamin E group, although not statistically significant (P= .06),
is also worrisome. However, this could have been due to a chance
elevation of the biopsy rate or biopsy sensitivity in that group,
and the absence of any prostate cancer increase in the combi-
nation group in SELECT or in PHS II is reassuring.
Second, single-agent interventions, even in combinations,
may bean ineffective approach toprimaryprevention in average-
risk populations. It may be time to give up the idea that the pro-
tective influence of diet on prostate cancer risk- which is clearly
observed in migrant studies and in populations transitioning
to a Western diet- can be emulated by isolated dietary mol-
ecules given alone or in combination to middle-aged and older
men. Evidence from the Prostate Cancer Prevention Trial sug-
gests that pharmacological inhibition of intraprostatic andro-
gen is capable of retarding the early growth of prostate cancers
9
Forcostandsafetyreasons, it is unreasonable, however, to rec-
ommend that all men begin takinga5a-reductase inhibitor daily
once they reach age 50 years. On the other hand, nonpharma-
cological dietary prevention of prostate cancer is probably more
complex and may involve certain inconvenient truths. Fortu-
nately, no dietary change this profound is likely to be benefi-
cial for prostate cancer alone. If it requires whole foods, extracts,
or dietary patterns, it may be necessary to give up the reduc-
tionist need to know which molecule is most responsible and
perhaps give up the notion of placebo controls as well. If it re-
quires starting eJ>rposure early in life and sustaining it for de-
cades, it may mean having to give up the idea of phase 3 trials
altogether. This does not mean that whole food or complex mix-
ture studies cannot be sound and biologically based. One can
start, for example, by asking how diet affects intra prostatic an-
drogen activity, a proven chemopreventive mechanism.
Third, it may be time to critically examine the methods
used to vet hypotheses for some phase 3 trials. In hind-
sight, and now with more recent information available, it
is easy to second-guess the rationale for these trials but more
difficult, and much more important, to determine how to
develop better phase 3 studies. Since the mid-l990s the tools
available for preclinical science, including new transgenic
models and genomic technologies, have vastly improved.
Phase 2 trials, which should be determining which few in-
E2 JAMA, Published online December 9, 2008 (Reprinted)
terventions make it through the pipeline, will improve as
intermediate end-point biomarkers become validated. More-
over, improvements in patient risk stratification, notably
through multi-single-nucleotide polymorphism genetic test-
ing,
10
will help design targeted trials with higher statistical
efficiency and more favorable risk-to-benefit ratios.
It now appears more likely that the unexpected results on
prostate cancer from the earlier selenium and vitamin E trials
were due to chance. In fact, the final report covering the entire
blinded treatment period of the Nutritional Prevention of Can-
cer Trial, published in 2003, found that for unexplained rea-
sons (chance), participants in theseleniumgroupwere less likely
than those in the placebo group, by 14% vs 35%, to undergo
biopsy when the PSA level was elevated.
11
After taking this into
account, the reduction in prostate cancer incidence was restricted
to the subgroup with low baseline PSA levels and the lowest
tertile of plasma selenium, a subgroup comprising fewer than
17 cases between treatment groups combined.
Epidemiology teaches that every statistical association bas
only 3 possible explanations: bias, chance, and cause. Regard-
ing nutritional prevention of prostate cancer, first-
generation phase 3 trials were too reliant on biased interpre-
tation of prior research; second-generation trials may have been
too reliant on chance; yet there is every reason to believe that
the next generation will have a firmer basis for causal hypoth-
eses. Until then, physicians should not recommend sele-
nium or vitamin E-or any other antioxidant supple-
ments-to their patients for preventing prostate cancer.
Published Online: December 9, 2008 (doi:10. 1001 /jama.2008.863)
Financial Disclosures: None reported.
REFERENCES
1. Clark LC, Combs GF Jr, Turnbull BW, et al; Nutritional Prevention of Cancer
Study Group. Effects of selenium supplementation for cancer prevention in pa
tients with carcinoma of the skin: a randomized controlled trial. lAMA. 1996:
276(24) 1957-1963.
2. The Alpha Tocopherol, Beta Carotene Cancer Prevention Study Group. The effect
of vitamin E and beta carotene on the incidence of lung cancer and other cancers
in male smokers. N Eng/ 1 Med. 1994;330(15):1029 1035.
3. Lippman SM. Klein EA, Goodman PJ, et al. Effect of selenium and vitamin Eon
risk of prostate cancer and other cancers: the Selenium and Vitamin E Cancer Pre-
vention Trial (SELECT). lAMA. doi:10.1001 /jama.2008.864.
4. Gaziano JM, Glynn RJ, Christen WG, et al. Vitamins E and C in the prevention
of prostate and total cancer in men: the Physicians' Health Study II randomized
controlled trial. lAMA. doi:10.1001/jama.2008.862.
5. Cooperberg MR, Cowan J, Broeri ng JM, Carroll PR. High-risk prostate cancer
in the United States. 1990-2007. World 1 Ural. 2008;26(3):211 -218.
6. Cancer statistics from the Surveillance, Epidemiology, and End Results (SEER)
program. National Cancer Institute, National Institutes of Health. http:/ /seer. cancer
.gov/statistics/. Accessed November 25, 2008.
7. Hamilton RJ, Goldberg KC, Platz EA, Freedland SJ. The influence ofstatin medi
cations on prostate-specific antigen levels. 1 Nat/ Cancer lnst. 2008;1 00(21):
1511-1 518.
8. Banez LL, Hamilton RJ, Partin AW, et al. Obesity-related plasma hemodil ution
and PSA concentration among men with prostate cancer. lAMA. 2007;298
(19):2275-2280.
9. Thompson IM , Goodman PJ, Tangen CM, et al. The infl uence of finasteride on
the development of prostate cancer. N Eng/ 1 Med. 2003;349(3):215-224.
10. Zheng SL. Sun J, Wiklund F, et al. Cumulative association of five genetic vari
ants with prostate cancer. N Eng/ 1 Med. 2008;358(9) :910-919.
11. Duffield-Lillico AJ, Dalkin BL, Reid ME, et al; Nutritional Prevention of Cancer
Study Group. Selenium supplementation. baseline plasma selenium status and in-
cidence of prostate cancer: an analysis of the complete treatment peri od of the
Nutritional Prevention of Cancer Trial. BlU Int. 2003;91 (7):608 612.
2009 American Medical Association. AU rights reserved.
J ~
Correction
Citations
the same issue
Subscribe
http://jama.com/ subscribe
Permissions
Effect of Selenium and Vitamin Eon Risk of Prostate
Cancer and Other Cancers: The Selenium and Vitamin E
Cancer Prevention Trial (SELECT)
Scott M. Lippman; Eric A. Klein; Phyllis J. Goodman; et al.
JAMA. publ ished online Dec 9, 2008; (doi :1 0.1 001/jama.2008.864)
http://jama.ama-assn.org/cgi/content/full/2008.864v1
Contact me if this arti cle is corrected.
This article has been cited 1 time.
Vitamins E and C in the Prevention of Prostate and Total Cancer in Men: The
Physicians' Health Study II Randomized Control led Trial
J . Michael Gaziano et al. JAMA. 2008;0(2008):2008862.
Randomized Trials of Antioxidant Supplementation for Cancer Prevention: First
Bias, Now Chance Next, Cause
Peter H. Gann. JAMA. 2008;0(2008) :2008863.
Email Alerts
http://jamaarchives.com/ alerts
http://pubs.ama-assn.org/misc/permissions.dtl
Repri nts/ E-prints
reprints@ama-assn.org
- ORIGINAL CONTIUBUTION JAMA-EXPRESS
Effect of Selenium and Vitamin Eon Risk
of Prostate Cancer and Other Cancers
The Selenium and Vitamin E Cancer Prevention Trial (SELECT)
Scott M. Lippman, MD
Eric A. Klein, MD
Phyllis J. Goodman, MS
M. Scott Lucia, MD
Tan M. Thompson, MD
Leslie G. Ford, MD
Howard L. T,arncs, MD
Lori M. Minasian, MD
J. Michael Gaziano, :MD, MPH
JoAnn Hartline, MPH
J. Kellogg Parsons, MD, MHS
James D. Bearden TTl, MD
E. David Crawford, 1\'ID
Gary E. Goodman, MD
Jaime Claudio, MD
Eric Winquist, MD, MSc
Elise D. Cook, MD
Daniel D. Karp, MD
Philip Walther, MD
M.ichael M. Lieber, 1\ID
Alan H. Kristal, DrPH
Amy K. Darke, MS
Kathryn B. Arnold, MS
Patricia A. Ganz, MD
Regina M. Santella, PhD
Demetrius Alhanes, MD
Philip H. Taylor, MD, SeD
Jeffrey L. Probstfield, MD
T. J. Jagpal, CCRP
John J. Crowley, PhD
FrankL. Meyskens Jr, MD
Laurence H. Baker, DO
Charles A. Coltman Jr, MD
Context Secondary analyses of 2 randomized controlled trials and supportive epi-
demiologic and precl inical data indicated the potential of selenium and vitamin E for
preventing prostate cancer.
Objective To determine whether selenium, vitamin E, or both could prevent pros-
tate cancer and other diseases with little or no toxicity in relatively healthy men.
Design, Setting, and Participants A randomized, placebo-controlled trial (Sele-
nium and Vitamin E Cancer Prevention Trial [SELECD) of 35 533 men from 427 par-
ticipating sites in the United States, Canada, and Puerto Rico randomly assigned to 4
groups (selenium, vitamin E, selenium + vitamin E, and placebo) in a double-bli nd
fashion between August 22, 2001, and June 24, 2004. Baseline eligibility included age
50 years or older (African American men) or 55 years or older (all other men), a serum
prostate-specific antigen level of 4 ng/mL or less, and a digital rectal examination not
suspicious for prostate cancer.
Interventions Oral selenium (200 (Jg/d from L-selenomethionine) and matched vi -
tami n E placebo, vitamin E (400 IU/d of all rac-a-tocopheryl acetate) and matched
selenium placebo, selenium + vitamin E, or placebo + placebo for a planned fol-
low-up of minimum of 7 years and a maximum of 12 years.
Main Outcome Measures Prostate cancer and prespecified secondary outcomes,
including lung, colorectal, and overall primary cancer.
Results As of October 23, 2008, median overall follow-up was 5.46 years (range,
4.17-7.33 years). Hazard ratios (99% confidence intervals [Cis]) for prostate cancer
were 1.13 (99% Cl , 0.95-1.35; n=473) for vitamin E, 1.04 (99% Cl, 0.87-1 .24; n=432)
for selenium, and 1.05 (99% Cl, 0.88-1 .25; n =437) for selenium+ vitamin E vs 1.00
(n =416) for placebo. There were no significant differences (all P> .15) in any other
prespecified cancer end points. There were statistically nonsignificant increased risks
of prostate cancer in the vitamin E group (P= .06) and type 2 diabetes mellitus in the
selenium group (relative ri sk, 1.07; 99% Cl , 0.94-1.22; P=.16) but not in the sele-
nium + vitamin E group.
Conclusion Selenium or vitamin E, alone or in combination at the doses and formula-
tions used, did not prevent prostate cancer in this population of relatively healthy men.
Trial Registration clinicaltrials.gov identifier: NCT00006392
lAMA. 2009;301(1):(doi:10.1001/jama.2008.864)
P
ROSTATE CANCER MORTALITY LN
the United States has declined
in recent years, but this can-
cer remains the most com-
mon nons kin epithelial malignancy in
US men, with 186 320 new cases and
28 660 deaths (the second leading cause
www.jama.com
Author Affiliations are listed at the end of this
article.
Corresponding Author: Scott M. Lippman, MD, De-
partment of Thoracic and Head and Neck Medical On-
cology. University of Texas M . D. Anderson Cancer
Center, 1515 Holcombe Blvd, Uni t432, Houston, TX
77030-4009 (slippman@mdanderson.org), or Eric A.
Klein, MD, Cleveland Clinic Lerner College of Medi
cine, Desk A100, 9500 Euclid Ave. Cleveland. OH
44195 (kleine@ccf.org) .
2009 American Medical Association. All rights reserved. (Reprinted) JAMA, Published online Dewnber 9, 2008 E1
SELENIUM AND VITAMIN E FOR CANCER PREVENTION
of cancer death) estimated for 2008.
1
An effective prevention strategy for
prostate cancer would have substan-
tial public health benefits, including the
potential to reduce the incidence ofbio-
logically indolent prostate cancer,
which is significantly overdetected by
widespread screening with prostate-
specific antigen (PSA) and for which
most newly diagnosed men still un-
dergo curative-intent therapy involv-
ing substantial morbidity despite sur-
gical and other advances.
2
6
Important secondary results of 2 ran-
domized controlled trials, the Nutri-
tional Prevention of Cancer (NPC)
s tudy and the Alpha-Tocopherol, Beta-
Carotene Cancer Prevention (ATBC)
study, showed prostate cancer risk re-
ductions of 63% for selenized yeast and
32% for a -tocopherol (or vitamin E) J-to
ln addition, a large-scale randomized
controlled trialu involving several dif-
ferent regimens found that a combina-
tion of selenium, vitamin E, and beta
carotene reduced overall cancer mor-
tality. These clinical data, supported by
epidemiologic and preclinical data,
12
-
19
led to the design of the Selenium and
Vitamin E Cancer Prevention Trial
(SELECT)
1 0
Investigators in the United States and
Canada from major cooperative groups
of the National Cancer Institute and De-
partment of Veterans Affairs used the
Prostate Cancer Prevention Trial
(PCPT) accrual infrastructure (200
clinical sites, with 18 882 randomized
men) in designing and activating
SELECT. We report herein the effects
of selenium and vitamin E, alone or in
combination, on the risk of prostate
cancer and secondary end points in
SELECT.
METHODS
Study Design
SELECT is a phase 3 randomized, pla-
cebo-controlled trial of selenium (200
pg/d from L-selenomethionine), vita-
min E ( 400 lU/d of all rac-a-tocoph-
eryl acetate), or both (planned fol-
low-up of minimum of 7 years and
maximum of 12 years) for prostate can-
cer prevention. The major eligibility re-
quirements included age 50 years or
older for African American men and 55
years or older for all other men, no prior
prostate cancer diagnosis, 4 ng/mL or
less of PSA in serum, and a digital rec-
tal examination (DRE) not suspicious
for cancer. No current use of antico-
agulant therapy other than 175 mg/d or
less of acetylsalicylic acid or 81 mg/d
or less of acetylsalicylic acid with clo-
pidogrel bisulfate, no history of hem-
orrhagic s troke, and normal blood pres-
sure were also required because of
antiplatelet effects of vitamin E andre-
lated findings of the ATBC study.
Participant characteristics were based
on self:. report, including self-identification
of race and ethnicity which were defined
by the US Census Bureau. Race and eth-
nicity data were collected mainly for the
generalizability of trial results. All poten-
tially eligible men were required to pro-
vide written informed consent before
being allowed to participate in the trial.
The local institutional review board of each
study site approved the study for activa-
tion and reviewed its progress annually.
The trial was activated in july 2001 and
follow-up blinded to the trial results ended
on October 23, 2008.
Baseline blood and toenail speci-
mens and a 5-year blood sample were
collected for future biological studies.
Prostate tissue samples collected dur-
ing the trial were submitted for confir-
mation by central pathology review (no
samples were collected at baseline). Par-
ticipants without prostate cancer had
clinic visits once every 6 months
throughout the trial; with prostate can-
cer, annually. Adherence and adverse
events were monitored every 6 months
and a limited physical exami nati on in-
cluding assessments of blood pres-
sure, weight, and smoking status was
conducted annually. Prespecified ad-
verse events known to be associated
with vitamin E or selenium were graded
according to the National Cancer In-
stitute Common Toxicity Criteria.
Although eligible PSA and DRE re-
sults were r equired at study entry, an-
nual prostate cancer screemng with PSA
and DRE was not mandatory because
the benefits of this screening were un-
der debate when the trial opened and
community screening standards were
expected to change during the trial. Par-
ticipants were recommended during an-
nual clinic visits to undergo a PSA test
and DRE according to the s tandard of
care at their study sites and the partici-
pant' s preferences. A formal preran-
domization period (28-90 days; no pla-
cebo run-in capsules) gave potential
participants time to decide if they would
agree to stop disallowed over-the-
counter supplements of selenium or vi-
tamin E throughout the study and to
demonstrate, by returning for random-
ization, their willingness to adhere to
the trial. Other adherence methods in-
cluded offering each participant a free
multivitamin containing no selenium
or vitamin E and assessi ng serum lev-
els of vitamin E and selenium in all par-
ticipants at a subset of study sites (22
sites representing 7.8% of the trial
population). These sites were chosen
a priori to be representative of the broad
range of sites in the trial.
End Point Assessment
Participants reported prostate cancers
to the study site staff. Study staff ob-
tai ned medical records supporting the
diagnosis and abstracted the diagnos-
tic method and clinical stage. Tissue and
the corresponding pathology report
were sent to the central pathology labo-
ratory for confinnation. Gleason Score
was based on central pathology review.
Men were asked at their first 6-month
clinic visit to report new events since
entering the trial and thereafter to re-
port new events since their last visit.
Cardiac-event data were collected in de-
tail from the trial beginning (2001); data
on diabetes were added through self-
reported glitazone medication use (be-
ginning in 2003) and self-report of dia-
betes (beginning in 2005) via the
following question at each clinic visit:
"Does the participant report having dia-
betes (either his doctor told him he has
diabetes or he is taking medication for
diabetes)7"
A general question regarding any
events considered severe or life-
threatening (grade 3 or 4), regardless
E2 JAMA, Published online December 9, 2008 (Reprinted) 2009 American Medical Association. AU rights reserved.
of attribution to the study supple-
ments, was also asked. A Social Secu-
rity Death Index search was con-
ducted in july 2008 for participants who
had a last contact date of more than 18
months before the search. Other spe-
cifically queried events (known at study
inception to be related to either of the
study supplements) included alope-
cia, dermatitis, fatigue, halitosis, nail
changes, and nausea.
Statistical Analysis
The prin1ary end point was prostate can-
cer incidence as determined by routine
clinical management Cancers that were
not confirmed centrally were included
in the analysis. SELECT was designed as
a 4 group trial with 5 prespecified com-
parisons (selenium vs placebo, vitamin
E vs placebo, selenium + vitamin E vs
placebo, selenium vs selenium + vita-
minE, and vitamin E vs selenium + vi-
tamin E). With a sample size of 32 400
men, using a 1-sided a =.005 level
(equivalent to a 2-sided a=.01 level) ,
there was 96% power to detect a 25% re-
SELENIUM AND VTTAMIN E FOR CANCER PREVENTION
duction in prostate cancer for either of
the single agents ( vs placebo), 89% power
to detect a 25% reduction for selenium
+ vitamin E ( vs an active single agent)
and more than 99% power to detect a
44% reduction of selenium + vitamin E
(vs placebo).
Design assumptions were based on
the PCPT, ATBC, and NPC trials.
The details of the statistical design
have been described elsewhere
20
Important elements included (1)
constant accrual over 5 years; (2)
prostate cancer incidence in the pla-
cebo group based on PCPT for the
first 3 years and the 1995 Puget
Sound SEER registry afterward; (3)
adherence to the study supplements,
which was assumed to decrease over
the course of the trial with a 5-year
rate of 68% and 12-year rate of 51%;
(4) a constant 10% drop-in rate.
defined as participants receiving pla-
cebo who are taking active supple-
mentation off-study; (5) loss to
follow-up of 0.5% per year; and (6)
deaths estimated from PCPT for
years 1 to 3 and from the 1995 US
standard rates of men aged 63 years
and all races for year 4 onward. The
sample size was calculated to be
32 400 men and the number of pros-
late cancers expected in the placebo
group was 533 over l2 years. Under
the assumed conditions, the required
median time under observation was
estimated to be 8.8 years.
The primary analysis consisted of the
5 prespecified comparisons detailed
above. These comparisons allowed for
a meaningful analysis of the study re-
sults whether an interaction between vi-
tamin E and selenium occurred. Each
individual test was conducted at a
1-sided a =.005 level (equivalent to a
2-sided a =.01 level) using a Bonfer-
roni factor of 5 to preserve an overall
1-sided a=.00251evel (equivalent to a
2-sided a=.05 level) .
An independent data and safety
monitoring committee met yearly and
reviewed data on safety, adherence, and
diagnosis of prostate cancer. In addi-
tion to the flnal analysis, interim analy-
Figure 1. Flow of Participants Included in Analysis by Intervention Group
c
35533 Men randomized at 427 participating sites
~
~ . . . . . . .
----
8856 Randomized to receive placebo 8904 Randomized to receive vitam1n E + I 8910 Randomized to receive selenium 8863 Randomized to receive selenium
+placebo placebo + placebo +VItamin E
8696 Received placebo + placebo 8737 Received vitamin E + placebo 8752 Received selenium+ placebo 8703 Received selenium + placebo
as randomized as randomized as randomized as randomized
160 Excluded 167 Excluded 158 Excluded 160 Excluded
155 Removed from 2 156 Removed from 2 155 Removed from 2 155 Removed from 2
participating Sites {poor participating sites {poor participating sites (poor participating sites (poor
data and participant data and participant data and participant data and participant
management and management and management and management and
regulatory issues) regulatory issues) regulatory Issues) regulatory issues)
5 Ineligible 11 Ineligible 3 Ineligible 5 Ineligible
1 Had prior prostate 5 Had prior prostate 1 Had prior prostate 2 Had prior prostate
cancer cancer cancer cancer
4 Randomized in error 6 Randomized in error
1
2 Randomized in error 3 Randomized in error
{never received prope< (neve< received proper (neve< received proper (never received proper
informed consent) informed consent) informed consent) informed consent)
I
133 Clinically Ineligible" 128 Chrucally 1nehgible" I 1 t 3 CliniCally Ineligible' 113 Cfinicatly ineligible
154 Insufficient baseline data to completely 151 lnsufficoent baseline data to completely 1 66 Insufficient baseline data to completely 169 Insufficient baseline data to completely
evaluate dinicat ellgibllitl' evaluate clinical eligibility evaluate clinical eligibility evaluate clinical etig,bilitl'
420 Lost to fotlowup (last contact data 385 Lost to follow up Qast contact data
I
434 Lost to follow-up (last contact data 379 Lost to follow-up (last contact data
>24 mo before analysis)" >24 mo before analysis)< >24 mo before anatysis)
0
>24 mo before anatysis)
0
I
8696 Included in primary analysis
I
8737 lnctuded in primary analysis
I
8752 tnctlXled in ptimary analysis 8703 Included in primary analysis
aoue to increased blood pressure, highgrade prostatic intraepithelial neoplasia, suspicious digi tal rectal examination (DRE) or i ncreased prostate-specific antigen (PSA),
aspirin dosage, prior cancer less than 5 years before randomization, participation in another clinical trial . or other clinical reason.
bslood pressure, PSA, and/or DRE not performed within required time f rame (but normal) or other data-related reason.
cAll data up until the last contact are included; these men also could have been either clinically ineligible or had insufficient baseline data. For time-to-event analyses,
these men were censored at their last follow up.
ses were planned for years 5, 7, 9, 10,
and 11 after the first participant was
randomized; the percentages of the ex-
pected total number of prostate can-
cer events in the placebo group at each
interval were 14%, 35%, 61%, 7 4%, and
88%, respectively. Each interim analy-
sis resulted in recommendations that
could have included modifications to
the study, including termination of ac-
cntal , modifications to data collec-
tion, or early reporting of results. Rec-
ommendations were made to the
steering conm1ittee, which made the fi-
nal decisions.
The interim analyses tested the null
hypothesis at a 1-sided <X=.0005 level
(equivalent to a 2-sided a =. 001 level)
using the Cox proportional hazards re-
gression model. In addition, the alter-
native hypothesis of a 25% reduction in
prostate cancer incidence was tested at
a 1-sided level of a =.0005 (equivalent to
a 2-sided a=.001level) using an exten-
sion of the Cox proportional hazards re-
gression model that allows for testing a
relative risk (RR) not equal to l. The pur-
pose of the second analysis was to al-
low for the study to stop if it was deter-
mined that the expected reduction in
prostate cancer would not be observed.
The frequencies of the number of car-
diovascular events and cases of diabe-
tes were tested with a X
2
test. For car-
diovascular event and diabetes analyses,
we did not capture the report of the date
Table 1. Baseline Characteristics of Study Participants
of the event, which thus was not incor-
porated into the analysis.
Participants were randomized in a
randomized block scheme, in which
the block was the study s i te. This
ensured a balance of the 4 interven-
tion groups within each study site. All
analyses were performed by using an
intention-to-treat analysis in which
men were classified according to the
group to which they were random-
ized. All men were followed up until
death or loss to follow-up. For cancer
end points, men were censored at the
time of their last follow-up or death.
The analysis did not incorporate
adjustments for baseline covariates.
Data were analyzed by using SAS ver-
No. (%) of Participants
Placebo Vitamin E Selenium
Characteristics (n = 8696) (n = 8737) (n = 8752)
Age. y
Median (interquartile range) 62.6 (58.1 -67.8) 623 (58.0-67.8) 62.6 (58.2-68.0)
50-54 355 (4) 402 (5) 337 (4)
55-64 5078 (58) 5143 (59) 5076 (58)
6574 2702 (31) 2641 (30) 2733 (31)
<=;75 561 (6) 551 (6) 606 (7)
Race/ethnicity
White 6863 (79) 6890 (79) 6942 (79)
African American 1078 (12) 1107 (13) 1053(12)
Hispanic (non-African American) 492 (6) 477 (5) 481 (5)
Hispanic (African American) 76 (1) 103 (1) 86 (1)
Qther
8
187 (2) 160 (2) 190 (2)
Education (highest level)
s:High school graduate or GED 1993 (23) 1875 (22) 1917 (22)
Some college/vocational school 2291 (26) 2387 (27) 2327 (27)
<=;College graduate 4317 (50) 4394 (51) 4430 (51)
Unknown/missing 95 (1) 81 {1) 78(1)
PSA, ng/ml
0.1-1 .0 4122(47) 4208 (48) 4218 (48)
1.1-2.0 2728 (31) 2653(30) 2661 (30)
2. 1-3.0 1168 (13) 1228 (14) 1211 (14)
3.1-4.0 666 (8) 634 (7) 652 (7)
>4.0 5(<1) 3(<1) 2 (<1)
Unknown/missing 7 (< 1) 11 (<1) 8(<1)
Smoking status
Never 3682 (42) 3752 (43) 3780 (43)
Current 655 (8) 659(8) 631 (7)
Former 4208 (48) 4194 (48) 4214 (48)
Ever (unknown status) 63 (1) 55 (1) 61 (1)
Unknown 88 (1) 77 (1) 66 (1)
Abbreviations: GED, general equivalency diploma; PSA, prostate-specific antigen.
Sl conversion: To convert PSA to !Jg/l, multiply by 1 .0.
a o ther race/ethnicity include Asian (n= 420), Native Amencan (n = 99), Pacific Islander (n= 39), muniple races (n =34), and unknown (n = 1 1 9).
Selenium +
Vitamin E
(n = 8703)
62.4 (58.1-67.8)
385(4)
5052 (58)
2731 (31)
535 (6)
6874 (79)
1076 (12)
484 (6)
95 (1)
174 (2)
1898(22)
2348(27)
4372 (50)
85(1)
4213(48)
2666 (31)
1149(13)
659(8)
1 (<1)
15(<1)
3666(42)
670 (8)
4242 (49)
56 (1)
69(1)
sion 9.1 (SAS Institute Inc, Cary,
North Carolina).
Supplement Quality Control
and Quality Assurance
The Pharmacy Coordinating Center
received the study supplements for
bottling as finished capsules in
shipments containing lots of ac-
tive capsules along with the ap-
propriate matching placebo. As
required by current good manufac-
turing practice,ll each lot of capsules
was quarantined upon receipt until
testing was perfom1ed to ensure that
capsules labeled "active" by the
manufacturer contai ned the ap-
propriate active agent and that cap-
sules labeled as "placebo" did
not contain an active agent. ln addi-
tion, each time the capsules were
bonled, production-run-verification
testing was performed to ensure that
bottles labeled as an active agent or
placebo contained the appropriate
material. To ensure that the quality
of the blind was maintained, cap-
sules received in each subsequent
lot were compared with the previous
lot and with matching capsules in
the current shipment for their char-
acteristics of weight, shape and size,
color and external marking, odor,
and comparability of contents of
opened capsules. Whether the par-
ticipant guessed or had an external
validation of whether he was getting
the active agent or placebo was not
assessed.
RESULTS
On September 15, 2008, the indepen-
dent data and safety monitoring com-
mittee met, reviewed data as of August
1, 2008, for the second formal interim
analysis, and recommended the discon-
tinuation of study supplements be-
cause the alternative hypothesis of no evi-
Tabl e 2. Adherence t o Study Supplements by Pill Counts and Bioadherence
% (Range)b
Pill Counts
8
Placebo Vitamin E Selenium Selenium + Vitamin E
Selenium/matching placebo
Year 1 (n=34 708) 85(76-85) 85 (77-85) 84 (76-84) 85 (77-84)
Year 2 (n =34163) 81 (72-81) 80 (72-81) 79 (71-80) 80 (72-80)
Year 3 (n=33616) 76 (68-77) 77 (69-77) 75 (68-76) 76 (69-77)
Year 4 (n=32976) 69(65-73) 73 (66-74) 71 (64-72) 72 (65-74)
Year 5 (n=23419) 69 (63-71) 71 (64-73) 69 (62-70) 70 (64-71)
Vitamin Elmatching placebo
Year 1 (n=34 708) 85(76-85) 85 (77-85) 85 (76-85) 85 (77-85)
Year 2 (n =34163) 80 (71-80) 80(71-80) 79 (70-79) 79 (71-80)
Year 3 (n=33616) 75 (67-75) 75 (67-76) 74 {67-75) 76 {69-77)
Year 4 (n=32976) 70 (63-72) 70 (63-72) 69 (62-71) 70 (63-72)
Year 5 (n=23419) 67 (61-69) 69 (62-71) 67 (61-69) 68 (61-70)
Median (lnterquartile Range)
Bioadherence (n = 285) (n = 290) (n = 277) (n = 257)
Serum selenium, IJ9/L
Baseline 137.6 (124.7-151 .8) 135.9 (122.4-148.4) 135.0 (123.4-145.9) 136.4 {122.9-150.0)
6-movisit 137.4 (123.3-152.0) 138.4 (124.1-154.0) 223.4 {198.6-251.8) 227.0 {199.4-251.2)
1st annual visit 138 1 (125.2-152.2) 137.7 (124.1-150.4) 232.4 (204.2261.4) 228.5 (205.5258.1)
2nd annual visit 132.0 (120.8-143.1) 129.8 (120.1-139.9) 228.0 (206.3256.9) 220.7 (194.0-249.5)
4th annual visit c 140.1 (124.3-150.8) 143.8 (126.2-158.6) 251.6 (218.7-275.0) 253.1 (21 0.5-283.0)
Cholesterol-adjusted a-tocopherol, J.lg/ml
Baseline 12.45 (10.70-1 4.95) 12.79 (10.69-15.37) 12.58 (10.43-14.75) 12.20 {10.12-15.35)
6-movisil 11.68 (1 0.09-13.61) 18.14 (15.21-22.45) 11.62 (10.1 0-13.44) 17.90 {15.11-20.84)
1st annual visit 11.68 (1 0.24-13.44) 18.50 (15.08-22.46) 11.69 (10.1013.03) 18.04 (14.77-22.35)
2nd annual visit 12.13 (10.80-13.72) 18.35 (15.1322.85) 11.80 (10.5713.58) 18.44 (15.32-22.89)
4th annual visit c 1209 (9.95-14.41) 16.57 (13.8622.61) 12.03 (9.57 - 13.53) 17.87 (14.68-22.31)
Cholesterol-adjusted -y-tocopherol, J-19/ ml
Baseline 1.31 (0.83-2.01) 1.43 (0.89-2.21) 1.50 (0.96-2.21) 1.44 (0.96-2.02)
6-movisit 1.50 (1.07 -1.97) 0.78 (0.51-1 .1 2) 1.64 (1.22-2.29) 0.74 (0.48-1.1 1)
1st annual visit 1.53 (1 .09-2.05) 0.75 (0.52 1.16) 1.69 (1.27 -2.33) 0. 70 (0.48- 1.04)
2nd annual visit 1.57 (1.13-2.13) 0.74 (0.49 1.08) 1. 76 (1.26-2.43) 0.66 (0.50-1.03)
4th annual visit c 1.69 (1.14-2.29) 0.80 (0.50-1 .23) 1.90 (1.48-2.70) 0.69 (0.47- 1.07)
Sl conversions: To convert serum selenium to IJmOI/L, multiply by 0.0127; a-tocopherol and -y-tocopherol to IJmoiiL. multiply by 23.22.
a Percentage of men adherent, defined as taking at least 80% of their study supplements. Denominators decrease over time reflecting the varying amounts of follow-up.
bThese ranges are estimates including those wfth missing data and assumes those missing were either all not adherent Qow estimate) or all adherent (high estimate).
cNumbers of participants for 4th annual visit are placebo (11= 79), vrtamin E (n= 78}, selenium (n= 72), and selenium+ vi tamin E (n= 71).
2009 American Medical Association. All rights reserved. (Reprinted) JAMA, Published online Dewnber 9, .WOS E5
dence of benefit from either study agent
was convincingly demonstrated
(P< .OOOl) and there was no possibil-
ity of a benefit to the plam1ed degree with
Table 3. Clinically Diagnosed Prostate Cancers
additional follow-up. Study sites were no-
tified to discontinue supplements on Oc-
tober 23, 2008, and the data presented
in this article are current as of this date.
Participants
A total of 35 533 men were accrued and
randomly assigned at 427 participat-
ing sites in the United States , Canada,
No. (%)of Participants
Placebo Vitamin E
(n = 8696) (n = 8737)
Total No. of prostate cancers diagnosed by study site 416 473
Method of diagnoses
Prostate biopsy 404 (97) 458 (97)
Other/unknown 12 (3) 15 (3)
No. of total prostate biopsies 1020 1011
PSA tests
8
Year 1 6708 (83) 6876(84)
Year2 6641 (86) 6652 (85)
Year3 6284 (8S) 6334 (8S)
Year 4 6043 (8S) 6087 (84)
Year S 426S (84) 4246(84)
DRE tests
8
Year 1 S766 (72) S936(73)
Year 2 SS67 (72) 5S63 (72)
Year 3 S180 (70) 5188 (70)
Year4 4862 (69) 4823 (67)
Year s 3420 (68) 3418 (68)
Reason for biopsy (positive biopsies)
Increased PSA 259 (64) 324 (71)
PSA prompting biopsy, median (IQR), ng/ ml 4.60 (4.00-S.SO) 4.60 (3.995.60)
PSA velocity 12 (3) 10(2)
Abnormal DRE 66 (16) 58(13)
Increased PSAIPSA velocity + abnormal DRE 55 (14) 49 (1 1)
Other 8 (2) 13(3)
Tstage
T1a-c 278 (70) 343 (75)
T2a-b 122 (30) 114 (25)
T3a-b 0 (0) 2 (0)
TX/not staged 16 14
N stage
NO 109 (100) 127 (100)
N1 0 (0) 0(0)
NX/not staged 307 346
Mstage
MO 124 (100) 134 (99)
M1a-b 0 (0) 2 (1)
MX/not staged 292 337
Gleason scoreb
No. graded by central laboratory 365 396
2-6 240 (66) 249 (63)
7 (grade 3 + grade 4) 80 (22) 97 (24)
7 (grade 4 + grade 3) 21 (6) 27 (7)
8-10 24 (7) 23(6)
Abbreviations: ORE, digital rectal examination: lOR, interquartile range; PSA, prostate-specific antigen.
Sl conversion: To convert PSA to J.Jg/L, multiply by 1.0.
Selenium +
Selenium Vitamin E
(n = 8752) (n = 8703)
432 437
419(97) 420(96)
13(3) l7 (4)
982 997
6807 (84) 6838 (84)
6635(8S) 6673(86)
6376(8S) 6349(8S)
606S(8S) 604S (84)
4271 (84) 42S7 (84)
S870(72) 5833(72)
5S61 (72) 5591 (72)
5198(70) 5190 (70)
4878(69) 4878(68)
3397 (68) 3425(68)
296 (71) 263(63)
4.83 (4.05-5.70) 4.70 (4.00-5.60)
13(3) 16 (4)
46(11) 56(13)
56(13) 72 (17)
12 (3) 17 (4)
301 (73) 286(69)
108(26) 128(31)
5 (1) 3 (1)
18 20
125 (99) 117 (100)
1 (1) 0(0)
306 320
122 (96) 119(98)
5(4) 2 (2)
305 316
361 365
217 (60) 220(60)
105 (29) 91 (25)
19(5) 24 (7)
20(6) 30(8)
a Percentages are based on alive participants who are prostate cancer-free and for whom the form was submitted.
bGieason score was based on central pathology review. The Gleason grade ranges from 1 to 5, with 5 having the worst prognoSis. The Gleason score ranges from 2 to 10, wrth 1 0
t1aving the worst prognosis.
and Puerto Rico between August 22,
2001 , and june 24, 2004. FIGURE l
s hows the SELECT randomization
scheme including participants who
were excluded from analyses; all 621
participants at 2 study sites were re-
moved from the analysis because of se-
vere problems that were detected early
on including poor data and partici-
pant management and regulatory is-
sues. These participants differed sub-
stantially from the rest of the SELECT
population in being from sites in the
south of the United States, 99% Afri-
can American, younger (median age 57
years), and of a lower education level
( 6 7% had < high school education), and
in having lower PSA levels (57% bad
< l.O nglmL) and a higher prevalence
of current smokers (33%) An addi-
tional9 participants were removed be-
cause they were found to have had pros-
tate cancer at randomization and 15
were removed because their informed
consent was never received. More men
were accrued (35 533 in 3 years) than
initially planned (32 400 in 5 years)
mainly because of a faster - than-
expected accrual rate and the admin-
istrative time it takes to close down
accrual.
The baseline characteristics of
SELECT participants by each of the 4
groups (placebo, vitamin E, selenium,
and selenium + vitamin E) are shown
in TABLE 1. All potentially important
risk factors were well balanced among
the groups. A total of 2.6% of SELECT
men were former PCPT men random-
ized to finasteride; during the trial,
4.8% of the non-PCPT participants
reported use of finasteride at 5 mg
(n=1602) or 1 mg (n=86).
The median overall follow-up was
5.46years (range, 4.17-7.33 years). The
percentages of participants with a re-
cent last-contact elate were more than
88% within 7 months and 92% within
l3 momhs of the SELECT data analy-
sis. loss to follow-up, defined as hav-
ing a last contact date of more than 24
months before analysis, involved 5.1%
of participants, which was higher than
had been estimated for the trial design
(3.5% at 7 years after trial activation).
SELENIUM AND VITAMI N E FOR CANCER PREVENTION
Figure 2. Cumulative Incidence of Prostate Cancer Detected Each Year by Intervention
Group
0.08
Placebo
0.07 ------Vitamin E .--
0.06
---Selenium , ./ ~
--Selenium+ vitamin E , --
j;-
0.05
'0
,/ /
.. ... ,.-'' ;
'
"'
0.04
.0
0.03
0.02
0.01
__ ... ' / . / , ..... !
0 2 3 5 6
Years After Randomization
No. at risk
Placebo 8689 8553 8328 8039 7389 4892 2516
Vitamin E 8732 8610 8373 8098 7401 4867 2537
Selenium 8750 8597 8341 8083 7393 4848 2558
Selenium + vitamin E 8700 8585 8371 8097 7428 4894 2580
Compared with placebo, there was a statistically nonsignificant increase in prostate cancer in the vitamin E
group (P=.06) and not in the selenium + vitamin E group (P=.52) or the selenium group (P=.62).
Adherence to both study agents as de-
termined by pill count was similar across
all study groups, and averaged 83% at
year land 65% at year 5. Adherence to
atleast1 of the 2 agents was 87% at year
1 and 72% at year 5 (the design-
estimated adherence rates were 90% at
year l and 68% at year 5) . Bi.oadher-
ence was measured in a subset of par-
ticipants by serum levels of selenium and
cholesterol-adjusted a-tocopherol and-y-
tocopherol (which is suppressed by a-
tocopherol) and showed a good separa-
tion in agent serum levels between the
groups (TABLE 2). The drop-in rate was
assessed by a direct question to the par-
ticipants about taking either of the
supplements. Positive responses were
3.1% or less for vitamin E and 1.8% or
less for selenium in each year (below the
design drop-in estimate of 10%). Pros-
tate tissue samples were sent to the cen-
tral pathology laboratory for confirma-
tion in 86% of cases. The central
laboratory agreed with the clinical site's
prostate cancer diagnosis in 99% of these
cases.
Prostate Cancer
There were no statistically significant dif-
ferences in the rates of prostate cancer
between the 4 groups (placebo, 416 cases
[5-year rate of 4.43%); selenium, 432
cases [4.56%]; vitamin E, 473 cases
[4.93%]; selenium + vitamin E, 437
cases [4.56%]) (TABLE 3 and fiGURE 2).
Compared with placebo, the hazard ra-
tios (HRs) for prostate cancer were 1.13
(99% confidence interval [ Cl], 0.95-
1.35; 95% CI, 0.99-1.29; P= .06) in the
vi tamin E-alone group, 1.05 (99% CI,
0.88-1.25; 95% CI, 0.91-1.20; P=.52)
in the selenium + vitamin E group, and
1.04 (99% Cl, 0.87-1.24; 95% Cl, 0.90-
L 18; P= .62) in the selenium-alone
group. The data and safety monitor-
ing committee had some concern over
the s tatistically nonsignificant in-
crease in prostate cancer in the vita-
min -alone group (P= .09 per interim
data of August 1, 2008) and over a non-
significant increase in diabetes melli-
tus associated with selenium (P= .08 per
interim data of August l , 2008).
The majority of prostate cancers di-
agnosed during the trial were early-
stage and low-grade, and cancer stage
and grade were similar across all groups
(Table 3). The percentage of patiems
who had an annual PSA examination
and DRE was similarly high and the bi-
opsy rate was similar across all groups,
indicating that the prostate cancer find-
ings were not due to screening-
associated detection bias. More than
95% of prostate cancers were cliag-
nosed by biopsy, the triggers for which
(based on PSA and other factors) are
shown in Table 3 and were similar
across all groups. The number of pros-
tate cancers in the placebo cohort was
higher than what was estimated at study
inception. This was due to the faster
than expected accrual, the larger than
expected sample size, and higher base-
line PSA levels than anticipated.
Secondary Outcomes
There were no significant differences
(all P> .15) in any prespecified second-
ary cancer end points (FIGURE 3 and
TABLE 4). At 5 years, the cumulative
death rate in the placebo group was 38
deaths per 1000 participants (95% Cl,
34 deaths per 1000 participants to 42
deaths per 1000 participants) ; the es-
timated rate at trial inception was 48
deaths per 1000 participants. The num-
bers of deaths from any cause were simi-
lar across the 4 groups (382 in pla-
cebo group, 358 in vitamin E group,
378 in selenium group, and 359 in se-
lenium + vitamin E group).
The study agents had no significant
effects on the overall incidence of car-
diovascular events (Table 4). A statis-
tically nonsignificant increase in type
2 diabetes mellitus (diagnosed after ran-
domization) occurred in the selenium-
alone group vs placebo group (n= 724;
10.0%; 99% Cl, 9.1%-11.0%; VS n=669;
9.3%; 99% Cl, 8.5%-10.2%, respec-
tively; RR, 1.07; 99% CI , 0.94-1.22;
P=.l6). The number (percentage) of
cases of diabetes mellitus was 700
(9.7%; 99% Cl, 8.8%-10.6%) in the vi-
tamin E group and 660 (9 1 %; 99% Cl,
8.2%-10.0%) in the selenium + vita-
min E group (P values of these data
compared with placebo were 0.47 for
vitamin E and 0.61 for selenium + vi-
tamin E). Data on known, clinically less
significant adverse effects of the study
agents (alopecia, dermatitis, halitosis,
Figure 3. Cumulative Incidence of Lung Cancer, Colorectal Cancer, All Other Primary Cancers, and Deaths by Intervention Group
Lung Cancer Colorectal Cancer
0 .020 0.020
0.018
Placebo
0.018
VitaminE
0.016 ---Selenium 0.016
0.014
--Selenium+ VItamin E
0.0 14
0.012 ,r 0.012
ii
r
'0
"'
0.010
"'
0.010
.0 .0
2 2
.:..:-
0.
0.008
0.
0.008
-
0.006 0.006 ::::::F-.r.r-;!;--3'-
0.004 0.004
0.002 0.002
0 2 3 5 6 0 2 3 5 6
Years Alter Randomization Years After Randomization
No. at risk No. at risk
Placebo 8689 8577 8440 8224 7602 5087 2626 Placebo 8689 8571 8431 8207 7582 5068 2613
Vitamin E 8732 8630 8491 8290 7647 5073 2649 Vrtamin E 8732 8628 8476 8277 7625 5062 2641
Selenium 8750 8621 8447 8257 7597 5010 2657 Selenium 8750 8615 8432 8240 7576 5000 2648
Selenium+ vitamin E 8700 8598 8457 8264 7647 5067 2674 Selenium + vitamin E 8700 8597 8447 8245 7629 5050 2655
All Other Cancers Deaths
0.08 0.08
O.D7 O.D7
0.06 0.06

0.05

0.05
n
"'
0.04
"'
0.04
.0 .0
2 2
0. 0.03 0. 0.03
0.02 0.02
0.01 0.01
0 2 3 4 5 6 0 2 3 4 5 6
Years Alter Randomization Years Alter Randomization
No. at risk No. at risk
Placebo 8689 8532 8362 8 113 7466 4977 2553 Placebo 8689 8585 8454 8236 7617 5100 2631
Vitamin E 8732 8602 8427 8190 7518 4954 2578 Vitamin E 8732 8639 8505 8310 7662 5086 2653
Selenium 8750 8587 8378 8160 7488 4913 2594 Selenium 8750 8628 8460 8277 7622 5029 2667
Selenium+ vitam<n E 8700 8557 8398 8176 7529 4972 2610 Selenium + vitBmJn E 8700 8610 8475 8285 7679 5089 2684
There were no significant differences in any prespecified secondary cancer or death end points (all P> .15). The blue portions of they-axes indicate 0 to 0.02 cancer probability.
ES JAMA, Published online December 9, 2008 (Reprinted) 2009 American Medical Association. AU rights reserved.
nail changes, fatigue, and nausea) are continued early (in year 7 of the overall rate, and loss to follow-up were largely
s hown in TABLE 5. The only statisti- 12-yearstudy) in accordance with a unani- met and gave the trial significant power
cally s ignificant di ffe rences (P<.Ol) mous recommendation of the data and to detect the estimated preventive ef-
were for selenium vs placebo for alo- safety monitoring committee stating that, fects . Furthermore, the large sample
pecia and grade.s 1 to 2 dermatitis. based on the evidence to date from the size, inclusion of a substantial propor-
COMMENT
7-year planned interim analyses, there was lion of non-white men, and equal dis-
no evidence of a benefit from either study tribution of known risk factors across
ln SELECT, neither 200 pg of selenome- agent and no possibility of a benefit to all trial gr oups make the conclusions
thionine or 400 IU of synthetic DL a- the pla1med degree with additional follow- drawn from SELECT especially ro-
tocopherol, given orally alone or combined up. Sensitivity analyses suggested that the bust and generalizable.
for a median of 5.5 years had significant prespecifiecl25% risk reduction was ex- Why were selenium and vitamin E in-
effects on the primary or secondary end tremely unlikely to be reached for either effective in preventing prostate cancer in
points. A statistically nonsignificant in- agent even with additional exposure. SELECT despite strong secondary evi-
creased incidence of prostate cancer The statistical assumptions made in dence suggesting efficacy?
7
8
Consider-
(P= .06) was observed in the vitamin E SELECT involving accrual rate, study ing selenium fi rst, the secondary reduc-
group but not in the selenium + vi tamin s uppl ement adherence and drop-in tion in prostate cancer incidence in the
E group. The trial supplements were dis- rates, prostate cancer incidence, death NPC study could have been subj ect to
Tabl e 4. Secondary O utcomes Including Diagnosis of Other Primary Cancers, Diabetes, Cardiovascular Events, and Deaths a
(n = 8696) (n = 8737) (n = 8752) (n = 8703)
No. HR No. HR No. HR No.
of Men (99% Cl) of Men (99% Cl) of Men (99% Cl) of Men
Any cancer (including prostate)b 824 1 (Reference] 856 1.03 (0.91-1.17) 837 1.01 (0.89-1.15) 846
Lung 67 1 (Reference) 67 1.00 (0.64- 1.55) 75 1. 12 (0.73- 1.72) 78
Colo rectal 60 1 (Reference] 66 1.09 (0.69-1. 73) 63 1.05 (0.66-1.67) 77
Other primary cancer
0
306 1 (Reference] 274 0.89 (0.72- 1.10) 292 0.95(0.77-1.1 7) 290
No. RR No. RR No. RR No.
of Men (99% CI) of Men (99% Cl) of Men (99% Cl) of Men
Diabetesd 669 1 (Reference] 700 1.04 (0.91-1.18) 724 1.07 (0.94-1.22) 660
Cardiovascular events
Any (including death) 1050 1 (Reference] 1034 0.98 (0.88-1.09) 1080 1.02 (0.92-1.13) 1041
Nonfatal strokes
Hemorrhagic 11 1 (Reference] 7 0.63 (0.18-2.20) 11 0 99 (0.33-2.98) 12
Ischemic 56 1 (Reference] 49 0.87 (0.53-1 .44) 51 0.90 (0.55-1.49) 67
Not specifiede 25 1 (Reference] 14 0.56 (0.24-1.32) 11 0.44 (0.17-1.1 1) 20
Ot her nonfatal (worst grade)
1
Grade3 626 1 (Reference] 642 1.02 (0.89-1. 17) 685 1.09 (0.95-1.25) 624
Grade4 190 1 (Reference] 203 1.06 (0.82- 1.38) 193 1.01 (0.78-1.31) 201
No. HR No. HR No. HR No.
of Men (99% CI) of Men (99% CI) of Men (99%CI) of Men
Deaths 382 1 (Reference] 358 0.93 (0.77-1.13) 378 0.99 (0.82-1.1 9) 359
Cancer 125 1 (Reference] 106 0.84 (0.60-1 . 18) 128 1.02 (0.74-1.41) 117
Prostate 0 1 (Reference] 0 NA NA 0
Lung 41 1 (Reference] 38 0.92 (0.52-1.65) 45 1.10 (0.63-1.91) 39
Colorectal 10 1 (Reference] 13 1.30 (0.44-3.83) 10 1.00 {0.32-3.1 6) 15
Other primary cancer
0
74 1 !Reference] 55 0.74 (0.47- 1.17) 72 0.97 (0.63 1.49) 63
Cardiovascular 142 1 !Reference] 119 0.84 (0.61-1.15) 129 0.91 (0.66-1.24) 117
Hemorrl1agic stroke 8 1 (Reference] 9 1.12 (0.32-3.92) 9 1.12 (0.32-3.93) 12
Other cardiovascular 134 1 (Reference] 110 0.82 (0.59-1.14) 120 0.89 (0.65- 1.24) 105
Other deaths 115 1 !Reference] 133 1. 15 (0.83-1.60) 121 1.05 (0.75-1.47) 125
Abbreviations: Cl. conlidence u1tervaJ: HR. hazard ratio; NA, not applicable; RR. relative risk.
a The HRs and RRs given for vitamin E, selenium, and selenium + vi tamin E groups are compared with the placebo group.
bNo. of participants that had more than 1 cancer for each gr01.1p are placebo (n=25), vitamin E (n=24), selenium (n=25), and selenium+ vitamin E (n=36).
c Excluding basal cell and squamous cell skin cancers.
d Based on self-report or reported use of diabetes medications of the glitazone class: excludes prevalent cases at randomization.
8
Not specifl8d as to whether an ischemic or hemorrhagic stroke.
!According to National Cancer Institute Common Toxicity Criteria.
HR
(99% CI)
1.02 (0.90-1.16)
1.16(0.76-1.78)
1.28 (0.82-2.00)
0.94 (0.76-1.16)
RR
(99%CI)
0.97 (0.85-1.1 1)
0.99 (0.89-1.10)
1.09 (0.37-3.19)
1.20 (0.75-1.90)
0.80 (0.37 - 1. 73)
1.00(0.87-1.15)
1.06 (0.82-1.37)
HR
(99% CI)
0.94 (0.77- 1.13)
0.93 (0.67-1 .30)
NA
0.95 (0.53- 1.69)
1.49 (0.52-4.28)
0.85 (0.551.32)
0.82 (0.60-1.13)
1 .49 (0.46-4.84)
0.78 (0.561.09)
1.08 (0. 78-1.51)
limitations inherent in secondary analy-
ses, such as chance findings due to mul-
tiple testing, especially because the over-
all NPC sample size was relatively small
( 1312 men and women vs 29 133 men
in the ATBC study). Second, the fonnu-
lation (high-selenium yeast) given in the
NPC trial may have been more active
than the l-selenomethionine given in
SELECT (both trials gave an equivalent
selenium dose) . In designing SELECT,
we carefully evaluated the choice of l-
selenomethionine vs high-selenium yeast
(and other forrnulations),
20
and our ra-
tionale for selecting 1.-selenornethio-
nine included the following consider-
ations: selenomethionine was the major
component of apparently active high-
selenium yeast; evidence indicated sub-
stantial batch-to-batch variations in spe-
cific organoselenium compounds in
samples of NPC yeast, making it un-
likely that we could duplicate the sele-
nium yeast fonnulation used in the NPC
study; potential genotoxicity of highly ac-
tive inorganic selenium compounds,
such as selenite, made them potentially
unsuitable for long-term prevention; low-
ering (vs selenomethionine) of overall
body selenium stores with selenite, which
is neither absorbed nor retained well;
practical and safety concerns over newer
selenium compounds, such as mono-
methylated forms (eg, lacking availabil-
ity, investigational new drug certifica-
tion, and clinical data); and in vitro data
indicating that selenomethionine was ef-
fective in suppressing malignant and not
normal prostate cells.
15
Despite this careful rationale, it is im-
possible to know now whether sele-
nized yeast would have been more ac-
tive than l-selenomethionine was in
SELECT. Finally, the NPC trial was con-
ducted in men chosen for deficient lev-
els of selenium, finding that selenium was
most preventive in the men with the low-
est baseline selenium levels
9
; SELECT
men generally were replete in selenium
at baseline, with median serum sele-
nium levels of 135 ng/mL vs 113 ng/rnl
in NPC. The NPC cutpoint for the low-
est 2 tertiles was 121.6 units; 78% of
SELECT men were above this level. The
NPC trial found a nonsignificant in-
crease in overall cancer rate in its high-
est tertile (HR, 1.20; 95% CI, 0.77-1.86)
12
There are potential reasons why vita-
minE did not prevent prostate cancer in
SELECT. First, the high dose ( 400 lU/d)
of the a-tocopherol form of vitamin E in
SELECT may have been less effective
than a lower dose such as the 8-fold lower
50 mg/d (roughly equivalent to 50 IU/d)
that produced the earlier positive sec-
ondary findings in the A TBC study.
7
(The
vitamin E fommlation, synthetic all rac-
a -tocopheryl acetate, was the same in
Tabl e 5. Adverse Events Known to Be Associated With the Study Supplements
Placebo Vitamin E
(n = 8696) (n = 8737)
No. RR No. No.
Adverse Event of Men {99% Cl) of Men RR(99% Cl) of Men
Alopecia 206 1 [Reference] 220 1.06 (0.83- 1.36) 265
Dermatitis
Grades 1-2 516 1 [Reference] 591 1.14 (0.98-1.32) 605
Grades 3-4 8 1 [Reference) 12 1.49 (0.46-4.83) 14
Halitosis 427 1 [Reference) 493 1.15 (0.97 -1.36) 503
Nail changes 1035 1 [Reference) 1041 1.00 (0.90-1.1 1) 1087
Fatigue
Grades 1-2 586 1 [Reference] 604 1.03 (0.89-1.19) 645
Grades 3-4 24 1 [Reference) 29 1.20 (0. 59-2.45) 21
Nausea
Grades 1-2 203 1 [Reference) 191 0.94 (0. 72-1.21) 244
Grade 3 9 1 [Reference) 3 0.33 (0.06-1.85) 9
Abbreviations: Cl, confidence interval; RR, relative risk.
SELECT and the ATBC study.) A sec-
ondary analysis of the HOPE trial
23
found
that a relatively high dose of natural vi -
tamin E did not reduce prostate cancer
incidence. Achieving higher plasma or
tissue levels of a-tocopherol within the
physiological range, such as through a
50-mg/d supplement, may have some
prostate cancer (or other) preventive
effect such as cell proliferation or tu-
mor growth inhibition
24
Furthermore,
high pharmacological doses of a-
tocopherol may have an adverse effect on
cytochrome p450 enzyme and other
regulatory mechanisms
25
that a lower
dose would not have. It is also possible
(but not certain) that the known effect
of a -tocopherol in suppressing poten-
tially beneficial plasma "{-tocopherol lev-
els would have been less with the lower
than higher dose of a-t.ocopherot2 Nev-
ertheless, men taking vitamin E with the
highest baseline (and thus total) serum
vitamin E levels in the ATBC study had
the highest reduction in prostate and
lung cancer/
6
which supported our
choice of the higher dose. A higher dose
also was associated with potential ben-
efits such as reductions in aging-related
Alzheimer disease and macular degen-
eration.
Second, several studies have sug-
gested that vitamin E is more protective
against prostate cancer in smokers, and
Selenium Selenium + Vitamin E
(n = 8752) (n = 8703)
No.
RR {99% Cl) of Men RR{99% Cl)
1.28 (1.01-1.62)b 238 1.15 (0.91-1.47)
1.17 (1.00-1.35)b 554 1.07 (0.92-1 .25)
1.74 (0.56-5.44) 16 2 00 (0.66-6.09)
1.17 (0.99-1.38) 531 1 .24 (1.06-1.46)
1.04 (0.94-1.16) 1075 1 .04 (0.93-1.15)
1.09 (0.95-1.26) 612 1 .04 (0.90- 1.20)
0 87 (0.40-1.88) 20 083 (0.38- 1.81)
1.19 (0.941.52) 202 0.99 (0.77 1.28)
0.99 (0.30-3.34) 8 0.89 (0.25-3.10)
a The RRs given for vitamin E, selenium, and selenium + vitamin E groups are compared with the placebo group. Maximum grade experienced by a participant are given. AlOpecia,
halitosis. and nail changes were only defi ned for grades 1 and 2. National Cancer Institute Common Toxicity Criteria were used for alopecia, nail changes, fatigue, and nausea.
Halitosis and dermatitis were defined in the study protocol. Generally, grade 1 =mild, grade 2= moderate, grade 3= severe, and grade 4 =
bp< .OI .
less than 60% of SELECT men were cur-
rent or former smokers (whereas all men
in the ATBC study were smokers). For
example, observational analyses in a trial-
based cohort of the Prostate, lung, Co-
lo rectal, and Ovarian Cancer Screening
Trial (PLC0),
27
a trial of screening vs
standard health care routines, showed a
71% reduction in the incidence of ad-
vanced prostate cancer associated \'llith
supplemental vitamil1 E use in current
and recent smokers. A subgroup analy-
sis of current and former smokers in
SELECT, however, did not show a smok-
ing-related benefit (placebo, 4.6% [223/
4863] vs vitamin E alone, 4.8% [232/
4853]). As with selenium in the NPC
study, vitamin E effects on prostate can-
cer incidence in the A TBC study could
have been due to chance findings in sec-
ondary analyses.
Selenium was not associated \'llith sig-
nificant effects on cardiovascular events,
lung cancer, other cancers, or overall
mortality in SELECT. One safety con-
cern \'lli th selenium is a potential asso-
ciation with increased risk for type 2
diabetes mellitus, for which there are
mixed data from prior studies.
28
29
Are-
cent analysis of the NPC study popu-
lation showed a significant increase in
type 2 diabetes mellitus (by self-
reportand medical records) , largely lim-
ited to the top tertile of plasma sele-
nium levels at baseline.
30
In SELECT, a nonsignificant in-
crease in risk (RR, 1.07; P= .16) of dia-
betes mellitus compared with placebo
was observed in the selenium group but
not in the selenium + vitamin E group
(RR, 0.97; P=.62). Concerns about the
safety of vitamin E supplementation
arose during SELECT. One meta-
analysis31 found that vitamin Eat doses
of at least 400 IU/d increased all-cause
mortality, and another study3
2
found evi-
dence that vitamin E supplementation,
alone or in combination with other an-
tioxidants, may increase mortality. Nei-
ther sn.1dy is directly relevam to the doses
and poptllation studied in SELECT;
many studies included in these meta-
analyses were in patients \'llith serious dis-
ease, and the finding of increased mor-
tality was driven by studies using doses
SELENIUM AND VTTAMIN E FOR CANCER PREVENTION
far higher than 400 lUid. In more rel-
evant, placebo-controlled trials com-
pleted in healthy men and women, there
were no associations of vitamin E supple-
mentation \'llith increased risks of either
cardiovascular disease or overall mor-
tality. 33 SELECT results support the
safety of vitamin Eat 400 lU/d in healthy
men, because there were no increases in
either cardiovascular disease or total mor-
tality in the vitamin E groups.
The 35 533 randomized men of
SELECT were needed because of the ro-
bust statistical design accommodating 4
study groups with 5 p1imary compari-
sons; this large trial population made
SELECT the largest cancer chemopre-
vention trial ever conducted to our
knowledge. African American men have
among the highest prostate cancer risks
in the world, and SELECT had the high-
est participation of African American
men (13%) of any large-scale cancer che-
moprevention trial to date.
The statistical rigor of the trial was
matched by the rigor of its implementa-
tion. Features of this inlplementation in-
cluded the SELECT Workbench, a se-
cure Web site administered by the
SELECT statistical center and used by
study-site staff and investigators. The
SELECT Workbench was used to ac-
cess participant and site-specific re-
ports, the study protocol, and a detailed
study manual and to submit data using
Web-based forms. Form submission in-
cluded detailed edit checks and a track-
ing system to idenLify all expected forms.
Training and monitoring consisted of
semi-annual workshops, quality assur-
ance audits at least once every 3 years,
and mentoring by trained statistical cen-
ter staff and experienced clinical re-
search associates. SELECT also main-
tained a public Web site initially designed
to recruit participants and later used to
promote participant adherence and to
keep SELECT in the public's eye.
20
Potential limitations of SELECT in-
clude that it did not test different for-
mulations or doses of selenium and vi-
tamin E and that it did not definitively
assess results in subgroups of men who
may have responded differently than did
the overall population. Because of ac-
tive annual screening (eg, PSA in 85%;
Table 3) and early detection (eg, 99.4%
s tage Tl orT2; Table3),SELECT could
not assess effects in reducing advanced
or fatal prostate cancer, which recent data
suggest may be a potential benefit of vi-
tamin E and selenium.
18
27
34
36
SELECT
also could not assess intervention ef-
fects in a population deficient in vita-
min E, selenium, or both since our trial
population was well-nourished at base-
line, or in current smokers since they rep
resented only 7.5% of the SELECT popu-
lation, a substantial difference from the
ATBC study in predominantly heavy
smokers.
Cancer chemoprevention is an im-
portant approach for reducing cancer
burden.
37
Several randomized con-
trolled trials have demonstrated sig-
nificant cancer or premalignancy risk
r eductions in the breast , colon -
rectum, prostate, and stomach.JS.
44
Pros.
tate cancer is a particularly attractive
target for chemoprevention because of
its clinical ubiquity, substantial treat-
ment-associated morbidity, and step-
wise molecular pathogenesis. In the
large-scale PCPT, which was reported
2 years after SELECT was activated, fi-
nasteride produced a 25% relative re-
duction in the 7 -year period preva-
lence of prostate cancer ( vs placebo) ,
43
and recent data suggest that finaste-
ride reduces the risk of clinically sig-
nificant disease and may not induce
high-grade cancers despite initial con-
cerns to the contrary.
4
H
9
CONCLUSION
In conclusion, SELECT has definitively
demonstrated that selenium, vitamin E,
or selenium + vitamin E (at the tested
doses and formulations) did not pre-
vent prostate cancer in the generally
healthy, heterogeneous population of
men in SELECT. These data under-
score the prudence that is needed in con-
sidering recommendations to use agents
for the prevention or comrol of disease
in the absence of convincing clinical trial
results. These findings also compel the
medical research community to con-
tinue the search for new, effective agents
for prostate cancer prevention.
2009 American Medical Association. All rights reserved. (Reprinted) JAMA. Published online December 9, 2008 E11
Published Online: December 9, 2008 (doi:1 0.1001/
jama. 2008.864)
Author Affiliations: Divisions of Cancer Medicine (Drs
Lippman and Karp) and Cancer Prevention and Popu-
lation Sciences (Drs Lippman and Cook), University of
Texas M.D. Anderson Cancer Center, Houston; Glick-
man Urological and Kidney Institute and Taussig Can-
cer Institute, Cleveland Clinic, Cleveland, Ohio (Or Klein);
Southwest Oncology Group Statistical Center, Seattle,
Washington (Dr Crowley and Mss P. Goodman. Hart-
line, Darke, and Arnold); Department of Pathology (Dr
Lucia) and Division of Urologic Oncology (Dr Crawford},
University of Colorado Health Sciences Center, Denver;
Departments of Urology (Dr Thompson) and Medicine/
Hematology and Medical Oncology (Dr Cottman), Uni-
versity of Texas Health Sciences Center, San Antonio; Di-
vision of Cancer Prevention (Drs Ford, Parnes. and Mi-
nasian) and Division of Cancer Epidemiology and Genetics
(Drs Albanes and Taylor), National Cancer Institute,
Bethesda, Maryland; Veterans Affairs Cooperative Stud-
ies Program and Massachusetts Veterans Epidemiology
Research and Information Center, Boston VA Healthcare
Center, Boston, Massachusetts (OrGaziano); Moores Can-
cer Center, La Jolla, California (Or Parsons); Upstate Caro-
lina CCOP, Spartanburg, South Carolina (Dr Bearden);
Division of Hematology and Oncology, Swedish Cancer
Institute. SeaWe. Washington (Dr G. Goodman); Alta-
mira Family Medicine, Rio Piedras. Puerto Rico (Dr Clau-
dio); London Regional Cancer Program, London Health
Sciences Center, London, Ontario, Canada (Or Winquist);
Department of Urologic Surgery, Duke University Medi-
cal Center, Durham. North Carolina (Or Walther); De-
partment of Urology, Mayo Clinic, Rochester, Minnesota
(Or Ueber); Departments ofEpidemioiOl,'Y (Or Krista!) and
Medicine and Cardiology (Or Probstfield), University of
Washington, Sea We; Division of Cancer Prevention and
Control Research, Jonsson Comprehensive Cancer Cen-
ter. University ofCalifomia, Los Angeles (Dr Ganz); Mail-
man School of Public Health, Columbia University, New
York, New York (Or Santella); Center for Clinical Epide-
miology and Evaluation, University of British Columbia,
Vancouver, Canada (Mr Jagpal); Chao Family Compre-
hensive Cancer Center. University of California at Irvine,
Orange (Dr Meyskens); and Division of Hematology and
Oncology, University of Michigan (Dr Baker), and South-
west Oncology Group (Drs Baker and Cottman), Ann Ar-
bor, Michigan.
Author Contributions: Drs Lippman and Klein had full
access to all of the data in the study and take respon-
sibility for the integrity of the data and the accuracy
of the data analysis. Both contributed equally to the
study.
Study concept and design: Lippman, Klein , P.
Goodman, Thompson. Ford, Parnes, Minasian,
Gaziano, Crawford, G. Goodman, Cook, Karp, Walther,
Lieber, Ganz. Santella, Albanes, Taylor, Probstfield,
Crowley, Meyskens, Cottman.
Acquisition of data: Klein, P. Goodman, Lucia, Hartline,
Parsons, Bearden. G. Goodman, Claudio, Winquist,
Cook. Lieber, Arnold, Jagpal, Crowley.
Analysis and interpretation of data: Lippman, Klein,
P. Goodman, Thompson, Ford, Parnes. Minasian,
Gaziano. Kri sta!, Darke. Arnold, Crowley, Baker.
Collman.
Drafting of the manuscript: Li ppman, Klei n,
P. Goodman, Thompson, Minasian, Darke.
Critical revision of the manuscript for important in-
tellectual content: Lippman, Klein, P. Goodman, Lucia,
Thompson, Ford, Parnes, Minasian, Gaziano, Hartline,
Parsons, Bearden, Crawford, G. Goodman, Claudio,
Winquist, Cook. Karp, Walther. Lieber, Krista!, Arnold,
Ganz, Santella, Albanes, Taylor. Probstfield. Jagpal,
Crowley, Meyskens, Baker, Collman.
Statistical analysis: P. Goodman, Darke, Arnold,
Crowley.
Obtained funding: Lippman, Cottman.
Administrative, technical, or material support:
Lippman, Klein, P. Goodman, Lucia, Thompson, Ford,
Minasian, Hartline, Bearden, Crawford. G. Goodman.
Claudio, Cook, Lieber, Ganz. Santella, Albanes, Taylor,
Probstfield, Meyskens.
Study supervision: Lippman, Klein, Thompson, Ford,
Minasian. Gaziano, Hartline, G. Goodman, Karp, Lieber.
Probstfield, Crowley, Baker, Cottman.
Financial Disclosures: Dr lucia reported serving as a con-
sultant for GlaxoSmithKiine and Veridex, and being a
member of the Advisory Board for GenProbe. Or Thomp-
son reported serving as a consultant torVeridex and Mis-
sion Pharmacal (with fees paid to University of Texas
Health Sciences Center, San Antonio). Dr Gaziano re-
ported receiving investigator-initiated research fund-
ing from Veroscience, Amgen, and BASF Corporation,
and research support in the form of study agents and
packaging from BASF Corporation, Wyeth Pharmaceu-
ticals, and DSM Nutritional Products Inc (formerly Roche
Vitamins); serving as a consultant or receiving hono-
raria from Bayer AG and Pfizer; and serving as an ex-
pert witness for Merck. Dr Meyskens reported being a
co-founder of Cancer Prevention Pharmaceuticals. No
other authors reported financial disclosures.
Funding/Support: This work was supported in part by
Public Health Service Cooperative Agreement grant
CA37429 awarded by the National Cancer Institute, Na-
tional Institutes of Health. Department of Health and
Human Services. and in part by the National Center for
Complementary and Alternative Medicine (National In-
stitutes of Health). Study agents and packaging were
provided by Perrigo Company (Allegan, Michigan), Sab-
insa Corporation (Piscataway, New Jersey), Tishcon Cor-
poration (Westbury, New York), and OSM Nutritional
Products Inc (Parsipanny, New Jersey).
Role of the Sponsor. The National Cancer Institute was
involved in the design and conduct of the study, in
the analysis and interpretation of the data, and in the
preparation, review, and approval of the manuscript.
Active SELECT Clinical Sites With ;;,:100 Participants
as of October23, 2008: San Diego, U ofCA: J. Kellogg
Parsons, principal investigator (PI) [1743 men); Up-
state Carolina CCOP: Jay Bearden Ill , PI (1201 men);
London Regional Cancer Program, London Health Sci-
ences Centre: Joseph l. Chin and Eric Winquist, Pis (981
men); University of Colorado: E. David Crawford, PI (964
men); Swedish Medical Ctr: Gary E. Goodman, PI (934
men); VAMC Jesse Brown: Thomas E. Lad, PI (749 men);
Harbor-UCLA: Rowan T. Chlebowski, PI (629 men); Le
Centre de Recherche: Yves Fradet, PI (628 men); Alta-
mira Family Med: Jaime Claudio, PI (610 men); Mayo,
Rochester: Michael M. Ueber, PI (606 men); Capital Re-
gion Prostate Centre: Gary Steinhoff, PI (543 men); Van-
couver Hospital: Mark FitzGerald, PI (423 men}; Rush
University Medical Center: Steven K. Rothschild, PI (385
men): MD Anderson Cancer Center: Elise D. Cook. PI
(381 men); VAMC San Juan: Luis Baez, PI (359 men);
SUNY Stony Brook: Iris A. Granek, PI (358); $her-
brooke University Hospital: Abdenour Nabid, PI (348
men); George Washington University: Richard J. Katz.
PI (342 men); William Beaumont Hospital: David A.
Decker, PI (321 men); Wilford Hall Medical Center: Kyle
J. Weld, PI (309 men); Cascadia Cancer Prevention at
St. Joseph Hospital: Frank E. James, PI (299 men); Day-
ton CCOP: Lawrence J. Litscher, PI (296 men); Grand
Rapids CCOP: Marianne K. Lange, PI (287 men); VAMC
Minneapolis: Timothy J. Wilt, PI (270 men); Carle Can-
cer Center CCOP: David l. Graham. PI (253 men); LOS
Hospital: Scott Chidester, PI (250 men); University of
Mississippi: Charles R. Pound, PI (238 men); Green-
ville CCOP: Jeffrey K. Giguere, PI (230 men); Metro-
Minnesota CCOP: Alice C. Shapiro, PI (229 men); VAMC
Cleveland: Donald R. Bodner, PI (227 men); Wichita
CCOP: Shaker R. Dakhil, PI (219 men); Arizona Can-
cer Center: Frederick R. Ahmann, PI (219 men); Marsh-
field Clinic Matthias Weiss, PI (215 men): University
of Iowa Hospital: Richard D. Williams, PI (207 men);
Baptist Hospital East: Keny Short, PI (202 men); Down-
state Medical Center: Richard J. Macchia, PI (197 men):
Kalamazoo CCOP: Raymond S. Lord Ill, PI (181 men);
Southern Nevada CCOP: John A. Ellerton. PI (173 men);
Sunnybrook Health Science Center: Laurence Klotz, MD,
PI (171 men); Missouri Baptist Medical Center: PaulK.
Schultz, PI (170 men); Geisinger Clinic Albert M. Ber-
nath, PI (165 men); VAMC Kansas City: Peter J. Van
Veldhuizen Jr., PI (163 men); Orocovis Med Ctr: Jose
S. Aponte, PI (163 men); Sutter Health Cancer Re-
search Group-Eastern Division: Vincent Caggiano, PI
(160 men); VAMC Washington, DC: Steven H. Kras-
now, PI (154 men): Bay Area CCOP: Norman R. Co-
hen, PI (153 men); Sentara Cancer Institute: Robert W.
Given, PI (152 men); VAMC Fargo: William K. Becker,
PI (151 men); Medical College of Wisconsin: Robert
F. Donnell, PI (149 men); VAMC Houston: Teresa G.
Hayes, PI (146 men); Baptist Regional Cancer fnsitute:
Neil Abramson, PI (136 men); Mount Sinai CCOP: Ro-
gerio C. Lilenbaum, PI (134 men); Methodist Hospi-
tals of Dallas: John V. Cox, PI (133 men); Miguel Sosa
Padilla: Miguel Sosa-Padilla, PI (133 men); Kaiser Perma-
nente: Nagendra R. Tirumali, PI (132 men); Duluth
CCOP: Steven A. Kuross, PI (131 men); Stormont- Vail
Health Care: Stanley J. Vogel, PI (130 men); Decatur
Memorial Hospital: James l. Wade Ill, PI (126 men);
VAMC Puget Sound: Daniel W. Lin, PI (124 men);
VAMC Boston: MaryT. Brophy, PI (122 men); Scott&
White CCOP: Scott Coffield, PI (119 men); Schiffler Can-
cer Center: Gregory S. Merrick, PI (116 men); Merit-
Care Hospital CCOP: Preston D. Steen, PI (115 men};
Gaston Memorial Hospital: Steven W. Yates. PI (114
men); VAMC Phoenix: James V. Felicetta. PI (113 men);
Lehigh Valley Hospital: Gregory R. Harper, PI (113 men);
Cancer Resource Ctr: Sushi! S. lacy, PI (112 men); Holy
Cross Hospital: Leonard J. Seigel, PI (112 men); Cleve-
land Clinic: Eric A. Klein, PI (111 men); Walter Reed
AMC: Rob Dean, PI (111 men); Kaiser Permanente-
GA: Joshua I. Barzilay. PI (110 men); Columbia River
CCOP: Keith S. Lanier, PI (110 men); Oregon Health &
Science University: Mark G. Garzotto,PI (1 10 men); H
Lee Moffitt Cancer Center: Julio M. Pow-Sang, PI (110
men); McGill University Health Center: Simon Tan-
guay. PI (11 0 men); Vanderbilt University: Michael S.
Cookson, PI (1 09 men); St Luke's Mountain State Tu-
mor Institute: Thomas M. Beck, PI (107 men); Wash-
ington University: Robert L. Grubb Ill, PI (107 men);
VAMC Southern Arizona: Maria C. Bishop, PI (1 06 men);
Andres Grillasca: Luis Baez, PI (106 men); VAMC Hines:
Nirmala Bhoopalam, PI (102 men): University of Okla-
homa: Daniel J. Culkin, Pl(1 02 men); Kaiser Permanente-
Oakland: Louis Fehrenbacher, PI (100 men); St Vin-
cent Hospital: Thomas J. Saphner, PI (100 men).
Intergroup Participants: Eastern Cooperative Oncol-
ogy Group: D. Karp (chief liaison): Cancer and Leu-
kemia Group 8: P. Walther (chief liaison); North Cen-
tral Cancer Treatment Group: M. Lieber (chief liaison);
Radiation Therapy Oncology Group: F. Khuri (chief
liaison); and Veterans Affairs Cooperative Studies Pro-
gram: M. Gaziano (chief liaison).
SELECT Steering Committee: Gary E. Goodman, MD,
Philip R. Taylor, MD, SeD, Powel H. Brown, MD. PhD,
Paul Godley, MD, PhD, Charles Bennett, MD, PhD.
Michael M. Lieber, MD, Lewis Musgrove. Ellen Rich-
mond, MS, RN, Alan R. Krista!, DrPH, Julia E. Ver-
trees, PharmD, Regina M. Santella, PhD, M. Scott Lu-
cia, MD, Demetrius Albanes, MD, Patricia A. Ganz,
MD, Jeffrey L. Probstfield, MD. Neil E. Fleshner, MD,
MPH, Isaac J. Powell, MD, T. J. Jagpal. CCRP, Wil-
liam R. Markesbery, MD, William Christen. SeD, Pa-
tricia A. Cassano, PhD, M. Peter Lance, MD, Carolyn
J. Hoban. DSc. Marjorie A. Godfrey, Abbie l. Brown,
Dana B. Sparks, MAT, Elaine Armstrong, MS, Frank
L. Meyskens Jr. MD, Cathy Tangen, OrPH, Garnet L.
Anderson, PhD, Amy Darke, MS, Katie Arnold, MS.
Karen Anderson, Monica Yee, Scott M. Lippman, MD,
Eric A. Klein, MD. Phyllis J. Goodman, MS. lan M.
Thompson, MD, Leslie G. Ford, MD, Howard L. Par-
nes, MD. J. Michael Gaziano, MD, MPH, Lori Mina
sian, MD, Jo Ann L. Hartli ne, MPH, J. Kellogg Par-
sons, MD, MHS, James D. Bearden, Il l, MD, Jaime
E1 2 JAMA, Published online December 9, 2008 (Reprinted) 2009 American Medical Association . AU rights reserved.
Claudio, MD, Elise D. Cook, MD, laurence H. Baker,
DO, John J. Crowley, PhD, Charles A. Coltman Jr. MD.
SELECT Committees and Subcommittees: Recruit-
ment and Adherence Committee: J. L. Probstfield
(chair); Minority and Medically Underserved Sub-
committee: E. D. Cook (chair); Health-related Qual-
ity of Life Committee: C. M. Moinpour and P. A. Ganz
(co-chairs); Pathology and Biomarkers Committee: M.
S. Lucia (chair); Molecular Epidemiology Commit
tee: R. M. Santella (chair); Diet and Nutrition Com-
mittee: A. R. Krista! (chair); Site Coordinators Com-
mittee: T. J. Jagpal (chair).
Disclaimer. Dr Gaziano, a contributing editor for lAMA,
was not involved in the editorial review of or decision
to publish this article.
Additional Contributions: We thank the 35 533 men
and many principal investigators and clinical research as-
sociates at our 427 clinical sites, whose participation in
SELECT has written an important chapter in the history
of cancer prevention. We also thank the many person
nel of the Southwest Oncology Group (the coordinat-
ing group of this Intergroup trial), whose tireless ef-
forts allowed SELECT to successfully complete the test
of its primary hypotheses. No compensation was re-
ceived.
REFERENCES
1. Jemal A, Siegel R, Ward E, et al. Cancer statistics,
2008. CA Cancer 1 C/in. 2008;58(2):7196.
2. Harlan SR. Cooperberg MR, Elkin EP, et al. Time
trends and characteristics of men choosing watchful
waiting for initial treatment of localized prostate can-
cer: results from CaPSURE. 1 Urol. 2003 ;1 70(5):
1804-1807.
3. Potosky AL, Legler J,Aibertsen PC, etal. Health out-
comes after prostatectomy or radiotherapy for prostate
cancer: results from the Prostate Cancer Outcomes Study.
1 Nat! Cancer lnst. 2000;92(19):1582-1592.
4. Stanford JL, Feng Z, Hamilton AS, et al. Urinary and
sexual function after radical prostatectomy for clini
cally localized prostate cancer: the Prostate Cancer Out-
comes Study. lAMA. 2000;283(3):354-360.
5. Cooperberg MR. Broering JM, Kantoff PW, Carroll
PR. Contemporary trends in low risk prostate cancer:
risk assessment and treatment. 1 Urol. 2007;178
(3 pt2):S14-S19.
6. Hoffman RM, Gilliland FD, Penson DF, Stone SN, Hunt
we, Potosky Al. Cross-sectional and longitudinal com-
parisons of health-related quality of lif e between pa-
tients with prostate carcinoma and matched controls.
Cancer. 2004;101 (9):2011-2019.
7. The Alpha-Tocopherol, Beta Carotene Cancer Pre-
vention Study Group. The effect of vi tamin E and beta
carotene on the incidence of lung cancer and other can-
cers in male smokers. N Eng! 1 Med. 1994;330(15):
10291035.
8. Clark LC, Combs GF Jr, Turnbull BW, et al; Nutri-
tional Prevention of Cancer Study Group. Effects of se
lenium supplementation for cancer prevention in pa-
tients with carcinoma of the skin: a randomized controlled
trial. lAMA. 1996;276(24):19571963.
9. Duffield-Lillico AJ. Dalkin Bl, Reid ME, et al; Nutri-
tional Prevention of Cancer Study Group. Selenium
supplementation, baseline plasma selenium status and
incidence of prostate cancer: an analysis of the com-
plete treatment period of the Nutritional Prevention of
Cancer Trial. BlU Int. 2003;91(7):608-612.
10. Heinonen OP, Albanes D. Virtamo J, et al. Pros-
tate cancer and supplementati on with alpha-
tocopherol and beta-carotene: incidence and mortality
in a controlled trial. 1 Nat! Cancer lnst. 1998;90(6):
440-446.
11. Blot WJ, Li JY, Taylor PR. et al. Nutrition interven-
tion trials in linxian, China: supplementation with spe
cific vitamin/mineral combinations, cancer incidence, and
disease-specific mortality in the general population. 1 Nat!
Cancer lnst. 1993;85(18):1483-1492.
12. Fleshner N, Fair WR. Huryk R, Heston WD. Vita-
minE inhibits the high-fat diet promoted growth of es-
tablished human prostate LNCaP tumors in nude mice.
1 Urol. 1999;161(5):1651-1654.
13. lpC, Thompson HJ.ZhuZ. Ganther HE.Invitroand
in vivo studies of methylseleninic acid: evidence that a
monomethylated selenium metabolite is critical for can-
cer chemoprevention. Cancer Res. 2000;60(11 ):
28822886.
14. Jiang C. Wang Z, Ganther H,Lu J. Caspases as key
executors of methyl selenium-induced apoptosis (anoi-
kis) of DU-145 prostate cancer cells. Cancer Res. 2001;
61 (7):3062-3070.
15. Menter DG, Sabichi AL, Lippman SM. Selenium ef-
fects on prostate cell growth. Cancer Epidemio/ Bio-
markers Prev. 2000;9(11):1171-1182.
16. Redman C, ScottJA, Baines AT, etal. Inhibitory effect
of selenomethionine on the growth of three selected hu-
man tumor cell lines. Cancer Lett. 1998;125(1-2):
103-110.
17. Taylor PR, Albanes D. Selenium, vitamin E. and pros-
tate cancer-ready for prime ti me? 1 Nat/ Cancer lnst.
1998;90(16):1184-1185.
18. Yoshizawa K. Willett WC, Morris SJ. et al. Study of
prediagnostic selenium level in toenails and the risk of
advanced prostate cancer. J Natl Cancer Jnst. 1998;
90(16):1219-1224
19. Zhong W, Oberley TD. Redox-mediated effects of
selenium on apoptosis and cell cycle in the LNCaP hu-
man prostate cancer cell line. Cancer Res. 2001;
61 (19):7071 7078.
20. Lippman SM, Goodman PJ, Klein EA, et al. Design-
ing the Selenium and Vitamin E Cancer Prevention Trial
(SELEG). 1 Nat/ Cancer lnst. 2005;97(2}:94-102.
21. Current good manufacturing practice regulations,
21 CFR 210, 21 1.
22. Duffield-l illico AJ, Reid ME, Tumbull BW. etal. Base-
line characteristics and the effect of selenium supple-
mentation on cancer incidence in a randomized clinical
trial: a summary report of the Nutritional Prevention of
Cancer Trial. Cancer Epidemiol Biomarkers Prev. 2002;
11 (7):630639.
23. Lonn E,BoschJ, YusufS.etai;HOPEandHOPE-TOO
Trial Investigators. Effects of long-term vitamin E supple-
mentation on cardiovascular events and cancer: a ran-
domized controlled trial. lAMA. 2005;293(11):1338-
1347.
24. Venkateswaran V, Fleshner NE. Klotz LH. Modu-
lation of cell proliferation and cell cycle regulators by vi-
tamin E in human prostate carcinoma cell lines. 1 Urol.
2002;168(4 pt 1):1578-1582.
25. Traber MG. How much vitamin E? just enough! Am
1 C/in Nutr. 2006;84(5):959960.
26. Weinstein SJ, Wright ME, Pietinen P, et al. Serum
alpha-tocopherol and gamma-tocopherol in relation to
prostate cancer risk in a prospective study. 1 Nat/ Can-
cer lnst. 2005;97(5):396-399.
27. Kirsh VA, Hayes RB. Mayne ST. et al; PLCO Trial.
Supplemental and dietary vitamin E, beta-carotene, and
vitamin C intakes and prostate cancer risk. 1 Nat/ Can-
cer lnst. 2006;98(4):245254.
28. Bleys J, Navas-Acien A, Guallar E. Serum selenium
and diabetes in U.S. adults. Diabetes Care. 2007;
30(4):829834.
29. Rajpathak S, Rimm E, Morris JS, Hu F. Toenail se-
lenium and cardiovascular disease in men with diabetes.
1 Am Col/ Nutr. 2005;24(4):250-256.
30. Stranges S, Marshall JR, Natarajan R, et al. Effects
of long- term selenium supplementation on the inci-
dence of type 2 diabetes: a randomized trial. Ann In-
tern Med. 2007;147(4):217-223.
31 . Miller ER Ill, Pastor-Barriuso R, Dalal D, Riemersma
RA, Appel U, Guallar E. Meta-analysis: high-dosage vi-
tamin E supplementation may increase all -cause
mortality. Ann Intern Med. 2005;142(1):3746.
32. Bjelakovic G. Nikolova D, Gluud ll, Simonetti RG,
Gluud C. Mortality in randomized trials of antioxidant
supplements for primary and secondary prevention: sys-
tematic review and meta-analysis. lAMA. 2007;
297(8):842-857.
33. Sessa HD, Suring JE, Christen WG. et al. Vitamins
E and C in the prevention of cardiovascular disease in
men: the Physicians' Health Study II randomized con
trolled trial. lAMA. 2008;300(18):2123-2133.
34. Lee IM, Gaziano JM, Suring JE. Vitamin E in the pre-
vention of prostate cancer: where are we today? 1 Nat/
Cancer lnst. 2006;98(4):225 227
35. Li H, Stampfer MJ. Giovannucci El , et al. A pro
spective study of plasma selenium levels and prostate
cancer risk. 1 Nat/ Cancer tnst. 2004;96(9):696-703.
36. Taylor PR. Parnes HL, Lippman SM. Science peels
the onion of selenium effects on prostate carcinogenesis.
J Nat/ Cancer lnst. 2004;96(9):645-647.
37. Lippman SM. Lee JJ, Sabichi Al. Cancer chemo
prevention: progress and promise. 1 Nat/ Cancer lnst.
1998;90(20):1514-1 528.
38. Bertagnolli MM, Eagle CJ, Zauber AG, et al; APC
Study Investigators. Celecoxibforthe prevention of spo-
radic colorectal adenomas. N Eng/ 1 Med. 2006;
355(9):873-884.
39. Fisher B, Costantino JP, Wickerham DL, et al. Ta-
moxifen for prevention of breast cancer: report of the
National Surgical Adjuvant Breast and Bowel Project P-1
study. 1 Nat/ Cancer lnst. 1998;90(18):1371-1388.
40. Floss mann E, Rothwell PM; British Doctors Aspirin
Trial and the UK-TIA Aspirin Trial. Effect of aspirin on
long-term risk of colorectal cancer: consistent evi-
dence from randomised and observational studies.
Lancet. 2007;369(9573): 1603-1613.
41. Fukase K, Kato M, Kikuchi S, et al; Japan Gast Study
Group. Effect of eradication of Helicobacter pylori on
incidence of metachronous gastric carcinoma after en-
doscopic resection of early gastric cancer: an open-
l abel. randomised controlled t ri al. Lancet. 2008;
372(9636):392-397.
42. Meyskens FL. Mclaren CE. Pelot D. et al. Difluo-
romethylornithine plus sulindacforthe prevention of spo-
radic colorectal adenomas: a randomized placebo
controlled, double-blind trial. Cancer Prev Res. 2008;
1:9-11.
43. Thompson IM, Goodman PJ, Tangen CM, etal. The
influence of finasteride on the development of pros
tate cancer. N Eng/ 1 Med. 2003;349(3):215-224.
44. Vogel VG, CostantinoJP, Wickerham DL. etal; Na
tiona! SurgicalAdjwantBreastand Bowel Project(NSABP).
Effects of tamoxifen vs raJoxifene on the risk of devel-
oping invasive breast cancer and other disease outcomes:
the NSABP Study of Tamoxifen and Raloxifene (STAR)
P-2 trial. lAMA. 2006;295(23):2727 2741.
45. Scardino PT. The prevention of prostate cancer: the
dilemma continues. N Eng/ 1 Med. 2003;349(3):
297-299.
46. Logothetis CJ, Schellhammer PF. High-grade pros-
tate cancer and the Prostate Cancer Prevention Trial.
Cancer Prev Res. 2008;1 (3):151-152.
47. Lucia MS. Darke AK, Goodman PJ, et al. Patho-
logic characteristics of cancers detected in the Prostate
Cancer Prevention Trial: implications for prostate can
cer detection and chemoprevention. Cancer Prev Res.
2008; 1 (3): 167 173.
48. Lucia MS, Epstein Jl, Goodman PJ, et al. Finaste
ride and high-grade prostate cancer in the Prostate Can-
cer Prevention Trial. 1 Nat! Cancer tnst. 2007;99
(18):1375-1383.
49. Redman MW, Tangen CM, Goodman PJ, Lucia MS.
Collman CA Jr, Thompson IM. Finasteride does not in
crease the risk of high-grade prostate cancer: a bias-
adjusted modeling approach. Cancer Prev Res. 2008;
1(3):174-181.
2009 American Medical Association. All rights reserved. (Reprinted) JAMA. Published online December 9. 2008 E13
J ~
Correction
Citations
the same issue
Subscribe
http://jama.com/ subscribe
Permissions
Vitamins E and C in the Prevention of Prostate and Total
Cancer in Men: The Physicians' Health Study II
Randomized Controlled Trial
J. Michael Gaziano; Robert J. Glynn; William G. Christen; et al.
JAMA. publ ished online Dec 9, 2008; (doi:1 0.1 001/jama.2008.862)
http://jama.ama-assn.org/cgi/content/full/2008.862v1
Contact me if this arti cle is corrected.
This article has been cited 1 time.
Randomized Trials of Antioxidant Supplementation for Cancer Prevention: First
Bias, Now Chance Next, Cause
Peter H. Gann. JAMA. 2008;0(2008) :2008863.
Effect of Selenium and Vitamin E on Risk of Prostate Cancer and Other Cancers:
The Selenium and Vitami n E Cancer Prevention Trial (SELECT)
Scott M. Lippman et al. JAMA. 2008;0(2008) :2008864.
Email Alerts
http://jamaarchives.com/ alerts
http://pubs.ama-assn.org/misc/permissions.dtl
Repri nts/ E-prints
reprints@ama-assn.org
- ORIGINAL CONTIUBUTION JAMA-EXPRESS
Vitamins E and C in the Prevention
of Prostate and Total Cancer in Men
The Physicians' Health Study II Randomized Controlled Trial
J. i\lliehael Gaziano, MD, MPH
Robert J. Glynn, SeD
William G. Christen, SeD
Tobias Kurth, MD, SeD
Charlene Belanger, J.VIA
Jean MacFadyen, BA
Vadim Bubes, PhD
JoAim E. Manson, MD, DrPH
Howard D. Sesso, SeD, MPH
Julie E. Buring, SeD
I
N SOME OBSERVATIONAL STUDIES, IN-
take or blood levels of vitamins E
and C have been associated with re-
duced risk of certain cancers.
1
Ba-
sic research has provided pl ausible
mechanisms by which antioxidant mi-
cronutrients such as vitamin E and vi-
tamin C may delay various steps in car-
cinogenesis. ~ However, definitive proof
that vitamins E and C can reduce the risk
of overall or site-specific cancers must
rely on large-scale randomized trials.
A number of trials have addressed the
potential rol e of vitamins in the pre-
vention of cancer; however, the re-
sul ts from these trials have not been
consistent. Some
5
'
8
but not all
9
-
16
have
supported a role for various antioxi-
dants in the prevention of total or site-
specific cancers. The most compelling
data supporting a role of vitamin E in
the prevention of prostate cancer have
come from the Finnish ATBC (a-
Tocopherol, Beta Carotene) Cancer Pre-
vention Trial.
9
This trial was designed
Context Many individuals take vitamins in the hopes of preventing chronic diseases
such as cancer, and vitamins E and C are among the most common individual supple-
ments. A large-scale randomized trial suggested t hat vitamin E may reduce risk of pros-
tate cancer; however, few trials have been powered to address this relationship. No pre-
vious t rial in men at usual risk has examined vitamin C alone in the prevention of cancer.
Objective To evaluate whether long-term vitami n E or C supplementation de-
creases ri sk of prostate and total cancer events among men.
Design, Setting, and Participants The Physicians' Health Study II is a random-
ized, double- blind, placebo-controlled factorial trial of vitamins E and C t hat began in
1997 and continued until its scheduled completion on August 31, 2007. A total of
14 641 male physicians in the United States ini tiall y aged 50 years or older, including
1307 men with a history of prior cancer at randomization, were enrolled.
Intervention Individual supplements of 400 IU of vitamin E every other day and
500 mg of vitami n C dai ly.
Main Outcome Measures Prostate and total cancer.
Results During a mean follow-up of 8.0 years, t here were 1008 confirmed incident
cases of prostate cancer and 1943 total cancers. Compared with placebo, vitamin E had
no effect on t he incidence of prostate cancer (active and placebo vitamin E groups, 9.1
and 9.5 events per 1000 person-years; hazard ratio [HR), 0.97; 95% confidence inter-
val [CI ). 0.85-1.09; P= .58) or t otal cancer (active and placebo vitamin E groups, 17.8
and 17.3 cases per 1000 person-years; HR, 1.04; 95% Cl, 0.95-1.13; P=.41)_ There
was also no signif icant effect of vitamin Con total cancer (active and placebo vitami n C
groups, 17.6 and 17.5 events per 1000 person-years; HR, 1.01; 95% Cl , 0.92-1 .10; P=.86)
or prostate cancer (active and placebo vitamin C groups, 9.4 and 9.2 cases per 1000
person-years; HR, 1.02; 95% Cl, 0.90-1.15; P=.80). Neither vitamin E nor vitamin C
had a significant effect on colorectal, lung, or other site-specific cancers. Adjustment for
adherence and exclusion of t he first 4 or 6 years of follow-up did not alter the results.
Stratification by various cancer risk factors demonstrated no signif icant modification of
the effect of vitamin E on prostate cancer risk or either agent on total cancer risk.
Conclusions In this large, long-term trial of male physicians, neither vitamin E norC supple-
mentation reduced t he risk of prostate or total cancer. These data provide no support for
the use of these supplements for the prevention of cancer in middle-aged and older men.
Trial Registration clinicaltrials.gov Identifier: NCT00270647
lAMA. 2009;301(1):(doi:10.1001/ jama.2008.862)
Author Affiliations: Divisions of Preventive Medicine
(Drs Gaziano, Gl ynn, Christen, Kurth, Bubes,
Manson, Sessa, and Buri ng and Mss Belanger and
MacFadyen), Aging (Drs Gazi ano, Kurth. Sesso.
and Buring), and Cardiovascular Di sease (Dr Gazi-
ano) , Department of Medicine, Brigham and Wom-
en's Hospital and Harvard Medi cal School, Boston,
Massachusetts; Massachuset ts Veterans Epi de-
miology Research and Information Center, VA Boston
www.jama.com
Healthcare System (Dr Gaziano), Boston; Depart
ment of Ambulatory Care and Prevention (Or Sur-
i ng), Harvard Medical School, Boston; and Depart-
ments of Epidemiology and Biostatistics (Drs Glynn,
Kurth, Manson, and Sesso), Harvard School of Public
Health, Boston.
Corresponding Author. J. Michael Gaziano, MD, MPH,
Brigham and Women's Hospital, 1620 Tremont St, Bos-
ton, MA 02120 (jmgaziano@partners.org).
VITAMINS E AND C AND CANCER PREVENTION IN MEN
to test the effect of vitamin E and beta
carotene on lung cancer risk among
current and past smokers. Although
there was no reduction in risk of lung
cancer with either agent, men as-
signed to active a-tocopherol had a 34%
reduction in the hazard of prostate can-
cer (HR, 0.66; 95% Cl, 0.52-0.86) .
17
The
HOPE-TOO (Heart Outcomes Preven-
tion Evaluation-The Ongoing Out-
comes) trial reported no reduction in
prostate cancer among those treated
with vitamin E compared with pla-
cebo for an average of 7 years (HR, 0. 98;
95% Cl, 0.76-1.26).
1
6
Other trials of vitamin E were not de-
signed specifically to address prostate
cancer risk, and most of these trials were
of just a few years' duration, which may
be too short to detect longer-term ef-
fects on cancer. Vitamin C alone has
been less well studied in large-scale trials.
One other recently completed study
evaluated vitamin C (500 mg daily)
supplementation and total and site-
specific cancer risk among wome.n.
18
Despite uncertainty about the long-
term health effects or benefits, more
than half of US adults take vitamin
supplements, and vitamins E and Care
among the most popular individual
supplements.
19
Given this widespread
use, the gaps in knowledge about the
role of these agents in cancer preven-
tion, and the uncertainties about other
long-term health effects of vitamins E
and C, we conducted the Physicians'
Health Study (PHS) II, a randomized,
double-blind, placebo-controlled fac-
torial trial designed to provide clini-
cally relevant information on the indi-
vidual effects of vitami n E and vitamin
C on total and prostate cancer among
14 641 male physicians treated and fol-
lowed up for an average of 8.0 years.
ln this article, we present the findings
on prostate, total, and other common
cancers. The effects of these agents on
cardiovascular events were recently
published.
20
METHODS
Study Design
The PHS ll is a randomized, double-
blind, placebo-controlled, 2 X 2 X 2 X 2
factorial trial evaluating the balance of
risks and benefits of vitamin E ( 400-IU
synthetic a -tocopherol or its placebo on
alternate days; BASF Corporation, Flo-
rham Park, New jersey), vitamin C
(500-mg synthetic ascorbic acid or its
placebo daily; BASF Corporation) , and
a multivitamin (Centrum Silver or its
placebo daily; Wyeth Pharmaceuti-
cals, Madison, New j ersey) in the pre-
vention of cancer and cardiovascular
disease among 14 641 male physi-
cians 50 years or older.
21
A fourth ran-
domized component, beta carotene
(50-mg Lurotin or placebo on alter-
nate days; BASF Corporation), waster-
minated on schedule in March 2003.
The multivitamin component is con-
tinuing at the recommendation of the
data and safety monitoring committee.
The PHS II study design has previ-
ously been describecl.
20 21
In brief, re-
cruitment, enrollment, and random-
ization of men into PHS II occurred in
2 phases (FIGURE l). Starting in july
1997, 18 763 PHS l participants
10
22
were
invited to participate in PHS IL Men
were ineligible if they reponed a his-
tory of cirrhosis or active liver disease,
were receiving anticoagulants, or re-
ported a serious illness that might pre-
clude participation. Men with a his-
tory of cancer as well as myocardial
infarction or stroke were eligible to en-
roll in PHS 11. Participants also must
have been willing to forgo during the
course of PHS II any current use of mul-
tivitamins or individual supplements
containing more than 100% of the rec-
ommended daily allowance of vitamin
E, vitamin C, beta carotene, or vita-
min A. A total of 7641 willing partici-
pants ( 41%) from PHS I were random-
ized into PHS ll and retained their
original beta carotene treatment
assignment.
ln 1999, invitational letters and base-
line questionnaires were mailed to
254 597 US male physicians 50 years or
older identified (rom a list provided by
the American Medical Association, ex-
cluding PHS I participants. Between July
1999 and july 2001, 42165men com-
pleted a baseline questionnaire. Of
these, ll 128 were willing and eligible
following the same eligibility criteria as
PHS I participanl'i. A 12-week placebo
run-in period excluded men who were
nonadherent, an attribute that typi-
cally emerges during the first several
months of participation.
21
23
Of 11128
physicians who entered the run-in
phase, 7000 willing and eligible men
(63%) took at least two-thirds of their
pills and were randomized into PHS II.
Thus, 14641 men (7641 from PHS
I and 7000 new physicians) were ran-
domized into PHS II in blocks of 16 and
snatified by age; prior diagnosis of can-
cer; prior diagnosis of cardiovascular
disease; and, for the 7641 PHS I par-
ticipants, their original beta carotene
treatment assignment. Men were ran-
domly assigned to receive vitamin E or
its placebo, to receive vitamin Cor its
placebo, and to receive active or pla-
cebo beta carotene and multivitamin.
With respect to vitamins E and C, ran-
domization yielded 4 nearly equal-
sized groups receiving active vitamin E
alone, active vitamin C alone, both ac-
tive agents, or both placebos. There
were 1307 men (8.9%) with a history
of cancer (excluding nonmelanoma
skin cancer) prior to randomization into
PHS 11. All participants provided writ-
ten informed consent, and the institu-
tional review board at Brigham and
Women's Hospital approved the re-
search protocoL
Study Treatment, Follow-up,
and Adherence
Participants were sent monthly calen-
dar packs containing vitamin E or pla-
cebo (taken every other day) and vita-
min Cor placebo (taken daily) every 6
months for the first year and annually
thereafter. Participants were also sent
annual questionnaires asking about ad-
herence, potential adverse events, the
occurrence of new end points, and up-
dated risk factors. Treatment and fol-
low-up continued in blinded fashion
through August 31, 2007, the sched-
uled end of the vitamin E and C com-
ponents of PHS 11. Analyses include fol-
low-up and validation through
September 2008 of reported end points
that occurred through the end of Au-
gust 2007. Morbidity and mortality fol-
low-up rates were extremely high, at
95.3% and 97.7%, respectively. Mor-
bidity and mortality follow-up as a per-
centage of person-time each exceeded
99.9%, with only 1055 and 289 person-
years of potential morbidity and mor-
talit)' follow-up lost through August 31,
2007.
Adherence was defined from partici-
pant. self-reports as taking at least two-
thirds of the study agents. For active vi-
tamin E and its placebo, adherence at
4 years was 78% and 77%, respec-
tively (P=.l2), and at the end of fol-
low-up (mean, 8.0 years) , 72% and 70%
(P= .004). For active vitamin C and its
placebo, adherence at 4 years was 78%
and 78%, respectively (P= .99), and at
the end of follow-up, 71% and 71%
(P =.54). There were no differences be-
tween groups in average rates of indi-
vidual nontrial vitamin E (3.2% ac-
tive, 3.1% placebo) or vitamin C
supplement use (3.8% active, 4.4% pla-
cebo) for 31 or more days per year
(drop-ins) at the end of the trial (each
P> .05).
Confirmation of End Points
For the vitamin E component, the pri-
mary cancer end point was total pros-
tate cancer; total cancer (excluding non-
melanoma skin cancer) was the primary
end point for the vitamin C compo-
nent and a prespecified secondary end
point for vitamin E. Incident colorec-
tal cancer was another prespecified sec-
ondary end point for the vitamin E com-
ponent. Other individual sites of cancer
were assessed and validated, as well as
total mortality and cancer mortality. For
each end point reported by partici-
Figure 1. Flow Diagram of Participants From Screening to Completion of the Vitamin E and Vitamin C Components of the Physicians' Health
Study II
18 763 Men from Physicians' Health
Study I invited to participate
348 Excluded
256 Incomplete response or
nonrespoose to onvitatiOfl
92 Dead
7641 Bigible
3656 Randomized to receive active
vitamin E (400 IU every other day)
and active vitamin C (500 mg/d)
68 Unknown vi tal status
I
3497 Included in analysis of prostate
cancer
3656 Included in analysis of total
cancer
3659 Randomized to rece<Ve active
\iitamin E (400 IU every
other day) and placebo vitamin C
-
Status on August 31, 2007
3169 Alive
399 Dead
91 Unknown vital status
I

cancer
I
254 597 Male physicians mvited to
I
participate
212 432 Excluded
182 439 Nonresponse to
invitation
23 261 UnfO<Wardable
5201 Unwilling to partocopate
1531 Dead
I 42 165 Completed baseline questionnaire I
r--- -
31 037 Excluded
20 591 Unwolling to participate
9908 Ineligible
538 Incomplete response
L
11 128 Entered run-in phase _j
4128 Excluded
2982 Noo-.adherent
978 Unwilling to participate
135 Ineligible
33 Dead
7000 Eligible
I
cancer
canoor
L
-
Status on August 31, 2007
3169 Alive
405 Dead
79 Unknown status
I
3491 Included on analysis of prostate
cancer
cancer
Those classified as "unforwardable" were not able to be contacted by mail. For the primary end point of prostate cancer, primary analyses were restricted to 13 983 men
without prostate cancer at baseline. The primary analysis of total cancer included all 14 641 men.
2009 American Medical Association. All rights reserved. (Reprinted) JAMA, Published online Dew nber 9, 2008 E3
pants by follow-up questionnaire, let-
ter, telephone call, and other corre-
spondence, we requested permission
from the participant to examine rel-
evant medical records. Once consent
was obtained, records were requested
from the hospital or attending physi-
cian and reviewed by an end-points
committee of physicians blinded to ran-
domized treatment assignment.
The vast majority of cancers were
confirmed with pathology or cytology
reports. Rarely, the committee con-
firmed a reported case of cancer based
on strong clinical and radiological or
laboratory marker evidence when pa-
thology or cytology review was not con-
ducted. Total mortality was con-
firmed by the end-points committee or
by obtaining a death certificate. A Na-
tional Death Index search was per-
formed for any participants with un-
known vital status. Only confirmed end
points are included in this report.
Statistical Analyses
All primary analyses classified study
participants based on the intention-to-
treat principle, in which randomized
participants were classified according
to their randomized vitamin E or vita-
min C treatment assignments and were
followed up until the occurrence of a
disease end point, death, loss to follow-
up, or the end of the vitamin E and vi-
tamin C components of PHS II on Au-
gust 31, 2007, whichever came first
SAS version 9.1 (SAS Institute Inc, Cary,
North Carolina) was used, with statis-
tical significance set at P < .05 using
2-sided tests.
The PHS li was designed to have
greater-than-80% power to detect a 13%
reduction in the hazard of total cancer
(excluding nonmelanoma skin can-
cer) and a 19% reduction in the haz-
ard of prostate cancer. Estimates of
study power relied on historical event
rates observed in PHS physicians that
predicted the trial would accrue 418 in-
cident prostate cancer cases and 877
total cancer cases in the half of partici-
pants randomized to a placebo treat-
ment. The actual number of cases of
prostate cancer and total cancer ac-
crued in the vitamin E placebo group
exceeded the numbers predicted for
power calculations by 23% and 9%,
respectively.
We first compared baseline charac-
teristics by vitamin E or C treatment as-
signment to evaluate whether random-
ization equally distributed baseline
characteristics. Cox proportional haz-
ards models were used to calculate the
hazard ratios (HRs) and 95% confi -
dence intervals ( Cls) comparing event
rates in the vitami n E and placebo
groups and the vitamin C and placebo
groups. For each prespecified end point,
models were stratified on the pres-
ence of cancer at randomization and ad-
justed for study design variables: age;
PHS cohort (original PHS I partici-
pant, new PHS U participant); and ran-
domized beta carotene, vitamin E or vi-
tamin C, and multivitamin assignments.
For total cancer analyses, all new can-
cers were included, regardless of
whether the participant had a baseline
history of cancer . For each site-
specific cancer analysis, participants
were excluded if they had a baseline his-
tory of cancer of the same site. Thus,
these analyses included 13 983 men ini-
tially free of prostate cancer,l4 520 ini-
tially free of colorectal cancer, and
14 610 initially free of lung cancer.
For analyses of the secondary end
points of total mortality, any cancer
mortality, and site-specific cancer
deaths, we included all participants. The
association between vitamin E and pros-
tate cancer mortality was also exam-
ined separately among the 13 334 men
without and 1307 with a baseline his-
tory of cancer. We tested the propor-
tional hazards assumption by model-
ing interaction terms separately for
vitamin E or C with the logarithm of
time, and these assumptions were not
violated (P> .05). We then investi-
gated whether vitamin E or C adher-
ence affected our primary results
through sensitivity analyses that cen-
sored follow-up when a participant re-
ported taking less than two-thirds of
either vitamin E or vitamin Cover the
previous year. To explore a possible late
benefit associated with vitamin E or C,
we analyzed separately the person-
time and outcomes in the first 4 years
of treatment and then after 4 years. Ad-
ditional exploratory analyses consid-
ered estimated treatment effects >vi thin
2-year time intervals and also whether
an effect was present if only events and
person-time after 6 years of treatment
were considered. Finally, we con-
ducted subgroup analyses stratified by
major cancer risk factors as well as pre-
vious cancer history and assessed effect
modification by using interaction tenns
between subgroup indicators and either
vitamin E or C assignment.
RESULTS
The PHS II randomized 14 641 men
with a mean (SD) age of 64.3 (9.2)
years. Randomization equally distrib-
uted all baseline characteristics be-
tween vitamin E or vitamin C and their
placebo groups (all P> .05) (TABLE 1).
During a mean follow-up of 8.0 years
(median, 7.6 years; interquartile range,
7.1-9.6 years; ma>.imum, 10.0 years),
total follow-up was 117 711 person-
years. There were 1943 confirmed total
cancer cases and 1008 prostate cancer
cases, with some men experiencing
multiple events. A total of 1661 men
died during follow-up.
Vitamin E and Cancer
The overall rates of prostate cancer were
9.1 and 9.5 per 1000 person-years in
the active and placebo vitamin E groups,
respectively. There was no effect of vi-
tamin E on prostate cancer incidence
(HR, 0.97; 95% Cl, 0.85-1.09; P= .58)
(TABLE 2). The cumulative incidence
curves indicate that this lack of effect
did not vary for up to 10 years of treat-
ment and follow-up Clog-rank P= .53)
(FIGURE 2). For total cancer, the over-
all rates were 17.8 and 17.3 per 1000
person-years in the active and placebo
vitamin E groups, respectively.
Compared with placebo, vitamin E
also did not reduce the incidence of total
cancer (HR, 1.04; 95% CI, 0.95-1.13;
P= .41). We found no effect for any si te-
specific cancers, including colorectal
(HR, 0.88; 95% Cl, 0.64-1.19; P=.40),
lung (HR. 0.89; 95% CI, 0.60-1.31;
P=.55), bladder (HR, 1.21; 95% Cl,
0.76-1.94; P= .43), and pancreatic (HR,
1.14; 95% Cl, 0.67-1.93; P= .63).ln ad-
dition, there was no significant effect
of vitamin E on total mortality (HR,
1.08; 95% CI, 0.98-l.l9; P=.l3) or can-
cer mortality (HR, 1.13; 95% CI, 0.95-
1.34; P = .16). Censoring participants at
the time of vitamin E nonadherence did
not impact the results for prostate can-
cer (HR, 0.95; 95% CI, 0.84-1.07;
P= .38) or total cancer (HR, 1.02; 95%
Cl, 0.93-l.ll; P=.68).
We next evaluated whether base-
line history of cancer, risk factors, and
other randomized interventions from
PHS II modified the effect of vitamin E
on prostate cancer or lOtal cancer
(TABLE 3). Among 13 334 men with-
out a baseline history of cancer, vita-
min E had no effect on the prevention
of total cancer, nor was there any sub-
stantial effect of vitamin Eon newly di-
agnosed cancer among 1307 men with
cancer at baseline or on prostate can-
cer among men with another cancer at
baseline. In addition, we found no other
significant effect modification by base-
line risk factors on prostate cancer. In
addition, there was no effect modifica-
tion by randomized beta carotene or vi-
tamin C or the ongoing multivitamin
treatment assignment. Analyses fo-
cused on the possibility that a number
of years of treatment were required be-
fore emergence of an effect found no
apparent relationship of vitamin E with
ei ther prostate or total cancer in data
restricted to person-time and events af-
ter 4 years of treatment. Further re-
striction to events and time after 6 years
of treatment similarly found no appar-
ent relationships.
Vitamin C and Cancer
The overall rates of total cancer for the
active and placebo vitamin C groups
were 17.6 and 17.5 per 1000 person-
years, respectively. There was no effect
of vitamin Con the primary end point
of total cancer (HR, 1.01; 95% Cl, 0.92-
1.10; P= .86) (Table 2) . The cumula-
tive incidence curves showed no dif-
ference between groups in the HRs over
time (log-rank P= .92) (Figure 2) . The
rates of prostate cancer were 9.4 cases
per 1000 person-years for the active vi-
tamin C group and 9.2 cases per 1000
person-years for the placebo group (HR,
1.02; 95% Cl, 0.90-1.15; P= .80). Vita-
min C also had no effect on other site-
specific cancers, including colorectal
(HR, 0.86; 95% Cl, 0.63-1..17; P= .35),
lung (HR, 0.95; 95% CT, 0.64-1..39;
P=. 78), bladder (HR, 0.85; 95% Cl,
0.53-1.36; P= .49), and pancreatic (HR,
0.97; 95% Cl, 0.57-1.64; P=.91). In
addition, no effect was found between
vitamin C and either total mortality
(HR, 1.07; 95% Cl, 0.97-1.18; P=.17)
or cancer mortality (HR, 1.06; 95% Cl,
0.89-1.25; P= .53). Censoring for non-
adherence with vitamin C did not ap-
Table 1. Baseline Characteristics According to Vitamin E and Vitamin C Treatment
Assignment in 14641 Men From the Physicians' Health Study II
Men, No. (%)
8
Vitamin Eb Vitamin Cb
Active Placebo Active Placebo
Self-reported Baseline Characteristics (n = 7315) (n = 7326) (n = 7329) (n = 7312)
Age, mean (SD), y 64.2 (9.1) 64.3(9.2) 64.3 (9.2) 64.3 (9.1)
Age, y
50-59 2940(40.2) 2951 (40.3) 2953 (40.3) 2938 (40.2)
60-69 2349(32. 1) 2347 (32.0) 2348 (32.0) 2348 (32.1)
=::70 2026(27.7) 2028(27.7) 2028 (27.7) 2026(27.7)
Body mass index, mean (SD)c 26.0(3.6) 26.0 (3.7) 26.0 (3.6) 26.0(3.7)
Cigarette smoking
Never 4104 (56. 1) 4148 (56.7) 4135 (56.5) 4117 (56.4)
Former 2967 (40.6) 2885 (39.4) 2908 (39.7) 2944 (40.3)
Current 239 (3.3) 285 (3.9) 280 (3.8) 244 (3.3)
Exercise =:: 1 time/ wk
No 2739 (38.4) 2766 (38.7) 2759 (38.5) 2746 (38.6)
Yes 4389 (61.6) 4383 (61.3) 4408 (61.5) 4364 (61.4)
Alcohol consumption
Rarely/never 1372 (18.9) 1358 (18.7) 1364 (18.7) 1366 (18.8)
=:: 1 drink/mo 5893 (81.1) 5923 (81.4) 5920 (81.3) 5896(81.2)
Current aspirin use
No 1627 (22.6) 1634 (22.6) 1638 (22.6) 1623 (22.6)
Yes 5578 (77.4) 5589 (77.4) 5605 (77.4) 5562 (77.4)
Parental history of cancerd
No 2906 (46.5) 2931 (46.5) 2927 (46.5) 2910 (46.5)
Yes 3344 (53.5) 3377 (53.5) 3371 (53.5) 3350(53.5)
Paternal history of prostate cancerd
No 5713(89.6) 5792 (89.8) 5755 (89.5) 5750(89.9)
Yes 663 (10.4) 661 (10.2) 678 (10.5) 646(10.1)
Parental history of colorectal cancerd
No 5492 (88.0) 5552 (88.5) 5535 (88.5) 5509 (88.1 )
Yes 748 (12.0) 719 (11.5) 721 (11.5) 746(11.9)
Self-reported history of cancer
No 6657 (91.0) 6677 (91.1) 6675 (91.1) 6659(91. 1)
Yes 658(9.0) 649 (8.9) 654 (8.9) 653 (8.9)
Self-reported history of prostate cancer
No 6983(95.5) 7000 (95.6) 7006 (95.6) 6977 (95.4)
Yes 332 (4.5) 326 (4.4) 323 (4.4) 335 (4.6)
Self-reported history of colorectal cancer
No 7253 (99.2) 7267 (99.2) 7270 (99.2) 7250 (99.2)
Yes 62 (0.8) 59(0.8) 59 (0.8) 62 (0.8)
a Unless otherwise indicated. The numbers do not always surn to group totals because of missing information for some
variables.
bp>.05 for all comparisons between active and placebo groups of vitamin E and vitamin C.
ccalculated as weight in kilograms divided by height in mews squared.
dExcludes 2083. 1812, and 2130 men with missing Information on parental hist01y of cancer, prostate cancer, and
colorectal cancer, respectively.
preciably affect our findings for total
cancer (HR, 1.00; 95% Cl, 0.92-1.09;
P=.98).
We then evaluated whether base-
line history of cancer, risk factors, and
other randomized interventions from
PHS ll or follow-up time modified the
effect of vitamin Con the primary end
point total cancer (TABLE 4) . There was
Tabl e 2. Association Between Randomized Vitamin E and Vitami n C Assignment and the Risk of Total Cancer, Site-Specific Cancer, and
Mortality in the Physicians' Health Study 11 a
Vitamin E Vitamin C
Men in Events, No. Events, No.
Analysis, Adjusted HR Adjusted HR
Outcome No.b Active Placebo (95% CI)C Active Placebo (95% Cl)c
Total cancer 14641 984 959 1.04 (0.95-1.13) 973 970 1.01 (0.92-1.10)
Total epithelial cell cancer 14641 876 877 1 .01 (0.92- 1.1 1) 873 880 1.00 (0.91-1.09)
Prostate cancer 13983 493 515 0 97 (0.85-1 .09) 508 500 1 02 (0.90-1 .15)
Prostate cancer death 14641 37 39 1.01 (0.64-1.58) 45 31 1.46 (0.922.31)
Colorectal cancer 14 520 75 87 0.88 (0.64-1.19) 75 87 0.86 (0.63-1.17)
Colorectal cancer death 14641 21 32 0.68 (0.391.18) 27 26 1.04 (0.611.78)
Lung cancer 14610 48 55 0.89 (0.60-1.31) 50 53 0.95 (0.64-1.39)
Lung cancer death 14 641 44 43 1.05 (0.691.60) 39 48 082 (0.53-1 .25)
Bladder cancer 14570 38 32 1.21 (0.76-1.94) 32 38 0.85 (0.53-1.36)
Bladder cancer death 14641 9 12 0.79 (0.33-1 .88) 10 11 0.92 (0.39-2.17)
Pancreatic cancer 14638 29 26 1.14 (0.67-1.93) 27 28 0.97 (0.57 1.64)
Pancreatic cancer death 14641 27 23 1 20 (0.69-2.09) 23 27 0 86 (0.49-1 .49)
Lymphoma 14595 73 60 1.23 (0.88-1.74) 69 64 1.08 (0.77-1.52)
Lymphoma death 14641 25 21 124 (0.69-2.21) 21 25 0 84 (0.47 -1 .50)
Leukemia 14613 46 34 1.38 (0.882.15) 44 36 1.23 (0.79-1 .91)
Leukemia death 14 641 20 12 1.71 (0.83-3.49) 18 14 1.30 (0.65-2.62)
Melanoma 14486 74 63 1.18 (0.85-1.66) 63 74 0.86 (0.611.20)
Melanoma death 14641 6 5 1.25 (0.38-4.09) 7 4 1.75 (0.51-5.98)
Total mortality 14 641 841 820 1 08 (0.981.19) 857 804 1.07 (0.97-1 .18)
Cancer mortality 14641 273 250 1.13 (0.95-1.34) 268 255 1 .06 (0.89-1.25)
Abbreviations: Cl, confidence interval; HR, hazard ratio; PHS, Physicians' Health Study.
a Mean follow-up of 8.0years for all 14641 men through August 31, 2007.
b For total cancer, site-specific mortality, total mortality. and cancer mortality, analyses included al114 641 participants. For the incidence of site-specific cancers, analyses were
restricted to men Without that site-specific cancer at baseline.
c Adjusted for age, PHS cohort (original PHS I participant. new PHS participant), and randomized treatment assignment (beta carotene. multMtamin, and either vitamin E or vitamin C)
and stratified on baseline cancer.
Figure 2. Cumulative Incidence Rat es of Total Prostate Cancer by Randomized Vitamin E Assignment or Total Cancer by Randomi zed
Vitamin C Assignment in the Physici ans' Health Study II
Vitamin E
0. 18
li;
0
c
0. 16
8 0. 14
- Placebo vitamin E

- ActivevitaminE
0. 12 {/)
e
Cl.
0 0.10
.,
,.,,.,..
..
0
c
008 <D
32 /
0
0.06
'
~
~ . . -
. ~ .. -
1i! 0.04
...... ~ .. "'"' :0
E
.-
:0
0.02 (.)
Log-rank P= .53
0 2 3 4 5 6 7 8 9
Follow-up, y
No. at Risk
Placebo 7000 6913 6799 6673 6539 6382 6251 5194 2965 2758
Active 6983 6871 6770 6650 6517 6357 6230 5204 2921 2750
E6 JAMA, Published online December 9, 2008 (Reprinted)
10
0. 18
0. 16
I;;
0
c
0. 14
"' 0
3.ii
{2
0. 12
0
0) 0.10
g
.,
008 :2
0

~
0.06
:0
0.04
E
8
0.02
0
No. at Risk
/ '
,.
VitaminC
Placebo vitamin C
- Ac11Vevitam1nC
Log-rank P= .92
3 4 5 6 7 8 9 10
Follow-up, y
Placebo 7312 7150 6997 6635 6674 6468 6301 5214 2958 2744
Active 7329 7197 7033 6848 6663 6461 6287 5195 2925 2709
2009 American Medical Association . AU rights reserved.
VITAMI NS E AND C AND CANCER PREVENTION IN MEN
no effect modification by any cancer
risk factor on the effect of vitamin Con
total cancer, including stratifying by
previous his tory of cancer. ln addi-
tion, there was no effect modification
b y randomized treatme nt ass ign-
ments, including beta carotene, vita-
min E, or the ongoing multi vitami n
component. To evaluate the effect of the
latency period, the effect of vitamin C
treatment assignment was examined by
years of foll ow-up in 2-year incre-
ments. Further, analyses restricted to
events and times after 4 years of treat-
ment, or after 6 years, found no asso-
ciation of vitamin C with total cancer.
When we examined the 2-way inter-
action between randomized vitamin E
and vitamin C assignments, we found
no significant interactions for either
total cancer (P for interaction= .87) ,
pros ta t e can cer (P for int erac-
tion = .55), colorectal cancer (P for in-
teraction = .59), or lung cancer (P for in-
teraction= .44) (FIGURE 3).
Adverse Effects
As previously published ,
20
we as-
sessed a number of potential adverse ef-
fects of vitamins E and C, and there
were no significant effects of either
agent on minor bleeding (including he-
maturia, easy bruising, and epis taxis)
or gastrointestinal tract symptoms (pep-
tic ulcer, constipation, diarrhea, gas-
tritis , and nausea), fatigue, drows i-
ness, skin discoloration or rashes, or
migraine. A greater number of hemor-
rhagic strokes was observed among
those assigned to vitamin E compared
with placebo (39 vs 23 events; HR, l. 74;
95% CI, 1.04-2.91) , a finding that was
observed in the ATBC Cancer Preven-
tion Trial
9
but not observed in other
trials of vitamin E.
Table 3. Association Between Randomized Vitamin E and t he Risks of Prostate Cancer and Total Cancer According to Baseli ne Characteristics,
Treatment Assignment. and Follow-up Time in the Physicians' Health Study 11 a
Prostate Cancer Total Cancer
Events, No. Events, No.
Adjusted HR Ptor Adjusted HR Pfor
Group Active Placebo (95% Cl)b Interaction Active Placebo (95% Cl)b Interaction
Age, y
50-59 118 120
0.99 (0.77-1.27) J
212 205 1.04 (0.86-1.26) J
60-69 209 237 0.88 (0.73-1.06) .37 382 381 1.01 (0.87-1.16) .84
;;::?Q 166 158 1.06 (0.86-1 .32) 390 373 1.06 (0.92-1 .22)
Body mass indexc
<25 215 211
1.01 (0.84-1 .23) J
420 396 1.05 (0.92-1 .21) J
25-29 231 252 0.95 (0.79- 1.13) .79 449 470 0.99 (0.87-1.13) .44
;;::3Q 47 51 089 (0.60-1.33) 114 92 1 .20 (0.91-1 .58)
Smoking status
Never 269 282
098 (0.83-1 .16) J
491 491
1.04 (0.91-1 .17) J
Former 207 218 0.92 (0.76-1.12) .47 451 424 1 .03 (0.90-1.17) .82
Current 17 14 1.50 (0.74-3.04) 42 43 1 .23 (0.80-1.88)
Exercise ;;,: 1 time/wk
No 194 198 0.98 (0.81-1.20) J
.99
401 374 1.10(0.95-1.26) J
.44
Yes 298 306 0.99 (0.84-1 .16) 569 562 1 02 (0.91-1 .15)
Alcohol consumption
Rarely/never 74 85 0.87 (0.64-1 .19) J
.43
173 163 1 .07 (0.86-1.32) J
.80
;;::1 drinklmo 415 424 0.99 (0.87-1.14) 806 790 1.03 (0.94-1.14)
Current aspirin use
No 97 99 1 .00 (0. 76-1.33) J
.88
218 198 1.14 (0.94-1.39) J .31
Yes 395 411 097 (0.85-1.1 1) 758 749 1 02 (0.92-1.13)
No 182 194 093(0.76-1.14) J
.73
357 347 1.04(0.90-1.21) J
.88
Yes 249 257 0.98 (0.83-1.1 7) 473 463 1.03 (0.91-1.17)
History of cancer
No 474 498 0.96 (0.85-1.09) J
.57
895 882 1.03 (0.94-1.13) J
.42
Yes 19 17 1 '1 0 (0.57 -2.13) 89 77 1 '1 4 (0.84-1.55)
Randomized to vitamin C
Placebo 255 245 1 .05 (0.88-1.25) J
.19
491 479 1 .03 (0.91-1 .1 7) J .89
Active 238 270 0 89 (0. 75-1 .06) 493 480 1.05 (0.92-1 .19)
Period of follow-up, y
< 4 211 242 095 (0.79-1.15) J
.39
437 444 1.01 (0.89-1 .16) J
.53
2 4 282 273 1 .04 (0.88- 1.23) 547 515 1 .08 (0.95-1.21)
Abbreviations: Ct. confidence interval; HR. hazard ratio; PHS, Physicians' Health Study.
a For prostate cancer, analyses were restricted to 13 983 men without prostate cancer at baseline. For total cancer, analyses included all 14 641 participants.
b Adjusted for age, PHS cohort (original PHS I participant. new PHS participant), and randomized treatment assignment (beta carotene. multivitamin, and vitamin C).
Calculated as weight in kilograms divided by hetght in meters squared.
d Excludes 2083 meo with missing infOf'mation on parental history of cancer.
COMMENT
ln this large-scale, randomized con-
trolled trial among middle-aged and
older men, neither long-term vitamin
E nor vitamin C supplementation re-
duced the risk of prostate or total can-
cer. Neither vitamin reduced the risk
of cancer death; major site-specific can-
cers, including colorectal, lung, blad-
der, pancreatic, lymphoma, leukemia,
or melanoma; or total mortality. There
was no suggestion of a latent effect for
either vitamin.
Vitamin E and Cancer
The most compelling data suggesting
that vitamin E may reduce the risk of
prostate cancer come from the Finn-
ishATBC Cancer Prevention Trial. The
ATBC trial was a randomized, double-
blind, placebo-controlled trial of ex-
tocopherol (50 mg daily) and beta caro-
tene (20 mg daily) among 29 133 male
smokers. Among those assigned to 50
mg of ex-tocopherol supplementation
daily, there was no overall reduction in
cancer risk; however, there was a 34%
Table 4. Association Between Randomized Vitamin C and the Risk of Total Cancer According
to Baseline Characteristics, Treatment Assignment, and Follow up Time in the Physicians'
Health Study 11 a
Total Cancer
Events, No.
Adjusted HR Pfor
Group Active Placebo (95% Cl)b Interaction
Age, y
50-59 205 212
0.96 (0.791.16) J
60-69 388 375 1.05 (0.91-1.21) .77
270 380 383 1.00 (0.87-1.16)
Body mass index
0
<25 402 414
0.98 (0.86-1. 13) J
2529 460 459 0.99 (0.87-1 .13) .56
2:30 110 96 1. 16 (0.88-1.52)
Smoking status
Never 506 476
1.08 (0.95-1.23) J
Former 426 449 0.95 (0.83-1.08) .15
Current 40 45 0. 76 (0.49-1.16)
Exercise 21 time/wk
No 403 372 1.09 (0.94-1.25) J
.15
Yes 551 580 0.95 (0.84- 1.07)
Alcohol consumption
Rarely/ never 168 168 1.00 (0.81 - 1.24) J
.99
2: 1 drink!mo 801 795 1.01 (0.92-1 .1 1)
Current aspirin use
No 205 211 0.97 (0.80-1 .18) J
.70
Yes 761 746 1.02 (0.92-1.13)
No 339 365 0.94 (0.81-1.09) J
.20
Yes 483 453 1.06 (0.93-1.21)
History of cancer
No 898 879 1.03 (0.94-1.13) J
. 17
Yes 75 91 0.82 (0.61 -1.12)
Randomized to vitamin E
Placebo 480 479 1.00 (0.88-1.14) J
.89
Active 493 491 1.01 (0.89-1.15)
Period of follow-up. y
< 4 443 438 0.96(0.84-1.10) J
.43
2 4 530 532 1.01 (0.891.13)
Abbreviations: Cf. confi dence interval; HR. hazard ratio; PHS. Physicians' Health Study.
a For total cancer, analyses included all 14 641 participants.
b Adjusted for age, PHS cohort (original PHS I participant, new PHS participant), and randomized treatment assign-
ment (beta carotene, multivitamin, and vitamin E).
Calculated as weight in kilograms divided by hetght tn meters squared.
d Excludes 2083 men with missing infOf'mation on parental history of cancer.
reduction in pros tate cancer inci-
dence during a median follow-up pe-
riod of 6.1 years.
9
This effect attenuated after several
years of posttrial follow-up. The effect
of vitamin E appeared to be s tronger on
more advanced tumors. A 41% reduc-
tion in prostate cancer mortality was also
observed. Since prostate cancer was not
a prespecified end point, it remains pos-
sible that this finding was due to chance.
Other completed trials of vitamin E, in-
cluding several among individuals at
high risk for cardiovascular disease, were
not powered to address the possible ben-
efit of vitamin Eon prostate cancer. The
HOPE-TOO prostate cancer results dem-
onstrated no clear harm or benefit of 400
lU daily of vitamin E.
16
In response to the A TBC finding, we
launched the PHS 11 trial to specifi-
cally test the hypothesis that vitamin E
might prevent pros tate cancer in
middle-aged or older men. Because the
effect in the A TBC trial appeared to be
stronger in later -stage cancers, we chose
not to screen for prostate cancer and
even included a small number of par-
ticipants with previously diagnosed can-
cers. The Selenium and Vitamin E Can-
cer Prevention Trial (SELECT) was also
designed to assess the role of vitamin
E in the prevention of incident pros-
tate cancer.
24
In contrast to PHS 11,
SELECT enrolled men initially free of
prostate cancer based on baseline pros-
tate-specific antigen values and digital
rectal examinations. Using a factorial
design, SELECT also tested selenium in
the prevention of prostate cancer.
One notable difference between the
PHS 1l and ATBC trials was the preva-
lence of smoking in A TBC and the very
low levels of smoking in PHS 11. lf the
effect of vitamin E was confined to
smokers, PHS II would likely miss this
effect. Another notable difference be-
tween the 2 trials was the lower dose
of vitamin E in ATBC (50 mg daily),
compared with 400 IU on alternate days
in PHS II. Alternatively, the results of
PHS 11 suggest that the observed ben-
eficial findings of vitamin Eon the de-
velopment of prostate cancer from the
A TBC trial may have been due to the
ES JAMA, Published online December 9, 2008 (Reprinted) 2009 American Medical Association. AU rights reserved.
play of chance. This illustrates the im-
portance of cautious interpretation of
findings on secondary end points.
Mixed results have been obtained
from trials of vitamin E supplementa-
tion and total cancer. In the ATBC trial,
there was no reduction in risk of total
cancer among those randomized to a-
tocopherol (50 mg daily) and/or beta
carotene (20 mg daily) , and vitamin E
alone had no effect on the primary end
points of lung or total cancer.
25
1n the
Chinese Cancer Prevention Trial, con-
ducted among 29 584 poorly nour-
ished residents of Linxian, China, those
assigned to a combined daily treat-
ment of vitamin E (30 mg), beta caro-
tene (15 mg), and selenium (50 pg) ex-
perienced statistically significant
reductions of 9% in total mortality, 13%
in cancer mortality, and 21% in gas-
tric cancer mortality after nearly 6 years
of treatment and follow-up.
5
How-
ever, these results may not be general-
izable to well-nourished populations.
Moreover, because 3 agents were tested
in combination, the specific benefit of
vitamin E, beta carotene, or selenium
cannot be determined. In the ATBC
trial, those assigned to vitamin E had
a nonsignificant 22% reduction (HR,
0.78; 95% CI, 0.55-1.09) in colorectal
cancer incidenceY
17
ln the \Nomen's
Health Study, there was no reduction
in the risk of colorectal cancer among
middle-aged and older women
(HR, 1.00; 95% CI, 0.77-1.31).
15
Our
findings do not support a role of vita-
min E in the prevention of total can-
cer, colorectal cancer, or other com-
mon cancers.
Vitamin C and Cancer
in contrast to vitamin E, which is
available in a limited number of
foods, vitamin C is derived from
many fruit and vegetable sources. ln a
review of data from more than 90 epi-
demiologic studies of dietary intake of
vitamin C (or foods that supply vita-
min C) and total cancer, Block
26
found that almost all showed a pro-
tective relationship, with a median
2-fold increased relative risk for low
compared with high intake. The
Figure 3. Hazard Ratios of Total Cancer, Prostate Cancer, Colorectal Cancer, and Lung
Cancer Compari ng Combinati ons of Active Vitami n E and Active Vitamin C Groups With
the Placebo Vitamin E and Placebo Vitamin C Groups in the Physicians' Health Study II
Favors Favors P l or
Outcome Events, No. Men, No. Active Placebo Interaction
Total cancer 1943 14641
Placebo vitamin E and C [reference) 479 3653

}87
Active vitamin E 491 3659
---
Active vitamin C 480 3673
--
Active vitamin E and C 493 3656
Prostate cancer 1008 13983
Placebo vitamin E and C [reference) 245 3491

} ss
--
Active vitamin C 270 3509
--
Colorectal cancer 162 14520
Placebo vitamin E and C [reference] 45 3624

J s9

Active vitamifl C 42 3643

Lung cancer 103 14610
Placebo vitamin E and C [reference] 32 3644

}44

Active vitamifl C 23 3667

0.4 0.6 08 1.0 1.2 1.4 1.6
Hazard Ratio (95% Cl)
CJ indicates confidence interval; HR. hazard ratio. Error bars indicate 95% Cis.
effects were statistically significant in
three-fourths of the studies. Both
dietary intake
27
-
29
and blood-based
studies have shown inverse relation-
ships. Epidemiologic evidence sug-
gests an inverse association between
dieta1y intake of vitamin C and risk of
a variety of specific cancers.
26
Pub-
lished reports show significant protec-
tive effects of vitamin C on breast,
oral, gastric, esophageal, pancreatic,
lung, cervical, and rectal cancer, while
none have reported elevated risk with
increasing intake.
2630
3
t
A major gap in the evidence regard-
ing a possible role of vitamin C in the
prevention of cancer is lack of data from
large-scale primary prevention trials.
Secondary prevention trials focusing on
vitamin C for the recurrence of colon
cancer or polyps have also yielded
mixed results, ranging from a nonsig-
nificant reduction (relative risk, 0.86;
95% Cl, 0.51-1.45) among individu-
als with colon polyps assigned to a com-
bination of vitami n E and C, com-
pared with placebo,
32
to no evidence
that either combined vitamins E and C
or beta carotene alone reduced the in-
cidence of subsequent colorectal ad-
enomas among patients with previous
adenoma.
13
The PHS II attempted to fill the gap
in the vitamin C literature with a long-
term trial of indi vidual vitamin C
supplements at a commonly used dose
in a large group of men. Our findings
of no reduction in risk of total cancer
and no dear evidence of a reduction in
site-specific cancers do not support the
use of vitamin C supplementation in the
prevention of cancer. lt remains pos-
sible that vitamin C intake is a marker
of other nutrients that are consumed
with vitamin C in the diet.
Strengths and Limitations
Major strengths of this study were the
high levels of adherence to the study
agents over a long period of Lime and
the high quality of reporting of health
information. Further, the conduct of
this trial entirely by mail greatly in-
creased the efficiency and reduced the
cost of this study. Recruitment costs
were about $200 per participant and an-
nual follow-up costs were about $100
per participant, a fraction of the cost of
similar studies that requi re establish-
ment of many study sites with dedi-
cated research personnel.
The study was conducted in a well-
nourished population, and thus, these
resulls may not preclude potential ben-
efits in less well-nourished popula-
tions. One concern is the choice of dose
used. lt is not feasible to test multiple
doses in these large-scale trials. The
doses of vitamin E and C in the PHS
were chosen because they were in the
range of those commonly in use, be-
cause they did not have known major
adverse effects that would impact ad-
herence, and because their safety data
were sound-a critical issue when con-
ducting a trial by mail. The form of vi-
tamin E chosen for our study was syn-
thetic a-tocopherol, the most abundant
component of natural vitamin E. How-
ever. in nature, vitamin E is composed
of both a- and ')'-tocopherol. ')'-
Tocopherol has been postulated to pos-
sibly play a more important role in pros-
tate cancer protection.
34
The duration of a large-scale trial is
also an issue of concern. We had to bal-
ance the desire to extend treatment and
follow-up as long as possible consid-
ering issues of cost and adherence; how-
ever, it remains possible that chemo-
prevention may require even longer
durations of treatment and follow-up
than is feasible in randomized trials. We
will continue to follow the PHS II co-
hort for emergent latent effects. Adher-
ence remains an issue of concern in any
long-term study, but levels of adher-
ence in the PHS 1I remained good af-
ter a mean follow-up of 8.0 years. lt re-
mains possible that these agents have
a role in chemoprevention only when
taken in the context of other micronu-
trients, a hypothesis we are testing in
the continuation of the multivitamin
component of PHS II.
CONCLUSION
In this long-term, large-scale, low-
cost trial, after a mean of 8.0 years of
treatment and follow-up in 14 641 men,
neither vitami n E nor vitamin C supple-
mentation reduced the risk of prostate
or total cancer. There was also no clear
effect of either agent on other site-
specific cancers. lt is reassuring that
there was not a clear signal of harm for
either agent. These data provide no sup-
port for the use of these supplements
in the prevention of cancer in middle-
aged and older men. Results of the mul-
tivitamin arm of the PHS II will be fonh-
coming in several years.
Published Online: December 9, 2008 (doi:10.1001/
jama.2008.862)
Author Contributions: Dr Gaziano had full access to
all of the data in the study and takes responsibility for
the integrity of the data and the accuracy of the data
analysis.
Dr Gaziano and Dr Glynn contributed equally and are
co-first authors of this article.
Study concept and design: Gaziano, Glynn, Belanger,
Manson, Sesso, Suring.
Acquisition of data: Gaziano, Glynn, Kurth, Belanger,
MacFadyen, Bubes, Manson, Sesso, Suring.
Analysis and interpretation of data: Gaziano, Glynn,
Christen, Kurth, Bubes, Manson, Sesso, Buring.
Drafting of the manuscript: Gaziano, Glynn, Sesso,
Buring.
Critical revision of the manuscript for important in-
telledual content: Gaziano, Glynn, Christen, Kurth,
Belanger, MacFadyen, Bubes, Manson, Sesso, Buring.
Statistical analysis: Glynn, Bubes, Sesso.
Obtained funding: Gaziano, Sesso, Buring.
Administrative, technical, or material support: Gaziano,
Kurth, Belanger, MacFadyen, Bubes, Manson, Sesso.
Study supervision: Gaziano, Kurth , Belanger,
MacFadyen, Bubes, Manson, Sesso.
Financial Disclosures: Dr Gaziano reported receiving
investigator-initiated research fundi ng from the Na-
tionallnstitutes of Health, the Veterans Administra-
tion, Veroscience, and Arngen: serving as a consul-
tant or receiving honoraria from Bayer AG and Pfizer;
and serving as an expert witness for Merck. Dr Glynn
reported receiving investigator-initiat ed research fund-
ing from the National Institutes of Health, Bristol-
Meyers Squi bb, and AstraZeneca. Dr Christen re-
ported receiving research funding support from the
Nationall nsututes of Health, Harvard University (Clini-
cal Nutrition Research Center), and DSM Nutritional
Products Inc (Roche). Dr Kurth reported receiving in-
vestigator-initiated research funding from Bayer AG,
the National Institutes of Health, McNeil Consumer
& Specialt y Pharmaceuticals, Merck, and Wyeth Con-
sumer Health care; being a consultant to 13 Drug Safety:
and receiving honoraria from Genzyme for educa-
tionallectures and Organon for contributing to an ex-
pert panel. Dr Manson reported receiving investigator-
initiated research funding from the National Institutes
of Health and assistance with study pills and packag-
ing from BASF and Cognis. Dr Sesso reported receiv-
ing investigator-initiated research funding from the Na-
tionallnsti tutes of Health, American Heart Association,
American Cancer Society, California Strawberry Com-
mission, Roche Vitami ns Inc (now DSM Nutri tional
Products Inc), and Cambridge Theranostics Ltd. Dr Bur-
ing reported receiving study agents and packaging from
Bayer Heaithcare and the Natural Source Vitamin E
Association, as well as research funding from the Na-
tionallnstitutes of Health. No other authors reported
financial disclosures.
Funding/Support: This work was supported by grants
CA 97193, CA 34944, CA 40360, HL 26490, and HL
34595 from the National Institutes of Health (Bethesda,
Maryland) and an investigator-initiated grant from
BASF Corporation (Florham Park, New Jersey). Study
agents and packaging were provided by BASF Cor-
poration, Wyeth Pharmaceuticals (Madison, New Jer
sey), and DSM Nutritional Products Inc (f ormerly Roche
Vitamins) (Parsippany, New Jersey).
Role of the Sponsor: BASF Corporation, Wyeth Phar-
maceuticals, and DSM Nutritional Products Inc had no
role in the design and conduct of the study: in the col-
lection, analysis, and interpretation of the data; or in
the preparation. review, or approval of the manu-
script.
Data and Safety Monitoring Board: Voting members
included Lawrence Cohen, Theodore Colton, I. Craig
Henderson, Ross Prentice, and Nanette Wenger (chair);
ex-officio members i ncluded Frederick Ferris, Peter
G reenwaid, Natalie Kurinij, Howard Parnes, Eleanor
Schron, and Alan Zonderman.
Disclaimer: Dr Gaziano, a contributing editor for lAMA.
was not involved in the editorial review of or decision
to publish this article.
Additional Contributions: We are indebted to the
14 641 physician participants for their long -standing
dedication and conscientious collaboration. We also
acknowledge the long-term contributions of Charles
Hennekens, MD, DrPH, Florida Atlantic University, to
the Physicians' Health Study. and the exemplary con-
tributions of the staff of the Physicians' Health Study
at Brigham and Women' s Hospital, under the lead-
ership of Charlene Belanger: Kenneth Breen, Mary
Breen, Mary G. Breen, Jose Carrion, Ivan Fi tchorov,
Natalya Gomelskaya, Beth Holman, Andrea Hrbek,
Tony laurinaitis, Chandra McCarthy, Geneva McNair,
leslie Power, Philomena Quinn, Harriet Samuelson,
Miriam Schvartz, Fred Schwerin, Michelle Sheehey,
Joanne Smith, Martin Van Denburgh, and Phyllis
Johnson Wojciechowski. Finally, we are grateful for
the efforts of the Physicians' Health Study Endpoint
Committee, including Samuel Goldhaber, Carlos Kase,
Mei r Stampfer, and James Taylor, over the course of
the Physicians' Health Study II. Each named indi-
vidual was compensated for his or her contribution as
part of the grant support.
REFERENCES
1. World Cancer Research Fund/American Institute
for Cancer Research. Food, Nutrition, Physical Activ-
ity, and the Prevention of Cancer. Washington, DC:
American Institute for Cancer Research; 2007.
2. Ames BN. Dietary carcinogens and anticarcino-
gens: oxygen radicals and degenerative diseases.
Science. 1983;221 (4617):1256-1264.
3. Dusinska M, Kazimirova A, Barancokova M, et al.
Nutritional supplementation with antioxidant s de-
creases chromosomal damage in humans. Mutagenesis.
2003;18(4):371-376.
4. Federico A, Morgillo F, Tuccillo C, Ciardiello F,
loguercio C. Chronic inflammation and oxidative stress
i n human carci nogenesis. lnl 1 Cancer. 2007;121
(11 ):2381 -2386.
5. Blot WJ, Li JY, Taylor PR, et al. Nutrition interven-
tion trials in linxian, China: supplementation with spe
cific vitamin/mineral combinati ons, cancer inci-
dence, and disease-specific mortalit y in the general
population. 1 Nat! Cancer lnst. 1993;85(18):1483-
1492.
6. Hercberg S, Galan P, Preziosi P, et al. The SU.VI
.MAX Study: a randomized, placebo-controlled trial
of the health effects of antioxidant vitami ns and
minerals. Arch Intern Med. 2004;164(21):2335-
2342.
7. Clark LC, Combs GF Jr, Turnbull BW, et al; Nutri-
tional Prevention of Cancer Study Group. Effects of
selenium supplementation for cancer prevention i n pa-
tient s with carcinoma of the skin: a randomized con-
trolled trial. lAMA. 1996; 276(24): 1957-1963.
8. Albanes D, Mallia N, Taylor PR, et al. Effects of
supplemental alpha-tocopherol and beta-carotene on
colorectal cancer: resul ts from a controlled trial
(Finland). Cancer Causes Control. 2000;11(3):
197 205.
E10 JAMA, Published online December 9 , 2008 (Reprinted) 2009 American Medical Association. AU rights reserved.
9. The Alpha-Tocopherol, Beta Carotene Cancer Pre-
vention Study Group. The effect of vitamin E and beta
carotene on the incidence of lung cancer and other
cancers in male smokers. N Eng/ J Med. 1994;
330(15):1029-1035.
10. Hennekens CH, Suring JE, Manson JE, et al. Lack
of effect of long-term supplementation with beta caro-
tene on the incidence of malignant neoplasms and car-
diovascular disease. N Eng/ J Med. 1996;334(18):
1145-1149.
11. Omenn GS, Goodman GE, Thornquist MD, eta!.
Effects of a combination of beta carotene and vi ta-
min A on lung cancer and cardiovascular disease.
N Eng/ J Med. 1996;334(18):1150-1155.
12. Gruppo Italiano per lo Studio della Soprawivenza
nell'lnfarto miocardico. Dietary supplementation with
n-3 polyunsaturated fatty acids and vitamin E after myo-
cardial infarction: results of the GISSI-Prevenzione trial.
Lancet. 1999;354(9177):447 -455.
13. Lee IM. Cook NR, Manson JE, Suring JE, Hennekens
CH. Beta-carotene supplementation and incidence of
cancer and cardiovascular disease: the Women's Health
Study. J Natl Cancer tnst. 1999;91 (24):2102-2106.
14. Heart Protection Study Collaborative Group. MRC
/BHF Heart Protection Study of antioxidant vitamin
supplementation in 20,536 high-risk individuals: a ran-
demised placebo-controlled trial. Lancet. 2002;
360(9326):23-33.
15. Lee IM, Cook NR, Gaziano JM, et al. Vitamin E in
the primary prevention of cardiovascular disease and
cancer: the Women's Health Study: a randomized con-
trolled trial. JAMA. 2005;294(1 ):56-65.
16. Lonn E, Bosch J, Yusuf S, et al; HOPE and HOPE-
TOO Trial Investigators. Effects of long-term vitamin
E supplementation on cardiovascular events and can-
cer: a randomized controlled tri al. JAMA. 2005;
293(11):1338-1347.
17. Virtamo J, Pietinen P. Huttunen JK, et ai;ATBC Study
Group. Incidence of cancer and mortali ty following
alpha-tocopherol and beta-carotene supplementa-
tion: a postintervention follow-up. JAMA. 2003;
290(4):476-485.
18. Manson JE, Gaziano JM, Spelsberg A, et al; The
WACS Research Group. A secondary prevention trial
of antioxidant vitamins and cardiovascular disease
in women: rationale, design. and methods. Ann
Epidemiol. 1995;5(4}:261-269.
19. Timbo BB, Ross MP, McCarthy PV, Lin CT. Di-
etary supplements in a national survey: prevalence of
use and reports of adverse events. J Am Diet Assoc.
2006;106(12):1966-1974.
20. Sessa HD, Suring JE, Christen WG, et al. Vitamins
E and C in the prevention of cardiovascular disease
in men: the Physicians' Health Study II randomized
controlled trial. lAMA. 2008;300(18):2123-2133.
21. Christen WG, Gaziano JM, Hennekens CH. De-
sign of Physicians' Health Study II: a randomized
trial of beta-carotene, vitamins E and C. and multi-
vitamins, in prevention of cancer, cardiovascular
disease, and eye disease, and review of results of
completed trials. Ann Epidemiol. 2000;1 0(2):125-
134.
22. Steering Committee of the Physicians' Health Study
Research Group. Final report on the aspirin compo-
nent of the ongoing Physicians' Health Study. N Eng/ J
Med. 1989;321(3):129-135.
23. LangJM, Suring JE, Rosner B, Cook N, Hennekens
CH. Estimating the effect of the run-in on the power
of the Physicians' Health Study. Stat Med. 1991 ;
10(10):1585-1593.
24. Klein EA, Thompson IM, Lippman SM, et al. SE-
LECT: the Selenium and Vitamin E Cancer Prevention
Tri al: rationale and design. Prostate Cancer Prostatic Dis.
2000;3(3):145-151.
25. Heinonen OP, Albanes D. Virtamo J. et al. Pros-
tate cancer and supplementation wi th alpha-
tocopherol and beta-carotene: incidence and mortal-
ity in a controlled trial. 1 Nat/ Cancer lnst. 1998;
90(6):440-446.
26. Block G. Vitamin C and cancer prevention: the epi-
demiologic evidence. Am J Clin Nutr. 1991 ;53(1)
(suppi):270S-282S.
27. Heilbrun LK, Nomura A, Hankin JH, Stemmermann
GN. Diet and colorectal cancer with special reference
to fiber intake. lntJ Cancer. 1989;44(1):1 -6.
28. Knekt P, Jarvinen R, Seppanen R, et al. Dietary an-
tioxidants and the risk of lung cancer. Am J Epidemic/.
1991 ;134(5):471-479.
29. Kromhout D. Essential micronutrients in relation to
carcinogenesis. Am J Clin Nutr. 1987;45(5)(suppl):
1361-1367.
30. Byers T, Guerrero N. Epidemiologic evidence for
vitamin C and vitamin E in cancer prevention. Am J Clin
Nutr. 1995;62(6)(suppl): 1385S-1392S.
31. Howe GR, Hirohata T, Hislop TG, et al. Dietary fac-
tors and risk of breast cancer: combined analysis of 12
case-control studies. J Nat/ Cancer tnst. 1990;82
(7):561-569.
32. McKeown-Eyssen G, Holloway C, Jazmaji V,
Bright-See E, Dian P, Bruce WR. A randomized trial of
vitamins C and E in the prevention of recurrence of co-
lo rectal polyps. Cancer Res. 1988;48(16):4701-
4705.
33. GreenbergER, Baron JA, Tosteson TD, eta!; Polyp
Prevention Study Group. A clinical trial of antioxidant
vitamins to prevent colo rectal adenoma. N Eng/ J Med.
1994;331 (3):141-147.
34. Helzlsouer KJ, Huang HY, Alberg AJ, et al. Asso-
ciat ion between alpha-tocopherol , gamma-
tocopherol, selenium, and subsequent prostate cancer.
1 Nat/ Cancer tnst. 2000;92(24):2018-2023.
2009 American Medical Association. All rights reserved. (Reprinted) JAMA. Published online December 9, 2008 E11
LETTER TO THE EDITOR
Recent Trends in Incidence of Cutaneous Melanoma
among US Caucasian Young Adults
Journal of Investigative Dermatology advance onl ine publ ication, 10 j ul y 2008i doi:l 0. 1 038/j id.2008.1 59
TO THE EDITOR
Recent findings suggest that non-mela-
noma ski n cancer incidence i n young
adults is risi ng, particularly among US
young women (Christenson et a/.,
2005). This raises the important ques-
tion of whether incidence of cutaneous
melanoma, the most lethal form of skin
cancer, is similarly increasing in young
adults. Although mel anoma incidence
among US older adults has been
increasing for several decades, there
have been indications that incidence
may be stabilizing for birth cohorts born
after 1945 (Dennis et a/., 1993; Hall
eta/., 1999). However, in an analysis of
melanoma trends between 1973 and
1997 in the Survei I lance, Epidemio-
logy, and End Results (SEER) Program,
Jemal et a/. (2001) noted evidence
of an increase among women born
after 1960. Since that analysis, an
additional 7 years of SEER data have
been collected. To better understand
recent trends in melanoma incidence
among young adults, we report findings
from a re-analysis of SEER data,
extended through 2004.
Our analysis was restricted to Cau-
casians from the nine registries that
have contributed data to the SEER
Program since 1973 (Atlanta, Connecti-
cut, Detroit, Hawaii, Iowa, New Mex-
ico, San Francisco-Oakland, Seattle,
Utah). We calculated annual age-ad-
justed incidence (SEER Program, 2007a)
and mortal ity rates (SEER Program,
2007b) of invasive cutaneous melano-
ma among men and women aged
15-39 years, standardized to the 2000
US population, using the software
SEER*Stat version 6.3.6 (National Can-
cer Institute; http:// www.seer.cancer.
gov/seerstatl). We assessed trends in
greater detail using joinpoint regression
models, which identify changes in
trends over successive segments of time
and describe the estimated annual
percent change (EAPC) i n incidence
within each segment (Kim et a/.,
2000), using the software Joinpoint
version 3.0 (National Cancer Institute;
http://www .srab .cancer .gov/joi npointl).
joinpoint analyses stratified by melano-
ma stage (locali zed vs regional/distant)
and thickness ( 1 vs > 1 mm) were
also performed. To describe age-speci-
fic trends by year of birth, we calcu-
lated incidence by 5-year age groups
and time periods, and plotted age-
specific incidence by calendar year of
birth (calculated from the age group
midpoint). Additionally, age-period-co-
hort modeling was used to simulta-
Q;
Cl.
Q)
'
'iii
::J
c
c
<(
20 .---------,

0 0

0.2
neously adjust age-specific incidence
trends for both calendar period and
birth cohort effects (Tarone and Chu,
2000). All P-values were two-sided.
Overall, the age-adjusted annual
incidence of melanoma among young
men increased from 4.7 cases per
100,000 persons (95% confidence
l imits: 3.8, 5.7) in 1973 to 7.7 per
100,000 in 2004 (6.8, 8.7). Among
women, age-adjusted annual incidence
per 100,000 increased from 5.5 (4.5,
6.6) in 1973 to 13.9 (12.7, 15.2) in
2004. Melanoma incidence increased
among young men (EAPC = 6.6; 95%
confidence limits (CL): 2.9, 10.4) and
women (9.2; 6.8, 11.7) during the
1970s (Figure 1, Table 1). Around the
start of 1980, this pattern changed. For
o Male incidence
Female incidence
Male mortality rate
Female mortality rate
1975 1985 1995 2005
Year of diagnosis
Figure 1. Trends in melanoma incidence and mortality among young adults. Age-adjusted (to 2000 US
population) annual cutaneous melanoma incidence and mortality rates among Caucasian males and
females aged 15-39 years in the Surveillance, Epidemiology, and End Results Program areas from 1973
through 2004. The segments of uni form trend from the best-fi tting Joinpoint models are also shown.
Abbreviation: SEER, the Surveillance, Epidemiology, and End Results Program
2008 The Society for Investigative Dermatology www.j idonline.org
MP Purdue et al.
Mel anoma Incidence among US Young Adults
Table 1. EAPC in incidence of melanoma and melanoma mortality among Caucasian males and females aged 15-39
years in the SEER Program from 1973 through 2004
Trend 1 Trend 2 Trend 3 Trend 4
Incidence
Overall
Males
Females
By stage
localized
Males
Females
Regional/distant
Males
Females
Years
1
EAPC (95% Cl )
1973-1980 6.6 (2.9, 1 0.4)
1973-1978 9.2 (6.8, 11.7)
1973-1980 9.6 (4.9, 14.5)
1973- 1978 15.8 (1 0.9, 21.0)
1973-2004 1.4 (0.6, 2.2)
1973-1994 -0.9 (-2.5, 0.8)
By thickness (1988+ only)
.;;; 1mm
Males
Females
>1 mm
Males
Females
Mortality
Males
Females
1988-2004 2.3 (1.2, 3.4)
1988-2004 3.1 (2.5, 3.6)
1988-2004 - 0.3 ( - 1.5, 1.0)
1988-2004 2.8 (1.6, 4.0)
1973-1981 5.0 (0.3, 10.1)
1973-2004 - 2.3 (- 3.1, - 1.5)
Years EAPC (95% Cl ) Years EAPC (95% Cl ) Years EAPC (95% CL)
1980-2004 0.4 (-0. 2, 0.9)
1978-1987 2.6 (1.5, 3.8) 1987-1992 -0.6 (-3.7, 2.6) 1992-2004 2.7 (2.1' 3.4)
1980-2004 0.5 (-0.2, 1.2)
1978-2004 1.9 (1.6, 2. 3)
1994-2004 9.2 (3.8, 14.9)
198 1-2004 - 3.6 (-4.5, - 2.7)
Cl, confidence l imits; EAPC, annual percent change in melanoma incidence within jointpoint segment.
Statistically signi ficant results in bold face type.
' Calendar period within j oinpoint segment. joinpoint modeling was carried out separately for males and females; hence, sex-specific joinpoint segments
may di ffer.
men, incidence leveled off between
1980 and 2004 (0.4; -0.2, 0.9). For
women, the rate of increase in inci -
dence declined from 1978 to 1987
(2.6; 1.5, 3.8) and stabil ized from
1987 to 1992 {-0.6; - 3.7, 2.6). After
1992, however, incidence began
increasing again (2.7; 2.1, 3.4). Inci-
dence among women from the 1990s
onward increased both for thinner and
thicker melanomas mm: 3.1; 2.5,
3.6 and > 1 mm: 2.8; 1.6, 4.0), and was
greater for regional and distant tumors
(9.2; 3.8, 14.9) compared with loca-
l ized lesions (1.9; 1.6, 2.3). Melanoma
mortality rates for men and women
2 Journal of Investigative Dermatology
decl ined from 1981 onward (men:
-3.6; -4.5, -2.7 and women: -2.3;
-3.1 , - 1.5).
Age-specific incidence patterns by
year of birth are presented in Figure 2.
Mal e age-specific incidence increased
steadily with each successive birth
cohort until 1950, at which time
incidence appeared to level off or
decrease slightly. Female age-specific
incidence by birth cohort increased
steadily until around 1950; thereafter,
i ncidence appeared to cli mb at a
slower pace until 1965, at which point
incidence appeared to begin accelerat-
ing. After adjustment for age and peri od
effects, age-period-cohort model ing
confirmed a change in the effect of
birth cohort for women born between
1960 and 1965 (Figure 51; slope
change parameter = 0.2146; 95% CL:
0.0576, 0.3716; P= 0.007).
It is important to consider whether
these trends may reflect changes in
data quality, diagnosis, or surveillance.
There is evidence of increased under-
reporti ng of melanoma over time with-
in the SEER program, with estimates
as high as 17% of all cases (including
in situ lesions) in two registries,
although such a trend in underreporting
cannot explain the observed increase
MP Purdue et al.
Melanoma Incidence among US Young Adults
Males Females
100 p "
A"
100
1965 birth cohort
-- 85+
.,. ...
l!!
R< -<> 80- 84
"'
w.-
.... . 75-79
Q)
--?-- 70- 74
?' , 'If ... . ...
c
4 <1
..... 65-69
0
l!!
-<1-' 60-64
Q) - ...
_,_ 55- 59
a.
10 "' o.c>.ooo.o
..... 50-54 10
0
0 ......,._........
. . 45- 49
0
o --6- 40- 44
0
...... . 35- 39
fi>..P,..O+
Q;
-<> . 30-34

a.
--- 25-29

-
...
Q)
- -1> - 20- 24
0
c
...
. . 15-19
Q)
'0
'(j
0.5 0.5

1900 1920 1940 1960 1980 1900 1920 1940 1960 1980
Figure 2. Age-specific melanoma incidence among Caucasians stratified by sex and birth cohort year in the SEER program from 1975-1979 through
2000-2004. The points vertical ly above each cohort year portray the cohort's age-specific incidence experience. The vertical l ine at 1965 on the x axis of the
plot for women represents the time point after which melanoma inci dence begins increasing in subsequent cohorts.
in incidence among women (Seiffert,
1992; Merlino et at., 1997). It is
unlikely that a change in melanoma
diagnosti c criteria would account for
our finding, because the effect of such a
change would not be expected to be
sex-specific. Changes in screening pat-
terns may have led to earlier detection
within this time period, with a higher
rate of increase seen among superficial
local ized tumors compared with thicker
lesions and regional or metastatic dis-
ease overall Oemal eta/., 2001; Welch
et a/., 2005). Indeed, the observed
decrease in melanoma mortal ity rates
after 1981 and previously reported
evidence of general improvement in
survival by stage over this time period
are consistent with a shift toward earli er
detect ion of disease through increased
surveillance Oemal eta/., 2001 ). How-
ever, in our analysis, the increasing
trend among young women from the
early 1990s onward was also observed
for thicker and regional/ distant tumors,
which are less susceptible to misclassi -
fication. Moreover, our
hort analysis suggested that, after
adjusting for age and peri od effects
(the latter of which is reflective of
changes in disease surveillance), the
observed increase in incidence among
women born after 1965 is consistent
with a bi rth cohort effect (indicative of
changes i n disease risk factor preva-
lence across birth cohorts; Tarone and
Chu, 2000). Thus, our findings are
compatible with a real increase in
incidence among young women,
although we cannot rule out the effects
of changes i n surveillance.
The recent increase i n incidence
among young women parallels reported
trends in exposure to UVR, the pri mary
environmental cause of melanoma
(Armstrong and Kricker, 2001 ). The
prevalence of sunburn is increasing
among US adult men and women
overall , although trends by age group
have not been reported (Robinson
et a/., 1997; Saraiya et a/., 2007).
Among adolescents aged 1 6-18 years,
both the prevalence of sunburn and
the average number of days spent at
the beach increased between sun sur-
veys conducted in 1998 and 2004
(Cokkinides et a/. , 2006). Tanning bed
usage, which has been recently evalu-
ated as a probable cause of melanoma
(International Agency for Research
on Cancer, 2007), is increasing among
US adults and is most prevalent among
young women (Robinson eta/., 1997;
lazovich and Forster, 2005).
In conclusion, our analysis of SEER
data suggests that melanoma incidence
is increasing among young women.
Additional studies are needed to clarify
whether the increasing trends for mel-
anoma and non-melanoma skin cancer
(Christenson eta/. , 2005) are the result
of changes in UVR exposure in this
population.
CONFLICT OF INTEREST
The authors state no confl ict of interest.
ACKNOWLEDGMENTS
This research was supported by the Intramural
Research Program of the National Institutes
of Health and the National Cancer Institute.
The SEER Program is operated by the National
Cancer Institute Surveillance Research Program.
Mark P. Purdue
1
, Laura f. Beane
Freeman
1
, William F. Anderson
1
and
Margaret A. Tucker
1
1
Division of Cancer Epidemiology and
Genetics, National Cancer Institute, Department
of Health and Human Services, National
lnstitlltes of Health Rockville, Maryland, USA
E-mail: purduem@mail.nih.gov
SUPPLEMENTARY MATERI AL
Figure 51. Sex-specific maximum li kel ihood
estimates of 5-year birth cohort effects for an
age-period-cohort model fit to cutaneous melano-
ma incidence data for Caucasians in the SEER
Program from 1973 through 2004.
REFERENCES
Armstrong BK, Kricker A (200'1) The epidemiology
of UV induced skin cancer. ) Photochem
Photobiol B 63:8-18
Christenson LJ, Borrowman TA, Vachon CM,
Tollefson MM, Otley CC, Weaver AL et a/.
(2005) Incidence of basal cell and squamous
cel l carcinomas in a popul ation younger than
40 years. lAMA 294:681-90
Cokkinides V, Weinstock M, Glanz K, Albano ),
Ward E, Thun M (2006) Trends in sun-
burns, sun protection practices, and attitudes
toward sun exposure protection and tanning
among US adolescents, 1998-:2004. Pedia
trics 118:853-64
Dennis LK, White E, Lee JA (1993) Recent cohort
trends in malignant melanoma by anatomic
site in the United States. Cancer Causes
Contro/4:93-1 00
Hall HI, Mi ll er OR, Rogers JD, Bewerse B (1999)
Update on the incidence and mortal ity from
www.j idonline.org 3
MP Purdue et al.
Melanoma Incidence among US Young Adults
melanoma in the United States. JAm Acad
Oermato/40:35-42
(2007) The association of use of sunbeds
with cutaneous malignant melanoma and
other skin cancers: a systematic review. lnt J
Cancer 120:11 I 6-22
Jemal A, Devesa SS, Hartge P, Tucker MA (2001 )
Recent trends in cutaneous melanoma inci-
dence among whites in the United States.
J Nat/ Cancer lnst 93:678-83
Kim HJ, Fay MP, Feuer EJ, Midthune DN (2000)
Permutation tests for joinpoint regression
with appli cations to cancer rates. Stat Med
19:335-51
Lazovich D. Forster J (2005) Indoor tanning by
adolescents: prevalence, practices and poli-
cies. Eur J Cancer 41 :20-7
Merl ino LA, Sullivan KJ, Whitaker DC, Lynch CF
(1997) The independent pathology laboratory
4 Journal of lnvcstig;ltivc Dermatology
as a reporting source for cutaneous melano-
ma incidence in Iowa, 1977-1994. J Am
Acad Oermatol 37:578-85
Robinson JK, Rigel DS. Amonene RA (19971
Trends in sun exposure knowledge, attitudes,
and behaviors: 1986-1996. J Am Acad
Oermatol37:179-86
Saraiya M, Balluz L, Wen XJ, Joseph DA (2007)
Sunburn prevalence among adults-United
States, 1999, 2003 and 2004. Morbidity and
Mortality Weekly Report 56:524-8
SEER Program (2007a) Surveillance, Epidemiol-
ogy, and End Results (SEER) Program
(www.seer.cancer.gov) SEERStat Database:
Incidence-SEER 9 Regs Limited-Usc,
November 2006 Sub (1973-2004)-Linked
To County Attributes- Total US, 1969- 2004
Counties, National Cancer Instit ute, DCCI>$,
Surveillance Research Program, Cancer
Statistics Branch, released April 2007, based
on the November 2006 submission
SEER Program (2007b) Surveillance, Epidemio-
logy, and End Results (SEER) Program
(www.seer.cancer.gov) SEERStat Database:
Mortality - All COD, Aggregated With
State, Total US (1969-2004), National
Cancer Institute, DCCPS, Surveillance
Research Program, Cancer Statistics Branch,
released April 2007. Underlying mortality
data provided by NCHS (www.cdc.gov/
nchs)
Seiffert J (1992) Underreporti ng of melanoma.
J Nat/ Cancer Ins/ 84:289
Tarone RE, Chu KC (2000) Age-period-cohort
analyses oi breast-, ovarian-, endometrial-
and cervical -cancer mortality rates for
Caucasian women in the USA. J Epidemiol
Biostat 5:221-31
Welch HG, Woloshin S, Schwartz LM (2005) Skin
biopsy rates and i ncidence of melanoma:
population based ecological study. BMJ
331:481
I
Do adolescent indoor tanners exhibit dependency?
Sarah Zeller, Mn,b DeAnn Lazovich, PhD,c j ean Forster, PhD, MPH," and Rachel Widome, MHSc
Minneapolis, Minnesota
Background: Indoor tanning is a common adolescent ri sk behavior that has been hypothesized to be
moti vated and maintained by the mood-altering effects of ul traviolet light.
Objective: Our purpose was to explore heretofore anecdotal reports that indoor tanning leads to
dependency.
Methods: A telephone interview was conducted among 1275 adolescents, ages 14 to 17 years. Self-
reported difficulty in quitting indoor tanning was assessed among 267 adolescents (20.9% of total) who
tanned indoors more than once in the previous year in relation tO age of initiation, frequency of use, and
positive or negative consequences of the practice.
Results: Difficulty in quitting was more likely wi th younger age at initiation (age 13 years or younger vs
ages 16 to 17; odds ratio= 4.3, 95% confidence interval 1.3-14.7) and higher frequency of use (P = .009),
even after accounting for positive or negative consequences of indoor tanning and other demographic
characteristics.
Limitations: This was a cross-sectional study design with a limi ted outcome measure.
Cotzclusio11: Although preliminary, our findings for age at initiation and frequency of use in relation to
difficulty in quitting indoor tanning are consistent with other potentially addictive behaviors taken up
during adolescence. (] Am Acad Dermatol 2006;54:589-96.)
T
he well-documented adverse health effects
of indoor tanning range from acute (photo-
sensitivity, cutaneous and corneal burns) to
cluonic (premature aging of the skin, cataract for-
mation). 1-4 Indoor tanning has also been implicated
in increasing the risk of skin cancer, the most com-
mon and increasingly prevalent fo rm of cancer in the
United States.
5
6
Modern tanning devices, first pop-
ul ari zed in the 1980s,
7
can emit a close of ultraviolet
A (UVA) per minimal erythema dose that is 10 times
greater than that of the sun.
8
Exposure to UVA
From the School of Public Health University of Minnesota," Hen-
nepin County Medical Center, Graduate Medical Educat ion,b
and the Division of Epidemiology and Community Health,
School of Public Health, University of Minnesota.<
Funding source: National Cancer Institute, Bethesda, MD, grant
1 R01 CA079593.
Conflicts of int erest: None identifi ed.
Accepted for publ ication December 16, 2005.
Reprint requests: DeAnn Lazovich, PhD, Division of Epidemiology
and Community Health, University of Minnesota, 1300 S 2nd St,
Suite 300, Minneapolis, MN 55454. E-mail: lazovich@epi.umn.
edu.
0190-9622/ $32.00
2006 by the American Academy of Dermatology, Inc.
doi:1 0.1 016/j.jaad.2005.12.038
Abhreuiaticms used:
CI: confidence interval
OR: odds ratio
UV: ultraviolet
radiation damages the DNA of epidermal kerati no-
cytes9 and has been associated with increased risk of
squamous cell carcinoma and basal cell carcinoma.
10
The relationship between melanoma and UV ra-
diation is complex, yet exposure to UV light remains
the primary environmental risk factor for me la-
noma.
11
Investigations of the association between
indoor tanning and the development of melanoma
have been hampered by methodological limitations
and the inclusion of older, lower UVA-emitting cle-
vices. 12-14 A recent study in Sweden has presented the
strongest evidence to date that indoor tanning in-
creases the risk of melanoma, particularly if initiated
at a younger age.
15
This study adds weight to the grow-
ing body of evidence that suggests that modern in-
door tanning is a risk factor for carcinogenesis
1 4
16
-
3
3
A number of advisory bodies warn against the
pract ice of indoor tanning,
34
-
37
especially for children
and adolescents, whose exposure places them at
589
590 Zeller et a!
increased risk of UV-i nduced carcinogenesis. Child-
hood exposure to environmental carcinogens has
been postulated to be especially carcinogenic to tissues
that undergo posmatal development, such as mela-
nocytes that undergo peak activity during early life.
38
39
The commercial tanning industry is growing from
a S5 billion a year industry in 2004 to a projected $7.5
billion annual revenue during the next 3 to 5 years
4 0
According to the indust1y, adolescents comprise
2 mill ion to 3 million of the US commercial tanning
industry's 30 mill ion annual "consumers. "<H The prev-
alence of indoor tanning among white US adolescents
aged 13 to 19 years has been estimated at 24o/o for
ever-use and 11.7o/o for frequent use ( 2:::10 times).
42
US female adolescent report prevalence rates of any
use from 16%
43
.4
4
to 51%,
45
whereas male adolescent
are 2 to 3 times less likely to tan indoors.
46
Many indoor tanners cite the mood-enhancing or
relaxing effects of indoor tanning as a major moti-
vating factor in seeking UV radiation stimulus.
17
48
Anecdotal evidence
49
"
51
suggests that a subset of
tanners may exhibit tanning "dependence," similar to
the compulsive need to use a substance to experience
its psychic effects or to avoid the discomfort of its
absence.
52
Adolescents who tan indoors frequently
are at higher risk for tobacco and other substance use,
which suggests that these teens may be susceptible
to addictive disorders
42
53
54
Recently, researchers
described an exercise in which the Diagnostic and
Statistical Manual of Mental Disorders, Fou1th
Edition, Text Revision (DSM-IV-TR) diagnostic crite-
ria for substance dependence (7 measures of toler-
ance, withdrawal, loss of control, preoccupation, and
continuance despite adverse consequences)
55
and
the CAGE questionnaire for alcohol dependence
were modified to screen for tanning dependence.
Of the 145 persons interviewed while sunbathing
on a beach, 77 (53%) met the modified DSM-IV-TR
diagnostic criteria for a substance-related disorder,
and 38 (26o/o) met the modified CAGE criteria 5
6
That indoor tanning may produce dependency
in some users has yet to be investigated in popula-
tion-based studies. Tn a survey of adolescents in
Minnesota and Massachusetts, we asked indoor
tanning users a single question regarding their per-
ceived difficulty to quit the practice. This measure is
similar to two items on a self-administered assess-
ment of nicotine dependency designed specifically
for adolescents.
57
-
59
In this report, we examine the
evidence that indoor tanning dependence may be
a real phenomenon by considering the association
between various characteristics commonly thought
to promote or limit dependent behavior and the
likelihood that adolescents would report difficulty
with quitting use of indoor tanning.
METHODS
Participants
J AM ACAD 0 ERMATOL
APRIL 2006
The Minnesota and Massachusetts Indoor
Tanning Study, initiated in January 2000, is cited
more completely elsewhere.
60
Briefly, adolescents
aged 14-17 years residing in the Boston, Massachu-
setts, and Minneapolis-St Paul, Minnesota metropol-
itan areas were identified from a commercial targeted
age list. A random selection of eligible households
was called and parental consent was obtained to
conduct a telephone interview with one randomly
chosen teenager per household. Girls were over-
sampled relative to boys in a 2:1 ratio. In the Boston
area, 33o/o of households (n = 2699) were found to
have an adolescent in the desired age range, whereas
in the Minneapolis-St. Paul area 44.7% of 1650
households had an age-eligible adolescent (37.4%
overall). Response rates were 73.1o/o (n = 651 ado-
lescents) in Boston and 87.8% (n = 647 adolescents)
in Minneapolis-St. Paul. To limit our study to com-
mercial tanning device users, we excluded 23 ado-
lescents (1.8%) who reported the exclusive use of
home tanning devices. Among 1275 remaining ado-
lescents, 386 (30.3o/o) had tanned indoors; 267
(20.9 o/o) had tanned more than once in the past
year and are the subject of this investigation.
Measures
We collected information about personal charac-
teristics including age, sex, smoking history, ac-
quaintance with someone with skin cancer, body
satisfaction (eg, how often are you happy with the
way you look), depression (eg, how often have
you been bothered by feeling unhappy, sad, and
depressed)
61
and skin cancer risk factors (eg, color of
hair, eyes, skin; propensity to sunburn; and history of
severe sunburns). A skin cancer risk summary score
was created from the risk factors for skin cancer
(range, 4-17; higher scores denoting higher risk).
Knowledge of the consequences of UV radiation
exposure was assessed by 5 questions including tans
as a marker of skin damage, indoor tanning as a
possible risk for skin cancer, whether indoor tanning
is safer than sunbathing, and drug-induced photo-
sensitivity; we assigned 1 point for each correct
answer to create a summary score (0-5 possible). We
assessed positive attitudes toward indoor tanni ng
by level of agreement with the statements that in-
door tanning lifts spirits or is relaxing. We likewise
assessed negative attitudes by agreement with state-
ments that indoor tanning is expensive, causes the
skin to smell, or feels claustrophobic.
Behaviors associated with indoor tanning were
assessed by asking the age at initiation and frequency
of indoor tanning. These measures are similar to
) AM ACAD DRMATOL
VOLUME 54. N UMBER 4
known predictors of dependency in other adoles-
cent risk behaviors, such as substance use.
62
63
We
asked teenagers to report the age at which they first
tanned and categorized responses into 4 groups:
13 and under, 14, 15, or 16-17. Frequency of indoor
tanning was determined by the response to two
related questions. First, we asked adolescents who
had ever tanned about use during each 3-month
season of the previous year. Adolescents who
reported tanning indoors each month wi thin that
season were assigned a frequency of 3 and those
who tanned none of the months were assigned a
frequency of zero; all others were given a value of 1.5
(the average of 1 or 2 months of tanning). Values
were then summed over the 4 seasons ro obtain the
total number of months in the previous year that the
adolescent had tanned. We asked a second question
about the number of times tanning occurred in the
months that the adolescent tanned. Frequency of
indoor tanning was determined by the product of the
number of sessions of indoor tanning per month and
the total number of months of indoor tanning in the
previous year and then divided into tertiles. We also
asked respondents to indicate whether or nor they
had ever experienced adverse effects from tanning
(from a list including itching, red/ painful skin, eye
problems, warts, medication reaction, and other).
Our measure of dependence was based on self-
report of perceived difficulty in quitting tanning
indoors. Adolescents were asked to rate "how hard
would it be for you to stop tanning indoors" on
a scale of 0 to 10, with 0 to 3 signifyi ng "not at all
hard" and from 7 to 10 signifying "extremely
hard." Adolescents who rated the measure from
0 to 3 (190 adolescents [71.2o/oD were classified as not
having any difficulty quitting tanning and therefore
nor ar high risk of being dependent on indoor
tanning. Adolescents who rated the measure ar 4 or
higher (77 adolescents [28.8%]) were considered ro
have at least some difficulty quitting indoor tanning
and therefore to be at some risk of dependence.
Using logistic regression, odds ratios (ORs) and
95o/o confidence intervals (Cis) were calculated to
estimate the association between adolescents' report
of perceived difficulty in quitting tanning indoors
and their personal characteristics, knowledge of risks
or attitudes wward tans, and indoor tanning behav-
ior. The relationship between each of these charac-
teristics and difficulty quitti ng were first examined
individually, while adj usting for age at the rime of
the interview (in categories 14-15, 16, or 17), sex,
and city (Boston or Minneapolis-St. Paul). The age
at which indoor tanning was initiated and the
Zeller et a! 591
frequency of indoor tanning use during the previous
year, the two factors hypothesized to be related to
dependence, emerged in support of rhe hypothesis
rhar adolescents may become dependent on indoor
tanning. Consequently, we considered the extent to
which these rwo characteristics were independently
associated with difficulty quitting, by estimating the
OR for each adjusted for the other, in addition to age,
sex, and city. To further explore these relationships,
we carried out additional anal yses to determine
whether any of the personal characteristics, knowl-
edge, or atti tudes changed the age-, sex-, and city-
adjusted ORs for age at initiation or frequency of
past year indoor tanning use and difficulty quitting.
If adjustment for the characteristic modified the
observed odds ratios for age at initiation or fre-
quency of indoor tanning use in a meaningful way,
or if the characteristic itself was found to be associ-
ated with the outcome independently of these two
measures, we retained the characteristic in a final
analysis that simultaneously considered all charac-
teristics selected in this manner to identify which
ones remained as important predictors of indoor
tanning dependence. We also assessed whether the
associations differed for younger versus older ado-
lescents, for boys versus girls, or for adolescents liv-
ing in Boston compared with those in Minneapolis.
However, the small sample size limited the interpre-
tation of these results, so we opted to present our
findings for the entire study sample.
RESULTS
Among adolescents who tanned more than once
in the past year, those who were 14 or 15 years old
at the time of the survey were more likely to report
difficulty quitting indoor tanning than 16-or 17-year-
olds (Table T) . Girls who tanned more than once in
the past year outnumbered boys by more than 8 tol
and were far more likely than boys to report difficulty
quitting indoor tanning (adjusted odds ratio [aOR] =
6.2; 95% CI, 1.4-26.8). Teenagers with higher skin
cancer risk scores were less likely ro report difficulty
quitting compared with those with lower skin cancer
risk scores (aOR = 0.9; 95o/o CI, 0.8-1.0). Those ado-
lescents who reported knowing someone with skin
cancer were less likely to repott difficulty quitti ng
than those who did not (aOR = 0.6; 95% CI, 0.3-1.1),
although the CI included 1.0. Other personal
characteristics such as city of residence (Boston or
Minneapolis-Sr Paul), degree of body satisfaction,
depression, and current smoking status were nor
associated with difficulty quitting.
A weak association was observed between level
of knowledge and difficulty quitting (Table II). On
the other hand, some attitudes were strongly related
592 Zeller et a!
Table I. Personal characteristics associated with
self-report of difficulty to quit indoor tanning
among adolescents aged 14 to 17 years
Sample Difficulty Adjusted OR'
Characteristic size quitting,% (95% Cl)
Age, y
14and 15 64 37.5 1.6 (0.8-3.0}
16 85 27.1 1.0 (0.5-1.9)
17 118 25.4 1.0
Sex
Girls 238 31.5 6.2 (1.4-26.8)
Boys 29 6.9 1.0
City
Boston 96 26.0 0.7 (0.4-1.3)
Minneapolis-St.Paul 171 30.4 1.0
Are you happy with
the way you look
Yes 180 26.7 1.0
No 87 33.3 1.3 (0.7-2.3}
Have you been bothered
by feeling unhappy,
sad, and depressed
in the last year
Yes 39 33.3 1.3 (0.6-2.6)
No 227 27.8 1.0
Skin cancer risk score 0.9 (0.8-1.0)
Do you know someone
with skin cancer
Yes 114 22.8 0.6 (0.3-1 .1)
No 153 33.3 1.0
Current smoker
Yes 77 31.2 1.3 (0.7-2.5)
No 190 27.9 1.0
Cl, Confidence interval; OR, odds ratio.
*Adjusted for city (Boston vs Minneapolis), sex, and age (a 3-level
variable).
ro the outcome. Of the two positive attitudes that we
considered-tanning lifts the spirits or is relaxing-
ORs were greater than 2.0, but only the association
between the effect of indoor tanning on spirits and
the likelihood of difficulty quitting reached statistical
significance (aOR = 2.7; 95% Cl, 1.5-4.7). Negative
attitudes toward tanning, such as agreeing that
it makes the skin smell badly (aOR = 2.1; 95% Cl,
1.2-3.8) or a history of symptoms (a OR = 2.8; 95% CI,
1.5-5.4), were also signifi cantly associated with diffi-
culty quitting, whereas find ing indoor tanning claus-
trophobic or expensive were not.
The mean age at which adolescents initiated
indoor tanning was 14.7 years; on average, adoles-
cents reported tanning 24.3 times during the prior
year (median, 15.0). These two characteristics from
Table II were the most strongly associated with dif-
ficulty quitting and a positive, increasing trend was
observed with younger age at initiation (P trend =
J AM ACAD 0 ERMATOL
APRIL 2006
Table ll. Indoor-tanning related knowledge,
attitudes, and behavior associated with
self-report of difficulty to quit indoor tanning
among adolescents aged 14 to 17 years
Sample Difficulty Adjusted OR
Characteristic size quitting,% (95% Cl)
Knowledge score 0.9 (0.8-1.0)
Tanning indoors helps
to lift my spirits
Agree 138 37.7 2.7 (1.5-4.7)
Disagree 129 19.4 1.0
Tanning indoors is
relaxing
Agree 219 31.1 2.1 (1.0-4.7)
Disagree 48 18.8 1.0
I feel claustrophobic in
the tanning bed or
booth
Agree 55 25.5 0.8 (0.4- 1.5)
Disagree 212 29.7 1.0
My skin smells bad after
I tan indoors
Agree 148 35.1 2.1 (1.2-3.8)
Disagree 119 21 .0 1.0
It's expensive to tan
indoors
Agree 209 28.7 0.8 (0.4- 1.5)
Disagree 58 29.3 1.0
Experienced any adverse
effects of indoor
tanning
Yes 171 35.7 2.8 (1.5-5.4)
No 96 16.7 1.0
Age first tanned, y
< 13 44 47.7 10.5 (3.5-31.5)
14 54 44.4 7.8 (2.8-21.5)
15 93 25.8 3.2 (1.3-8.0)
16-17 76 10.5 1.0
P trend .0119
Frequency (tertiles)
High 95 44.2 6.9 (3.0-1 5.9)
Medium 89 28.1 2.6 (1.1-6.0)
Low 83 12.1 1.0
P trend .0018
Cl, Confidence interval; OR, odds ratio.
*Adjusted for city (Boston vs. Minneapolis), sex, and age (a three
level variable).
.0119) and higher frequency of use (Ptrend = .0018).
We examined whether these characteristics were
independent predictors of difficulry quitting by
adjusting one for the other in addition to age, sex,
and city (Table III). Although somewhat attenuated,
the ORs for the association between age at initiation
or frequency of use and difficulty quitting remained
strong and the significance of the trends did not
change.
) AM ACAD DRMATOL
To determine which characteristics in Tables I and
II might explain the associations observed for age
at initiation and frequency, or were independently
related to difficulty quitting, we identified the fol -
lowing factors for our final analysis: skin cancer risk
score, knowing someone with skin cancer, having
experienced symptoms from indoor tanning, and
two attitudes (indoor tanning lifts spirits, indoor
tanning makes the skin smell) . Even after taking
into account characteristics theoretically postulated
to increase the likelihood of diffi culty quitting (eg,
positive effects of indoor tanning on one's spirits),
or ro decrease the likelihood of difficulty quitting
because of negative consequences (eg, experiencing
symptoms), we continued to find that age at initia-
tion was related to difficulty quitting, especially
among the youngest adolescents. Similarly, our ad-
justment for these characteristics barely changed the
ORs associated with increasing frequency of indoor
tanning use, and the P value remained signifi cant.
Adjusting the other characteristics in Table III for
age at initiation and difficulty quitting did not affect
the magnitude and significance of the associations
previously observed for most of the other factors,
such as skin cancer risk, knowing someone with
cancer, agreement that indoor tanning lifts spirits,
or the experience of symptoms from indoor tanning.
Agreement that indoor tanning makes the skin
smell, however, was no longer associated with the
outcome.
DISCUSSION
This study explores the implications of a novel
measure, self-report of difficulty in quitting indoor
tanning among adolescents in the Minneapolis-
St.Paul and Boston metropolitan areas. Our results
suggest that adolescents who initiate indoor tanning
at a younger age or who tan more frequently are
more likely to report clif.ficulry in quitting indoor
tanning. We did not find these relationships to be
diminished by the perceived negative consequences
of indoor tanning, such as being at higher risk for
skin cancer or acquaintance with someone with skin
cancer. In fact, some negative consequences such as
experience of acute adverse effects of indoor tanning
and acknowledgment that indoor tanning gives skin
an unpleasant odor were actuall y positively associ-
ated with difficulty quitting, perhaps because these
measures reveal a familiarity that accompanies ex-
cessive indoor tanning. These findings would sug-
gest an element of dependence, that is, continuation
of a behavior despite acknowledgment of its nega-
tive effects.
Our study should be interpreted in light of its
limitations. First, because this study was cross-
Zeller et a! 593
Table III. Personal characteristics and indoor-
tanning related knowledge, attitudes, and
behavior associated with self-report of difficulty to
quit indoor tanning among adolescent tanners
Characteristic
Age first tanned, y
:::; 13
14
15
16-17
P trend
Frequency (tertiles)
High
Medium
Low
P trend
Skin cancer risk score
Do you know someone with
skin cancer
Yes
No
Tanning indoors helps to lift
my spirits
Agree
Disagree
My skin smells bad after
I tan indoors
Agree
Disagree
Experienced any adverse
effects of indoor
tanning
Yes
No
Adjusted OR
(95% CI)
6. 1 (3.019.2)
4.4 (1.5 12.7)
2.5 (1.0-6.5)
1.0
.0108
4.5 (1.9 10.9)
2.1 (0.9-4.9)
1.0
.0016
C/, Confidence interval; OR, odds ratio.
Adjusted ORt
(95% Cl)
4.3 (1.3-14.7)
2.9 (0.9-9.0)
1.8 (0.7-5.0)
1.0
.119
4.1 (1.6-1 0.3)
2.0 (0.8-4.8)
1.0
.009
0.8 (0.7-1.0)
0.5 (0.2-0.9)
1.0
2.3 (1 .2-4.3)
1.0
1.3 (0.7-2.5)
1.0
2.9 (1.4-6.0)
1.0
*Age first tanned and frequency of indoor tanning adjusted
for each other and for city (Boston vs Minneapolis), sex, and age
(a 31evel variable).
t Adj usted for all other variables in table and for city (Boston vs
Minneapolis), sex, and age (a 3-level variable).
sectional, a temporal relationship between self-
report of difficulty quitting and high frequency of
indoor tanning or early age at initiation could not
be established. Although the close association of
the 3 measures (early age at initiation, frequent use,
and difficulty quitti ng) may indicate the existence
of a profile of dependency, the exact relationship
between the measures remains unclear. \Y/e chose
"difficulty to quit'' as the better marker for depen-
dency because it implies loss of control and contin-
uance despite adverse consequences, whereas it is
possible to frequently engage in a behavior without
becoming dependent. Furthermore, we acknowl-
edge that dependency is complex and may not be
adequately captured by our measures. Another
594 Zeller et a!
limitation is that our measures of indoor tanning
frequency and age at initiation were based on self-
report and therefore were subject to misclassifica-
tion. However, a previous study of the accuracy of
self-report found a strong correspondence (r = 0.78)
between self-report of indoor tanning and events
recorded in diaries.
64
The use of the targeted age list
may have introduced some selection bias, but the
response bias common to survey studies was likely
minimized by a high response rate in both Minnesota
(88%) and Bosron (73%). Finall y, our results are
based on a relatively small number of participants,
\vhich likely affected the precision of our estimates.
Despite the relatively small sample size, our results
demonstrate a strong association between difficulty
in quirring and characteristics similar to those shown
to predict dependency for other high-risk behaviors
among adolescents, and, to our knowledge, ours is
the first population-based examination of these
heretofore anecdotal repo1ts.
We also found that adolescents who agree that
tanning improves their mood were more li kely to
find the practice difficult to quit. Relaxation is a side
effect of indoor tanning noted in more than half of
users in previous studies
48
65
66
and was reported as
a major reason for indoor tanning for 42% (n = 225)
of respondents in one series.
47
A number of mecha-
nisms have been proposed to describe the possibly
mood-altering-effect of UV light exposure. It has
been posited that the UV-induced cutaneous release
of neuropeptides may contribute to the feeling of
well-being that in part motivates tanning behavior
67
Tn vivo studies of human melanocytes and kera-
tinocytes have demonstrated that UVB radiation
stimulates increased expression of the proopiomel-
anocortin gene (POMC) and subsequent production
and release of endorphins.
68
However, significant
increases in plasma opioid levels after tanning
have yet to be demonstrated in human subjects
69
70
Other investigations of the effect of UV light on
mood have focused on the photosensitivity of the
pineal melatonin-generating system. The enzyme
N-acetyltransferase, which initiates the conversion
of serotonin to melatonin, is reported to be inhibited
by UVA radiation?
1
72
This mechanism could theo-
retically result in increased serotonin production and
depletion of circulating melatonin in UV-exposed
subjects.
Studies in humans to elucidate these proposed
mechanisms have been scant. A recent randomized
clinical trial found that, although those participants
exposed to UV radiation showed positive changes
in psychological parameters compared with unex-
posed control subjects, there was no significant
difference in plasma seroronin or melatonin levels
J AM ACAD 0 ERMATOL
APRIL 2006
between the two groups
73
The potentially "rein-
forcing effect" of t.N light was further investigated in
a recent placebo-controlled clinical trial
74
in which
14 young adult frequent tanners were given access
to identical tanning beds with UVand non-tN filters.
After initial exposure to both beds, participants were
allowed to select which bed to use for an additional
session. Under blinded conditions, participants
chose the UV bed for 95% of the 41 additional
sessions, suggesting possible psychologic or physi-
ologic effect of UV light.
Many parall els exist between indoor tanning
behavior and other adolescent risk behaviors, such
as smoking, substance use, and pathological gam-
bling. Not only are adolescents who tan indoors
at increased risk of using tobacco and other sub-
stances,425354 but early age at onset of these behav-
iors and increased dose of exposure are each
associated with greater rates of alcohof
5
-80and
nicotine

Adolescence is a neuro-
logical developmental period that is speculated to
render young people more vulnerable to addictive
disorclers,
8 1
although these adolescent risk behaviors
may be psychologicall y or physiologicall y deter-
mined. Although our findings lend support to our
hypothesis, further explorations of the possibility of
indoor tanning dependency are needed. In partic-
ular, a longitudinal design would address potential
problems related to accuracy of self-reported indoor
tanning use as well as establish a temporal relation-
ship between "at-risk" characteristics and tanning
dependency.
REFERENCES
1. Spencer JM, Amonette RA. Indoor tanning: risks, benefits, and
future trends. J Am Acad Dermatol 1995;33:288-98.
2. Walters BL, Kelley TM. Commercial tanning facilities: a new
source of eye inj ury. Am J Emerg Med 1987;5:386-9.
3. Lichtenstein J, Sherertz EF. Harmful effects of indoor tanning.
Am Fam Physician 1985;32:142-6.
4. Matsumura Y, Ananthaswamy HN. Toxic effects of ultraviolet
radiation on t he skin. Toxicol Appl Pharmacol2004;195:298-308.
5. American Cancer Society. Cancer facts and figures 2005.
Atlanta (GA): American Cancer Society; 2005.
6. Jemal A. Clegg LX, Ward E, Ries LA. Wu X, Jamison PM, et al.
Annual report to the nation on the status of cancer, 1975-
2001, with a special feature regarding survival. Cancer 2004;
101:3-27.
7. Spencer JM, Amonette RA. Tanning beds and skin cancer:
artificial sun + old sol = real risk. Cli n Dermatol 1998;16:
487-501.
8. Miller SA. Hamilton SL, Wester UG, Cyr WH. An analysis of UVA
emissions from sunlamps and the potential importance for
melanoma. Photochem Photobiol 1998;68:63-70.
9. Whitmore SE, Morison WL, Potten CS, Chadwick C. Tanning
salon exposure and molecular alterations. JAm Acad Dermatol
2001;44:77S-80.
10. Karagas MR, Stannard VA, Mott LA, Slattery MJ, Spencer SK,
Weinstock MA. Use of tanning devices and risk of basal cell
) AM ACAD DRMATOL
VOLUME 54. NUMBER 4
and squamous cell skin cancers. J Natl Cancer lnst
2002;94:224-6.
11. Gandini S, Sera F, Cattaruzza MS, Pasquini P, Picconi 0, Boyle P,
et al. Meta-analysis of risk factors for cutaneous melanoma:
II. sun exposure. Eur J Cancer 2005;41 :45-60.
12. Lim HW, Cyr WH, DeFabo E, Robinson J, Weinstock MA,
Beer JZ, et al. Scient ific and regulatory issues related to indoor
tanning. J Am Acad Dermatol 2004;51:781-4.
13. Swerdlow AJ, Weinstock MA. Do tanning lamps cause mela-
noma? An epidemiologic assessment. J Am Acad Dermatol
1998;38:89-98.
14. Autier P. Perspectives in melanoma prevention: the case of
sunbeds. Eur J Cancer 2004;40:2367-76.
15. Westerdahl J, lngvar C, Masback A, Jonsson N, Olsson H. Risk
of cutaneous malignant melanoma in relation to use of
sunbeds: further evidence for UV-A carcinogenicity. Br J
Cancer 2000;82:1593-9.
16. Garland CF, Garland FC, Gorham ED. Epidemiologic evidence
for different roles of ultraviolet A and B radiation in melanoma
mortality rates. Ann Epidemiol 2003;13:395-404.
17. Agar NS, Halliday GM, Barnetson RS, Ananthaswamy HN,
Wheeler M, Jones AM. The basal layer in human squamous
t umors harbors more UVA than UVB fingerprint mutations: a
role for UVA in human skin carcinogenesis. Proc Natl Acad Sci
U S A 2004;1 01:4954-9.
18. Larsson P, Andersson E, Johansson U, Ollinger K, Rosdahl I.
Ultraviolet A and B affect human melanocytes and keratino-
cytes differently. A study of oxidative alterations and apopto-
sis. Exp Dermatol 2005;14:117-23.
19. Linge C. Relevance of in vitro melanocytic cell studies to the
understanding of melanoma. Cancer Surv 1996;26:71-87.
20. Wang SQ, Setlow R, Berwick M, Polsky D, Marghoob AA, Kopf
AW, et al. Ultraviolet A and melanoma: a review. J Am Acad
Dermatol 2001 ;44:837-46.
21 . Bachelor MA, Bowden GT. UVA-mediat ed activation of signal-
ing pathways involved in skin tumor promotion and progres-
sion. Semin Cancer Bioi 2004;14:131-8.
22. Runger TM. Rol e of UVA in the pathogenesis of melanoma and
non-melanoma skin cancer. A short review. Photodermatol
Photoimmunol Photomed 1999;1 5:212-6.
23. Woollons A, Kipp C, Young AR, Petit-Frere C, Arlett CF,
Green MH, et al. The 0.8% ultraviolet B cont ent of an ultra-
violet A sunlamp induces 75% of cyclobutane pyrimidine
dimers in human keratinocytes in vitro. Br J Dermatol 1999;
140:1023-30.
24. Marrot l, Belaidi JP, Meunier JR, Perez P, Agapakis-Causse C.
The human melanocyte as a particular target for UVA radiation
and an endpoint for photoprotection assessment. Phot ochem
Photobiol 1999;69:686-93.
25. Burren R, Scaletta C, Frenk E, Panizzon RG, Applegate LA.
Sunlight and carcinogenesis: expression of p53 and pyrimidine
dimers in human skin following UVA I, UVA I + II and solar
simulating radiations. lnt J Cancer 1998;76:201 -6.
26. Kvam E, Tyrrell RM. Induction of oxidative DNA base damage
in human skin cells by UV and near visible radiation. Carcino-
genesis 1997;18:2379-84.
27. Ley RD. Ultraviolet radiation A-induced precursors of cutane-
ous melanoma in monodelphis domestica. Cancer Res 1997;
57:3682-4.
28. Setlow RB, Grist E, Thompson K, Woodhead AD. Wavelengths
effective in induction of malignant melanoma. Proc Natl Acad
Sci USA 1993;90:6666-70.
29. Lindelof B, Sigurgiersson B, Tegner E, Larko 0, Johannessen A,
Berne B, et al. PUVA and cancer risk: t he Swedish follow-up
study. Br J Dermatol 1999;141 :108-12.
Zeller et a! 595
30. Katz KA, Marcil I, Stern RS. Incidence and risk factors associ-
ated with a second squamous cell carcinoma or basal cell
carcinoma in psoralen + ultraviolet A light-treated psoriasis
patients. J Invest Dermatol 2002;118:1 038-43.
31 . Stern RS, Nichols KT, Vakeva LH. Malignant melanoma in
patients treated for psoriasis wit h methoxsalen (psoralen) and
ultraviolet A radiat ion (PUVA). The PUVA follow-up study.
N Engl J Med 1997;336:1041-5.
32. Roest MAB, Keane FM, Agnew K, Hawk JLM, Griffiths WAD.
Multiple squamous skin carcinomas following excess sunbed
use. J R Soc Med 2001;94:636-7.
33. Lever LR, Lawrence CM. Nonmelanoma skin cancer associated
with use of a tanning bed [letter). N Engl J Med
1995;332:1450-1 .
34. Glanz K, Saraiya K, Wechsler H. Guidelines for school programs
to prevent skin cancer. MMWR Morb Mortal Wkly Rep 2002;
51:1-18.
35. Ferrini RL, Perlman M, Hill L. American College of Preventive
Medicine practice policy statement: skin prot ection from
ultraviol et light exposure. Am J Prev Med 1998;14:83-6.
36. Robinson JK, Amonette RA, Wright SW, Bewerse BA, Bergfeld
WF, Farris PK. Executive summary of the national "Sun Safety:
Prot ecting Our Fut ure" Conference: American Academy of
Dermatology and Centers for Disease Control and Prevention,
New York, New York, May 1 and 2, 1997. JAm Acad Dermatol
1998;38:774-80.
37. Council on Scientific Affairs. Harmful effects of ultraviolet
radiation. JAMA 1989;262:380-4.
38. Khlat M, Vail A, Parkin M, Green A. Mortality from melanoma in
migrant s to Australia: variation by age at arrival and duration
of stay. Am J Epidemiol1992;135:1 103-13.
39. Whiteman DC, Whiteman CA, Green AC. Childhood sun expo-
sure as a risk factor for melanoma: a systematic review of
epidemiologic studies. Cancer Causes Control 2001 ;12:69-82.
40. Indoor Tanning Association, Inc. ITA estimates industry over-
view 2004. Available at: http://www.theita.com/pdf/estimates.
pdf. Accessed February 4, 2005.
41. Deveney J. Understanding consumer base demographics.
Sun Business 2000 2000;3:51-3.
42. Demko CA, Borawski EA, Debanne SM, Cooper KD, Stange
KC. Use of indoor tanning facilities by white adolescents in
the United States. Arch Pediatr Adolesc Med 2003;157:854-60.
43. Cokkinides VE, Weinstock MA, O'Connell MC. Thun MJ. Use of
indoor tanning sunlamps by US youth, ages 11-18 years, and
by their parent or guardian caregivers: prevalence and corre-
lates. Pediatrics 2002;1 09:1 124-30.
44. Robinson JK, Rademaker AW, Sylvester JA, Cook B. Summer
sun exposure: knowledge, attit udes, and behaviors of Midwest
adolescents. Prev Med 1997;26:364-72.
45. Oliphant JA, Forster JL, McBride CM. The use of commercial
tanning facilities by suburban Minnesota adolescents. Am J
Public Health 1994;84:476-8.
46. Lazovich D, Forster J. Indoor tanning by adolescents: preva-
lence, practices and policies. Eur J Cancer 2005;41:20-7.
47. Knight JM, Ki rincich AN, Farmer ER, Hood AF. Awareness of the
risks of t anning lamps does not influence behavior among
college students. Arch Dermatol 2002;138:1311 -5.
48. Mawn VB, Fleischer AB Jr. A survey of attitudes, beliefs, and
behavior regarding tanni ng bed use, sunbathing, and sun-
screen use. J Am Acad Dermat ol 1993;29:959-62.
49. Netburn D. Young, carefree, and hooked on sunlamps. New
York Times May 26, 2002;9.
50. Chedru MF, Dumont C, Jourdain F, Postei-Vinay N, Dab W.
[Cutaneous melanoma prevention. Modify behaviors.) Rev Prat
2004;54:1 1656. French.
596 Zeller et a!
51. WCCO-TV. Hooked on tanning? Available from: htt p://wcco.com/
specialreports/local_story_052181257.ht ml. Accessed February
10, 2005.
52. World Health Organization. WHO Expert Committee on Drug
Dependence, Technical Report Series. 1992;28.
53. Lazovich D, Forst er J, Sorensen G, Emmons K, Stryker J,
Demierre MF, et al. Characteristics associated with use or
intention to use indoor tanning among adolescents. Arch
Pediatr Adolesc Med 2004;1 58:918-24.
54. Boldeman C, Jansson B, Nilsson B, Ullen H. Sunbed use in
Swedish urban adolescents relat ed t o behavioral characteris-
tics. Prev Med 1 997;26:114-9.
55. American Psychiatric Association. Diagnostic and Statistical
Manual of Mental Disorders. 4t h ed. Text revision. Washington
(DC): American Psychiatric Association; 2000.
56. Warthan MM, Uchida T, Wagner RF Jr. UV light tanning as a type
of substance-related disorder. Arch Dermatol2005;141:963-6.
57. O'Loughlin J, DiFranza J, Tarasuk J, Meshefedjian G, McMillan-
Davey E, Paradis G, et al. Assessment of nicotine dependence
symptoms in adolescents: a comparison of five indicators. Tob
Control 2002;1 1:354-60.
58. O'Loughlin J, Tarasuk J, DiFranza J, Paradis G. Reliability of
selected measures of nicot ine dependence among adoles-
cents. Ann Epidemiol 2002;12:353-62.
59. Wheeler KC, Fletcher KE, Wellman RJ, DiFranza JR. Screening
adolescent s for nicotine dependence: the hooked on nicotine
checklist. J Adolesc Health 2004;35:225-30.
60. Stryker JE, Lazovich D, Forster JL, Emmons KM, Sorensen G,
Demierre MF. Maternal/female caregiver influences on adoles-
cent indoor tanning. J Adolesc Health 2004;35:528.e 1 -528.e9.
61. Mayhew KP, Flay BR, Mott JA. Stages in t he development of
adolescent smoking. Drug Alcohol Depend 2000;59(Suppl 1 ):
S61-81 .
62. O'Loughlin J, DiFranza J, Tyndale RF, Meshefedian G, McMillan-
Davey E, Clarke PB, et al. Nicotine-dependence symptoms are
associated with smoking frequency in adolescents. Am J Prev
Med 2003;25:219-25.
63. Robinson ML, Berlin I, Moolchan ET. Tobacco smoking trajec-
t ory and associat ed ethnic differences among adolescent
smokers seeking cessation treatment. J Adolesc Health 2004;
35:217-24.
64. Hillhouse JJ, Turrisi R, Kastner M. Modeling tanning salon
behavioral tendencies using appearance mot ivation, self-
monitoring and the t heory of planned behavior. Health Educ
Res 2000;15:405-14.
65. Diffey BL. Use of UV-A sunbeds for cosmetic tanning. Br J
Dermatol 1986;1 15:67-76.
66. Dougherty MA, McDermott RJ, Hawkins MJ. A profile of
commercial tanning salons. Health Values 1988;12:21-9.
67. Wintzen M, Yaar M, Burbach JP, Gilchrest BA. Proopiomelano-
cort in gene product regulation in keratinocytes. J Invest
Dermatol 1996;1 06:673-8.
J AM ACAD 0 ERMATOL
A PRIL 2006
68. Chakraborty AK, Funasaka Y, Slominski A, Ermak G, Hwang J,
Pawelek JM, et al. Production and release of proopiomelano-
cortin (POMC) derived peptides by human melanocytes and
keratinocytes in culture: regulation by ultraviolet B. Biochim
Biophys Acta 1 996;1 313:130-8.
69. Gambichler T, Bader A, Vojvodic M, Avermaete A, Schenk M,
Altmeyer P, et al. Plasma levels of opioid peptides after
sunbed exposures. Br J Dermatol 2002;147:1207-1 1.
70. Wintzen M, Ostijn DM, Polderman MC, le Cessie S, Burbach JP,
Vermeer BJ. Total body exposure to ultraviolet radiation does
not inf luence plasma levels of immunoreactive beta-endor-
phin in man. Photodermatol Photoimmunol Photomed 2001;
17:256-60.
71. Zawilska JB, Rosiak J, Nowak JZ. Effects of near-ultraviolet
(UV-A) light on melat onin biosynthesis in vertebrat e pineal
gland. Bioi Signals Recept 1 999;8:64-9.
72. Fraikin GY, Strakhovskaya MG, lvanova EV, Rubin AB. Near-UV
activation of enzymatic conversion of 5-hydroxytryptophan to
serotonin. Photochem Photobiol 1989;49:475-7.
73. Gambichler T, Bader A, Vojvodic M, Bechara FG, Sauermann K,
Altmeyer P, et al. Impact of UVA exposure on psychological
parameters and circulating serotonin and melatonin. BMC
Dermat ol 2002;2:6.
74. Feldman SR, Liguori A, Kucenic M, Rapp SR. Fleicher AB Jr, Lang
W, et al. Ultraviolet exposure is a reinforcing stimulus in frequent
indoor tanners. J Am Acad Dermatol 2004;5 1:45-51.
75. DeWit OJ, Adlaf EM, Offord DR, Ogborne AC. Age at first
alcohol use: a risk factor for the development of alcohol dis-
orders. Am J Psychiatry 2000;1 57:745-50.
76. Grant BF, Dawson DA. Age at onset of alcohol use and its
associat ion with DSM-IV alcohol abuse and dependence:
results from the National Longitudinal Alcohol Epidemiologic
Survey. J Subst Abuse 1997;9:103-10.
77. Grant BF, Stinson FS, Harford TC. Age at onset of alcohol use
and DSM-IV alcohol abuse and dependence: a 1 2-year follow-
up. J Subst Abuse 2001 ;13:493-504.
78. Wagner FA, Anthony JC. From first drug use to drug depen-
dence; developmental periods of risk for dependence upon
marijuana, cocaine, and alcohol. Neuropsychopharmacology
2002;26:479 88.
79. Vega WA, Aguilar-Gaxiola S, Andrade L, Bijl R, Borges G,
Caraveo-Anduaga JJ, et al. Prevalence and age of onset for
drug use in seven int ernat ional sites: results f rom t he inter-
national consortium of psychiatric epidemiology. Drug Alco-
hol Depend 2002;68:285-97.
80. Warner LA, White HR. Longitudinal effects of age at onset and
f irst drinking situat ions on problem drinking. Subst Use Misuse
2003;38:1983-2016.
81. Chambers RA, Taylor JR, Potenza MN. Developmental
neurocircuitry of motivation in adolescence: a critical
period of addiction vulnerability. Am J Psychiat ry 2003;160:
1041-52.
What's new online?
Click on www.eblue.org today to find out how you can link from cited references to abstracts and full text of
other participating journals.
124:5 MAY 2005 LETIERS TO THE EDITOR 1079
Proportion of Lifetime UV Dose Received by Age 18, What Stern
et a/ Actually Said in 1986
To the Editor:
I read with interest Thieden et a/'s (2004) study that as-
sessed UV exposure among children, teenagers, and adults
in Copenhagen. As have many "campaigns" and most of
the more than 200 publications that cite our work that I have
read, the authors appear to have failed to understand our
findings and misquote our work (Stern et a/, 1986).
We developed a mathematical model of the relation of
UV exposure, age, and other factors to non-melanoma skin
cancer (NMSC) risk. We used this model to estimate the risk
reduction that could be achieved with the regular use of a
high SPF sunscreen for NMSC. Our model predicted that
consistent use of high SPF sunscreen to age 18 would re-
duce the lifetime risk of NMSC by 78% (base case analysis).
In our base case analysis, we assumed that children living in
the northern United States spend three times as much time
in the sun as adults annually. This assumption is consistent
with slightly less than half of the total lifetime sun exposure
occurring by age 18.
Because we understood that our base case assumption
about sun exposure was based on scant data and the ratio
of childhood to adult sun exposure would affect the pre-
dicted effect of childhood sun protection on lifetime skin
cancer risk reduction, our model included an independent
variable to account for the variability in the ratio of annual
child to adult sun exposure, which we termed the sun af-
finity ratio (SAR) (Table I) (Fig 1 ).
Our sensitivity analyses examined the impact of various
assumptions about sun exposure patterns over a lifetime
(i.e., varying SAR) on the primary endpoint: lifetime NMSC
risk reduction with effective UV protection up to age 18. One
full journal page of our paper as well as a figure and a table
(reproduced here from Stern, 1986) detailed these concerns
and the range of assumptions about sun exposure habits
over a lifetime that we utilized in our sensitivity analysis. The
extreme low and high assumptions of SAR that we utilized
in our sensitivity analysis are consistent: 16% and 78% of total
lifetime UV exposure occurring up to age 18, respectively. The
extreme high case would describe a youthful sun worshiper
who becomes a compulsive sun avoider after age 18.
The greater importance of sun exposure early in life than
in adult years for NMSC lifetime risk, particularly basal can-
cer risk, is supported by epidemiologic studies performed
subsequent to our study (Gallagher, 1995; Corona et a/,
2001). Our model, which assumed a multistage model of
carcinogenesis, accounted for the latency between expo-
sure to a carcinogen and the actual development of the
cancer. Such a latency period is observed for most cancers.
As a result, the same UV exposure many years in the past
(i.e., as a youth} is likely to have been a greater contributor
to the development of a skin cancer that develops in an
Abbreviations: NMSC, non-melanoma skin cancer; SAR, sun af-
finity ratio
adult than a comparable quantity of recent exposure. Be-
cause of latency between exposure and cancer, when we
assumed annual sun exposure is the same throughout life
(SAR = 1% and 21% of lifetime exposure occurring to age
18), the reduction in lifetime skin cancer risk predicted with
high levels of sun protection to age 18 decreased to 62%
from our base case prediction of 78% risk reduction (with
SAR = 3% and 45% of exposure occurring up to age 18).
Of course, the pattern of an individual's sun exposure
over a lifetime is likely to be an even more complex one than
described by the SAR. We are pleased that our early work
has stimulated studies such as Thieden's, which attempts
to quantify sun exposure patterns. But, an assessment of
the robustness of Thieden's findings and their relevance to
our base case assumptions should consider changes in
habits over the last two decades as well as the generaliz-
ability and precision of Thieden's findings.
Since the 1980s, sun exposure habits may have
changed. Perhaps the message that we and others subse-
quently emphasized, decreasing childhood sun exposure is
particularly important for NMSC risk reduction, has actually
reduced sun-seeking behavior in children more than adults
(i .e., decreased the SAR).
Data from Denmark may not reflect US conditions. My
lucky adult Scandinavian friends seem to have nearly as
much opportunity for sun exposure as their children
(SAR = 1) and along with my children have at least three
times the summer vacation I have (i.e., SAR = 3). These dif-
ferences from my small and admittedly biased sample
highlight the cultural and social determinants of SAR and
the importance of unbiased sampling if estimates that ac-
curately reflect a large population too are achieved. Thi-
eden's sample of Danish health care workers and golfers
seems unlikely to be an ideal one to assess sun exposure
patterns in the general population.
Thieden does not detail the power or precision of her
estimates. But, the groups she studied are quite small (less
than five subjects per year of age for those less than 20 and
less than two subjects per year of age over age 20). Even if
the authors were fortunate enough to have recruited a sam-
Table I. Sun affinity ratios (SAR)
used in the analysis of skin cancer risk"
Adult sun affinity
Child sun affinity LoW' Average
Lowd
6 2
Average 9 3
Highe 12 4
Highc
2/ 3
1
11
2
"An average ch1ld is assumed to have three times the annual sun ex-
posure of an average adult.
bOne-third times average.
"Three times average; nine times low.
dTwo-thirds times average.
eone and one-third times average; two times below.
Reference: Stern (1986).
1080 LETIERS TO THE EDITOR
d
18
Age,yr
Death
Figure 1
Age pattern of ultraviolet-B exposure. SPF indicates sun protective
factor; SAR, sun affinity ratio; and d, dose affiliations
pie that is truly representative of the Danish population, Thi-
eden's findings are likely to be more applicable to 21 st-cen-
tury Denmark, a quite northern country (latitude 55) blessed
with far more generous vacations than those who worked in
not quite tropical Boston (latitude 42) in the 1980s.
Robert S. Stern,
Department of Dermatology,
Beth Israel Deaconess Medical Center,
Harvard Medical School , Boston,
Massachusetts, USA.
THE JOURNAL OF INVESTIGATIVE DERMATOLOGY
DOl: 1 0.11 11/j.0022-202X.2005.2371 O.x
Manuscript received January 18, 2004; revised January 14, 2005;
accepted for publication January 19, 2005
Address correspondence to: Robert S. Stern, Department of Dermatol-
ogy, Beth Israel Deaconess Medical Center, Harvard Medical School,
Boston, Massachusetts, USA. Emails: rthornto@bidmc.harvard.edu,
rstern@caregroup.harvard.edu, sjanus@bidmc.harvard.edu
References
Corona R, Dogliotti E, D'Errico M, et al: Risk factors for basal cell carcinoma in a
Mediterranean population: Role of recreational sun exposure early in life.
Arch Dermatol 137:1162-1168, 2001
Gallagher RP, Hill GB, Bajdik CD, Fincham S, Goldman AJ, Mclean Dl,
Threlfall WJ: Sunlight exposure, pigmentary factors, and risk of non-
melanocytic skin cancer. I. Basal cell carcinoma. Arch Dermatol 131:
157-1 63, 1995
Stern RS, Weinstein MC, Baker SG: Risk reduction for nonmelanoma skin
cancer with childhood sunscreen use. Arch Dermatol 122:537- 545,
1986
Thieden E. Philipsen PA, Sandby-Moeller J, Heydenreich J, Wulf HC: Proportion
of lifetime UV dose received by children, teenagers and adults based on
time-stamped personal dosimetry. J Invest Dermatol 123: 1147- 1150,
2004
See related Commentary on page vii
Vitamin D Induces the Antimicrobial Protein hCAP18
in Human Skin
To the Editor:
Cathelicidins are a class of mammalian antimicrobial pep-
tides expressed in leukocytes and at epithelial surfaces
(Zanetti, 2004). Human cathelicidin antimicrobial protein
hCAP18 is encoded by CAMP (Ensembl Gene ID ENS
G00000164047) on chromosomal location 3p21 and is the
sole cathelicidin protein in humans. Recent studies have
shown that cathelicidins, in addition to being antimicrobial,
are multifunctional proteins with receptor-mediated effects
on eukaryotic cells and activity in chemotaxis, angiogene-
sis, and wound healing (Zaiou et at, 2003; Zanetti, 2004). In
the skin, there is low constitutive expression of hCAP18 in
the basal layer of keratinocytes but rapid upregulation upon
inflammation and injury (Frohm Nilsson et at, 1999; Dors-
chner et at, 2001; Heilborn et at, 2003).
Molecular mechanisms controlling the expression of
CAMP are still poorly understood. We have investigated
whether its expression could be influenced by agents that
affect the proliferation and differentiation of skin keratin-
ocytes. Human neonatal epidermal keratinocytes (Cascade
Biologics, Portland, Oregon) were cultured in Epilife serum-
free keratinocyte growth medium (Cascade Biologics) con-
taining growth supplements and a calcium concentration of
0.06 mM. At 60% confluency, the agents assayed were
added to the medium and cells were harvested after 24 h.
RNA was extracted and reverse transcribed by standard
methods, and the expression was quantified by Real-Time
RT-PCR on an ABI Prism 7700 (Applied Biosystems, Foster
City, California) using 5 ng of eDNA according to standard
protocols. Sequences were 5' -GTCACCAGAGGATIGTGA-
CTICAA-3' and 5' -TIGAGGGTCACTGTCCCCATA-3' for
the primers, and 6-FAM-5'-CCGCTICACCAGCCCGTC-
CTI-3'-BHQ1 for the fluorigenic probe.
An upregulation of CAMP of about one order of magni-
tude was achieved by treatment with 1 00 nM MC903/ca-
lcipotriol, a vitamin D analog applied for psoriasis treatment
(Kragballe, 1995). Calcium is known to regulate major func-
tions of the epidermis including terminal differentiation. Pre-
treatment of cells by 1.5 mM calcium for 48 h increased the
expression by about 1.5-fold, and was synergistic to the
effects of MC903 (Fig 1a). Based on these findings, we as-
sayed the effect of vitamin D and its metabolites (all from
Fluka, Buchs, Switzerland). Both biologically active forms
of vitamin 0
3
, i.e., 1 ,25(0HhD
3
and 25(0H)D
3
, stimulated
CAMP expression at the same magnitude as MC903. The
corresponding vitamin 0
2
analogs were slight ly less effi -
cient. All compounds were active down to levels of 1 0 nM
(shown for 1 ,25(0HhD
3
). The precursor of vitamin D bio-
synthesis, 7 -dehydrocholesterol (7 -DHC), was ineffective.
Western blot analysis confirmed that the elevated transcrip-
tion was reflected on the protein level (Fig 1b).
An in silica analysis revealed two putative vitamin D re-
sponsive elements (VDRE) of the DR3 type, and one puta-
tive heterodimer site of the DR5 type, within 1 kb upstream
of the transcription start (Table 1). This region was subcloned
Update on sun protection and tanning in children
Robert J. MacNeal and James G.H. Dinulos
Purpose of review
In this article we examine the latest information regarding
pediatric sun protection and specifically focus on the
impact of indoor tanning on children.
Recent findings
We examine and discuss the prevailing views and trends on
ultraviolet radiation exposure, sun protection and the
attendant adverse health effects among children. We
identify the challenges related to limiting ambient ultraviolet
exposure in children. As children, particularly adolescents,
gain an understanding of sun protection, their use of
ultraviolet radiation protection does not increase
concordantly. We discuss health policies regarding
ultraviolet radiation protection, compliance with these
policies and potential methods to increase compliance. We
also examine addictionlike behavior and other psychosocial
factors affecting those who are regular indoor tanners.
Finally, we consider possible alternatives to ultraviolet
radiation induced tanning and whether encouraging these
options leads to a confusing, mixed message from
healthcare providers.
Summary
In this article we examine and discuss current practices,
attitudes and recent trends in sun protection. We discuss
indoor tanning, with special emphasis on the potential
addictive qual ities of ultraviolet radiation. Our goal is to
identify opportunities and future directions to promote sun
safety and protect children from harmful ultraviolet radiation
exposure.
Keywords
indoor tanning and vitamin D, pediatric, sun protection,
sunburn
Curr Opin Pediatr 19:425- 429. 2007 Lippincott Williams & Wilkins.
Section of Dermatology, Dartmouth- Hitchcock Medical Center, Lebanon. New
Hampshire, USA
Correspondence to James G.H. Dinulos, MD, Section of Dermatology,
Dart mouth-Hitchcock Medical Center, One Medical Center Drive, Lebanon,
NH 03756, USA
Tel : + 1 603 653 9400; fax: + 1 603 650 0921 ;
e- mail: j ames.g.h.dinulos@hitchcock.org
Current Opi nion i n Pedi atrics 2007, 19:425- 429
Abbreviations
SPF sun protection factor
UVR ultraviolet radiation
2007 Lippincott Williams & Wilkins
1040-8703
Introduction
The incidence of skin cancer surpasses all other human
cancers. In the US, the incidence of malignant melanoma
is increasing approximately 4% per year. The principal
risk factor for all types of skin cancer is ultraviolet
radiation (UVR), particularly cumulative exposure during
childhood. Hence, decreasing exposure to UVR in child-
hood has the potential to significantly lower the incidence
of most forms of skin cancer. In this article we examine
the latest information regarding pediatric sun protection
and specifically focus on the impact of indoor tanning
on children.
Pediatric sun protection
Skin cancer, especially basal and squamous cell carci-
noma, has been clearly linked to early and intense
exposure to UVR. Educating parents and young children
about the harmful effects of UVR should lead to healthy
sun protection behavior and skin cancer prevention in
adulthood. Large-scale interventions have attempted to
change behavior by educating children, parents and
communities about the deleterious effects of UVR; how-
ever Olson eta/. (1] found that this may not be as simple as
it seems. They showed that substantial barriers exist in
educating adolescents on sun safe behavior. In this
randomized controlled trial of 10 US communities in
New England, multiprong, long-term, educational and
active counseling programs were implemented. This
study found that at least 2 years of intervention are
required to observe a positive change in adolescent
behavior towards UV exposure. As this study demon-
strates, impacting widely held societal beliefs and beha-
viors is possible and can lead to large-scale improvement
in public health bur may take much longer than con-
ventionally thought. Antismoking campaigns arc another
example. Through these massive efforts, teen smoking
rates have declined in the US. Continued governmental
support, regular messages from the media and direct
education efforts will be required to maintain and
improve positive outcomes [2].
Many adults and adolescents regularly expose them-
selves to UVR for the purpose of tanning. Adolescents
show a disturbing trend of increased exposure to artificial
and natural UVR as they progress into adulthood. In the
US, most adolescents are aware of the deleterious effects
of UVR, yet many continue this unsafe practice. One
principal factor driving tanning behavior is the cultural
belief that equates tanned skin with beauty and afflu-
ence. The media have a strong influence over societal
425
Copyright Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
426 Dermatology
trends and beliefs. A recent study [3] points to television
as the major influence on teens' definition of attractive-
ness and their source of sun protection information.
Images of tanned celebrities are constant and form
the basis for the look to which adolescents aspire.
Tanning is no longer the mark of the common man,
toiling outdoors for his living. T anned skin is now
viewed as a luxury: a sign that one can afford time away
from the office. A recent study supports this. After
looking at responses from over 200 000 people from
the Centers for Disease Control (CDC) Behavioral Risk
Factor Surveill ance System a positive correlation
between increased rates of sunburn and higher socio-
economic class was found. This was felt to be secondary
to participants' ability to afford vacations in locations
with more intense UVR as well as regular indoor tanning
sessions [4. ].
Ultraviolet radiation and vitamin D
The skin serves an essential role in vitamin D biosyn-
thesis (Fig. 1). A few epidemiologic srudies show lower
vitamin D levels are associated with higher cancer rates
[5,6]. Proponents of these studies believe higher vitamin
D levels through UVR exposure will prevent many of
these cancers. Encouraging exposure to UVR for the
purpose of elevating vitamin D levels is not wise and
should not be recommended for several reasons. First, t he
action spectra for burning, DNA damage and vitamin D
production are all virtually identical, occurring at UV
wavelengths of around 300 nm [7,8]. Higher wavelengths
in the UV A range (such as those found in many tanning
beds) have minimal effect on cutaneous vitamin D pro-
duction [9]. Second, maximal cutaneous production of
vitamin D occurs in a very short period of time. Prolonged
exposure to UVR results in the accumulation of inactive
vitamin D metabolites. For example, a child with a very
fair complexion will reach maximal vitamin D synthesis
after 2- 8 min of sun exposure in t he New York or Boston
areas. Climates further north require only minimally
Figure 1 Vitamin D biosynthetic pathway
Vit amin 0 synthesi s
Cholecalciferol {vitami n 0
3
)

25-Hydroxycholecalciferol
1,25-0 i hydroxy vitamin D (calcitriol)
more time [10]. Children wi t h dark complexions have
highly effective, natural UV protection and therefore get
very little of their required vitamin D from UVR [11].
Third, supplementation with multivitamins and a
balanced diet, for the vast majority of individuals, mini-
mizes the importance of UVR for maintenance of ade-
quate levels of vitamin D. Finally, despite the reported
anti -tumorigenic effects of UVR-induced vitamin D,
increased sun exposure obviously fails to suppress skin
cancer [12,13]. It is ironic that the tanning industry
aggressively pursues healthy whi te adolescent girls,
extolling the 'benefits' of higher vitamin D levels and
overall health in those that tan. These girls need supple-
mental UV induced vitamin D the least. They already
receive ample amounts of vitamin D through diet and
incidental ambient UV exposure and are at greatest risk
for the negative effects of UVR [14,15].
Despite the many adverse effects of long term UVR
exposure, physicians often treat difficult skin conditions
with controlled doses of UVB and UVA radiation. UVR
improves inflammatory disorders such as psoriasis and
eczema and neoplastic disorders such as cutaneous T-cell
lymphoma. Adolescents are often confused when UVR is
prescribed for therapeutic purposes, yet medical provi-
ders discourage tanning. One must emphasize to patients
and parents that even UVR utilized for medical purposes
carries an increased risk for skin cancer and premature
skin aging. Moreover, exposure ro natural sunlight and
indoor tanning is done in a relatively uncontrolled man-
ner, potentially leading tO a greater risk for photoaging
and skin cancer.
Pediatric indoor tanning
Despite ongoing efforts to educate the public about the
deleterious effect of UVR [16- 18], many adolescents
continue to frequent indoor tanning salons. Risk-taking
behavior is common among adolescents and tan seekjng
behavior is no exception. Recently, O'Riordan eta!. [19]
examined the association of tanning with other high-risk
adolescent behaviors, such as smoking and substance
abuse. This prospective cohort study referred tO as Grow-
ing Up Today Study (GUTS) evaluated over 6000 ado-
lescent girls who are daughters of theN urses Health Study
II. These investigators found significant associations
between tanning and other common high-risk behaviors
such as binge drinking, drug use and unhealthy weight loss
techniques like laxative use and self-induced vomiting.
Multiple other studies of adolescent behavior have con-
firmed a positive correlation between indoor tanning
and unhealthy behaviors such as tobacco, alcohol and drug
use (20- 301.
A recent study reported a high percentage of adolescents
who tan have a difficult time quitting. These adolescents
showing addictive tendencies began tanning at an early
age and were more likely to have substance abuse pro-
blems [31"). Although the ability of UVR to induce
addiction requires further investigation, when blinded
individuals are exposed to tanning beds with and without
UVR filters, 95% prefer tanning beds with unfiltered
UVR [32]. In addition to addiction, poor body image
may impact an adolescent's desire to tan. Many adoles-
cents wid1 body dysmorphic disorder frequent indoor
tanning salons 133]. As with many other maladaptive
behaviors, involvement in positive, healthful activities
often decreases the rate of addiction and substance abuse.
A recent study found adolescents attending a gym and
participating in recreational hobbies had lower rates of
tanning bed use, suggesting that an improved sense of
well being is an important factor to curb indoor tanning
[34].
To discourage indoor tanning, 29 states have passed
legislation limiting a minor's access ro indoor tanning
facilities. Most of these states bar children of a certain
age, usually around age 15, from using any indoor tanning
facility. Older adolescents, age 16- 18, require parental
permission before tanning. Despite these important regu-
lations, studies show many tanning facilities repeatedly
provide access to underage children, patrons with photo-
sensitive skin, and often fail to provide adequate eye
protection [35",36]. The US Food and Drugs Adminis-
tration (FDA) is charged with the duty of overseeing
regulation of equipment specifications, UV exposure,
eyewear and warnings. The Federal Trade Committee
is charged with moniroring the tanning industry's adver-
tising practices to ensure that there are no deceptions,
particularly regarding the supposed heal th benefits of
tanning [35"]. Parents must be made aware that laws
governing the use of indoor tanning are currently inef-
fective. They should be encouraged to monitor and
educate their children regarding the dangers of excessive
UVR exposure, as well as continue to support further
policing of the tanning industry by involving their local
government officials.
Sunless tanners offer a safe alternative to tanning. There
are numerous over-the-counter products containing
the skin darkening agent found in sunless tanners, dihy-
droxyacetone. This compound combines with amino
acids in the skin producing a brown/yellow/orange color
simulating a tan. In previous years, sunless tanners pro-
duced an artificial and streaky appearance. Newer formu-
lations provide natural appearing colors and have gained
wider acceptance [37]. Although sunless tanners are a
safe alternative to UVR tanning, a study from the US
[38"] found that a large percentage of individuals aged
18- 30 years who used sunless tanners were likely to
have sustained recent sunburn and used tanning beds.
A similar study from Australia supports these findings.
Hence, sunless tanners may be acting more as a supple-
Sun protection and tanning in children MacNeal and Dinulos 427
ment for those who are frequent 'tan seekers' rather than
a substitute for those who are considering UV exposure
for tanning purposes. Many tanning centers offer both
UVR tanning and sunless tanning booths enforcing the
association between sunless tanners and UV derived
tanning. Nonetheless, many patients replace UVR tans
with sunless tanners [39) and anecdotally, we have noted
that many patients who desire a tan do report replacing at
least some of there tanning sessions with sunless tanning
lotions.
Many sunless tanners contain sunscreen and carry sun
protection factor (SPF) labeling. For the SPF to be
effective, the sunless tanner would have to be applied
frequentl y and liberally like any other sunscreen. Many
users of sunless tanners equate the darker skin color with
sun protection and subject themselves to more UVR.
Proper labeling and education by healtbcare providers
should help clear this potentially dangerous confusion.
Sunscreens
When properly applied, sunscreens are effective in the
prevention of UVR-relatecl damage and skin cancer [40-
44]. Developed in the 1930s, sunscreens have historically
provided protection solely against UVB. Within the last
20 years, UVA (wavelengths 320- 400 nm) has become
better recognized for its ability to act as a carcinogen,
mediator of skin aging, and producer of photoallergy and
phorocoxicity (17,45,46]. Sunscreen manufacturers have
targeted the UVA spectrum in newer formulations.
These newer UVA blocking agents include avobenzone
(parsol 1789), benzophenones and dicamphor sulfonic
acid (Mexoryl SX). Typically, OVA blocking agents are
combined with a UVB (wavelengths 290- 320 nm) block-
ing agent such as octyl methoxycinnamate (octinoxate).
Sunscreen products that offer UVA protection in addition
to UVB are advertised as 'full-spectrum' sun protection.
l\tlany of these broad-spectrum sunscreens provide
improved UVR protection, but gaps continue to exist
at some wavelengths (47,48]. Unlike UVB, which has the
SPF assessment of efficacy, there is no currently wholly
agreed upon standard for assessing UVA protection.
Physical agents such as ti tanium dioxide and zinc oxide
have been gaining popularity due to the cosmetically
acceptable newer 'micronized' formulations that improve
on the white paste appearance that plagued these pro-
ducts in the past. Physical blockers provide good broad-
spectrum protection and act synergistically to block UVR
when combined. The FDA has historically been very
slow to adopt new sunscreen agents. US physicians
typically look to Europe and Canada for what new
sunscreens are on the horizon. One sunscreen available
in Europe is methoxyphenyltriazene (Tinosorb S).
This broad spectrum fil ter stabilizes avobenzone-
containing sunscreens and offers excellent broad spec-
trum protection [49).
428 Dermatology
Despite continuous improvements, all sunscreens suffer
from several major drawbacks. First, most users do not
apply the 2 mg/cm
2
amount of sunscreen used in the
laboratory testing used to determine SPF. This amount
corresponds to about 1 oz per full body application.
Second, the 15 min required drying time prior to UV
exposure is usually not observed. Finall y, real world
usage of sunscreen involves environmemal forces such
as water, sweat and clothing that act to remove sunscreen
from the skin. Regular use is important. Imermittent use,
even of hi gher SPF sunscreens, does not provide ade-
quate or consistent protecti on over time [SO]. Manufac-
turers have attempted to address these issues with
improved water resistance/proofing; however neither is
perfect. Water resistant is defined as maintenance of SPF
after 40 min of immersion in water and waterproof is
defined as effective after 80 min of immersion. Despite
large-scale education campaigns showing the benefits
of sunscreens [51], use among adolescents has remained
low and unchanged [sz ]. l'vlany adul ts and adol escents
rely on sunscreens as their primary or only form of sun
protection [3]. There is st ill much room for improvement
regarding our patients' compliance with proper sunscreen
use.
Conclusion
Children and their parents should continue to be the focus
of large-scale sun protection campaigns as there are still
clear knowledge deficits throughout all socio-economic
and educational classes. Studies show increasing knowl-
edge translates into improved UVR protection in early
childhood, however as children enter adolescence rates of
tanning increase and sunscreen use diminishes. Changing
adolescent high-risk behavior cont inues to be challenging
and will likely require intense and age-specific public
health initiatives as well as governmental regulations.
References and recommended reading
Papers of particular interest, published within the annual period of review, have
been highlighted as:
of special interest
of outstanding interest
Additional references related to this topic can also be found in the Current
Wortd Literature section in this issue (p. 523) .
Olson AL, Gaffney C, Starr P, eta/. SunSafe in the middle school years: a
communitywide intervention to change earlyadolescent sun protection.
Pediatrics 2007; 119:e247- e256.
2 Cigarette smoking among adults: Uni ted States, 2004. MMWR Morb Mortal
Wkly Rep 2005; 54:11 21 -1 124.
3 El Sayed F, Ammoury A, Nakhle F, et a/. Photoprotection in teenagers.
Photodermatol Photoimmunol Photomed 2006; 22:18- 21 .
4 Brown TT, Quain RD, Troxel AB, Gelfand JM. The epidemiology of sunburn
in the US population in 2003. J Am Acad Dermatol 2006; 55:577 -
583.
Thi s study examines risk factors for sunburn in the US. Like others, they found
high risk behaviors (drinking, tobacco) but they also found an association wi th
higher incomes and education.
5 Oh K, Willett WC, Wu K, eta/. Calcium and vitamin D intakes in relation to risk
of distal colorectal adenoma in women. Am J Epidemiol 2007; 165:1 178-
1186.
6 Knight JA, Lesosky M, Barnett H, et at. Vitamin D and reduced risk of breast
cancer: a populationbased case-control study. Cancer Epidemiol Biomarkers
Prev 2007; 16:422-429.
7 Freeman SE, Hacham H, Gange RW, el at. Wavelength dependence of
pyrimidine dimer formation in DNA of human skin irradiated in situ with
ultraviolet light. Proc Natl Acad Sci US A 1989; 86:5605 - 5609.
8 Parri sh JA, Jaenicke KF, Anderson RR. Erythema and melanogenesis action
spectra of normal human skin. Photochem Photobiol1982; 36:1 87-191.
9 Adams JS, Clemens TL, Parrish JA, Holick MF. VitaminD synthesis
and metabolism after ultraviolet irradiation of normal and vitaminDdeficient
subjects. N Eng I J Med 1982; 306:722 - 725.
10 Holick MF, Jenkins M. The UV advantage: new medical breakthroughs reveal
powerful health benefits from sun exposure and tanning. 1st ed Roseville,
California: Prima Pub; 2003.
11 Clemens TL, Adams JS, Henderson SL, Holick MF. Increased skin pigment
reduces the capacity of skin to synthesise vitamin 0
3
. Lancet 1982; 1:74-76.
12 Wolpowitz D, Gilchrest BA. The vitamin D questions: how much do you need
and how should you get i t? JAm Acad Dermatol 2006; 54:301 - 3 17.
13 Gilchrest BA. Sun protection and vitamin 0: three dimensions of obfuscation.
J Steroid Biochem Mol Bioi 2007; 103:655- 663.
14 Bell NH. Vitamin D metabolism, aging, and bone loss. J Clin Endocrinol Metab
1995; 80:1051.
15 Need AG, Morris HA, Horowi tz M, Nordin C. Effects of skin thickness, age,
body fat, and sunlight on serum 25hydroxyvitamin D. Am J Clin Nutr 1993;
58:882-885.
16 Levine JA, Sorace M, Spencer J, Siegel OM. The indoor UV tanning industry: a
review of skin cancer risk, health benefit claims, and regulation. J Am Acad
Dermatol 2005; 53:1038-1044.
17 Wang SO, Setlow R, Berwick M, eta/. Ultraviolet A and melanoma: a review.
J Am Acad Dermatol 2001; 44:837- 846.
18 Westerdahl J, lngvar C, Masback A, et a/. Risk of cutaneous malignant
melanoma in relation to use of sunbeds: further evidence for UVA carcino
genicity. Br J Cancer 2000; 82:1593 - 1599.
19 O' Riordan DL, Field AE, Geller AC, eta/. Frequent tanning bed use, weight
concerns, and other health risk behaviors in adolescent females (United States).
Cancer Causes Control 2006; 17:679-686.
This study supports the link between indoor tanning and other highrisk, health
threatening behaviors such as binge drinking, tobacco, recreational drug use and
vomiting and laxative use for weight control.
20 Demko CA, Borawski EA, Debanne SM, eta/. Use of indoor tanning facilities
by white adolescents in the United States. Arch Pediatr Adol esc Med 2003;
157:854- 860.
21 Lazovich D, Forster J, Sorensen G, eta/. Characteristics associated with use
or intention to use indoor tanning among adolescents. Arch Pediatr Adolesc
Med 2004; 158:918- 924.
22 Boldeman C, Jansson B, Nilsson B, Ullen H. Sunbed use in Swedish urban
adolescents related to behavioral characteristics. Prev Med 1997; 26:114-
119.
23 O'Loughlin J, DiFranza J, Tyndale RF, eta/. Nicotine-dependence symptoms
are associated with smoking frequency in adolescents. Am J Prev Med 2003;
25:219- 225.
24 Robinson ML, Bertin I, Moolchan ET. Tobacco smoking trajectory and as so
ciated ethnic differences among adol escent smokers seeking cessation
treatment. J Adolesc Health 2004; 35: 217-224.
25 DeWit OJ, Adlai EM, Offord DR, Ogborne AC. Age at first alcohol use: a risk
factor for the development of alcohol disorders. Am J Psychiatry 2000;
157:745- 750.
26 Grant BF, Dawson DA. Age at onset of alcohol use and its association
with DSMIV alcohol abuse and dependence: results from the National
Longitudinal Alcohol Epidemiologic Survey. J Subst Abuse 1997; 9:103-
110.
27 Grant BF, Stinson FS, Harford TC. Age at onset of alcohol use and DSMIV
alcohol abuse and dependence: a 1 2year followup. J Subst Abuse 2001;
13:493- 504.
28 Wagner FA, Anthony JC. From first drug use to drug dependence; devel op
mental periods of risk for dependence upon marijuana, cocaine, and alcohol.
Neuropsychopharmacology 2002; 26:479- 488.
29 Vega WA, AguilarGaxiola S, Andrade L, Bijl R, eta/. Prevalence and age of
onset for drug use in seven international sites: results from the international
consortium of psychiatric epidemiology. Drug Alcohol Depend 2002;
68:285-297.
30 Warner LA. White HR. Longitudinal effects of age at onset and first drinking
situations on problem drinking. Subst Use Misuse 2003; 38:1 983- 2016.
31 Zeller S, Lazovich D, Forster J, Widome R. Do adolescent indoor tanners
exhibit dependency? J Am Acad Dermatol 2006; 54:589-596.
These authors conducted a telephone survey of a cross section of Minnesota and
Boston adol escents. They found that many teens who tanned had difficulty
quitting.
32 Feldman SR, Liquori A, Kucenic M, et al. Ultraviolet exposure is a reinforcing
stimulus in frequent indoor tanners. J Am Acad Dermatol 2004; 51 :45-
51.
33 Phillips KA, Conroy M, Dufresne RG, et al. Tanning in body dysmorphic
disorder. Psychiatr Q 2006; 77: 129- 138.
34 DanoffBurg S, Mosher CE. Predictors of tanning salon use: behavioral
alternatives for enhancing appearance, relaxing and socializing. J Health
Psychoi 2006; 11:51 1- 518.
35 Forster JL, Lazovich D, Hickle A, et al. Compliance with restrictions on sale of
indoor tanning sessions to youth in Minnesota and Massachusetts. J Am Acad
Dermatot 2006; 55:962- 967.
An aggressive study that sent underage patrons into tanning salons to buy
sessions in areas where there are laws forbidding sale to minors. Most were
successful.
36 Dobbinson S, Wakefield M, Sambell N. Access to commercial indoor tanning
facilities by adults with highly sensitive skin and by underage youth: com
pliance tests at solarium centres in Melbourne, Australia. Eur J Cancer Prev
2006; 15:424 - 430.
37 Muizzuddin N, Marenus KD, Maes DH. Tonality of suntan vs sunl ess tanning
with dihydroxyacetone. Skin Res Techno! 2000; 6: 199-204.
38 Brooks K, Brooks D, Dajani Z, el al. Use of artificial tanning products among
young adults. J Am Acad Dermatol 2006; 54: 1060- 1066.
An interesting study that looks at the link between sunless tanners and increased
tanning bed use and recent sunburns.
39 Sheehan DJ, Lesher JL Jr. The effect of sunless tanning on behavior in the sun:
a pilot study. South Med J 2005; 98: 1192- 1195.
40 Thompson SC, Jolley D, Marks R. Reduction of solar keratoses by regular
sunscreen use. N Eng! J Med 1993; 329:114 7- 1151.
Sun protection and tanning in children MacNeal and Dinulos 429
41 Naylor MF, Boyd A, Smith OW, et a/. High sun protection factor sunscreens
in the suppression of actinic neoplasia. Arch Dermatol 1995; 131 :1 70-
175.
42 Boyd AS, Naylor M, Cameron GS, et al. The effects of chronic sunscreen use
on the histologic changes of dermatoheliosis. J Am Acad Dermatol 1995;
33:941 - 946.
43 Green A, Williams G, Neale R, el a/. Daily sunscreen application and
betacarotene supplementation in prevention of basalcell and squamous-cell
carcinomas of the skin: a randomised controlled trial. Lancet 1999;
354:723- 729.
44 Darlington S, Williams G, Neale R, et a/. A randomized controlled trial
to assess sunscreen application and beta carotene supplementation
in the prevention of solar keratoses. Arch Dermatol 2003; 139:451 - 455.
45 Allen JE. Drug-induced photosensiti vi ty. Clin Pharm 1993; 12:580- 587.
46 Gould JW, Mercurio MG, Elmets CA. Cutaneous photosensitivity diseases
induced by exogenous agents. JAm Acad Dermatol 1995; 33:551-573; quiz
574- 576.
47 O' Neill JJ. Effect of film irregularities on sunscreen efficacy. J Pharm Sci 1984;
73:888- 891 .
48 Ferrero L, Pissavini M, Marquerie S, Zastrow L. Efficiency of a continuous
height distribution model of sunscreen film geometry to predict a realistic sun
protection factor. J Cosme! Sci 2003; 54:463- 481 .
49 Tuchinda C, Lim HW, Osterwalder U, Rougier A. Novel emerging sunscreen
technologies. Dermatol Clin 2006; 24:105- 117.
50 Phillips TJ, Bhawan J, Yaar M, et a/. Effect of daily versus intermittent
sunscreen application on solar simulated UV radiation-induced skin response
in humans. J Am Acad Dermatoi 2000; 43:610- 618.
51 Cokkinides V, Weinstock M, Glanz K, el a/. Trends in sunburns, sun protection
practices, and attitudes toward sun exposure protection and tanning among
US adolescents, 1998- 2004. Pediatrics 2006; 118:853- 864.
52 Jones SE, Saraiya M. Sunscreen use among US high school students, 1999-
2003. J Sch Health 2006; 76:150- 153.
This analysis of CDC data showed that the minority of high-school students use
sunscreen and that this did not change from 1999 to 2003.
SYMPOSIUM ON SOLID TUMORS
Malignant Melanoma in the 21st Century, Part 1:
Epidemiology, Risk Factors, Screening, Prevention, and Diagnosis
SvETOMlR N. MARKovrc, MD, PHD; LoRl A. ERiCKSON, MD; RAVI D. RAo, MBBS; RoGER H. WEENIG, MD;
BARBARA A. PocKAJ, MD; ADITYA BARDIA, MD; CELrNE M. VAcHoN, PHD; STEVEN E. ScHILD, MD;
RoBERT R. McWILLIAMs, MD; JENNIFER L. HAND, MD; SusAN D. LAMAN, MD; LisA A. KorrscHADE, RN;
WILLIAM J. MAPLES, MD; MARK R. PITTELKow, MD; JosE S. PuLIDO, MD, MS, MPH, MBA;
J. DouGLAS CAMERON, MD; AND EDWARD T. CREAGAN, MD,
FOR THE MELANOMA STUDY GROUP OF THE MAYO CLINIC CANCER CENTER
Malignant melanoma is an aggressive, therapy-resistant malig-
nancy of melanocytes. The incidence of melanoma has been
steadily increasing worldwide, resulting i n an increasing public
health problem. Exposure to solar UV radiation, fair skin, dysplas-
tic nevi syndrome, and a fami ly history of melanoma are major risk
factors for melanoma development. The interactions between
genetic and environmental ri sk factors that promote melanoma-
genesis are currently the subject of ongoing research. Avoidance
of UV radiation and surveil lance of high-ri sk patient s have the
potential to reduce t he population burden of melanoma. Biopsies
of the primary tumor and sampling of draining lymph nodes are
required for optimal diagnosis and stagi ng. Several clinically rel-
evant pathologic subtypes have been identified and need to be
recognized. Therapy for early disease is predomi nantly surgical,
with a mi nor benefit noted with t he use of adjuvant therapy.
Management of systemic melanoma is a challenge because of a
paucity of active treatment modalities. In the first part of this 2-
part review, we discuss epidemiology, risk factors, screening, pre-
vention, and diagnosis of mal ignant melanoma. Part 2 (which will
appear in the April 2007 issue) will review melanoma staging,
prognosis, and treatment.
Mayo Clin Proc. 2007;82(3):364-380
T
he current spectrum of malignant melanoma includes
2 clinical extremes. At one end of the spectrum, thin
primary cutaneous melanoma is characterized by a rela-
tively w1iform treatment and a high cure rate. At the oppo-
site end, metastatic melanoma is characterized by no
proven effective therapy and poor outcomes. In recent
From the Division of Hematology (S.N.M.). Department of Oncology (S.N.M .
R.D.R., R.R.M . L.A.K., E.T.C.), Department of Immunology (S.N.M.), Depart
ment of Laboratory Medicine and Pathology (L.A. E.), Department of Dermatol-
ogy (R.H.W . J.L.H., M.R.P.). Department of Internal Medicine (A. B.), Depart-
ment of Health Sciences Research (C.M.V.). and Department of Ophthalmology
U.S.P.,j.D.C.), Mayo Clinic College of Medicine, Rochester. Minn; Department
of Surgery (B.A.P.). Department of Radiation Oncology (S. E.S.). and Depart-
ment of Dermatology (S.D.L.), Mayo Clinic College of Medicine, Scottsdale,
Ariz; and Division of Hematology/Oncology (W.J .M.). Mayo Clinic College of
Medicine, jacksonville, Fla. A complete list of the current members of the
Melanoma Study Group of the Mayo Clinic Cancer Center appears at the end of
this article.
Address correspondence to Svetomir N. Markovic, MD, PhD. Division of
Hematology, Mayo Clinic College of Medicine. 200 First St sw. Rochester. MN
55905. Individual reprints of this article and a bound reprint of the entire
Symposium on Solid Tumors will be available for purchase from our Web site
www.mayoclinicproceedings.com.
2007 Mayo Foundation for Medical Education and Research
decades, a heightened awareness of melanoma has led to an
increased rate of diagnosis of early-stage disease. Notwith-
standing the fact that treatment of advanced melanoma still
remains in the realm of expetimental and early-phase clini-
cal trials, tremendous research efforts in melanocyte biol-
ogy, melanoma genetics, and tumor immunology are pro-
viding hope for a cure.
Although more than 95% of tumors are found in the
skin, melanoma is not exclusively a skin cancer. Sites of
primary extracutaneous melanoma include ocular, mu-
cosal, gastrointestinal, genitourinary, leptomeninges, and
lymph nodes (melanoma of unknown ptimary cancer).
With the exception of neuroectodermally derived melano-
cytes that give rise to the retinal pigment epithelium, mel-
anocytes originate from neural crest cells. During the early
weeks of gestation, melanocyte precursors differentiate
and migrate from the neural crest to numerous tissues
(including the skin).
The function of melanocytes is best studied and defined
in the skin. Within the basal layer of the epidermis, each
melanocyte develops an intimate relationship with many
keratinocytes, sending out slender dendritic processes that
transfer packets of melanin pigment (melanosomes) to
keratinocytes. The melanin granules oti ent in an umbrella-
like fashion over the keratinocyte nucleus parallel to the
skin surface. Melanin absorbs ultraviolet radiation (UVR)
and possesses potent antioxidant properties capable of neu-
tralizing UVR-generated free radicals. Paradoxically, mel-
anocytes are injured and transformed by the same agent
that melanocytes are programmed to defend against, UVR.
Moreover, the cancers that melanocytes help prevent
(squamous cell and basal cell carcinoma) are less lethal
than the cancer that melanocytes become (melanoma).
Melanoma is rare among deeply pigmented ethnic
groups relative to indi viduals of Northern European decent.
The type of melanin pigment (eumelanin vs pheomelanin),
as well as the number, size, and density of melanosomes,
determines skin pigmentation. Anthropological and ge-
netic studies trace human pigment dilution to the migra-
364 Mayo Clin Proc. March 2007;82(3):364-380 www.mayoclinicproceedings.com
f-or persona J ~ o ; ; ; 1\aa.>S I efJrvdLC on.y With rJt .nt.>Sa0o1 frolli Mayo L 'mic Proceedmgs.
tion of early human groups to northern latitudes. Mutations
of genes that govem skin pigmentation and/or are associ-
ated with increased melanoma risk were likely propagated
by selective pressures (sexual selection) and/or a founder
effect.
The causal relationship of UV exposure to melanoma is
complex, and acute, intense, intermittent sun exposure in
youth equates to higher risk of melanoma. The 1isk of
melanoma also increases with proximity to the equator.
The intensity of UVR increases considerably at midday,
around which time (10 AM to 3 PM) sun avoidance is
strongly encouraged. Fueled in part by public education
campaigns, stories of melanoma diagnoses among celebri-
ties, and marketing efforts by the manufacturers of sun-
screen and photoprotective clothing, melanoma awareness
and sun avoidance behaviors have substantially increased.
However, sunbed use, low sun protection factor sun-
screens, and sun-seeking behaviors remain popular among
young people.
Every clinician encounters patients at 1isk for mela-
noma. Additionally, depending on the tumor location and
stage, melanoma is a disease that requires the unique exper-
tise and collaboration of medical, surgical, and pathology
specialists. This review is intended to provide clinicians with
a comprehensive, multi specialty overview of melanoma.
EPIDEMIOLOGY
Melanoma is the fifth most common cancer in men and the
sixth most common cancer in womenL
2
in the United
States. An estimated 62, 190 new cases of melanoma were
diagnosed in the United States during 2006.
12
According to
data from Surveillance, Epidemiology, and End Results
(National Cancer Institute), in 2000 approximately 629,822
people (304,097 men and 325,725 women) alive had a
history of melanoma.
3
The annual age-adjusted incidence
of and mortality from melanoma in the United States is
18.3 per 100,000 persons per year and 2.7 per 100,000
persons per year, respectively? Internationally, the inci-
dence rates vary 100-fold among different populations,
with Australia having the highest rates worldwide; in
Queensland, Australia, the cumulative incidence in the
population older than 50 years is 1 in 19 for men and 1 in 25
for women.
4
The incidence of malignant melanoma is increasing rap-
idly worldwide, including in countries with histmically low
incidence rates. This increase is occurring at a faster rate
than for any other neoplasm, with the exception of lung
cancer in women.
3
In the United States, the incidence of
malignant melanoma from 1973 to 2002 increased by
270%.
3
Currently, 1 in 63 Americans will develop mela-
noma during their lifetime; histmically, the 1isk was 1 in
MALIGNANT MELANOMA IN THE 21ST CENTURY
1500 in 1935 and 1 in 250 in 1980.
35
6
Parallel with this
increasing incidence is an increase in melanoma-related
mortality, albeit to a lower degree. In the United States, the
mortality rate from malignant melanoma increased by
1.4% every year between 1977 and 1990. Since 1990, the
mortality rate has shown a small downward trend and
decreased by 0.3% per year from 1990 to 2002. Cunently,
1 person dies each hour from metastatic melanoma.
3
The
annual direct cost of treating newly diagnosed melanoma in
the United States was estimated to be $563 million in
1997_7-8
Melanoma is notorious for affecting young and middle-
aged people, unlike most other solid tumors, which mainly
affect older adults. The median age at diagnosis of mela-
noma is 57 years, and the median age at death is 67 years.
3
The incidence of melanoma increases linearly after the age
of 15 years until the age of 50 years and then slows,
especially in females.
3
Approximately half the incidence is
in people between the ages of 35 and 65 years, with ap-
proximately 80% occuning in those in the age range of 20
to 74 years.
3
The age distribution is likely to show a left-
ward shift with a trend toward an increase in the incidence
of melanoma among children, as well as due to an increase
in skin biopsy rates.
9
10
Males are approximately 1.5 times
more likely to develop melanoma than females.
3
The distri-
bution of favored sites of occurrence of melanoma is sex
dependent: the most common areas are the back for men
and the arms and legs for women.
1
1.1
2
White populations have an approximately 10-folcl
greater risk of developing cutaneous melanoma than black,
Asian, or Hispanic populations? This presumably relates to
the higher sensitivity of white skin to sun exposure. How-
ever, both white and African American populations have a
similar 1isk of developing plantar melanoma.
13
Noncutane-
ous melanomas (eg, mucosal) are more common in non-
white populations.
RISK FACTORS
Several factors have been hypothesized to be responsible
for the worldwide increase in the incidence of melanoma.
Of these, the major factors that are responsible for most of
this increase are increased exposure to UVR, predomi-
nantly due to the depletion of the ozone layer, allowing
penetration of UV rays into the atmosphere but possibly
also due to behavioral change (such as increased use of sun
or tanning beds), and increased surveillance.
10
14
These fac-
tors are swnmarized in Table 1 and reviewed subsequently.
ENVIRONMENTAL FACTORS
Sunlight. A large body of evidence supports the role of
solar UVR exposure as the most important environmental
Mayo Clin Proc. March 2007;82(3):364-380 www.mayoclinicproceedings.com 365
For personal .... se. Mass reprodLce c.,r.; WI t pe. n .. ssh.H. Proceeding,:,.
MAUGNANT MELANOMA IN THE 21ST CENTURY
TABLE 1. Summary of Risk Factors for Malignant Melanoma by Evidence
Strong evidence Weak evidence
Inconclusive or
no increased risk
Environmental and lifestyle factors
Exposure to sunlight
Geographical location
Tanning beds. sunlamps
Obesity
Exogenous hormones
Alcohol , smoking
Coffee industrial occupation
Vitamins A and E
Host factors
Number of nevi
Dysplastic nevi
Family history of melanoma
lnmlUnOsuppression
Personal history of skin cancer
Higher socioeconomic status
Brown hair
Pregnancy
Male sex
Sun sensitivity or inability to tan
Blue or green eyes; blond or red hair
Endogenous hormones
(age at menarche, parity)
tisk factor for developing malignant melanoma. Multiple
studies have shown that intermittent sun exposure, assessed
indirectly by history of sunburns, appears to be a major
determinant of risk for melanoma (in contrast to non-
melanomatous skin cancers, which are linked more to cu-
mulative sun exposure).
15
This etiological link is further
supported by the finding of higher melanoma incidence in
populations that li ve in lower latitudes, higher incidences
in white populations, lower incidences among dark-
skinned populations, and data from genetic and migration
smdies.
16
21
The tisk of melanoma developing in people
who have had sunburns is double that in people who have
never had a sunburn. Moreover, the age at which sunbums
occur appears to be imp011ant; sunburns in childhood are
associated with the highest risk.
15
22
Further evidence for
the role of UVR comes from srudies of immigrant popula-
tions in Australia, which have shown that the risk of mela-
noma is proportional to the length of stay and inversely
to age at arrival.
4
Likewise, further support for the role
of UVR as a risk factor for melanoma comes from clin-
ical evidence. Patients with xeroderma pigmentosum (a
genetic disorder characterized by deficient DNA repair
mechanisms and a corresponding hypersensitivity to
UVR damage) have a substantially higher risk of develop-
ing melanoma, particularly in sun-exposed skin.
18
20
23
24
The use of psoralen-UV -A radiation photochemotherapy
for psoriasis has also been associated with an increased
1isk of melanoma.
25
Other Sources of UVR. A relationship between in-
creased risk of melanoma from nonsolar sources of UVR
(eg, fluorescent light, sunlamps, sunbeds and tanning beds)
has been postulated, but data have been inconclusive.
1
L
26
27
A recent meta-analysis of 12 case-control studies and 1
cohort study suggested that sunbed and sunlamp exposure
might lead to a slightly higher risk of melanoma (odds
ratio, 1.25; 95% confidence interval, 1.05-1.49)_27 The risk
was higher with longer duration of exposure and earlier age
at exposure.
Protection From UVR Exposure. The relationship es-
tablished between UVR exposure and melanoma risk has
prompted researchers to hypothesize that sunscreen use
may reduce the risk of melanoma developing. However,
studies to prove this hypothesis have been inconclusive.
28
30
Proper use of sunscreens with broad-spectrum sun protec-
tion factor 30 has been shown to reduce the number of
acquired nevi that develop in children by 30% to 40%_31
Because acquired nevi are markers for sun exposure (and
thereby for increased risk of developing melanoma), sun-
screen use may have a role in melanoma prevention. How-
ever, other studies found that by protecting against sun-
burns, sunscreen use results in an increase in the actual
duration of intentional (ie, recreational) sunlight exposure
and may increase the total exposure to UVR.
32
36
Therefore,
reliance on sunscreen as a sole method of sun protection is
not recommended, and other methods (eg, long-sleeved
clothing) should be used in conjunction with sunscreen.
Occupation. Because exposure to sunlight is a major
risk factor for melanoma, a concem has been that people
occupationally exposed to sun, such as farmers, would
have a higher risk of developing melanoma. However,
although farmers are indeed at a higher risk of developing
skin cancers such as squamous cell carcinoma, they appear
to have no increased risk for melanoma.
37
38
Swedish airline
pilots, compared with the general male Swedish popula-
tion, have been identified to have an increased incidence of
melanoma. The same study reported that Swedish military
pilots did not have an increased risk of melanoma but only
of nonmelanoma skin cancer.
39
These findings support the
hypothesis that intermittent (rather than cumulative) sun
exposure increases the risk of melanoma. Occupational
exposure to ionizing radiation, vinyl chloride, polychlori-
nated biphenyls, and petrochemicals has been linked to a
possible increase in the risk of melanoma; however, the
risk attributable to these factors appears to be small. More-
over, this link has not been shown consistently in different
studies.
f-or persona. J ~ o ; ; ; 1\aa.>S I efJrvdLC on.y wrth rJt.,nt.;>s.o.1 frorr, Mayo L.'tnic Proceedmgs.
Reproductive Factors and Oral Contraceptive Use.
Estrogen and progesterone have been postulated to in-
crease the risk of melanoma by stimulating melanocyte

Furthermore, patients with breast cancer
have a slightly higher risk (odds ratio, 1.4) of developing
melanoma,
42
which has led to speculation that the develop-
ment to melanoma is influenced by the hormonal milieu. A
Jew early studies indicated that oral contraceptive pill or
hormone replacement therapy use increased the Iisk of
melanoma.
43
However, more recent studies and a pooled
analysis failed to validate these findings.
4445
A link be-
tween an increased risk of melanoma and several reproduc-
tive factors (eg, age at menarche, menopausal status, age at
menopause, number of live births) has been noted, again
suggesting the influence of estrogen on melanogenesis. A
recent pooled analysis of 10 case-control studies suggested
that women with both earlier age at first birth ( <20 years)
and higher parity live births) had a lower risk than
women with later age at first birth and lower parity (odds
ratio, 0.33).
4446
Presumably, endogenous rather than exog-
enous honnones act as risk factors. Of note, pregnancy
does not seem to significantly affect the risk of malignant
melanoma.
47
Anthropometric, Dietary, and Lifestyle Factors.
Obesity has been postulated to increase the tisk of mela-
noma, plausibly because of a larger body surface area
exposed to sun or differential hormonal profiles.
454 8
Epide-
miological studies evaluating the role of dietary vitamins
have been inconclusive.
48
.
50
Similarly, the relationship with
other dietary factors, including intake of fat, vitamin E,
alcohol, coffee, and smoking, has been inconsistent.
48
58
Hosr FACTORS
Melanocytic Nevi. Melanocytic nevi are benign accu-
mulations of melanocytes or nevus cells and may be con-
genital or acquired. The risk of melanoma varies on the
basis of type, size, number, and location of nevi. In ap-
proximately 25% of cases, melanoma occurs in conjunc-
tion with a preexisting nevus. 5
9
The risk of melanoma has been directly correlated to the
total number of benign nevi (both dysplastic and non-
dysplastic) on the body. The Iisk is approximately 1.5-
times higher in people with 11 to 25 nevi (compared with
s10 nevi) and appears to be doubled with every increase of
25 nevi.
411
16
6
o.us Similarly, larger nevi (particularly >5 mm)
are associated with a higher risk of melanoma.
66
68
Giant nevi
(>20 em) are associated with a significantly higher risk of
melanoma.
69
70
Dysplastic nevi (ie, melanocytic nevi with cytological
atypia) are associated with an increased risk of mela-
noma.6 .. 6271 Nonfamilial melanoma may occur in the setting
of a preexisting dysplastic nevus in up to 29% to 49% of
cases.
65
In a person with a family or a personal history of
melanoma, the presence of dysplastic nevi is a marker for a
significantly increased Iisk of melanoma.
72
Moreover, the
presence of dysplastic nevi is a Iisk factor for the develop-
ment of multiple primary melanomas. Dysplastic nevi often
cluster in families, particularly in families with melanoma.
73
81
A familial syndrome variously referred to as dysplastic ne-
vus syndrome, atypical mole syndrome, or familial atypical
multiple mole and melanoma syndrome
81
has been recog-
nized. Patients with this syndrome have almost a 100-times
higher Iisk of developing melanoma, and approximately
50% of these patients develop melanoma by the age of 50
years.
75
Melanomas that develop in the setting of previous nevi
are more likely to be on the trunk, occur in younger pa-
tients, and belong to the superficial spreading variety.
5982
Family History. A family history of melanoma is a
strong tisk factor for melanoma. Patients with a history of
melanoma in a first-degree relative have approximately a 2-
times higher Iisk of developing melanoma than those with-
out a family history; the risk increases in the presence of
other Iisk factors.
83
84
Familial melanoma is thought to ac-
count for 10% of all melanoma cases.
85
Presence of a familial
melanoma syndrome should be considered in patients with a
family history of pancreatic cancer or astrocytoma. Some
families with inherited melanoma demonstrate a clear pat-
tern of autosomal dominant inhetitance with multiple family
members affected in more than l generation. Mutations in
CDKN2A (or pl6) are the most common genetic abnOtmali-
ties found in d1ese fan1ilies. A second mutation, CDK4, has
been found much more rarely.
86
87
Features in a patient with
melanoma that indicate an underlying generic predisposition
include occurrence at a younger age ( <40 years ),
88
92
multiple
primary melanomas, or a history of precursor lesions such as
dysplastic nevi. Patients with a genetic predisposition to
melanoma are more likely to have tumors that are superfi-
cially invasive and have a better prognosis.
93
94
Additionally,
melanoma is known to cluster in families with family cancer
syndromes such as familial retinoblastoma, Li-Fraumeni
cancer syndrome, and Lynch syndrome type II.
A genetic predisposition to melanoma may also occur in
a patient without a suggestive family history. Such patients
may have a new mutation, including an alteration in the
CDKN2A gene or CDK4 gene. Once acquired, this predis-
position can be passed on to offspring in an autosomal
dominant fashion.
Immunosuppression. Immunosuppression has been
associated with a higher 1isk of melanoma. This has been
noted in patients with acquired immunodeficiency syn-
drome, those with hematological malignancies, and those
receiving immunosuppressive agents after solid organ
transplantation.
95
"
100
For personal .... se. Mass reprodLce c.,r.; WI t pe. n .. ssh.H. Proceeding.;;,.
Phenotypic Characteristics and Variations in Mel-
anocortin-1 Receptor. Certain phenotypic charactetistics,
such as eye color (blue or green), hair color (red > blond >
brown compared with black), presence of freckles, sun sensi-
tivity, and an inability to tan, have been associated with
approximately a doubling of the risk of developing mela-
noma.22.4571. 101102 One possible mechanism for the increased
1isk relates to va1iation in melanocortin-1 receptor. Variation
in melanocortin-1 receptor alleles has been found to result in
the development of both sporadic cutaneous melanomas and
an increased 1isk of melanoma among those predisposed to
familial melanoma. Studies have found that red hair, pres-
ence of freckles, and sun sensitivity are associated with
specific genetic variations in melanocortin-1 receptor and
possibly account to some extent for the increased tisk seen
with the phenotype.
Other Factors. Higher socioeconomic status has been
associated with a higher tisk of melanoma. A personal his-
tory of melanoma (and other skin cancers) can increase the
1isk of subsequent melanoma developing. However, both of
these associations appear to be confounded by an increased
exposure to sunlight and better surveillance.
15
93
94
10
3-
112
RISK PREDICTION MODELS
Risk prediction models have been developed to estimate an
individual's lifetime risk of melanoma.
83
84113
Risk factors
that have been consistently demonstrated to increase the 1isk
of melanoma include family history of melanoma, higher
number of nevi, history of severe sunburn (>3 before the age
of 20 years), marked freckling on the upper back, and light
hair color. When compared with the 1isk in the general pop-
ulation, the presence of 1 or 2 of these risk factors is associ-
ated with a 2- to 4-times higher tisk, and occurrence of 3 or
more factors results in approximately a 20-fold risk.
83
84
113
SCREENING AND PREVENTION
Prevention and public health measures are essential to de-
crease the risk of melanoma. Melanoma prevention is well
divided into primary, secondary, and tertiary prevention.
Primary prevention (prevention of occurrence) of mela-
noma entails reduction of known risk factors in high-risk
populations. Reducing UV exposure, especially intense,
intermittent sun exposure, is the most important modifiable
behavior for melanoma prevention. Reducing UV exposure
would include avoiding excessive sun exposure (midday
sun), coveri ng skin with clothing, weruing broad-btimmed
hats, using detergent that increases the photoprotective
abi lity of one's clothing, and applying sunscreen.
11 4
These
measures should also be emphasized in individuals with a
personal or family history of melanoma. Sunburn avoid-
ance in childhood is paramount.
115
Educating the popula-
tion and especially parents about sun protection, risk fac-
tors for skin cancer, and the skin self-examination (ABCD
of melanoma) is essential.
116
117
Photoprotective measures
(sun-safe behavior) should also be emphasized for all indi-
viduals, especially those with a personal or family history
of melanoma. The most successful exrunple of such popu-
lation-based cancer education programs is the effort of
the Anti-Cancer Council of Victoria in Australia. The
council has been organizing sun protection programs
(Slip! Slap! Slop! and SunSmart) for more than 20 years.
These programs have resulted in marked reductions in sun
exposure and have helped change society's approach to
the sun.
118
Also noteworthy are primary chemoprevention
efforts using statins and fibrates in patients at risk for
melanoma. As yet, there does not appear to be conclusive
evidence of the benefit of such interventions.
119
-
121
Secondary prevention of melanoma is accomplished by
diagnosis and treatment of early-stage (highly curable)
melanoma. People at high risk for development of mela-
noma should be identified and evaluated. Screening of
high-risk patients is cost-effective and likely to be associ-
ated with an improved survival.
122
Individuals with xero-
derma pigmentosum, giant congenital nevi, immunosup-
pression, familial atypical multiple mole and melanoma
syndrome, unusual-apperuing nevi, numerous (>50) nevi,
changing nevi, and a family history of melanoma and men
older than 50 years should receive complete baseline and
petiodic follow-up skin examinations by a physician.
123
125
Additionally, such people should be encouraged to conduct
skin self-examinations. Patients with a history of mela-
noma should be educated on the need for continued surveil-
lance for subsequent additional ptimary melanomas and for
recurrence and metastasis.
126
Melanoma screening includes
complete skin exami nation as part of a general medical
examination by primary care physicians, clming evalua-
tions for other skin problems by a dermatologist, and com-
munity-based screening programs. Each of the foregoing
screening methods has afforded increased rates of melanoma
detection.
127
-
129
Both the value and the need for sustained
melanoma education were demonstrated by a nationwide
melanoma awareness campaign in Australia, resulting in
reduction of the median melanoma thickness overall and a
higher percentage of thin ( <0.75 mm) melanomas diagnosed
eluting the crunpaign year. After the campaign, both mea-
surements rebounded to precampaign year values. Tertiru-y
prevention of melanoma involves limiting morbidity and
extending survival in patients with advanced disease.
DIAGNOSIS OF MELANOMA
The diagnostic clinical features of melanoma range from
subtl e to obvious characteristics. A well-established guide
f-or persona J ~ o ; ; ; 1\aa->S I efJrvdLC on.y With rJt .nt.;>Sa0o1 frolli Mayo L 'mic Proceedmgs.
FIGURE 1. Primary cutaneous melanoma demonstrating pigmented
variation.
to examine and interpret pigmented lesions for both health
care professionals and patients has been the ABCDE acro-
nym. This list of features, which is easy to memorize and
use, stands for Asymmetry, Border irregularity, Color
variation, Diameter (>6 mm), and the recently added E, for
Evolving (or changing), has been widely popularized dur-
ing the past 2 decades. Evolving encompasses any signifi-
cant change in size, shape, surface (raised, bleeding, crust-
ing), shades of color, or symptoms (itching, tenderness)
and was only recently added to the list since most patients
with melanoma have been found to observe these com-
ponent changes occmTing within suspicious pigmented
lesions.
130
Va1ious assisti ve optical devices for a more detailed
examination of the skin and, in particular, pigmented le-
sions are becoming part of routine clinical practice. These
devices include high-resolution optical handheld devices
that have been designated dermoscopes (or dermatoscopes
orepiluminescent microscopes). In addition, portable scan-
ning units using visible, infrared, and UV sources create
images or spectroscopic outputs that can be electronically
captured, archived, retrieved, and analyzed. A number of
these devices are continually being developed, refined, and
tested in clinical studies aimed at improving the diagnostic
accuracy and sensitivity of melanoma detection. Several
excellent reviews are recommended for further reading.
131
.u
4
During evaluation of suspicious pigmented lesions, cli-
nicians must heed the observation of the patient or the
patient's fami ly member of a changing or new pigmented
skin lesion. Pigment variation (Figure 1), an irregular or ill-
defined border, asymmetry in color or shape, ulceration,
development of a nodule within a preexisting lesion (Fig-
ure 2), and pigment loss (Figure 3) are clinical clues for
melanoma. However, a definitive diagnosis or an exclusion
of melanoma requires biopsy and experienced pathological
FIGURE 2. Primary cutaneous melanoma demonstrating irregularly
shaped lesion.
review of the specimen. An established diagnosis of malig-
nant melanoma can be made only after histopathologic
analysis. Even then, the diagnosis can be challenging. Ma-
lignant melanoma must be distinguished from conventional
melanocytic nevi, proliferation nodules in congenital nevi,
atypical or dysplastic nevi, and unusual nevus variants such
as Spitz nevi, pigmented spindle cell nevi, deep penetrating
nevi, plexiform spindle cell nevi, clonal nevi, genital nevi,
acral nevi , flexural nevi, cellular blue nevi, and recurrent
melanocytic nevi, among many others. Unusual morpho-
logic variants of malignant melanoma must be recognized
to prevent misdiagnosis as an epithelial or mesenchymal
neoplasm. The histologic evaluation of the tumor also pro-
vides prognostic infonnation needed by the clinician and
surgeon to formulate the treatment plan.
Ciiteria for the histologic diagnosis of cutaneous malig-
nant melanoma are based on architectural and cytologic
features and need to be interpreted in the context of the
FIGURE 3. Primary cutaneous melanoma demonstrating focal pig
ment loss.
clinical situation. A junctional melanocytic proliferation
that shows areas of pagetoid spread from an acral site of a
pediatric patient would likely be interpreted differently
than a similar proliferation on sun-damaged skin of the face
of an elderly patient. Additionally, partial or incisional
biopsies of larger lesions are well-known pitfalls in the
diagnosis of malignant melanoma. There is no absolute
criterion for the diagnosis of malignant melanoma. A con-
stellation of architectural and cytologic features is evalu-
ated along with clinical infom1ation.
Architectural features of malignancy include large size,
asymmetry, poor circumscliption, pagetoid spread, con-
fluence of growth, effacement of the rete ridges, scalloping
of the clem1al-epicleimal junction, marked cellularity, sheet-
like growth pattern, effacement of underlying dermal archi-
tecture, lack of architectural and cytologic maturation with
depth, and perineural invasion, among many others. Cyto-
logic features of malignancy include enlarged nuclei,
prominent nucleoli, thick and irregular nuclear membranes,
abnormal cytoplasmic melanization, dermal mitotic activ-
ity, and atypical mitotic figures.
Complicating the diagnosis of melanoma is the under-
standing that many of these histologic features can be seen
in benign melanocytic nevi. Generally, the larger the size of
a lesion the more suspicious for melanoma; however, con-
genital melanocytic nevi can be large and melanomas can
be small. Small melanomas, measuring only 2 to 3 mm in
diameter, are well documented.
135
136
Pagetoid spread is
thought to be one of the most important features for the
diagnosis of malignant melanoma; however, pagetoid
spread is seen in most nevi of palms and soles, recurrent
nevi, and genital nevi and is seen in a proportion of Spitz
nevi, pigmented spindle cell nevi, and nevi of infancy and
childhood.
137
Confluence of growth along the deimal-epi-
dermal junction is often seen in malignant melanoma, but a
confluence of junctional theques (aggregations of nevus
cells) can also be seen in genital nevi.
138
Mitotic figures,
particularly deep and atypical mitotic figures, are often
seen in malignant melanoma, but mitotic figures can also
be seen in benign melanocytic nevi. Nevoid melanomas
"mimic ordinary compound or intradermal melanocytic
nevi when the melanoma cells are small, or Spitz's nevi
when the cells are large." ..
19
Thus, the diagnosis of malig-
nant melanoma requires careful evaluation of numerous
histologic features in each patient.
Conventional melanomas are classified as superficial
spreading, nodular, lentigo maligna, and acral lentiginous.
However, this classification may be difficult to apply in
small incisional biopsy specimens. Malignant melanomas
can show such heterogeneity that they may not always be
classifiable.
140
141
A number of studies suggest that different
types of melanoma (eg, nodular, acral lentiginous) may
have a different prognosis,l
42
-
145
but the importance of the
histologic type of melanoma may decrease with multivari-
ate analysis.
146
147
This classification is not included in the
American Joint Committee on Cancer staging system but is
generally included on pathology reports.
Superficial spreading is the most common type of ma-
lignant melanoma, accounting for approximately 70% of
cases. Superficial spreading melanomas most often occur
on the back of the legs of women and on the backs of
men.
148
These tumors are commonly found on sun-exposed
skin, particularly areas of intermittent sun exposure. Such
tumors usually show prominent pagetoid spread and are
composed of epithelioid melanoma cells.
149
By convention,
superficial spreading melanomas extend 3 rete ridges be-
yond the dermal component.
150
Superficial spreading mela-
nomas may arise de novo or in association with a nevus.
Nodular melanoma accounts for 5% of melanomas and
most often occurs on the trunk and limbs of patients in the
fifth or sixth decade of life, with males more commonly
affected than females.
146
151
Nodular melanomas are often
ulcerated and when amelanotic may be mistaken for a
vascular neoplasm.
152
Nodular melanomas do not have a
radial growth phase; they have only a vertical growth
phase. Nodular melanomas are not frequently associated
with regression or an associated nevus.
149
Polypoid mela-
nomas are considered a variant of nodular melanoma.
153
Acral lentiginous melanoma is uncommon, accounting
for 5% of melanomas among white people. Although acral
lentiginous melanoma is uncommon in all ethnic groups, it
is the most common type of melanoma an1ong Asian pa-
tients, Hispanic patients, and patients of African de-
scent. 14J.154-157 The overall incidence of melanoma in these
ethnic groups is low compared with that in white patients.
Acral lentiginous melanoma occurs on glabrous skin and
adjacent skin of the digits, palms, and soles.
143
151
.1
58
-
165
On
the soles, acral lentiginous melanomas often involve
heels.
141159
Digit melanomas usually involve the nail bed of
the great toe or thumb and more commonly affect elderly
people, with a female predominance.
154
157
159
166
173
Superfi-
cial spreading and nodular melanomas may also involve
acral sites, including the nail bed.
167
172
In a study of 100
subungual melanomas, approximately half were acral
lentiginous type.
Lentigo maligna accounts for 4% to 15% of cutaneous
melanoma and correlates with long-term sun exposure and
increasing age.'
46
149
174
-
177
Additionally, lentigo maligna is
known as Hutchinson melanotic freckle
178
and precancer-
ous melanosis of Dubreuil h.
179
Lentigo maligna melano-
mas are composed of single melanocyt ic cells and
dyshesive nests along an atrophic dermal-epidermal junc-
tion. The tumor cells often show extension clown into hair
follicles. These tumors usually mise in the skin, showing
f-or persona. J ~ o ; ; ; 1\aa->S I efJrvdLC on.y wrth rJt.,nt.;>s.o.1 frorr. Mayo L.'tnic Proceedmgs.
marked solar elastosis and other feamres of long-term sun
damage. Among the types of in situ melanoma, lentigo
maligna can be difficult to distinguish from melanocytic
hyperplasia in sun-damaged skin.
180
181
This lesion may
evolve for decades before invading into the papillary der-
mis.174 To distinguish between lentigo maligna and malig-
nant melanoma in situ, the term lentigo maligna has been
proposed.
175
182
In the precursor lesion lentigo maligna,
there is an increase in atypical melanocytic cells singly
disposed along the dermal -epidermal junction, whereas in
malignant melanoma in situ, the lentigo maligna type
shows nesting along the j unction, confluence, and pagetoid
spread. Progression to invasive melanoma from malignant
melanoma in situ of the lentigo maligna type is thought to
be similar to other forms of melanoma in situ, whereas
progression from lentigo maligna, as a precursor lesion, is
hypothesized to occur over a much longer time.
175
182
UNUSUAL VARIANTS OF MALIGNANT MELANOMA
Unusual vatiants of melanoma (Table 2 and Figure 4) are
rare and can be confused with epithelial or mesenchymal
neoplasms. These unusual variants of malignant melanoma
generally behave similarly to conventional types of mela-
noma when prognostic factors such as depth of invasion are
taken into account. The most important aspect of these
variants is knowledge of their existence to avoid misdiag-
nosis as another tumor or as a scar in the case of desmoplas-
tic melanoma.
DESMOPLASTIC MELANOMA
Desmoplastic malignant melanoma often arises on the head
and neck
183
-
186
but can occur on a variety of cutaneous and
mucosal areas, including the conjunctiva, gingiva, genitalia,
and perianal areas.
183
187
304
312
-
314
Desmoplastic melanoma oc-
curs in older individuals from the sixth to eighth decades of
life and is slightly more common in men. 161.162,183.184.18&-193
Clinically, desmoplastic melanoma may be amelanotic and
present as an erythematous or pale or flesh-colored nodule
or plaque arising in sun-damaged skin or with pigmentation
overlying a dermal nodule. Some desmoplastic melanomas
arise in association with a lentigo maligna melanoma, but
they may also arise in association with superficial spread-
ing melanoma, mucosal melanoma, or acral lentiginous
melanoma.
Desmoplastic melanomas are positive for S 100, but
many commonly used markers of melanocytic differentia-
tion, including HMB45, Mart l, and Melan A, are usually
negative.
185
193
315
Differentiating desmoplastic malignant
melanoma from scar tissue in reexcised specimens can be
particularly difficult because S 1 00-positive cells can also
be seen in dermal scars. Desmoplastic melanomas often
TABLE 2. Unusual Vari ants of Melanoma
Melanoma variant
Desmoplastic melanoma
Mucosal melanoma
Malignant melanoma arising in
blue nevus (malignant blue nevus)
Nevoid melanoma
Minimal deviation melanoma
Small cell melanoma
Spitzoid melanoma
Dermal melanoma
Amelanotic melanoma
Myxoid melanoma
Signet ring cell melanoma
Balloon eel\ melanoma
Rhabdoid melanoma
Pigment-synthesizing animal or
equine melanoma
Osteoid melanoma
Chondroid and cartil aginous melanoma
Basomelanocytic tumor
References
161, 162, 183-213
214-219
220-228
l39, 162. 229-233
162, 210, 234-257
141 ' 234, 258-261
LSI , 162, 237, 238,245
262, 263
264-268
234, 269-277
234,258,264, 278-288
151.161. 234.258,289-292
234, 258, 293-297
234. 298
169. 299-303
169,304-309
310.31 1
show nerve infiltration. When this occurs, they have been
referred to as desmoplastic neurotropic melanomas. When
matched for depth of invasion, desmoplastk melanomas
have a lower tisk of metastases than conventional melano-
mas of similar depth.
187
18931
s..
317
Desmoplastic melanomas
may show immunopositivity for actin and can be confused
with smooth muscle tumors. Thus, immunoperoxidase
studies need to be interpreted with caution in cases of
desmoplastic melanoma.
Desmoplastic melanomas have high recurrence rates
due to their highly infiltrative growth and frequent peri-
neural invasion and are usually deeply invasive at diagno-
sis. 195 A large retrospective study from the Mayo Clinic
demonstrated that desmoplastic melanomas have a high
incidence of local recunence, rarely metastasize to the
lymph nodes, and have a propensity to metastasize to the
lungs. The same clinical behavior has been confirmed by
other groups, with the incidence of lymph node me-
tastases ranging from 0% to 8%. A distinction must be
made between pure desmoplastic melanomas and mela-
nomas with a desmoplastic component. The pure des-
moplastic melanomas have unique clinical behaviors,
whereas melanomas with desmoplastic components be-
have as other cutaneous melanomas; this may explain the
conflicting reports of older clinical studies that found
a higher rate of lymph node metastases. The differ-
ence between pure desmoplastic melanomas and cutane-
ous melanomas has been further distinguished by gene
profiling in which desmoplastic melanomas have down-
regulation of melanocyte differentiation and up-regula-
tion of neurotropic factors. On the basis of the predilec-
tion for local recurrence, the North Central Treatment
Group has embarked on a phase 2 study evaluating the
For personal .... se. Mass reprodLce <.,r,; WI t pe. n .. ssh.H. Proceeding,:,.
FIGURE 4. Unusual variants of melanoma. A, Desmoplastic melanoma showing prominent
perineural invasion. 8, Balloon cell melanoma metastatic to a lymph node. C, Myxoid
melanoma with marked cytologic atypia and mitotic activity. D, Osteogenic melanoma from
the foot. E. Rhabdoid melanoma. F, Basomelanocytic melanoma.
utility of adjuvant radiation for the treatment of desmo-
plastic melanoma.
When matched for depth of invasion, desmoplastic
melanomas have a lower risk of metastases than conven-
tional melanomas of similar depth.
195
Compa1ing desmo-
plastic melanomas with conventional types with a depth
of invasion more than 4 mm, the 5-year survival for des-
moplastic melanoma is 72% compared with 37% to 48%
for conventional melanoma.
189
Overall survival for both
groups was similar to that for patients with other cutaneous
melanomas. Factors associated with higher rates of local
recurrence were surgical margin clearance less than I em
and neurotropism. However, desmoplastic melanomas
showed a lower rate of regional lymph node metastasis at
presentation than conventional melanomas.
189
BALLOON CELL MELANOMA
Balloon cell melanoma is a rare histologic variant of malig-
nant melanoma that resembles balloon cell nevus but ef-
faces the dermal architecture with sheets of tumor cells and
shows cytologic atypia, nuclear pleomorphism, and mito-
ses.290291 Balloon cell melanomas must be differentiated
from cutaneous clear cell tumor and metastases. The clear
cytoplasm has been thought to represent degenerating mel-
anosomes, lipid, or glycogen.
292
Immunohistochemical and
electron microscopic tindings suggest that the tumor cells
are likely metabolically active melanocytic cells.
191
The
prognosis of these rare histologic vmiants usually COITelates
with tumor thickness and is similar to that in other types of
melm1oma. These melanomas are often deeply invasive at
diagnosis, which reflects in the overall poor survival.
291
f-or persona J ~ o ; ; ; 1\aa.>S I efJrvdLC on.y wrth rJt .nt.>Sa0o1 frolli Mayo L 'mic Proceedmgs.
MYXOID MELANOMA
Myxoid melanoma can easily be confused with other myx-
oid and mucinous neoplasms. Prominent myxoid change
can be seen in primary melanomas,234.2S8.269273.3Js.Jzo recur-
rences, and metastases.
274
319
321
'
323
Benign melanocytic neo-
plasms can also show myxoid and mucinous change.
324
327
In primary cutaneous tumors, the presence of a junctional
component helps to confirm a diagnosis of myxoid mela-
noma. Melanomas with only foci of myxoid change are
also less difficult to diagnose than those that are predomi-
nantly myxoid. Myxoid melanomas often show a vaguely
lobular architecture with pushing margins and paraseptal
and perivascular accentuation of cells. The twnor cells
may be small, stellate, spindled, or large and epithelioid
and are arranged singly, in cords, or rarely as pseudo-
glandular structures.
258
Myxoid change can be seen in
metastases even when not present in the primary tumor.
321
Thus, a high index of suspicion is needed to diagnose
melanoma in a metastasis that shows myxoid change.
Cutaneous tumors in the differential diagnosis include
myxoma, nerve sheath myxoma, malignant chondroid sy-
ringoma, myxoid dermatofibrosarcoma protuberans,
myxoid atypical fibrous histiocytoma, mucinous eccrine
carcinoma, and metastases.
321
'
323
327
'
336
Immunohis-
tochemical studies can be helpful for diagnosis because
myxoid melanomas are positive for SlOO, NKIC3, vimen-
tin, and HMB45Y
1273
321
Keratin immunopositivity and
negative staining for S 100 are helpful in ruling out a
myxoid melanoma.
SIGNET R tNG CELL MELANOMA
Signet ring cell melanoma is a rare but well-recognized
variant of malignant melanoma.
234
258
264
278
288337
Signet ring
cell features are more common in metastases than in pri-
mary melanomas. The presence of signet ring cells in a
melanocytic neoplasm is not diagnostic of malignancy
because signet 1ing cells have been described in melano-
cytic nevi.
284
The importance of a signet ring cell pheno-
type is unknown, but it is important to consider this vari-
ant in the differential diagnosis of a primary cutaneous
tumor or a metastasis with signet ring cell features. Signet
ring cells are most common in mucin-secreting adenocar-
cinomas but may be seen in lymphomas, liposarcomas,
basal and squamous cell carcinomas, ovarian stromal tu-
mors, epithelioid smooth muscle tumors, hidradenomas,
and cylindromas, among others.234.2ss.278.279.284.287 Metas-
tases of malignant melanoma with signet ring cell cyto-
morphologic features have been described in lymph nodes,
in the lungs, and as peritoneal effusions.
258
278
279280
284
287
The
immunohistochemical profile can clarify the diagnosis, but
a high index of suspicion, particularly in a metastasis, is
necessary.
OSTEOGENIC MELANOMA
Osteogenic melanomas usually occur on the digits (particu-
larly subungual) and in the nose and sinuses and are often
deeply invasive at the time of diagnosis.
169
299
303
Melano-
mas may also show cartilaginous differentiation in these
sites.
169
304
309
Men and women appear to be equally af-
fected, and the tumor is most common in middle-aged
and elderly patients. Osteogenic melanomas mimic osteo-
sarcomas, chondrosarcomas, and metaplastic carcinomas.
Osteogenic melanomas must be differentiated from osteo-
sarcomas because their treatment is different.
301
Strong im-
munoreactivity for S 100 protein, HMB45, and Mel an A is
helpful in differentiat.ing osteogenic melanomas from os-
teosarcomas and tumors with osteoid metaplasia.
RHABDOID MELANOMA
Malignant melanoma with rhabdoid cytomorphologic fea-
tures has been described in both primary and metastatic
lesions.
234
258
293
296
Primary rhabdoid melanomas usually
show junctional activity or a background of more conven-
tional cells, which are helpful in recognizing the differen-
tiation of the tumor. Although rhabdoid melanomas are
positive for S 100, they often lose I-IMB45 expression.
295
296
Additionally, focal aberrant expression of both keratin and
smooth muscle actin has been noted in these tumors.
295
The
rhabdoid phenotype may represent further evolution to a
more primitive dedifferentiated fonn.
295
The prognosis for
patients with melanoma metastases with rhabdoid features
is similar to that of conventional melanoma metastases.
Not enough data are available to evaluate whether rhabdoid
morphologic findings have prognostic significance in the
primary tumor.
BASOMELANOCYTIC TuMORS
Biphasic tumors with epithelial and melanocytic compo-
nents are rare.
310
311
Four squamomelanocytic tumors have
been described, but none of the patients were reported to
die of metastatic melanoma or squamous cell carcinoma.
338
In a recent report, a patient with a basomelanocytic tumor
died of metastatic malignant melanoma, suggesting that at
least a subset of biphasic tumors may have a more aggres-
sive biologic potential than previously known.
311
Biphasic
tumors show an intimate admixture of melanocytic cells
and basaloid cells or squamous cells, in contrast to collision
tumors in which there are usually distinctly separate com-
ponents of melanocytic and epithelial cells.
339
However, a
recent report of a malignant melanoma metastatic to a basal
cell carcinoma showed an intimate association of the mela-
noma cells within the basal cell carcinoma.
340
Another case
of a malignant melanoma colonizing a basal cell carcinoma
has been reported, but the malignant melanoma was limited
to the epidermis.
341
The cells in basomelanocytic tumors
Mayo Clin Proc. March 2007;82(3):364380 www.mayoclinicproceedings.com 373
may arise from a common precursor and show biphasic
immunophenotypic differentiation. Overall, these are rare
tumors in which a subset appears to have a more aggressive
biologic potential than previously known.
CONCLUSION
This review represents a condensed summary of the cunent
state of knowledge regarding the epidemiology, risk fac-
tors, screening, prevention, and diagnosis of malignant
melanoma with emphasis on clinical practice. A discussion
of the extensive basic science works that have provided the
mechanistic interpretations of the clinical observations in
malignant melanoma was thought to be beyond the scope
of this review. Maintaining a clinical emphasis, part 2 of
this review (in the April issue of Mayo Clinic Proceedings)
will summarize the current state-of-the-art in the staging,
prognosis, and treatment of melanoma.
Members of the Melanoma Study Group. Iftikhar Ahmed, MD;
Jacob B. Allred, MS; Keith H. Baratz, MD; Uldis Bite, MD;
Elizabeth A. Bradley, MD; Leslie J. Christenson, MD; Rjcky P.
Clay, MD; Edward T. Creagan, MD; Gary A. Croghan, MD, PhD;
Mark Denis P. Davis, MD; Allan B. Dietz, PhD; John H.
Donohue, MD; Lori A. Erickson, MD; David R. Farley, MD;
Evanthia Galanis, MD; Yolanda I. (Nina) Garces, MD; James A.
Ganity, MD; Dennis A. Gasti neau, MD; BobbieS. Gostout, MD;
Clive S. Grant , MD; Jennifer L. Hand, MD; James N. Ingle, MD;
Amer N. Kalaaji, MD; Jan L. Kasperbauer, MD; Judith S. Kaur,
MD; Li sa A. Kottschade, RN; Kathleen M. Liffrig, BA; Noralane
M. Lindor, MD; Charles L. Loprinzi, MD; Val Lowe, MD; Leo J.
Maguire, MD; Svetomir N. Markovic, MD, PhD; Colin A.
McCannel, MD; Elizabeth S. McDonald, MD, PhD; Robert R.
McWilliams, MD; Jane M. fvlilbum, MBA; Steven L. Moran,
MD; Wendy K. Nevala. BS: Kerry D. Olsen, MD; Clark C. Otley,
MD: Animesh Pardanani, MBBS, PhD; Mark R. Pi uelkow, MD;
Karl C. Podratz, MD, PhD; Jose S. Pulido, MD; Ravi D. Rao,
MBBS; Michael G. Rock, MD; Randall K. Roenigk, MD; Diva R.
Salomao, MD; Aleksandar Sekulic, MD, PhD; Thomas C. Shives,
MD; Franklin H. Sim, MD; C. Robert Stanhope, MD; W. P.
Daniel Su, MD; Vera J. Suman, PhD; Nho V. Tran, MD; Celine
M. Vachon, PhD; Stanimir Vuk-Pavlovic, PhD; Roger H. Weenig,
MD; Ryan A. Wilcox, MD, PhD; Timothy 0. Wilson, MD;
Gregory A. Wiseman, MD.
REFERENCES
I. Jemal A, Siegel R, Ward E. et al. Cancer statistics. CA Cancer J Clin.
2006;56: I 06-130.
2. Jemal A. Devesa SS. Hartge P, Tucker MA. Recent trends in cutaneous
melanoma incidence among whites in the United J Nat/ Cancer f11st.
200 I ;93:678-683.
3. Ries LA, Wingo PA, Miller DS, et al. The annual report to the nation on
the status of cancer. 1973- 1997, with a special section on colorectal cancer.
Ca11cer. 2000:88:2398-2424.
4. Goldstein AM, Tucker MA. Etiology, epidemiology, risk factors, and
public health issues of melanoma. Curr Opin Oneal. 1993;5:358 363.
5. Rigel DS. Naylor M, Robinson J. What is the evidence for a sunscreen
and melanoma controversy? Arch Dermatol. 2000; 136:1447-1449.
6. Rigel DS. Carucci JA. Malignant melanoma: prevention, early detection,
and treatment in the 21st century. CA Cancer} Clin. 2000:50:215-236.
7. Tsao H, Sober AJ. Ultraviolet radiation and malignant melanoma. Cli11
Dermatol. 1998: 16:67-73.
8. Tsao H. Rogers GS. Sober AJ. An estimate of the annual direct cost of
treating cutaneous melanoma. J Am Acad Dermatol. 1998;38(5. pi I ):669 680.
9. Strouse JJ, Fears TR, Tucker MA, Wayne AS. Pediatric melanoma: risk
factor and survival analysis of the surveillance. epidemiology and end results
database. J Cli11 Oncol. 2005;23:4735-4741 .
10. Welch HG, Woloshin S. Schwartz LM. Skin biopsy rates and incidence
of melanoma: population based ecological study. BMJ. 2005 Sep 3:331:481.
Epub 2005 Aug 4.
11. Boyle P, Maisonneuve P, Dore JF. Epidemiology of malignant mela-
noma. Br Med Bull. I 995;51 :523-547.
12. Tsai T, Yu C. Henson DE. Cutaneous. ocular and visceral melanoma in
African Americans and Caucasians. Melanoma Res. 2005; 15:213-217.
13. Stevens NG, Liff JM, Weiss NS. Plantar melanoma: is the incidence of
melanoma of the sole of the foot really higher in blacks than whiles? 1m .I
Cancer. 1990;45:691 -693.
14. Hiatt RA. Fireman B. The possible effect of increased surveillance on
the incidence of malignant melanoma. Prev Med. 1986;15:652-660.
15. Elwood JM. Jopson J. Melanoma and sun exposure: an overview of
published studics. /nt J Cancer. 1997:73: 198-203.
16. Goldstein AM, Dracopoli NC. Ho EC. et al. Further evidence for a locus
for cutaneous malignant melanoma-dysplas tic nevus (CMM/DN) on
chromosome lp, and evidence for genetic heterogeneity. Am J f/um Genet.
1993;52:537-550.
17. Goldstein AM, Dracopoli NC, Engelstei n M, Fraser MC, Clark WH .lr,
Tucker MA. Linkage of cutaneous malignant melanoma/dysplastic nevi to
chromosome 9p, and evidence for genetic heterogeneity. Am .I Hum Genet.
1994;54:489-496.
18. Kraemer KH. Dysplastic nevi, melanoma and sunlight. Photoderma
tology. 1987;4:227-229.
19. Kraemer KH. Lee MM. Andrews AD. Lambert WC. The role of
sunlight and DNA repair in melanoma and nonmelanoma skin cancer: the
xerodenna pigmentosum paradigm. Arc/1 Demwtol. 1994: 130: I 018- 1021.
20. Spatz A, Giglia-Mari G. Benhamou S, Sarasin A. Association between
DNA repair-deficiency and high level of p53 mutations in melanoma of
Xeroderma pigmentosum. Cancer Res. 200 I ;61 :2480-24&6.
21. Cummins DL, Cummins JM, Pantle H, Silverman MA, Leonard AL,
Chanmugam A. Cutaneous malignant melanoma. Mayo Clin Proc. 2006;8 1:
500-507.
22. Yeierod MB. Weiderpass E, Thorn M. et al. A prospective study of
pigmentation, sun exposure, risk of cutaneous malignant melanoma in
women. J Nat/ Cancer Ins/. 2003:95:1530-1538.
23. Greene MH, Tttcker MA, Clark WH Jr. Kraemer KH, Elder DE, Fraser
MC. Hereditary melanoma and the dysplastic nevus syndrome: the risk of
cancers other than melanoma. JAm Acad Dermatol. 1987:16:792-797.
24. Kraemer KH. Dysplastic nevus syndrome and cancer risk- response. J
Am Acad Oermatol. 1987:17(5, pt 1):850-851.
25. Stern RS. The risk of melanoma in association with long-tenn exposure
to PUVA. J AmAcad Dennato/. 2001 ;44:755-761.
26. Shah CP, WeisE, Lajous M, Shields JA, Shields CL. Intermittent and
chronic ultraviolet light exposure and uveal melanoma: a meta-analysis.
Opilthalmology. 2005; 112: 1599- 1607.
27. Gallagher RP, Spinelli JJ. Lee TK. Tanni ng beds. sunlamps. and risk of
cutaneous malignant melanoma. Cancer Epidemiol Biomarkers Prev. 2005;
14:562-566.
28. Yai nio H, Bianchini F. Cancer-preventive effects of sunscreens are
unce1tain. Scand J Work Environ Health. 2000:26:529-531.
29. Vaiuio H. Miller AB. Bianchini F. An i1uernarional evaluation of the
cancer-preventive potential of sunscreens. /111 J Cancer. 2000;88:838-842.
30. Huncharek M, Kupelnick B. Use of topical stmscreens and the risk of
malignant melanoma: a meta-analysis of 9067 patients from I I case-control
studies. Am J Public Health. 2002:92: 1173-1 177.
31. Gallagher RP. Rivers JK, Lee TK. Bajdik CD, McLean DI. Coldrnan
AJ. Broad-spectrum sunscreen usc and the development of new nevi in white
children: a randomized controlled trial. JAMA. 2000:283:2955-2960.
32. Autier P, Dore JF, Negrier S, et al. Sunscreen use and duration of sun
exposure: a double-blind. randomized trial. J Nat/ Cancer Ins/. 1999;91: 1304-
1309.
f-or persona 1\aa.>S I efJrvdLC on.y With rJt .ni.>Sa0o1 frolli Mayo .. 'mic Proceedmgs.
33. Autier P. Dore JF, Reis AC. et al. Sunscreen use and intentional
exposure to ultraviolet A and B radiation: a double blind randomized trial using
personal dosimeters. Br J Cancer. 2000:83: 1243-1248.
34. Autier P. Dore JF, Renard F, et al. Melanoma and sunscreen use: need
for st11dies representative of actual behaviours. Melanoma Res. 1997:7(suppl
2):S 115-S 120.
35. Autier P, Severi G, Boniol M, de Vries E, Coebergh JW, Dore JF. Re:
sun exposure and mortality from melanoma [leller]. J Nat/ Cancer /n.rt. 2005;
97: 1159; author reply 1159-1160.
36. Dupuy A. Shamsaldin A. Qui11iou E. et al. Risk of melanoma following
adulthood cancer: a case-control sl1dy. Eur J Can.ca. 2005 Dcc;41 :2904-
29 10. Epub 2005 Oct 21.
37. Linet MS. Malker HS. Chow WH. et al. Occupational risks for
cutaneous melanoma among men in Sweden . .I Occup Environ Med. 1995;
37:1127- 11 35.
38. Alavanja MC, Sandler DP, Lynch CF. et al. Cancer incidence in the
agricultural health study. Scand J Work Environ Health. 2005:3 1(suppl I ):39-45.
39. Hammar N. Linnersjo A. Alfredsson L. Dammstrom BG. Johansson M,
Eliasch H. Cancer incidence in airline and military pilots in Sweden 1961 -
1996. Aviat Space Environ Med. 2002;73:2-7.
40. Snell RS, Bischitz PG. A study of the melanocytes and melanin in the
skin of the immarurc, mature and pregnant female guinea-pig. Z Zellforsch
Mikrosk A11a1. l 960:51:225-242.
41. Snell RS. E!Iect of alpha M.S.H. and estrogen on melanin pigmentation
in the albino. J lnvesr Derm(llo/. 1965;44: 17-21.
42. Levi F, Te VC. Randimbison L, La Vecchia C. Cancer risk in women
with previous breast cancer. Ann Oncol. 2003:14:71-73.
43. Feskanich D, Hunter DJ, Willett WC, et al. Oral contraceptive use and
risk of melanoma in premenopausal women. Br J Cancer. 1999:81:9 18-923.
44. Karagas MR. Zens MS, Stukel TA, et al. Pregnancy history and
incidenc.e of melanoma in women: a pooled analysis. Cancer Causes Comrol.
2006:17:1 1- 19.
45. Naldi L. Altieri A. Imberti GL, Giordano L, Gallus S, La Vccchia C.
Oncology Smdy Group of the Italian Group for Epidemiologic Research in
Dermatology. Cutaneous malignant melanoma in women: phenotypic
characteristics, sun exposure, a11d hormonal factors: a case-control study from
Italy. Ann Epidemiol. 2005:15:545-550.
46. Karagas MR. Stukel TA, Dykes J, et al. A pooled analysis of 10
conuol studies of melanoma and oral contraceptive use. Br J Cancer. 2002:86:
1085- 1092.
47. Wiggins CL, Berwick M, Bishop JA. Malignant melanoma in preg-
nancy. Obsret Gyneco/ C/in North Am. 2005;32:559-568.
48. Veierod MB, Theile DS, Laake P. Diet and risk of cutaneous malignant
melanoma: a prospective study of 50.757 Norwegian men and women. ftll J
Cancer. 1997:71: 600-604.
49. Stryker WS, Stampfer MJ. Stein EA. et al. Diet. plasma levels of beta-
carotene and alpha-t.ocopherol, and risk of malignant melanoma. Am J
Et'idemiol. 1990:131 :597-61 J.
50. Miller AJ. Du J, Rowan S. Hershey CL. Widlund Fisher DE.
Transcriptional regulation of the melanoma prognostic marker melastatin
(TRPMI) by MITF in melanocytes and melanoma. Cancer Res. 2004:64:509
516
51. Osterlind A. Tucker MA. Stone BJ. Jensen OM. The Danish case-control
swdy of cutaneous malignant melanoma, IV: no association with nutritional
factors, alcohol, smoking or hair dyes. lnt J Cancer. 1988:42:825-828.
52. Osterlind A, Tucker MA, Stone BJ, Jensen OM. The Danish case-
control study of c utaneous malignant melanoma, Ill: hormonal and
reproductive factors in women.//11 J Cancer. 1988:42:821-824.
53. Osterlind A, Engholm G, Jensen OM. Trends in cutaneous malignant
melanoma in Denmark 1943-1982 by anatomic site. APM!S. 1988;96:953-963.
54. Osterlind A. Tucker MA, Stone BJ, Jensen OM. The Danish case-
control study of cutaneous malignant melanoma, II : imp01tance of UV- Iight
exposure. Ill/ ./ Cancer. 1988;42:3 19-324.
55. Osterlind A, Hou-Jcnsen K, Moller Jensen 0. Incidence of cutaneous
malignant melanoma i.L1 Denmark 1978- 1982: anatomic site distribution.
histologic types, and comparison with non-melanoma skin cancer. Br J
Cancer. 1988;58:385-391.
56. Osterlind A. Tucker MA. Hou-Jensen K. Stone BJ, Engholm G, Jensen
OM. The Danish cae-control study of cutaneous malignant melanoma, 1:
importance of host factors. lm J Cancer. 1988;42:200-206.
57. Kirkpatrick CS. White E, Lee JA. Case-control study of malignant
melanoma in Washington State. II: diet alcohol. and obesity. Am J Epidemiol.
1994:139:869-880.
58. Westerdahl J, Olsson H. Masback A. lngvar C. Jonsson N. Risk of
malignant melanoma in relation to drug intake, alcohol, smoking and hormonal
factors. Br J Cancer. l 996;73: 1126-1131.
59. Bevona C, Goggi ns W. Quinn T, Fullerton J, Tsao H. Cutaneous
melanoma' associated with nevi. Arch Dermato/. 2003;139: 1620-1624.
60. Holly EA. Kelly JW, Shpall SN. Chiu SH. Number of melanocytic nevi
as a major risk factor for malig1ant melanoma. J Am Acad Dermarol.
1987: 17:459-468.
61. Garbe C, Buttner P, Weiss J. et al. Associated factors in the prevalence
of more than 50 common melanocytic nevi. atypical melanocytic nevi. and
actinic lentigines: mult icenter case-control study of the Central Malignant
Melanoma Registry of the German Dermatological Society. J lnvesr Derma rot.
1994:102:700-705.
62. Garbe C, Buttner P, Weiss J, et at. Ri sk factors for developing cutaneous
melanoma and criteria for identifying persons at 1isk: multicenter case-control
study of the Central .Malignant Melanoma Registry of the German
Dermatological Society. J lnvesr Dermatol. 1994:102:695-699.
63. Goldstein AM, Bale SJ, Tucker MA. Linkage analysis of melanoma
alone and chromosome lp markers PND, DIS47. and LJ1;1YC. Cytogenet Cell
Gener. 1992;59:203-205.
64. Goldstein AM, Tucker MA. Cmtcher WA. Harrge P. Sagebiel RW. The
inheritance pattern of dysplastic naevi in families of dysplastic naevus patients.
Melanoma Res. 1993;3:15-22.
65. Tucker MA. Halpern A. Holly EA, et al. Clinicall y recogr1ized
dysplastic nevi: a central risk factor for cutaneous melanoma. JAMA. 1997;
277: 1439-1444.
66. Naldi L. Lorenzo lmberti G. Parazzini F, Gallus S. La Veccbia C.
Pigmentary traits, modalities of sun reaction. history of sunburns, and
melanocytic nevi as risk factors for cutaneous malignant melanoma i11 the
Italian population: results of a collaborative case-control study. Cancer.
2000:88:2703-27 I 0.
67. Naldi L. Gall us S, lmberti GL. Cainelli T, Negri E. La Vecchia C.
Sunscreens and cutaneous malignant melanoma: an Italian case-control study.
blf J Cancer. 2000;86:879-882.
68. Naldi L, Gallus S, lmbe11i GL, Cainelli T, Negri E, LaVecchia C, Italian
Group for Epidemiological Research in Dermatology. Sunlamps and sunbcds
and the risk of cutaneous melanoma. Eur J Cancer Prev. 2000:9:133-134.
69. Wall AJ, Kotsis SV. Chung KC. Risk of melanoma arising in large
congenital melanocytic nevi : a systematic review. Plasr Reconstr Surg.
2004:113:1968- 1974.
70. Tannous ZS, Mihm MC Jr, Sober AJ, Duncan LM. Congenital
rnelanocytic nevi: cli nical and histopathologic features. risk of melanoma. and
clinical management. JAm Acad Dermatol. 2005:52:197-203.
71. Titus-Emstoff L. Perry AE. Spencer SK, Gibson JJ. Cole BF, Ernstoff
MS. Pigmentary characteristics and moles in relation to melanoma risk. lm J
Cancer. 2005: 116:144- 149.
72. Halpero AC, Guerry D IV. Elder DE. Track B. Synnestvedt M. A cohort
study of melanoma in patients with dysplastic oevi. J /nvesr Dermatol. 1993:
100:346S-349S.
73. Piepkorn MW. A perspective on the dysplastic nevus controversy.
Parllology (Phi/a). 1994;2:259-279.
74. Piepkorn MW, Barnhill RL, Cannon-Albright LA, et a!. A multi-
observer, population-based analysis of histologic dysplasia in melanocytic
nevi. JAm A cad Dermatol. I 994:30:707-714.
75. Tucker MA, Crutcher WA. Hartge P. Sagebiel RW. Familial and
cutaneous features of dysplastic nevi: a case-control study. J Am Acad
Dennatol. 1993:28:558-564.
76. Tucker MA, Fraser MC. Goldstein AM, Elder DE, Guerry D IV.
Organic SM. Risk of melanoma and other caJJcers in melanoma-prone families.
J Invest Dennatol. 1993:100:350S-355S.
77. Greene MH, Clark WH .lr, Tucker MA, et al. Acquired precursors of
cutaneous malignant mel anoma: the fami lial dysplastic nevus syndrome. N
Eng/J Med. 1985;312:91 -97.
78. Greene MH, Clark WH Jr, Tucker MA, Kraemer KH. Elder DE, Fraser
MC. High Jisk of mal ignant mehmoma in melanoma-prone famil ies with
dysplastic nevi . Ann Intern Me d. 1985: I 02:458-465.
79. Greene MH. Wilson J. Second cancer following lymphatic and
hematopoietic cancers in Connecticut, 1935-82. Nat/ Cancer /nsf Monogr.
1985:68:191-217.
80. Greene MH. The dysplastic nevus syndrome: precursors of hereditary
and nonfami lial cutaneous melanoma. lmporrwu Adv Oncol. 1986: 173- .1 91.
81. Greene MH. The genetics of hereditary melanoma and nevi: 199R
update. Cancer. 1999;86( 11, suppl):2464-2477.
82. Purdue MP. From L. Armstrong BK. et at. Genes. Environment. and
Melanoma Study Group. Etiologic and other factors predicting nevus-
associated cutaneous malignant melanoma. Cancer Epidemiol Biomarkers
Prev. 2005; 14:2015-2022.
83. Cho E, Rosner BA. Colditz GA. Risk factors for melanoma by body
site. Cancer Epidemio/ Biomarkers Prev. 2005; 14: 1241-1244.
84. Cho E. Rosner BA, Feskanich D. Colditz GA. Risk factors and individual
probabi lities of melanoma for whites. J Cli11 Oncol. 2005;23:2669-2675.
85. Meyer LJ, Zone JH. Genetics of cutaneous melanoma. J /t1vest
Dermatol. 1994: 103(5. suppl): 112S-ll6S.
86. Tsao H. Niendotf K. Genetic testing in hereditary melanoma. J Am
Acatl Dermatol. 2004:51:803-808.
87. Tsao H. Goel V, Wu H, Yang G. Haluska FG. Genetic interaction
between NRAS and BRAF mutations and PTEN/MMACI inactivation in
melanoma. J Invest Derma to/. 2004; 122:337-341.
88. Newton Bishop JA. Harland M, Bishop DT. The genetics of melanoma:
the UK experience. Clifl E\p Dermato/. 1998;23: 158-161.
89. Harland M. Meloni R. Gruis N. et al. Germline muiations of the
CDKN2 gene in UK melanoma famil ies. Hum Mol Genet. 1997:6:2061-2067.
90. Harland M. Mistry S, Bishop DT, Bishop JA. A deep intronic mutation
in CDKN2A is associated wilh disease in a subset of melanoma pedigrees.
Hum Mol Genet. 200 I ; I 0:2679-2686.
91. Harland M, Taylor CF. Bass S, et al. lntronic sequence variants of the
CDKN2A gene in melanoma pedigrees. Genes Chromosomes Concer. 2005:
43: 128-136.
92. Harland M. Taylor CF, Chambers PA. et al. A mutation hotspot at the
p14ARF splice site. Oncogene. 2005;24:4604-4608.
93. Titus-Emstoff L, Perry AE, Spencer SK, et al. Multiple primary
melanoma: two-year results from a population-based study. Arch Demw/0/.
2006; 142:433-438.
94. Ferrone CR, Ben Po rat L. Panageas KS, et al. CJ inicopathological
features of and risk factors for multiple pri m;uy melanomas. lAMA. 2005;
294: 1647-1654.
95. Hollenbeak CS. Todd MM. Billingsley EM, Harper G. Dyer AM.
Lcngcrich EJ. Increased incidence of melanoma in renal transplantation
recipients. Coucer. 2005: 104: 1962-1967.
96. Euvrard S. Kanitakis J, Claudy A. Skin cancers after organ transplanta-
tion. N Eng/ J Met/. 2003;348: 1681- 1691.
97. Euvrard S. Kanitakis J. Pouteii-Noble C, Claudy A, Tourai ne JL. Skin
cancers in organ transplant recipients. Arm Transplant. 1997;2:28-32.
98. Euvrard S. Kan.itakis J, Cochat P. Claudy A. SkiLl cancers following
pediatric organ transplantation. Dermatol Surg. 2004;30(4, pt 2):616-621.
99. Smith KJ. Skelton HG. Heimer W. et al, Military Medical Consortium
forthe Advancement ofRetroviral Research. Melanocytic activation in HIV-1
disease: HMB-45 staining in common acquired nevi. JAm Acad Dermatol.
1993;29:539-544.
100. Smith KJ. Skelton HG. Yeager J, Angritt P, Wagner KF. Mibtary
Medical Consortium for the Advancement of Retroviral Research. Cutaneous
neoplasms in a military population of HlV- 1-positive patients. J Am Acad
Dermatol. 1993:29:400-406.
101. Holly EA, Aston DA. Cress RD. Ahn DK. Kristiansen JJ. Cutaneous
melanoma in women, ll : phenotypic characteristics and other host-related
factors. Am J Epidemiol. 1995; 141:934-942.
102. Holly EA, Aston DA. Cress RD, Ahn DK, Kristiansen JJ. Cutaneous
melanoma .in women. 1: exposure to sunlight, ability to tan, and other risk
factors related to ultraviolet light. Am J Epitlemiol. !995; 141:923-933.
103. Kirkpatrick CS. Lee JA, White E. Melanoma risk by age and socio-
economic status./m J Cancer. 1990;46:1-4.
104. Elwood M, Gallagher R. Skin cancer epidemiology in 1996: the Third
Symposium on the Epidemiology of Melanoma and Non-melanocytic Skin
Cancer. Melanoma Res. 1997:7:74-77.
105. Elwood JM, Koh HK. Etiology, epidemi ology, risk factors. and public
heallh issues of melanoma. Curr Opin Onco/. 1994;6: 179-187.
106. Elwood .JM, Hislop TG. Solar radiation in the etiology of cutaneous
malignant melanoma in Caucasians. Nat/ Cancer lnst Monogr. 1982:62:167-171.
107. Elwood JM, Gallagher RP, Hill GB. Pearson JC. Cutaneous melanoma
in relation to intermittent and c-onstant sun exposure-the Western Canada
Melanoma Study. lm J Ca11cer. 1985;35:427-433.
J08. Elwood JM, Gallagher RP, Davison J, Hi ll GB. Sunburn, suntan and the
risk of cutaneous mal ignant melanoma-the Western Canada Melanoma
Study. Br J Cancer. 1985:51:543-549.
109. Elwood JM. Melanoma and sun exposure. Semin Oncol. 1996:23:650-
666
110. Elwood JM. Melanoma and ultraviolet radiation. Cliu Dermatol. 1992:
10:41-50.
I II. Elwood J M. Mela noma and sun exposure: contrasts between
intermittent and chronic exposure. World J Surg. 1992:16: 157-165.
112. Elwood JM. Health and the ozone layer [letter]. BMJ. 1988:297:918.
113. Rigel DS. Epidemiology and prognostic factors in maligmmt mel<moma.
Ann Plast Surg. 1992:28:7-8.
114. Maier T, Kort ing HC. Sunscreens-which and what for? Skin
Pharmacal Plnsiol. 2005 Nov-Dec;l8:253-262. Epub 2005 Aug 19.
115. Kaske1 P, Sander S, Kron M, Kind P, Peter RU, Krahn G. Outdoor
activities in childhood: a protective factor for cutaneous melanoma? results of a
case-control study in 271 matched pairs. Br J Dermatol. 2001; 145:602-609.
116. Grob JJ. Guglielmioa C, Gouvernet J, Zarour H. Noe C. Booeraodi JJ.
Study of sunbathi ng habits in children and adolescenLs: application to the
prevention of melanoma. Dermatology. 1993; 186:94-98.
117. Knowles JH. The responsibility of the individual. Daedalus. Winter
1977; 106:57-80.
118. Montague M. Borland R. Sinclair C. Sl ip! Slop! Slap! and SunSmart.
1980-2000: skin cancereontrol and 20 years of population-based campaigning.
Health Educ Behav. 200 l ;28:290-305.
119. Dellavalle RP, Drake A, Graber M. et a! . Statins and librates for
preventing melanoma. Cochrane Database Syst Rev. 2005:4:CD003697.
120. Shellman YG. Ribble D. Miller L. et al. Lovastatin-induccd apoptosis in
human melanoma cell lines. Melauoma Res. 2005:15:83-89.
121. Dellavalle RP, Nicholas MK. Schilling LM .. Melanoma chemo-
prevention: a role for statins or fibrates? Am J Ther. 2003: I 0:203-2 10.
122. Freedberg KA, Geller AC, Miller DR. Lew RA. Koh HK. Screeni ng for
malignant melanoma: a cost-effectiveness analysis. J Am Acad Dermatol.
1999;41(5, pt 1):738-745.
123. Gell er AC. Sober AJ. Zhang Z. et al. Strategies for improvi ng
melanoma education and screeni ng for men age ;;,:50 years: findings from the
American Academy ofDem1atology National Skin Cancer Screening Program.
Cancer. 2002;95: 1554-1561.
124. Geller AC. Zhang Z. Sober AJ. ct a!. The first 15 years of the American
Academy of Dermatology skin cancer screening programs: 1985-1999. J Am
Acad Dermmol. 2003;48:34-41.
125. Geller AC, Emmons K, Brooks DR, et al. Skin cancer prevemion and
detection practices among siblings of patients with melanoma. J Am Acatl
Dermatol. 2003;49:631 -63&.
126. DiFronw LA, Wanek LA, Morton DL. Earlier diagnosis of second
primary melanoma confims the benefi ts of patient education and routine
postoperative follow-up. Cancer. 2001:91:1520-1524.
127. Aitken .IF. Janda M. Elwood M. Youl PH. Ring IT. Lowe JB. Clinical
outcomes from skin screening clinics within a community-based mehmoma
screening program. JAm Acud Dermatol. 2006 Jan;54: 105-114. Epub 2005 Nov
28.
128. Vandaele MM. Richert B, Van der Endt JD, et al. Melanoma screening:
results of the first one-day campaign in Belgium ( 'melanoma Monday'). J Eur
Acad Dermatol Venereal. 2000;14:470-472.
129. Rigel DS, Friedman RJ , Kopf A W, et al. Importance of complete
cutaneous examination for the detection of malignant melanoma. J Am Acad
Dermatol. 1986:14:857-860.
130. Abbasi NR. Shaw HM, Rigel DS. et a!. Early diagnosis of cutaneous
melanoma: revisiting the ABCD critelia . .lAMA. 2004:292:2771 -2776.
131. Voigt H, Classen R. Computer vision and digital imagi ng technology in
melanoma detection. Semin Oncol. 2002:29:308-327.
132. Voigt H. Classen R. Topodermatographic image analysis for melanoma
screening and Lhe quantitative assessment of tumor dimension parameters of
the skin. Cancer. 1995:75:981-988.
133. Marchesini R, Bono A, Bartoli C, Lualdi M, Tomatis S. Cascinell i N.
Optical imaging and automated melanoma detection: questions and answers.
Melanoma Res. 2002; 12:279-286.
134. Marghoob AA. Swindle LD. Moricz CZ, et al. Instruments and new
technologies for Lhe in vivo diagnosis of melanoma. J Am Acad Dermatol.
2003;49:777-797.
135. Gonzalez A. West AJ. Pitba JV. Taira .JW. Small-diameter invasive
melanomas: clinical and pathologic characteristics. J Cutan Patho/. 1996;23:
126-132.
136. Kamino H. Kiryu H, Ratech H. Small malignant melanomas: clinico-
pathologic correlation and DNA ploidy analysis. J Am Acad Dermatol. 1990;
22(6, pt I): 1032-1038.
137. Stern JB. Haupt HM. Pagetoid melanocytosis: tease or tocsin? Semin
Diagn Pathol. 1998:15:225-229.
f-or persona. J ~ o ; : : ; 1\,a.>S I efJrvdLC on.y wrth rJt.,nt.:>S.0.1 frorr, Mayo L.'tnic Proceedmgs.
138. Clark WH Jr. Hood AF. Tucker MA. Jampel RM. Atypical melanocytic
nevi of the geni llll type with a discussion of reciprocal parenchymal-stromal
imeractions in the biology of neoplasia. Hum Pmhol. 1998;29( 1, suppl I):Sl -
S24.
139. McNutt NS. Urmacher C. Hakimian J. Hoss DM. Lugo J. Nevoid
malignant melanoma: morphologic patterns and immunohistochemical
reactivity. J Cutan Patlwl. 1995:22:502-5 17.
140. Heenan PJ, Matz LR, BlackweiJ JB, et al. I nter-observer variation
between pathologists in the classification of cutaneous malignant melanoma in
western Australia. Histopathology. 1984:8:717-729.
141. Bamhill RL, Picpkorn M, Busam KJ. Pathology of Melanocytic Nevi
and Malignant Melanoma. 2nd ed. New York. NY: Springer; 2004:238-394.
142, Balcb CM. Cutaneous melanoma: prognosis and treatment results
worldwide. Semin Surg Oneal. 1992:8:400-4 14.
143. Casci nell i N, Zurrida S, Galimberti V, et al. Acral lentiginous
melanoma: a histological type without prognostic. significance. J Demwtol
Surg Oncol. 1994;20:817-822.
144. Chang AE, Karnell LH. Menck HR. The National Cancer Data Base
report on cutaneous and noncutaneous melanoma: a summary of 84,836 cases
from the past decade. Cancu. 1998:83:1664-1678.
14S. Kuno Y. Ishihara K, Yamazaki N. Mukai K. Clinical and pathological
features of cutaneous mal ignant melanoma: a retrospective analysis of 124
Japanese patients. Jpn J C/in Oneal. 1996;26:144- 151.
146. Cox NH. Aitcllison TC, Sirel JM. MacKie RM. Scottish Melanoma
Group. Comparison between lentigo maligna melanoma and other histogenetic
types of malignant melanoma of the head and neck. Br J Cancer. 1996:73:940-
944
147. Clark WH Jr, Elder DE, Guerry D IV. et al. Model predicting survivaJ in
stage I melanoma on tumor progression . .I Nat/ Cancer lnst. 1989;81:
1893- 1904.
148. Newell GR. Sider JG, Bergfelt L, Kripke ML Incidence of cutaneous
melanoma in the United States by histology with special reference to the face ..
Cancer Res. I 988:48:5036-5041 .
149. Crowson AN. Magro CM. Mihm MC. Tile Me/onocytic Proliferations:
A Comprehensive Textbook of Pigmented Lesions. New York, NY: Wiley-Liss:
2001:539.
ISO. Clark WH Jr, From L, Bernardino EA. Mihm MC. The histogenesis and
biologic behavior of primary human malignant melanomas of the skin. Cancer
Res. 1969:29:705-727.
l SI. Bamhi ll RL, Mihm MC Jr. The histopathology of cutaneous malignant
mehmoma. Semin Diagn Pathol. 1993;10:47-75.
1S2. Bondi EE. Clark WH Jr. Cl inical differentiation of adult-onset
cavernous angioma from nodular malignant melanoma. Arch Derma to/. 1980:
116:299-300.
1S3. Manci EA, Balch CM, Murad TM, Soong SJ. Polypoid melanoma, a
virulent variant of the nodular growth pattcm. Am J C/in Patlwl. 1981:75:810.815.
154. Kato T, Suetake T, Sugiyama Y, Tabata N, Tagami H. Epidemiology
and prognosis of subungual melanoma in 34 Japanese patients. Br .I Dermatol.
1996:134:383-387.
ISS. Chen YJ, Wu CY, Chen JT , Shen JL, Chen CC, Wang HC.
Cli nicopathologic analysis of malignant melanoma in Taiwan. J Am Acad
Dermatol. 1999:4 1:945-949.
1S6. Luk NM, Ho LC. Choi CL. Wong KH. Yu KH. Yeung WK.
Cli nicopathological features and prognostic factors of cutaneous melanoma
among Hong Kong Chinese. C/in Exp Dermatol. 2004;29:600-604.
1S7. MacKie RM. Malignant melanocytic tumours. J Cutcm Pmhol. 1985;
12:251-265.
I S8. Boyd AS, Rapini RP. Acral melanocytic neoplasms: a histologic
analysis of 158 lesions. J Am A cad Demwtol. 1994:3.1(5, pt 1):740-745.
159. Feibleman CE. Stoll H. Ma.ize JC. Melanomas of the palm. sole. and
nail bed: a clinicopathologic study. Cancer. 1980;46:2492-2504.
160. Paladugu RR. Winberg CD, Yonemoto RH. Acrallentiginous melanoma:
a clinicopathologic study of 36 patients. Ccmcer. 1983:52: 161-168.
161. Perniciaro C. Dennatopathologic variants of malignant melanoma.
Mayo Clin Proc. 1997;72:273-279.
162. Reed RJ, Martin P. Variants of melanoma. Semin Cutan Med Surg.
1997:16: 137- J 58.
163. SlingluffCLJr. Vollmer R, Seigler HF. Acral melanoma: a review of 185
patients witb identification of prognostic variables. J Surg Oncol. 1990;45:91-
98.
164. Sondergaard K, Olsen G. Malignant melanoma of the foot: a cl inico-
pathological study of 125 primary cutaneous malignant melanomas. Acto
Pathol Microbial Scand [A]. 1980;88:275-283.
165. Wells KE. Reintgen OS. Cruse CW. Tile current rnanagement and
prognosis of acral lentigi nous melanoma. Ann Pfost Surg. 1992;28: 100- 103.
166. Barnhill RL. Pathology and prognostic factors. Curr Opin Oncol. 1993:
5:364-376.
167. Blessing K. Kemohan NM, Park KG. Subungual malignant melanoma:
cl inicopathological features of I 00 cases. Histopathology. 1991 ; 19:425-
429.
168. Finley RK Ill, Driscoll DL, Blumenson LE, Karakousis CP. Subungual
melanoma: an eighteen-year review (published correction appears in Surgery.
1994;116:471]. Surgery. 1994:116:96- 100.
169. Gicle H. Hollowood K. Gibbons CL, Wi lson DJ. Athanasou NA.
Subungual melanoma with osteocarti laginous differentiation. Skeletal Radio/.
2003 Dec:32:724-727. Epub 2003 Oct 8.
170. Heaton KM, ei-Naggar A, Ensign LG, Ross Ml , Balch CM. Surgical
management and prognostic factors in with subungual melanoma. Ann
Surg. 1994;219: 197-204.
171. O' Leary JA, Berend KR, Johnson JL, Levin LS, Seigler HF. Subungual
melanoma: a review of 93 cases with identification of prognostic variables.
C/in Orthop Relat Res. 2000:378:206-212.
172. Panizzon R, Krebs A. Malignant subungual melanoma [in German) .
Hauwrv. 1980;3 1 :132-140.
173. Quinn MJ. Thompson JE. Crotty K . .McCarthy WH, Coates AS.
Subungual melanoma of the hand. J Hand Surg [Am/. 1996:21:506-5 11.
174. Clark WH Jr, Mihm MC Jr. Leotigo maligna and lentigo-maligna
melanoma. Am J Pat hoi. 1969:55:39-67.
175. Flotte T.l . Mihm MC Jr. Lentigo maligna and mal ignant melanoma in
situ, lentigo maligna type. Hum Pathol. 1999:30:533-536.
176. Koh HK, Michal ik E, Sober AJ, etal. Lentigo maligna melanoma has no
better prognosis than other types of melanoma. J Clin Oncol. 1984;2:994- 100 I .
177. McGovern VJ, Cochran AJ. Van der Esch EP, Little JH, MacLennan R.
The classification of malignant melanoma. its histologicaJ reporting and
registration: a revision of the 1972 Sydney c lass ification. Pathology.
1986: 18: 12-21.
178. Hutchinson J. Notes on the cancerous process and on new growths in
generaL Arch Surg. 1890:2:83-86.
179. Dubreui lh MW. Lentigo malin des viei!Jards. Ann D<!mzatol Syplril.
1894:5: 1092- 1099.
180. Acker SM. Nicholson JH, Rust PF, Maize JC. Morphometric dis-
crimination of melanoma in situ of sun-damaged skin from chronically sun-
damaged skin. J Am Acad Derm<llol. 1998;39(2, pt I ):239-245.
181. Weyers W, Bonczkowitz M. Weyers I. Bittinger A, Schi ll WB.
Melanoma in situ versus melanocytic hyperplasia in sun-damaged skin:
assessment of the signi fi cance of histopathologic criteria for di fferential
diagnosis. Am J Dermatopatlrol. 1996:18:560-566.
182. Tannous ZS, Lerner LH, Duncan LM, Mihm MC Jr, Flotte TJ.
Progression to invasive melanoma from mal ignant melanoma in situ, lentigo
maligna type. Hum Pathol. 2000;31:705-708.
183. Carlson JA, Dickersin GR. Sober AJ. Barnhill RL. Desmoplastic
neurotropic melanoma: a cl inicopathologic analysis of28 cases. Co11cer. 1995:
75:478-494.
184. Egbert B, Kempson R. Sagcbicl R. Desmoplastic malignant melanoma:
a clinicohistopathologic study of 25 cases. Cancer. 1988;62:2033-204 1.
185. Skelton HG. Smith KJ. Laskin WB. et al. Desmoplastic malignant
melanoma. J Am At(u/ Dermotol. 1995:32:717-725.
186. Conley J, Lanes R, Orr W. Desmoplastic malignant melanoma (a rare
variant of spindle cell melanoma). Cancer. 197 1;28:9 14-936.
187. Anstey A. Cerio R. Ramnarain N, Orchard G, Smith N. Jones EW.
Desmoplastic malignant melanoma: an imu1uuocytochem.ical study of 25
cases. Am J Dermatopatlrol. I 994:16: 14-22.
188. Jain S, Allen PW. Desmoplastic malignant melanoma and its variants: a
study of 45 cases. Am J Surg Patlr.ol. 1989;13:358-373.
189. Quinn MJ, Crotty KA. Thompson JF. Coates AS, O'Brien CJ.
McCarthy WH. Desmoplastic and desmoplastic neurotropic melanoma:
experience with 280 patients. Cancer. 1998:83:1128-1135.
190. Reiman HM. Goellner JR, Woods JE. Mixter RC. Desmoplastic
melanoma of the head and neck. Cancer. 1987;60:2269-2274.
191. Shendrik I, Silvers DN. Desmoplastic and desmoplastic neurotropic
melanoma: experience wit h 280 patients [letter]. Cancer. 1999:85:2491 -2492.
192. Walsh NM, Rober1S JT, Orr W, Simon GT. Desmoplastic mal ignant
melanoma: a clinicopathologic study of 14 cases. Arclr Patlrol Lab Med. 1988;
112:922-927.
193. Anstey A, McKee P. Jooes EW. Desmoplastic malignant melanoma: a
cli nicopathological study of 25 cases. Br J Derma to/. I 993; I 29:359-371.
Mayo Clin Pmc. March 2007;82(3):364-380 www.mayoclinicproceedings.com 377
194. Batsakis JG. Raymond AK. Desmoplastic melanoma. Anu 0101 Rhino/
Laryngol. 1994; )()3:77-79.
195. BM, McLeod GR, Little JH. Desmoplastic, neural trans-
fomlingand neurotropic melanoma: a review of 45 cases. Ausr N ZJ Surg. 1990;
60:967-972.
196. Brandt TP. Desmoplastic malignant melanoma. J Am Acad Dermatol.
1 996;34:712-713.
197. Bruijn JA, Mihm MC Jr. Barnhill RL. Desmoplastic melanoma.
Histopathology. 1992:20:197-205.
198. Bruijn JA. Sa!ascbe S. Sober AJ, Milun MC. Barnhill RL. Desmoplastic
melanoma: cl inicopathologic aspects of six cases. Dermatology. 1992:185:3-8.
199. Busam KJ. Cutaneous desmoplastic melanoma. Adv Anm PMhol. 2005;
12:92- 102.
200. Carlson JA. Egbert BM. Desmoplastic neurotropic malignant melano-
ma. Pathology (Phi/a). 1994;2:339-357.
201. Hawkins WG, Busam KJ , Ben-Porat L, eta!. Desmoplastic melanoma:
a pathologically and cl inically distinct form of cutaneous melanoma. Ann Surg
Oneal. 2005 Mar; 12:207-213. Epub 2005 Mar 3.
202. Frolow GR. Shapiro L. Brownstei n MH. Desmoplastic malignant
melanoma. Arch Demwtol. 1975; Il l :753-754.
203. Kossard S, Doherty E, Murray E. Neurotropic melanoma: a variant of
desmoplastic melanoma. Arch Dermatol. 1987: 123:907-912.
204. Lens MB, Newton-Bishop JA, Boon AP. Desmoplastic malignant
melanoma: a systematic review. Br J Dermatol. 2005; 152:673-678.
205. Livestro DP, Muzikansky A, Kaine EM, et al. BiolO!,'}' of desmoplastic
melanoma: a case-control comparison with other melanomas. J Clin Oncol.
2005:23:6739-6746.
206. Longacre T A, Egben BM, Rouse RV. and spindle-cell
malignant melanoma: an immunohistochemical study. Am J Surg Patlwl.
1996;20: 1489- 1500.
207. McCatt hy SW. Scolyer RA. Palmer AA. Desmoplastic melanoma: a
diagnostic trap for the unwary. Pathology. 2004;36:445-451.
208. Payne WG. Kearney R. Wells K. et al. Desmoplastic melanoma. Am
Surg. 2001:67:1004- 1006.
209. Posther KE, Selim MA, Mosca PJ, et al. Histopathologic characteristics,
recunence patterns, and survival of 129 patients with desmoplastic melanoma.
Ann Surg Oncol. 2006 May:l3:728-739. Epub 2006 Mar 17.
210. Reed RJ, Leonard DO. Neurotropic melanoma: a variant of desmo-
plastic melanoma. Am J Surg Pathol. 1979;3:30 1-31 1.
211. Scolyer RA, Thompson JF. Desmoplastic melanoma: a heterogeneous
entity in which subclassification as '"pure" or 'mixed" may have important
prognostic signi fi cance [editorial j. Ann Surg Oneal. 2005 Mar; 12:197-199.
Epub 2005 Mar 3.
212. Tsao H, Sober AJ, Barnhill RL. Desmoplastic neurotropic melanoma.
Semin Cutan Med Surg. 1997;16: 131- 136.
213. Weinzweig N, Tuthill RJ, Yetman RJ. Desmoplastic malignant mel-
anoma: a clinicohistopathologic review. Plast Reconstr Surg. !995:95:548-555.
214. Regauer S. Anderhuber W, Richtig E. Schachenrcitcr J. 011 A. Beham
A. Primary mucosal melanomas of the nasal cavity and paranasal sinuses: a
clinicopathological analysis of 14 APM/S. 1998:106:403-410.
215. Bongiorno MR, Arico M. Primary malignant melanoma of the oral
cavity: case report. lnt J Dermatol. 2002;41 :178-181.
216. DeMatos P, Tyler OS. Seigler HF. Malignant melanoma of the mucous
membranes: a review of I 19 cases. Anti Surg Oncol. 1998:5:733-742.
217. Saida T. Kawachi S. Takata M, ct al. Histopathological characteristics
of malignant melanoma affecting mucous membranes: a uni fying concept of
histogenesis. Pathology. 2004:36:404-413.
218. Tomicic J. Wanebo HJ. Mucosal melanomas. Surg C/in North Am.
2003;83:237-252.
219. Gutman M, Inbar M, Chaitchik S, et al. Malignant mehmoma of the
mucous membranes. Eur J Surg Oneal. 1992:18:307-312.
220. GranterS, McKee P, Calonge E . .Mihm MC Jr. Busam K. Melanoma
associated wid1 blue nevus and melanoma mimcking cellular blue nevus: a
clinicopathologic study of l 0 cases on the spectrum of so-called "malignant
blue nevus .. , Am J Surg Pathol. 2001:25:316-323.
221. Mehregan DA, Gibson LE, Mehregan AH. Mal ignant blue nevus: a
repon of eight . ./ Dermarol Sci. 1992;4: 185-192.
222. Boi S, B;ubareschi M. Vigl E, Cristofolini M . .Malignant blue nevus:
report of four new cases and review of the literature. Histol Histoparhol.
1991 ;6:427-434.
223. Fisher ER. Malignant blue nevus. AMA Arch Denn. 1956;74:227-231.
224. Pich A. Chiusa L. Margaria E. Aloi F. Proliferative activity in the
malignant cellular blue nevus. Hum Pathol. 1993;24: 1323 1329.
225. Hernandez FJ. Malignant blue nevus: a light and electron microscopic
study. Arch Demwtol. 1973; I 07:741-744.
226 Connelly J. Smith JL Jr. Malignant blue nevus. Cancer. 1991 ;67:2653-
2657.
227. Aloi F. Pich A, Pippione M. Malignant cell ular blue nevus: a
cl inicopathological study of 6 cases. Denuatology. 1996; 192:36-40.
228. Duteille F, Duport G. Larregue M. Neau A, Duriez P, Herve MC.
Malignant blue nevus: three new cases and a review of the literature. Ann Pfost
Surg. 1 998;41 :674-678.
229. Zembowicz A. McCusker M. Clliarelli C. et al. Morphological analysis
of nevoid melanoma: a study of 20 cases with a review of the literature. Am J
Dermatopathol. 2001:23:167-175.
230. Cassarino OS. Fullen DR. Sondak VK. Duray PH. Metastatic nevoid
melanoma in a 4 In -year-old child. J Cutan Pat/to/. 2003;30:647-651 .
231. Pereira F, Carey W, Shihata H. Burnier MN Jr, Wang B. Multiple
nevoid mal ignant melanomas in a patient with AIDS: the role of proliferating
cell nuclear antigen in the diagnosis . .I Am Acad Dermato/. 2002;47(2,
suppi):S 172-S 174.
232. Wong TY, Suster S. Duncan LM. Mi hm MC Jr. Nevoid melanoma: a
clinicopathological study of seven cases of malignant melanoma mi micking
spindle and epithelioid cell nevus and verrucous dermal nevus. Hum Patlwl.
1.995;26: 171 -179.
233. Schmoeckel C, Castro CE, Braun-Falco 0. Nevoid malignant mela-
noma. Arch Dermatol Res. 1985;277:362-369.
234. Magro CM, Crowson AN, Mihm MC. Unusual variants of malignant
melanoma. Mod Parhol. 2006; 19(suppl 2):S41 -S70.
235. Mones JM, Ackerman AB. "Atypical' " Spitz's nevus. "malignant"
Spitz's nevus. and '"metastasizing'' Spitz's nevus: a critique in historical
perspective of three concepts tlawed fatally. Am J Dermatopathol. 2004;26:
310-333.
236. Kaul ich TW. Zurheide J, Haug T. Budach W. Nusslin F. Bamberg M.
On the actual state of industrial quality assurance procedures with regard to
(106)Ru ophthalmic plaques. Srrahlenther Onkol. 2004;!80:358-364.
237. Chorny JA, Barr RJ, Kyshtoobayeva A, JakowatzJ, Reed RJ. Ki-67 and
p53 expression in minimal deviation melanomas as compared with other
nevomelanocytic lesions. Mod P(lfho/. 2003: 16:525-529.
238. Podnos YD. Jimenez JC, Zainabadi K. Jakowatz JG. Barr RJ. Minimal
deviation melanoma. CMcer Trem Rev. 2002;28:219 221.
239. Reed RJ. Mi nimal deviation melanoma: borderli ne and intermediate
melanocytic neoplasia. Clin Lab Med. 2000;20:745-758.
240. Kim JC, Murphy GF. Dysplastic melanocytic nevi and prognosticall y in-
determinate nevomelanomatoid proliferations. C/in Lab Med. 2000;20:691-
712.
241. Bohn J. Svensson H. Aberg M. Dermabrasion of large congenital
melanocytic naevi in neonates. Scand J Plasr Reconsrr Surg Hand Surg.
2000:34:321-326.
242. Stas M. van den Ood JJ. Garmyn M. Degreef H, De Wever l. De Wolf-
Peeters C. Minimal deviation and/or naevoid melanoma: is recognition
worthwhile? a clinicopathological study of nine cases. Melanoma Res.
2000; I 0:371 -380.
243. Reed RJ. Dimensional ities: borderline and intermediate melanocytic
neopla5ia. Hum Pat hoi. 1999;30:521-524.
244. LeBoit PE. Minimal deviation melanoma: concept or quagmire? Adv
Dermarol. 1997: 13:289-304.
245. Reed RJ, Webb SV, Clark WH Jr. Minimal deviation melanoma (halo
nevus variant). Am J Surg Pathol. 1990;14:53-68.
246. Donner LR. Manriquez M. Greene JF Jr. Minimal deviation spindle cell
melanoma: unusual histologic pattern in an 11-year-old black girl. Pediatr
Pmhol. 1988:8:401-407.
247. Reed RJ. Minimal deviation melanoma. Monogr Patlwl. 1988;30: 110
152.
248. Voll mer RT. Minimal deviation melanoma (letter]. Hum Patho/.
1987; 18:869-870.
249. Phillips ME, Margolis RJ, Merot Y. et al. The spectrum of minimal
deviation melanoma: a clinicopathologic study of 21 cases. Hum Parhol.
1986;17:796-806.
250. Briggs JC. Melanoma precursor lesions and borderline melanomas.
Hisropmhology. 1985:9:1251-1 262.
251. Merot Y, Mihm MC Jr. Unusual and unknown aspect of cutaneous
malignam melanoma: mi ni mal deviation malignam melanoma: retrospective
study of 45 cases [in French]. Ann Dermatol Venereol. 1985;112:325-336.
252. Barr LH. Goldman Ll . Solomon JA, Sanusi Dl , Reed RJ. Minimal
deviation melanoma. Surg Gyneco/ Obstet. 1984:159:546-548.
f-or persona. 1\,a.>S I efJrvdLC on.y wrth rJt.,ni.>S!Ot1 frorr, Mayo L.'tnic Proceedmgs.
253. Muhlbauer JE. Margolis RJ. Mihm MC Jr. Reed RJ. Mittimal deviation
melanoma: a histologic variant of cutaneous mal ignant melanoma in its vertical
growth phase. J fnves1 Dermarol. I983:80(suppl):63s65s.
254. Hendrickson MR, Ross JC. Neoplasms arising in congenital giant nevi:
morphologic study of seven cases and a review of the literature. Am J Surg
Pmho/. 1981:5: 109-135.
255. Warner TF, Seo IS, Bennett JE. Minimal deviation melanoma with
epidermotropic metastases arising in a congenital nevus. Am J Surg Pmhol.
1980:4: 175-183.
256. Becker DW Jr, Miller CJ, Keller HB. Metastatic minimal-deviation
melanoma. Am1 Plosr Surg. 1980:4:230..237.
257. Reed RJ. Minimal deviation malignant melanoma arisi ng in a
congenital nevus. Am J Surg Po rho/. 1978;2:215-220.
258. Banerjee SS. Harris M. Morphological and immunophenotypic
variations in mal ignant melanoma. Hisropmhology. 2000;36:387-402.
259. Blessing K, Grant JJ, Sanders DS, Kennedy MM, Husain A, Cobum P.
Small cell malignant melanoma: a variant of naevoid melanoma: clinico-
pathological features and histological differential diagnosis. J C/in Parho/.
2000;53:591-595.
260. Cuellar FA, Vilalta A, Rull R, et al, Malignant Melanoma Group. Small
cell melanoma and ulceration as predictors of positive sentinel lymph node in
malignant melanoma patients. Melanoma Res. 2004:14:277-282.
261. Miura K, Harada H. Tsutsui Y. Small cell type malignant melanoma
which developed in a 16-ycar-old female with a congenital dermal nevus and
metastasized 12 years after excision. Parho/111/. 1999;49:247-252.
262. Swetter SM. Ecker PM. Johnson DL. Harvell JD. Primary dermal
melanoma: a distinct subtype of melanoma. Arch Dermc11ol. 2004;140:99- 103.
263. Taj ima Y, Nakajima T, Sugano I, Nagao K, Kondo Y. Malignant mel-
anoma within an intradermal nevus. Am J Dermatoparhol. 1994; 16:3() 1- 306.
264. Nakhleh RE, Wick MR, Rocamora A, Swanson PE, Dehner LP.
Morphologic diversity in malignant melanomas. Am.! C/in !'athol. 1990:93:731 -
740.
265. Allan SJ, Dicker AJ. Tidman MJ. McLaren Klvl. Hunter JA. Amclanotic
lentigo maligna and amelanotic lentigo maligna melanoma: a report of three
cases mimicki ng intraepidermal squamous carcinoma . .I Eur Acad Dermato/
Venereal. 1998:11:78-81.
266. Oguri H, Izumiya C, Maeda N. Fukaya T. Moriki T. A primary
amelanotic melanoma of the vagina, diagnosed by immunohistochemical
staining with HMB-45, which recurred as a pigmented melanoma. J Clin
Pat hoi. 2004:57:986-988.
267. Ulmer A, Dietl J. Schaumburg-Lever G, Fierlbeck G. Amelanotic
malignant melanoma of the vulva: case report and review of the literature. Arch
Gynecol Obster. 1996;259:45-50.
268. Clark WI-I Jr. Four types of cellular fine structure associated with
human amelanotic melanoma. Yale J Bioi Med. 1973;46:428.
269. Urso C. Giannotti B, Bondi R. Myxoid melanoma of the skin. Arch
Patlwl Lab Med. l 990;114:527-528.
270. Sarode VR. Joshi K. Ravichandran P. Das R. Myxoid variant of primary
cutaneous malignant melanoma. Histopathology. 1992;20: 186-187.
271. Patel P, Levin K, Waltz K. Helm KF. Myxoid mel<moma: immunohisto-
chemical studies and a review of the literature. JAm Ac(ul Dennarol. 2002:46:
264-270.
272. Olinici CD. Vasiu R. Myxoid malignant melanoma: repmt of three
cases and review of the literature. Morpho/ Embryo/ (Bucur). 1989;35:285-
287.
273. Hitchcock MG, McCal mont TH, White WL. Cutaneous melanoma with
myxoid features: twelve cases with di fferential diagnosis. Am J Surg Pmhol.
1999:23:1506- 1513.
274. Auger M. Caraway NP, Wojcik EM, OrdonezNG. Fine needle aspiration
biopsy of metastatic myxoid melanoma. CyJOpmhology. 1994;5: 118-122.
275. Hitchcock MG, White WL. Malicious masquerade: myxoid melanoma.
Semi11 Diagn Par hoi. 1998; 15: 195-202.
276. Prieto VG, Kanik A, Salob S. McNutt NS. Primary cutaneous myxoid
melanoma: immunohistologic clues to a difficult diagnosis. J Am Acad
Dermaro/. 1994:30:335-339.
277. Urso C. Cutaneous melanoma with pseudomyxoid features. J Cman
Pmho/. 2006:33:312-314.
278. Breier F, Feldmann R. Fellenz C, Neuhold N, Gschnait F. Primary
invasive signet-ring cell melanoma. J Cu1a11 Par hoi. 1999;26:533-536.
279. Rutten A, Huschka U, Requena C. Rodriguez-Peralto JL, Requena L.
Primary cutaneous signet-ring cell melanoma: a clinico-pathologic and
immunohistochemical study of two cases. Am J Dermaropmhol. 2003:25:418-
422.
280. Niemann TH. Thomas PA. Melanoma with signet-ring cells in a
peritoneal effusion. Diagn Cy10parlrol. 1995:12:241-244.
281. Won JH, Aim SK, Lee SH, Lee WS. Kim SC. Signet-ring cell
melanoma: poor prognostic factor? [letterJ. B1 J Dermarol. 1994; 131: 135-137.
282. Botev IN. Metastatic signet-ring cell melanoma [letter!. J Am Acad
Dennmol. 1993;29:51 0-511.
283. Tsang WY, Chan JK, Chow LT. Signet-ring cell melanoma mimicking
adenocarcinoma: a case report. Acw Cyro/. 1993;37:559-562.
284. LiVolsi VA. Brooks JJ. Soslow R, Johnson BL. Elder DE. Signet cell
mclanocytic lesions. Mod Parlwl. 1992:5:515520.
285. Eckert F, Baricevic B. Landthaler M. Schmid U. Metastatic signet-ring
cell melanoma in a patient with an unknown primary tumor: histologic,
immunohistochemical, and ultrastructural tindiogs. J Am Acad Dermarol.
1992;26(5. pt 2):870-875.
286. al-Talib RK, Theaker JM. Signet-ring cell melanoma: light micro-
scopic, immunohistochemical and ultrastntctural feat ures. Hisroparhology.
1991; 18:572-575.
287. Sheibani K, Battifora H. Signet-ring cel l melanoma: a rare morphologic
variant of malignant melanoma. Am J Surg Pathol. 1988; 12:28-34.
288. Bonelli F, Colombari R, Zamboni G. Chilosi M. Signet ring melanoma.
S- 100 negative neuerl. Am J Surg Pmhol. 1989;13:522-523.
289. Adamek D, Kaluza J. Stachura K. Primary balloon cell malignant
melanoma of the right temporo-parietal region arising from meningeal naevus.
Clin NeuropMhol. 1995:14:29-32.
290. Gardner WA Jr. Vazquez MD. Balloon cell melanoma. Arch Parhol.
1970:89:470-472.
291. Kao GF. Helwig EB, Graham JH. Balloon cell malignant melanoma of
the skin: a clinicopathologic study of 34 cases with histochemical. immuno-
histochemical, and ultrastructural observations. Cancer. 1992;69:2942-2952.
292. Mmtinez F. Merenda G. Bedrossian CW. Lipid-rich metastatic balloon-
cell melanoma: diagnosis by a multimodal approach to aspiration biopsy
cytology. Diagn Cylopathol. 1990;6:427-433.
293. Abbott JJ. Amirkhan RH, Hoang MP. Malignant melanoma with a
rhabdoid phenotype: histologic. immunohistochemical, and ultrastructural
study of a case and review of the literature. Arch Patho/ Lab Med.
2004:128:686-688.
294. Bittesini L. Dei Tos AP. Fletcher CD. Metastatic malignant melanoma
showing a rhabdoid phenotype: further evidence of a non-specific histological
pall em. HisiOpathology. 1992:20: 167-170.
295. Borek BT, McKee PH. Freeman JA. Maguire B, Brander WL, Calonje
E. Primary malignant melanoma with rhabdoid features: a histologic and
immunocytochemical study of three cases. Am J Dermatoparhol. 1998;20: 123-
127.
296. Chang ES, Wick MR, Swanson PE, Dehner LP. Metastatic malignant
melanoma with "rhabdoid" features. Am .I C/in Patlwl. 1994; I 02:426-43 1.
297. Laskin WB. Knittel DR, FrameJN. S100 protein and HMB-45 negative
''rhabdoid" malignant melanoma: a totally dedifferentiated malignant mel-
anoma? Petter]. Am J Clin Pathol. 1995; 103:772-773.
298. Crowson AN, Magro CM, Mihm MC Jr. Malignant melanoma with
prominent pigment synthesis: animal type'' melanoma- a clinical and
histological study of six cases with a consideration of other melanocytic
neoplasms with prominent pigment synthesis. Hum Patfwl. 1999;30:543-550.
299. Friedmann I. Osteoid and bone formation in a nasal mucosal melanoma
and its metastasis [letter]. Histopatl1ology. 1998:33:88.
300. Homweg JJ, Loftus BM, Hilgers FJ. Osteoid and bone formation in a
nasal mucosal melanoma and its metastasis. Histoparlwlogy. 1997;3 1 :465-468.
301. Lucas DR, Tazclaar HD. Unni KK. et al. Osteogenic melanoma: a rare
variant of malignant melanoma. Am J Surg Pmhol. 1993; 17:400-409.
302. Moreno A, Lamarca J, Martinez R, Guix M. Osteoid and bone
format ion in desmoplastic malignant melanoma. J Curan Par/wl. 1986;13: 128-
134.
303. Pellegrini AE, Scalamogna PA. Malignant melanoma with osteoid
formation. Am J Demwropathol. 1990;12:607-611.
304. Prasad ML, Patel SG. Busam KJ. Primary n1ucosal desmoplastic
melanoma of the head and neck. Head Neck. 2004:26:373-377.
305. Katenkamp K, Henke RP, Katenkamp D. Dermal chondroid melanoma
[in German]. Porhologe. 2005:26:149-152.
306. Ackley CD. Prieto VG, Bentley RC, Horenstein MG. Seigler HF. Shea
CR. Primary chondroid melanoma. J Curan Parhol. 200 I ;28:482-485.
307. Grunwald MH, Rothem A. Desmoplastic cartilagi nous formation in
malignant melanoma [letter]. J Cwan Parhol. 1987;14:255.
308. Cachia AR. Kedziora AM. Suhungual malignant melanoma with
differentiation. Am J Dermmoparhol. 1999:21:165-169.
For personal .... se. Mass reprodLce c.,r.; WI t pe. n .. ssh.H. Proceeding.;;,.
309. Banetjee SS. Coyne JD. Menasce LP. Lobo C.J , Hirsch PJ. Diagnostic
lessons of mucosal melanoma with osteocartilaginous differentiation.
Histopathology. 1998:33:255-260.
310. Rodriguez J, Nonaka D. Kuhn E. Reichel M, Rosai J. Combined high-
grade basal cell carcinoma and malignant melanoma of the skin (''malignant
basomelanocytic tumor''): report or two cases and review or the literature. Ami
Dermaropmhol. 2005;27:314-318.
311. Erickson LA, JL. Mihm MC, Markovic SN, Pittelkow MR.
Malignant basomelanocytic tumor manifesting as metastatic melanoma. Am J
Surg Parlwl. 2004:28:1393- 1396.
312. Benda JA. Neurotropic desmoplastic melanoma of the vulva Lletter].
Oynecol Oncol. 1 997;64: 180.
313. Kurihara K. Sanada E, Yasuda S, Yamasaki H. Desmoplastic malignant
melanoma of tJ1e gingiva. Oral Surg Oral Med Oral Pothol. 1992:74:201-
205.
314. Rogers RS lll, Gibson LE. Mucosal, genital, and unusual cl inical
variants of melanoma. Mayo Cli11 Proc. 1997:72:362-366.
315. Orosz Z. Melan-A/Mart-1 expression in various melanocytic lesions
and in non-melanocytic soft tissue tumours. Histopathology. 1999;34:517-
525.
316. Kaneishi NK. Cockerell CJ. Histologic differentiation of desmoplastic
melanoma from cicatrices. Am./ Dermalopathol. 1998:20:128- 134.
317. Robson A. Allen P, HoJJowood K. S 100 expression in cutaneous scars:
a potential diagnostic pitfall in the diagnosis of desmoplastic melanoma.
Histoparlwlogy. 2001 ;38: 135-140.
318. Bhuta S, Mi JTa JM, Cochran AJ. Myxoid malignant melanoma: a
previously undescribed histologic pattern noted in metastatic lesions and a
report of four cases. Am J Surg Pmhol. 1986: I 0:203-211.
319. Garcia-Caballero T, Fraga M, Antunez JR, Perez-Becerra E, Beiras A.
Forteza J. Myxoid metastatic melanoma. Histopathology. 1991; 18:371-373.
320. McCluggage WG, Shah V, Toner PG. Primary cutaneous myxoid
malignant melanoma. HistOf>athology. 1996;28: 179-182.
321. Collina G, Losi L, Taccagni GL, Maiorana A. Myxoid metastases of
melanoma: report of three cases and review of the literature. Am J Dermaro
pathol. 1997;19:52-57.
322. Gupta D. Deavers MT, Silva EG, Malpica A. Malignant melanoma
involving the ovary: a clinicopathologic and immunohistochemical study of 23
cases. Am J Surg Parhol. 2004:28:771 780.
323. Lodding P. Kindblom LG, Angervall L. Metastases of malignant
melanoma simulating sort tissue sarcoma: a clinico-pathological, light- and
electron microscopic and immunohistochemical study of 21 cases. Virchows
Arch A Pmlwl Anal Histoparhol. 1990;417:377-388.
324. Biernat W, Kordek R, Wozniak L. Cellular blue nevi with myxoid
change-diagnostic difficulties and the review of the literamre. Pol J Pathol.
1995;46:83-86.
325. Hoang MP. Myxoid Spitz nevus . .I Cwan Pathol. 2003:30:566-568.
326. Michal M. Mucinous naevocellular naevi. Dermatology. 1994; 188:232-
235.
327. Zeiger BG, Steiner H, Wambacher B, Zeiger B. Malignant melanomas
simulating various types of soft tissue tumors. Dennatol Surg. 1997;23: 1047-
1054.
328. Rosenberg AS, Langee CL. Stevens GL. Morgan MB. Malignant
peripheral nerve sheath tumor with perineurial differentiation: "malignant
perineurimna." J Cwcm Pathol. 2002:29:362-367.
329. Reis-Filho JS. Silva P. Milanezi F. Lopes JM. Hyaline cell-rich
chondroid syringoma: case report and review of the literatu re. Parhol Res
Pract. 2002; 198:755-764.
330. Patel SR, Burgess MA. Papadopoulos NE. Linke KA, Benjamin RS.
Extraskeletal myxoid chondrosarcoma: long-term experience with chemo
therapy. Am J C/in Oncol. 1995;18: 161-163.
331. Patel SR. Plager C, Papadopoulos NE, Benjamin RS. Myxoid malignant
fibrous histiocytoma: experience with chemotherapy. Am J Clin Oneal. 1995:
18:528-53].
332. Patel SR, Burgess MA, Plager C, Papadopoulos NE, Linke KA,
Benjamin RS. Myxoid liposarcoma: experience with chemotherapy. Cancer.
1994;74: 1265-1269.
333. Nasiell K, Tani E, Skoog L. Fine needle aspiration cytology and im-
munocytochemistry of metastatic melanoma. Cytopathology. 1991 :2: 137-147.
334. Lindholm K. de Ia Torre M. Fine needle aspiration cytology of royxoid
metastatic ma.lignam melanoma. A era C.wol. 1988;32:719-721.
335. Kim YC, Vandersteen DP, Jung HG. Myxoid clear cell sarcoma. Am .I
Dermmopathol. 2005;27:51-55.
336. EUiott D, Pitman MB. Mal ignant melanoma with a myxoid stroma: a
diagnostic pitfall on fine-needle aspiration biopsy. Diagn C:wopmhol. 2001:
25:185-190.
337. Orosz z. Pitfalls in the diagnosis of malignant melanoma [in Hungarian].
Magy Onkol. 2003;47:27-39. Epub 2003 Apr 18.
338. Pool SE. Manieei F. Clark WH Jr, Harrist TJ. Dermal squamo-
melanocytic nnnor: a unique biphcnotypic neoplasm of uncertain biological
potential. Hum Pachal. 1999:30:525-529.
339. Pierard GE. Fazaa B, Henry F. Kamoun MR, Pierard-Fnmchimont C.
Collision of primary malignant neoplasms on the skin: the connection between
malignant melanoma and basal cell carcinoma. Dennarology. 1997:194:378-
379.
340. Busam KJ , Halpern A, Marghoob AA. Malignant melanoma metastatic
to a basal cel l carcinoma simulating the pattern of a basomclanocytic tumor.
Am J Surg Pathol. 2006:30: 133-136.
341. Burkhalter A, White WL. Malignant melanoma in situ colonizing basal
cell carcinoma: a simulator of i11vasive melanoma. Am .I Dermatopathol.
1997;1 9:303-307.
The Symposium on Solid Tumors will continue in the April issue.
f-or persona 1\aa.>S I efJrvdLC on.y wrth rJtant.:>SaOt1 frolli Mayo L 'mic Proceedmgs.
VOLUME 5: NO. 4 OCTOBER 2008
ORIGINAL RESEARCH
Enforcement of State Indoor Tanning Laws
in the United States
Joni A. Mayer, PhD, Katherine D. Hoerster, BA, Latrice C. Pichon, MPH, Debra A. Rubio, BA,
Susan I. Woodruff, PhD, Jean L. Forster, PhD, MPH
Suggested citation for this article: Mayer JA. Hoerster
KD, Pichon LC. Rubio DA, Woodruff SI, Forster JL.
Enforcement of state indoor tanning laws in the United
States. Prev Cluonic Dis 2008;5(4). http://ll\rww.cdc.gov/
pcd/issues/2008/oct/07 _ 0194. htm. Accessed [date].
PEER REVIEWED
Abstract
Introduct ion
Twenty-eight US states have passed legislation for
indoor tanning facilities. To our knowledge, whether
these state laws are actually enforced has not been
evaluated previously in all 28 states. Therefore, we
interviewed key informants in these states to assess
enforcement practices.
Methods
Two trained interviewers used a structured survey
instrument to interview 28 key informants who were
knowledgeable about enforcement practices for laws
regarding indoor tanning. Respondents provided informa-
tion specific to the most populous city in their states.
Results
Licensure for indoor tanning businesses was required
in 22 of the 28 cities. Slightly less than half of the cities
gave citations to tanning facilities that violated state law.
Approximately 32% of the cities did not inspect indoor tan-
ning facilities for compliance with state law, and another
32% conducted inspections less t han annually. Of t hose
cities that inspected at all, most conducted unannounced
inspections.
Conclusion
The relatively low rates of annual inspections and cita-
tions are of concern. We recommend that future studies
assess whether legislation, enforcement practices, or a
combination of the 2 affects the practices of indoor tanning
facilities or of consumers.
Introduction
Indoor tanning with UV radiation lamps has been
linked to melanoma (1), squamous cell carcinoma (1),
molecular damage associated with skin cancer (2), and
other acute damage to eyes and skin (3,4). Commercial
indoor tanning facilities are prevalent in the United
States (5), and "all-you-can-tan" discount pricing pack-
ages make indoor tanning inexpensive (6). The rates of
indoor tanning for teen girls in the United States are
high (7-10); in a national sample, approximately 40% of
17- to 18-year-old girls had used indoor tanning in the
past year (7).
Some US states have passed legislation regulating
indoor tanning facilities, with the intent of reducing risks
to consumers. Ongoing systematic updates on the number
and content of these laws have been provided, focusing on
youth access restrictions (11-13). A recent report quanti-
fied the stringency of state indoor tanning legislation in
the 28 states that had a state law as of2006 (14). However,
in order to assess the level at which the laws are imple-
mented and the effect of these laws on the industry and
consumers, information about enforcement practices is
needed. Consequently, we conducted telephone interviews
of key informants in states with indoor tanning legislation
to assess enforcement practices.
The opinions expressed by authors contributing to this journal do not necessarily reflect the opinions of the US Department of Health and Human Serv1ces,
the Public Health Serv1ce, the Centers for D1sease Control and Prevention, or the authors' affiliated mstitut1ons. Use of trade names IS for identification only
VOLUME 5: NO. 4
OCTOBER 2008
Methods
Settings and participants
The CITYlOO (Correlates of Indoor Tanning in Youth)
project assesses factors that may influence use of indoor
tanning by adolescents (10,14); 1 objective is to better
understand current legislation that pertains to indoor tan-
ning. In the current study, we targeted the most populous
city in each of the 28 states to evaluate how state laws are
enforced at the local level.
Our goal was to interview, by telephone, the person who
was t he most knowledgeable about enforcement prac-
tices in each city or county. From a list of contacts for each
state's legislation presented on a Web site operated by the
tanning industry (15), we telephoned these contacts (typi-
cally at the state or county health department) and asked
them to identify the best key informant. The process for
identifying each city's respondent continued until we found
a knowledgeable potential 1espondent. We then mailed
an introductory letter to each potential respondent that
explained the purpose ofthe study and its voluntary nature
and assured anonymity of t he respondent and that data
would not be linked to the city's name in any published
reports. Approximately 1 week after mailing the letter, we
attempted to contact informants until we reached them
and they completed the interview. The 2 interviewers
(K.D.H. and L.C.P.) had previous experience in conducting
telephone interviews and received training for this study.
Survey
The survey questions were based on a combination of
previous study in this area (16), expert opinion about mea-
suring enforcement activities in the tobacco control area
(a good model for indoor tanning), and select enforcement
and monitoring activities mentioned in the indoor tanning
laws (14). Initially, we developed a longer version of the
survey that asked for specific data on various activities
(eg, number of facilities inspected in the previous year,
number of complaints received). That version assumed a
high level of inspection and other enforcement activities,
assumed that enforcement agencies kept detailed records
of those activities, and requested that informants obtain
that information before the interview and provide it during
the interview. During the informant identification process,
we became aware that the level of enforcement activities
was fairly low. Therefore, to better match the depth of our
assessment to actual practices, reduce the amount of work
required of respondents, and achieve a higher response
rate, we retained only the basic items and eliminated the
more elaborate, labor-intensive items.
The following factors were assessed: number of staff allo-
cated to carry out enforcement activities in the city/county;
whether indoor tanning businesses were required to have
a license; frequency of inspections (in absence of a com-
plaint); whether inspections were announced in advance;
whether inspection included review of customer records
and, if so, whether customer's age, parental consent forms,
number and dates of tanning sessions, and duration of
sessions were examined; and whether businesses received
citations when they violated the law. We also assessed the
types of penalties for selling sessions to underage youth or
not obtaining parental permission for minors and whether
graduated penalties (more severe penalties for each suc-
cessive violation) were used. A copy of the survey is pro-
vided in the Appendix.
A draft of the survey was reviewed for clarity by 2 public
health department professionals. All survey procedures
and materials were approved by t he institutional review
borud at San Diego State University. Interviews were con-
ducted in April and May of 2007.
Descriptive statistics, including frequencies or means,
were computed for each vruiable. Additionally, we con-
ducted bivariate tests (x
2
and correlations) to assess the
associations between the stringency of the written law (14)
and reported enforcement practices. Specifically, we exam-
ined the relationship between reported inspection fiequen-
cy and overall law stringency score, youth access subscore,
enforcement subscore, and 1 individual inspection item.
For the items about penalties specific to youth access, we
computed frequencies for only the 21 cities in states with
youth access laws. We did not perform multivariate tests
because of the small sample size. All analyses were con-
ducted in SPSS 13.0 (SPSS Inc, Chicago, Illinois).
Results
Response rates and respondent characteristics
We identified 28 respondents (1 for the most populous
VOLUME 5: NO. 4
OCTOBER 2008
city in each of the 28 targeted states) who confirmed that
they were knowledgeable about enforcement practices.
Data were obtained for all 28 cities. If respondents told
the interviewer that their states had no law (n = 5) or that
the cities engaged in no enforcement activities (n = 2), the
interviewer contacted additional informants to confirm
that laws were not enforced. The interviewer then coded
the remaining survey items to indicate nonperformance of
enforcement activities.
Enforcement resources and practices
More than three-fourths of the respondents were
employed by a state or local health agency (Table 1). The
organizations that employed the respondents also consti-
tuted the primary enforcement entity for the state indoor
tanning legislation in the designated city.
The number of full-time employees available for inspec-
tions and other enforcement activities ranged from 0 to 15,
with a mean of 3.29 (standard deviation 3.89) staff and a
median of 2. Approximately 29% of the cities had no full-
time enforcement staff (Table 2). Licensure for indoor tan-
ning businesses was required in most cities. Slightly less
than half of the cities gave citations (ie, penalties) to tan-
ning facilities that violated the state law. Approximately
32% of the cities did not inspect indoor tanning facilities
for compliance with the state law, and another 32% con-
ducted inspections less than annually. Of those cities that
inspected, most conducted unannounced inspections.
Of the 19 cities that conducted inspections, most reviewed
customer records as part of the inspection process. Of
these, most reviewed information about customers' ages,
parental consent forms, number and dates of tanning ses-
sions, and tanning session duration (Table 3).
Of the 21 cities in states that had youth access laws,
approximately half penalized these violations (Table 4).
Warnings, monetary fines, and license suspensions were
used for both kinds of youth access violations, with no
strong predominance by type of penalty. Of the cities that
penalized violations, most gave graduated penalties for
each of the youth access- related violations, in which each
additional violation results in a larger penalty.
Bivariate associations
We conducted Pearson correlational tests between
inspection frequency and the variables from an earlier
assessment of state indoor tanning laws (14). These cor-
relations (N = 28) were 0.51 (P = .006) for enforcement
subscore, 0.34 (P = .075) for minor's access stringency sub-
score, and 0.58 (P = .001) for overall law stringency score.
Reported inspection frequency was positively correlated
with the number of full-time enforcement staff reported
by the respondent (r = 0.48, P = .Oll). We then dichoto-
mized reported inspection frequency (less than annually
vs at least annually) and the individual inspection item
score from the earlier analysis of state laws (less strict vs
more strict) . These variables were significantly associated
(x
2
= 5.18, P = .023) . Of cities whose laws on inspections
were less strict (n = 21), only 23.8% conducted inspections
at least annually. Of those whose inspection requirement
was stricter (n = 7), 71.4% conducted inspections at least
annually. A license requirement in the written law was
significantly associated with actual (self-reported) license
requirement (x
2
= 5.06, P = .024). In cities in which the
state law did not mention licensure (n = 6), 50% required
licensure, whereas in cities whose law mentioned licensure
(n = 21), 90.5% required licensure.
Discussion
To our knowledge, this article is 1 of only 3 to report
actual enforcement practices related to state indoor tan-
ning laws (16, 17) and the only article to date that pro-
vides enforcement information for all 28 states. Our data
indicate that routine annual inspections, which are a pre-
requisite for other enforcement activities such as levying
penalties for violations, are not conducted in 64% of the cit-
ies. However, for those cities that conduct regular inspec-
tions, most conduct unannounced inspections, which likely
increases their effectiveness. Additionally, the annual
inspections routinely included review of client records and
encompassed information that may reflect UV radiation
exposure levels (eg, duration and frequency of sessions)
and youth access (eg, customer age and parental consent
forms) . Thus, the annual inspections appear to be of high
quality. The relationship between inspection frequency
and staffing level suggests that cities that do not conduct
annual inspections need more resources. However, we can-
not infer causality because of the study's cross-sectional
design. Results from a study of sanitarians within 1 large
metropolitan area in both Massachusetts (21 munici-
palities of Boston) and Minnesota (21 jurisdictions of the
Twin Cities) indicated the rates of routine inspections
VOLUME 5: NO. 4
OCTOBER 2008
by Massachusetts agencies were higher than those in
Minnesota (89.9% vs 28.6%) (16). Responses from state
health department staff in Texas, lllinois, and Wisconsin
showed a large amount of variability (for each state as a
whole) on both penalties to facilities for youth access viola-
tions and facility inspection/auditing practices (17).
The low rates of penalizing any violation and youth
access violations are also cause for concern. As is the case
with enforcement activities regarding tobacco and alcohol
control, businesses rue less likely to comply with age-of-
sale laws if noncompliance is not penalized (18, 19).
The strong association between various aspects of the
law and enforcement activities is encouraging. However,
in the 7 states with more stringent inspection require-
ments, the inspection requirements tended to overestimate
the actual level of inspections. Moreover, 5 respondents
incorrectly reported that their states, at the time of the
interview, had no law on indoor tanning. Therefore, even
though some states had laws, they were not being enforced
even minimally.
One limitation of this study is that the data were based
on the report of enforcement professionals, and we did
not attempt to verify them with other measures such as
interviews with tanning facility managers. Although we
assured their anonymity, respondents may have over-
reported enforcement practices. Second, we used the most
populous city in each state to represent enforcement of the
state law, so our findings may not generalize to other cit-
ies and rural areas in each state. Initially, in 18 of the 28
states, we interviewed participants about the enforcement
practices of at least 1 additional large city in that state.
However, because the key enforcement practices were
almost perfectly consistent between cities in each state, we
ultimately used the "largest city in state" approach. That
experience leads us to believe that the data for each city
generalize to other large cities in each state. Finally, we
neglected to ask about each state's licensing fees, if any;
such fees could help fund enforcement activities (20).
Strengths of our study included an optimal response
rate and our access to reliable data on the stringency of
each state's law (14) . These data facilitated compruisons
between "ideal" and "real" enforcement activities. As noted
earlier, we promised respondents that we would not link
published data to individuals or cities; this assurance of
anonymity probably improved both our response rate and
the accuracy of the data. Even though we are unable to
reveal which states were enforcing at lower levels, we will
be providing each respondent with feedback on how the
city's enforcement level compares with enforcement for all
other cities combined. For states in which enforcement is
low, the feedback may increase enforcement practices.
In tobacco control, antitobacco organizations historically
focused their efforts on passing new laws, but enforcement
of existing laws is viewed by many to have been critical to
reducing tobacco use (21-26). In the field of indoor tanning
legislation, we cannot say whether legislation, enforce-
ment practices, or a combination of the 2 has any effect on
the practices of indoor tanning facilities or of consumers.
Therefore, we recommend that future evaluations of indoor
tanning legislation measure not only the Wl"itten law but
also its implementation and enforcement; research should
also attempt to assess the relationship between legislation
strictness/enforcement level and the practices of business-
es and consumers, especially adolescent consumers.
Acknowledgments
This study was supported by the National Cancer Institute,
grants R01 CA093532, R01 CA093532S1, and K05 100051.
The following were partially responsible for study concept
and design: Dr George Belch, Dr James Sallis, Dr Donald
Slymen, Ms Elizabeth Clapp, and Ms Rebecca Garrow,
San Diego State University; Dr Todd Gilmer, University
of California Medical School, San Diego; and Dr Mrutin
Weinstock, Veterans Administration Medical Center and
Brown University. We thank Ms Lorri Freitas of the San
Diego County Health and Human Services Agency and Dr
Donald Bishop of the Minnesota Department of Health for
valuable feedback on the survey instrument and the survey
participants for their time and effort.
Author Information
Corresponding Author: Joni A. Mayer, PhD, Graduate
School of Public Health, San Diego State University, 9245
Sky Pruk Ct, Ste 220, San Diego, CA 92123. Telephone:
619-594-7916. E-mail: jmayer@mail.sdsu.edu.
Author Affiliations: Katherine D. Hoerster, J oint Doctoral
Program in Clinical Psychology, Latrice C. Pichon, Joint
Doctoral Program in Public Health, Health Behavior, San
VOLUME 5: NO. 4
OCTOBER 2008
Diego State University/University of California, Debra
A. Rubio, Susan I. Woodruff, School of Social Work, San
Diego State University, San Diego, California; J ean L.
Forster, School of Public Health, University of Minnesota,
Minneapolis, Minnesota. At the time of this 1esearch,
Susan I. Woodruff was affiliated with the Graduate School
of Public Health, San Diego State University, San Diego,
California.
References
1. International Agency for Research on Cancer, working
group on artificial ultraviolet (UV) light and skin can-
cer. The association of use of sunbeds with cutaneous
malignant melanoma and other skin cancers: a sys-
tematic review. [Published erratum in: Int J Cancer
2007;120(11):2526]. Int J Cancer 2007;120(5):1116-
22.
2. Whitmore SE, Morison WL, Potten CS, Chadwick C.
Tanning salon exposure and molecular alterations. J
Am Acad Dermatol 2001;44(5):775-80.
3. Centers for Disease Control and Prevention. Injuries
associated with ultraviolet tanning devices- Wisconsin.
MMWR Morb Mortal Wkly Rep 1989;38(19):333-5.
4. Walters BL, Kelley TM. Commercial tanning facili-
ties: a new source of eye injury. Am J Emerg Med
1987;5(5):386-9.
5. Palmer RC, Mayer JA, Woodruff SI, Eckhardt L, Sallis
JF. Indoor tanning facility density in eighty US cities.
J Community Health 2002;27(3):191-202.
6. Kwon HT, Mayer JA, Walker KK, Yu H, Lewis EC,
Belch GE. Promotion of flequent tanning sessions
by indoor tanning facilities: two studies. J Am Acad
Dermatol 2002;46(5):700-5.
7. Cokkinides VE, Weinstock MA, O'Connell MC, Thun
MJ. Use of indoor tanning sunlamps by US youth,
ages 11-18 years, and by their parent or guard-
ian caregivers: prevalence and correlates. Pediatrics
2002; 109(6): 1124-30.
8. Demko CA, Borawski EA, Debanne SM, Cooper KD,
Stange KC. Use of indoor tanning facilities by white
adolescents in the United States. Arch Pediatr Adolesc
Med 2003;157(9):854-60.
9. Geller AC, Colditz G, Oliveria S, Emmons K, Jorgensen
C, Aweh GN, et al. Use of sunscreen, sunburning rates,
and tanning bed use among more than 10,000 US chil-
dren and adolescents. Pediatrics 2002;109(6):1009-14.
10. Hoerste1 KD, Mayer JA, Woodruff SI, Malcarne V,
Roesch SC, Clapp E. The influence of pruents and
peers on adolescent indoor tanning behavior: find-
ings flom a multi-city sample. J Am Acad Dermatol
2007;57(6):990-7.
11. Dellavalle RP, Parker ER, Cersonsky N, Hester EJ,
Hemme B, Burkhardt DL, et al. Youth access laws:
in the dark at the tanning parlor? Arch Dermatol
2003; 139( 4):443-8.
12. Francis SO, Burkhardt DL, Dellavalle RP. 2005: a
banner year for new US youth access tanning restric-
tions. Arch Dermatol2005; 141(4):524-5.
13. McLaughlin JA, Francis SO, Burkhardt DL, Dellavalle
RP. Indoor UV tanning youth access laws: update
2007. Arch Dermatol 2007;143(4):529-32.
14. Woodruff SI, Pichon LC, Hoerster KD, Forster JL,
Gilmer T, Mayer JA. Measuring the stringency of states'
indoor tanning regulations: instrument development
and outcomes. JAm Acad Dermatol 2007;56(5):774-
80.
15. A guide to state radiation control offices. Phoenix (AZ):
National Tanning Training Institute. http://www.
tanningtraining .com/reginfo/sta te.h tml. Accessed
September 25, 2006.
16. Hickle A, Forster J , Lazovich D, Allwood P, Remba
N, Grossmeier J, et al. Sanitarians' work with indoor-
tanning businesses: findings from interviews in
two major metropolitan areas. J Environ Health
2005;67(8):30-6, 54.
17. Hester EJ, Heilig LF, D'Ambrosia R, Drake AL,
Schilling LM, Dellavalle RP. Compliance with youth
access regulations for indoor UV tanning. Arch
Dermatol 2005;141(8):959-62.
18. Dent CW, Grube JW, Biglan A. Community level alco-
hol availability and enforcement of possession laws as
predictors of youth drinking. Prev Med 2005;40(3):355-
62.
19. Forster JL, Wolfson M. Youth access to tobacco: poli-
cies and politics. Annu Rev Public Health 1998;19:203-
35.
20. Demierre MF. Time for national legislation of
indoor tanning to protect minors. Arch Dermatol
2003;139:520-4.
21. Forster JL, Komro KA, Wolfson M. Survey of city ordi-
nances and local enforcement regarding commercial
availability of tobacco to minors in Minnesota, United
States. Tob Control1996;5(1):46-51.
22. Jacobson PD, Wasserman J. The implementation and
enforcement of tobacco control laws: policy implica-
tions for activists and the industry. J Health Polit
VOLUME 5: NO. 4
OCTOBER 2008
Policy Law 1999;24(3):567-98.
23. Jason LA, Porkorny SB, Schoeny ME. Evaluating the
effects of enforcements and fines on youth smoking.
Critical Public Health 2003; 13:33-45.
24. Lazovich D, Forster J, Widome R, VanCoevering P.
Tobacco possession, use, and purchase laws and pen-
alties in Minnesota: enforcement, tobacco diversion
programs, and youth awareness. Nicotine Tob Res
2007;9(Suppl1):S57-64.
25. Ma GX, Shive S, Tracy M. The effects of licensing
and inspection enforcement to reduce tobacco sales
to minors in greater Philadelphia, 1994-1998. Addict
Behav 2001;26(5):677-87.
26. Woodhouse LD, Sayre JJ, Livingood WC. Tobacco
policy and the role of law enforcement in prevention:
the value of understanding context. Qual Health Res
200 1; 11(5):682-92.
Tables
Table 1. Respondent and Organization Characteristics (N =
28), CITY100 Enforcement Survey, April and May 2007
Variable n (%)
Sex
-
Women 7 (25.0)
Men 21 (75.0)
Organization
State health, environmental health, or radiologic health 15 (53.6)
agency
City or county health, environmental health, or radiologic 7 (25.0)
health agency
State cosmetology board 4 (14.3)
Other state agency 2 (7.1)
Occupation/title
Department head or director 6 (21.4)
- -- -
Supervisor or manager 6 (21.4)
Health physicist 4 (14.3)
Environmental health or industrial hygiene 3 (10.7)
-
Sanitarian 3 (10. 7)
Inspector 2 (7.1)
L Other"
-
4 (14.3)
Abbreviat ion: CITY100, Correlates of Indoor Tanning in Youth.
a These titles included investigator, board administrator, regulatory enforce
ment, and compliance officer.
PUBLIC HEALTH RESEARCH , PRACTICE, AND POLICY
VOLUME 5: NO. 4
OCTOBER 2008
Table 2. Indoor Tanning Legislation Enforcement Resources and General Practices (N = 28), CITY100 Enforcement Survey,
April and May 2007
Variable
No. of full-time enforcement staff
0
1-2
3-4
2:5
1-
Ucensure required
No
1-
Yes
Don't know
Give citation if facility violates law
No
Yes
Don't know
Inspection schedule
Never
Less than annually
Annually
1-
Twice a year
Announce inspection in advance
8
1-
Never/rarely
Sometimes
Often/always
a Based on the 19 cities that ever conducted inspections.
n (%)
8 (28.6)
7 (25.0)
7 (25.0)
6 (21.4)
-
5 (17.9)
22 (78.6)
1 (3.6)
r
14 (50.0)
13 (46.4)
1 (3.6)
-
9 (32.1)
9 (32.1)
6 (21.4)
-
4 (14.3)
17 (89.5)
2 (10.5)
1
0
-
PUBL I C HEALT H RESEARCH, PRACTICE, AND POLICY
VOLUME 5: NO. 4
OCTOBER 2008
Table 3. Indoor Tanning Facility Inspection Practices Related to Customer Records, CITY100 Enforcement Survey, April and
May 2007
Frequency
--------------------------------------------
Practice
Inspection includes customer record review"
Review includes customer's ageb
Review includes parental consent formsb
1--
Review includes number and dates of sessionsb
Review includes session duration!>
Abbreviation: CI1Y100, Correlates of Indoor Tanning in Youth.
a Based on the 19 cities t hat ever conducted inspections.
b Based on the 17 cities t hat included customer record review when inspecting.
Never/Rarely
n (%)
2 (10.5)
1 (5.9)
2 (11.8) l
1 (5.9)
2 (11.8)
Sometimes Often/Always
n (%) n (%)
2 (10.5) 1 15 (78.9) 1
0 16 (94.1)
0 15 (88.2)
-
3 (17.6) 13 (76.5)
1 (5.9) 14 (82.4)
Table 4. Penalties for Youth Access Violations in States With Indoor Tanning Access Laws, CITY100 Enforcement Survey, April
and May, 2007
Violation
-------------------------------------------------------------------
Selling Sessions to Underage Youth Not Obtaining Parental Consent for Minors
Penalty n (%)" n (%)b
Penalty type
Any type 11 (55.0) 10 (47.6)
Warning 10 (50.0) 9 (42.9)
Monetary fine 8 (40.0) 8 (38.1)
License suspension 7 (35.0) 7 (33.3)
Other 3 (15.0) 1 (4.8)
Graduated penalties given
I
No 4 (36.4) I 3 <3o.o> I
Yes
-
a Based on 20 respondents because of missing data.
b Based on 21 respondents.
c Based on those respondents whose cities gave any type of penalty for the violation.
7 (63.6) I 7 (70.0)
the Public Health Serv.ce, the Centers for D1sease Control and Prevention, or the authors' affiliated mstitutlons. Use of trade names is for identification only
VOLUME 5: NO. 4
OCTOBER 2008
Appendix: CI1Y100 Enforcement Survey
Study Introduction
This interview is part of the CITYiOO Indoor Tanning Project. CITYiOO
(Correlates of Indoor Tanning in Youth) is a project funded by the National
Cancer Institute that will help us better understand the factors that influ
ence teens to use indoor tanning. The project is based at the Graduate
School of Public Health at San Diego State University and is focusing on
over iOO cities in the US. One goal of CITYiOO is t o evaluate enforcement
activities in cities located in states with indoor tanning laws. Your state has
a law governing indoor tanning.
If you decide to participate, I will ask you questions about the enforcement
activities in [city/county]. such as inspections of indoor tanning facilities.
These are activities at the city or county level to enforce the state
law. We would like you to participate, irrespective of whether your city has
many or few enforcement activities. The interview will take only around 5
to 7 minutes. The researcher in charge of this study is Dr Joni Mayer; you
may have her phone number if you wish to write it down. Collect calls are
accepted. She will be able to answer any questions you have. Or if you
have any questions now, I can answer them for you. I want to assure you
that your responses in this interview wil l not be linked with your narne.
city, or county in any written reports or publications. All data will be kept in
locked file cabinets, and only the CITYiOO research staff will have access
to the data. Participation is voluntary, and you are free to end the interview
at any point.
Resources (R)
I'd like to first ask you about your employment, and [city'S/county's)
resources related to regulating indoor tanning businesses.
Ria. What is the agency you work for?
Ri b. What department within that agency?
Ric. Is your job at the
City level?
County level?
State level?
None of these? Describe.
R2. What is your occupation and/or j ob title?
R3. How many agencies are responsible for enforcing the state's indoor
tanning law in [city/county]?
None/no enforcement
One
Two or more
Don' t know
Refused
No state law
R4. The primary enforcement agency-is it at the
City-level? Name:
County-level? Name:
Statelevel? Name:
None of these? Name:
R5. How many full-time staff (or FTE) are allocated to carry out the inspec-
tions and/or other enforcement activities in [city/county]?
. Gave a number:
Don't know
Refused
Not applicable
Licensure (L)
Li. Are indoor tanning businesses in [city) required to have a license?
No
Yes
Don't know
Refused
Not applicable
Inspections (I)
Now I'd like to ask you about inspection procedures.
li. In the absence of a complaint, how often is a tanning facility inspected
in [city/county]?
Never (go to question P1)
Less than once a year
Once a year
Twice a year
. More than twice a year
Other (describe)
Don't know
Refused
Not applicable
12. How often is the inspection announced in advance to the tanning facility
t hat will be inspected?
Never/ rarely
Sometimes
Often/always
Don't know
Refused
Not applicable
13. Please tell me how often an inspection includes the review of customer
records.
Never/ rarely
. Sometimes
Often/always
Don't know
Refused
Not applicable
14. (If the answer to 13 is sometimes, often, or always) When customer
records are reviewed, how often are the following looked at?
14a. Age of customers
Never/ rarely
Sometimes
The opinions expressed by authors contributing to th1s j ournal do not necessarily reflect the opinions of the US Department of Health and Human Serv1ces,
VOLUME 5: NO. 4
OCTOBER 2008
Often/always
Don't know
Refused
Not applicable
14b. Parental consent forms
Never/rarely
Sometimes
Often/always
Don't know
Refused
Not applicable
14c. Number and dates of tanning sessions
Never/rarely
Sometimes
Often/always
Don't know
Refused
Not applicable
14d. Duration of tanning sessions
Never/ rarely
Sometimes
Often/always
Don't know
Refused
Not applicable
Penalties/Fines (P)
The next questions are about penalties for violations of laws regulating
indoor tanning.
Pl. If a tanning facility in [city/county] violates a law regulating this type of
business, will the facility receive a citation?
No (skip the remaining items}
Yes
Don't know
Refused
Not applicable
P2. In general, what is the penalty if a facility is cited for selling tanning
sessions to underage youth?
P2a. A warning
No
Yes
Don't know
Refused
Not applicable
No
Yes
Don't know
Refused
Not applicable
. No
. Yes
Don't know
Refused
Not applicable
P2d. Other (probe}
P3. If a facil ity is cited more than once for selling tanning sessions to
underage youth, does [city/county] use graduated penalties, in which
each repeat violation results in a larger penalty?
No
. Yes
. Don't know
Refused
Not applicable
P4. In general, what are the penalties if a facility is cited for not obtaining
parental permission for minors?
P4a. A warning
No
Yes
Don't know
Refused
Not applicable
No
Yes
. Don't know
Refused
Not applicable
No
. Yes
. Don't know
Refused
Not applicable
P4d. Other (probe}
P5. If a facility is cited more than once for not obtaining parental permis-
sion, does your city/county use graduated penalties, in which each
repeat violation results in a larger penalty?
No
Yes
. Don't know
Refused
Not applicable
That concludes all my questions. I really appreciate all of your time! Do
you have any questions or comments about this survey? Thanks again.
Goodbye.
Medications
That Increase
Sensitivity To Light:
A 1990
Listing
prepared by
Jerome I. Levine, M.S., R.Ph.
December 1990
U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES
Public Health Serriee/Food ancl Drul Aclmtntstratlon
Center for Device and Radlolotleal Health
HHS PnbUeatfon FDA 918280
PREFACE
The Office of Science and Technology is a major operating
component of the Food and Drug Administration's Center for Devices
and Radiological Health. The Office conducts laboratory and field
research on the human heal th effects of medical device technologies
and radiation. This publication provides a listing of medications and
other agents that may increase sensitivity to ultraviolet exposure from
either sunlight or artificial light. It is intended to alert both consumers
and health professionals to the risk of photosensitization and the need
for protective measures.
Igor Cerny, Pharm.D., of the Food and Drug Administration's
Center for Drug Evaluation and Research, made a valuable contribution
to this publication by performing a comprehensive review of the
manuscript.

Director
Office of Science and Technology
ii
CONTENTS
Introduction
0 0. 0 0 0 0 0 0 0 0 .. .. ..... . .
1
Primary classes_ of medications responsible
for photosensitizing reactions . . . . . . . . . . . . . . . . . . . . . . . . . 3
Reported photosensitizing medications . . . . . . . . . . . . . . . . . . . 6
Other photosensitizing agents . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
iii
The mention of commercial products, their sources, or their use
in connection with material reported herein is not to be construed
as either an actual or implied endorsement of such products by
the Department of Health and Human Services (HHS).
iv
MEDICATIONS THAT INCREASE SENSITIVITY TO
LIGHT: A 1990 LISTING
INTRODUCTION
Many me<:lications and other agents contain ingredients that may
cause photosensitivity, which is defined as a chemically induced
change in the skin that makes an individual unusually sensitive to light.
An individual who has been photosensitized may develop a rash,
sunburn, or other adverse effect from exposure to light of an intensity
or duration that would normally not affect that individual.
Reactions to photosensitizing agents involve both photoallergy
(allergic reaction of the skin) and phototoxicity (irritation of the skin)
after exposure to ultraviolet radiation from natural sunlight or artificial
lighting (particularly from tanning booths). This photosensitization of
the skin may be caused by creams or ointments applied to the skin, by
medications taken orally or by injection, or by the use of prescription
inhalers.
In addition to an exaggerated skin burn, itching, scaling, rash,
or swelling, exposure to ultraviolet light in combination with
certain medications may result in:
Skin cancer
Premature skin aging
Skin and eye burns
Allergic reactions
Cataracts (clouding in the eyes)
Reduced immunity
Blood vessel damage
1
The following list of photosensitizing medications was taken from
the current literature. The list is not complete; if you are using a
product that is not listed, you should consult your physi cian or
pharmacist for guidance.
Before using the list, you should be aware of the following:
1. NOT all individuals who use or take these medications will
experience a photosensitive reaction. Also, an individual
who experiences a photosensitive reaction on one
occasion will NOT necessarily experience it again or every
time.
Most entries in this listing do not indicate how often a
photosensitive reaction has been reported. If an entry
includes a percentage in the Brand Name column (in bold
type and parentheses), this number is the frequency of
reaction reported in reference number 2 in the
Bibliography.
2. A medlcation will NOT cause the same degree of skin
reaction in all individuals.
3. Brand names of products should be considered only as
examples; they do NOT represent all names under which
the generic product may be sold.
This listing will require periodic updating as new Information
becomes available.
2
PRIMARY CLASSES OF MEDICATIONS RESPONSIBLE
FOR PHOTOSENSITIZING REACTIONS
(Examples by Generic Name)
Note: If you are using a medication in any of these classes,
you should take precautions even if your particular medica
tion is not :listed.
1. Antihistamines
examples: Astemizole
Azatadine
Brompheniramine
Buclizine
Carbinoxamine
Chlorpheniramine
Clemastine
Cyclizine
Cyproheptadine
Dexchlorpheniramine
Dimenhydrinate
Diphenhydramine
Diphenylpyraline
Doxylamine
Hydroxyzine
Meclizine
Methapyrilene
Methdilazine
Orphenadrine
Pheniramine
Promethazine
Pyrilamine
Terfenadine
Trimeprazine
Tripelennamine
Triprolidine
2. Coal Tar and Derivatives (examples by brand name)
examples: Alphosyl
Aquatar
Denorex Medicated
Shampoo
DHS Tar Gel
Shampoo
DOAK Shampoo
Estar
3
lonil T Plus
!.AVATAR
Medotar
T/Derm Tar Emollient
Tegrin Shampoo
T/Gel Therapeutic Shampoo
Zetar Shampoo
3. Contraceptives, Oral & Estrogens (birth control pills, female sex
hormones)
examples: Estrogens
Chlorotrianisene
Diethylstilbestrol
Estradiol
Estrogens, conjugated
Estrogens, esterified
Estropipate
Progestogens
Ethinyl estradiol
Medroxyprogesterone
Megestrol
Norethindrone
Norgestrel
Quinestrol
4. NSAID: Non-Steroidal Anti-Inflammatory Drugs (antiarthritics)
examples: Diclofenac
Diflunisal
Fenoprofen
Flurbiprofen
, Ibuprofen
Indomethacin
Ketoprofen
Meclofenamate
Naproxen
Phenylbutazone
Piroxicam
Sulindac
Suprofen
Tolmetln
5. Phenothiuines (major tranquilizers, anti-emetics)
examples:
6. Psoralens
examples;
Acetophenazine
Butaperazine
Carphenazine
Chlorpromazine
Ethopropazine
Fluphenazine
Mesoridazine
Methdilazine
Methotrimeprazine
Perphenazine
Methoxsalen
4
Piperacetazine
Prochlororperazine
Promazine
Promethazine
Propiomazine
Thiethylperazine
thioridazine
Trifluoperazine
T riflupromazine
Trimeprazlne
Trioxsalen
7. Sulfonamides (
11
sulfa
11
drugs, antimicrobials, anti-infectives)
examples: Acetazolamide
Sulfacytine
Sulfadiazine
Sulfadoxine
Sulfamethizole
Sulfamethoxazole
Sulfapyridine
Sulfasalazine
Sulfinpyrazone
Sulfisoxazole
8. Sulfonylureas (oral anti-diabetics, hypoglycemics)
examples: Acetohexamide
Chlorpropamide
Glipizide
9. Thiazide Diuretics (
11
water-pills
11
)
examples: Bendroflumethiazide
Benzthiazide
Chlorothiazide
Chlorthalidone
Cyclothiazide
Glyburide
Tolazamide
Tolbutamide
Hydrochlorothiazide
Hydroflumethiazide
Methyclothiazide
Polythiazide
Trichlonnethiazide
10. Tetracyclines (antibiotics, anti-infectives)
examples: Chlortetracycline
Demeclocycline
Doxycycline
Methacycline
11. Tricyclic Antidepressants
examples: Amitriptyline
Amoxapine
Desipramine
Doxepin
5
Minocycline
Oxytetracycline
Tetracycline
Imipramine
Nortriptyline
Protriptyline
Trimipramine
REPORTED PHOTOSENSITIZING MEDICATIONS
Therapeutic
Generic Name Brand Name Class
Acetazolamide Diamox Anticonvulsant
Antiglaucoma
Diuretic
Amiloride + Moduretic Antihypertensive
Hydrochlorothizide Thiazide diuretic
Amiodarone Cordarone ( 15%) Antiarrhythmic
Amitriptyline Elavil Antidepressant (tricyclic)
II
Endep Antidepressant (tricyclic)
Amoxapi ne Asendin ( < 1%) Antidepressant (tricyclic)
Astemizole Hismanal Antihistamine
Atenolol + Tenoretic Beta-adrenergic blocker
Thiazide diuretic
Auranofin Ridaura Antiarthritic
Gold compound
Azatadine Optimine Antihistamine
Azatidine + Trinalin Antihistamine
Pseudoephedrine Repetabs Decongestant
Bendroflumethiazide Naturetin Antihypertensive
Thiazide diuretic
Benzthiazide Exna Antihypertensive
Thiazide diuretic
Bromodiphenhydramine Ambenyl Antihistamine
Brompheniramine Dimetane Antihistamine
Captopril Capo ten Antihypertensive
Captopril + Capozide Antihypertensive
Hydrochlorothiazide Thiazide diuretic
Carbamazepine Tegretol Analgesic
Anticonvulsant
6
Therapeutic
Chlordiazepoxide + Limbitrol Antidepressant (tricyclic:
Amitriptyline Tranquilizer
Chlorothiazide Diuril Antihypertensive
Thiazide diuretic
Chlorpheniramine Chlorphenlramine Antihistamine
Chlorpheniramine + Deconamlne Antihistamine
D-Pseudoephedrine Decongestant
Chlorpheniramine + Ru-Tuss II Antihistamine
Phenylpropanolamine Decongestant
Chlorpromazine Thorazine Antiemetic
Tranquilizer
Chlorpropamide Diabinese Antidiabetic (oral)
Sulfonylurea
Chlorprothixene Taractan Antiemetic
Tranquilizer
Chlorthalidone Hygroton Antihypertensive
Thiazide diuretic
"
Thalitone Antihypertensive
Thiazide diuretic
Chlorthalidone + Demi-Regroton Antihypertensive
Reserpine
Thiazide diuretic
"
Regroton Antihypertensive
Thiazide diuretic
Ciprofloxacin Cipro (<1%) Anti-infective
Clemastine Tavist Antihistamine
Clofazime
lamprene ( < 1%) Antibacterial
Antileprosy agent
Clonidine + Combipres Antihypertensive
Chlorthalidone
Thiazide diuretic
Coal tar
Estar Gel Anti psoriatic
Eczema
7
Therapeutic
Coal Tar Baine tar Antipsoriatic
Eczema
Contraceptive, oral Estrogen Birth control pill
Female sex hormone
Cromolyn lntal Inhaler Antiasthmatic
Cyclobcnzaprine Flexeril Anti-skeletal
muscle spasms
Cyproheptadine Periactin Antihistamine
Antiserotonergic
Dacarbazine OTIC-Dome Anti-Hodgkin's disease
Antimetabolite
Danazol Danocrine Gonadotropin inhibitor
Demeclocycline Declomycin Antibiotic
Desipramine Norpramin Antidepressant (tricyclic)
" Pertofrane Antidepressant (tricyclic)
Dexchlorpheniramine Polaramine Antihistamine
Diclofenac Voltaren ( < 1%) NSAIO, antiarthritic
Diflunisal Dolobid (1%) NSAID, antiarthritic
Diltiazem Cardizem ( < t%) Antianginal
Antihypertensive
Calcium channel
blocker
Diphenhydramine Benadryl Antihistamine
Diphenylpyraline Hispril Spansule Antihistamine
Doxepin Sinequan Antidepressant (tricyclic)
Doxycycline Vibramycin Antibiotic
Doxycycline Hyclate Ooryx Antibiotic
Non-Steroidal Anti-Inflammatory Drug
8

Generic Name Class
Enalapril Vasotec Antihypertensive
Enalapril + Vaseretic Antihypertensive
Erythromycin Pediazole Antibiotic
Ethylsuccinate +
Sulfisoxazole
Estrogens Contraceptive, Female sex hormone
oral
" Estrogens Female sex hormone
Ethionamide Trecator-SC Anti-infective
Antituberculosis
Etretinate Tegison Antipsoriatic
Floxuridine FUDR Injectable Antimetabolite
Antineoplastic
Flucytosine Ancobon Antifungal
Fluorouracil Adrucil Antineoplastic
"
Efudex Antineoplastic
" Fl uorouracil Antineoplastic
Fluphenazine Prolixin Antipsychotic
Tranquilizer
" Permitil Antipsychotic
Tranquilizer
Flurbiprofen Ansaid NSAID, antiarthritic
Flu tam ide Eulexin Antimetastatic
(prostatic carcinoma)
Furosemide Lasix Antihypertensive
r
Diuretic
1
9
TherApeutic
Gentamicin Caramycin Antibiotic
Glipizide Glucotrol Antidiabetic (oral)
Sulfonylurea
Glyburide Diabeta Antidiabetic (oral)
Sulfonylurea
"
Micronase Antidiabetic (oral)
Sulfonylurea
Gold Salts (Compounds) Solganal Antiarthritic
Cold compound
Gold Sodium Thiomalate Myochrysine Antiarthritic
Cold compound
Griseofulvin Fulvicin U/F Antibiotic
Antifungal
Griseofulvin Crisactin Ultra Antibiotic
Ultramicrosize Antifungal
Guanethidine + Esimil Antihypertensive
Haloperidol Haldol Antipsychotic
Tranquilizer
Hexachlorophene Phisohex Antibacterial
Hydralazine + Apresazide Antihypertensive
"
Antihypertensive
Thiazide diuretic
Hydrochlorothiazide Esidrix Antihypertensive
Thiazide diuretic
" Hydrodiuril Antihypertensive
Thiazide diuretic
" Ore tic Antihypertensive
Thiazide diuretic
Hydrochlorothiazide + Oreticyl Antihypertensive
Deserpidine
Thiazide diuretic
10
Generic Name Brand Name
Therapeutic
Class
Hydrochlorothiazide + Dyazide Antihypertensive
Triamterene Thiazide diuretic
"
Maxzide Antihypertensive
Thiazi de diuretic
Hydroflumethi azide Diucardin Antihypertensive
Thiazide diuretic
"
Saluron Antihypertensive
Thiazide diuretic
Hydroflumethi azide + Salutensin I Antihypertensive
Reserpine Salutensin-Demi Thiazide diuretic
Ibuprofen Advil ( <1%) NSAID, antiarthritic
"
Motrin ( < t%) NSAID, antiarthritic
Imipramine Tofranil Antidepressant (tricyclic)
lndapamide Lozol Antihypertensive
Diuretic
Interferon ALFA-2B lntron A ( < t%) Antiviral agent
l socarboxazid Marplan Antidepressant
MAO inhibitor
l sotretinoin Accutane Antiacne
Ketoprofen Orudis < < t %) NSAID, antiarthritic
Labetalol Normodyne Antihypertensive
Beta- and alpha-
adrenergic bl ocker
Labet al ol + Normozide Beta- and alpha-
Hydrochlorothiazide adrenergic blocker
Thiazi de diuretic
"
Trandate HCT Beta- and alpha-
adrenergic blocker
Thiazide diuretic
11
--.... '
Therapeutic
Generic Name Brand Name Class Generic Name Brand 1"1
lisinopril + Prinzide Antihypertensive Metolazone Diulo
Hydrochlorothiazide
Thiazide diuretic
i
"
Zestoretic Antihypertensive
"
Mykrox
i
l
.
Thiazide diuretic
i

" Zaroxol)
I
Lovastatin Mevacor
Anticholesterol
!
I
Maprotiline Ludiomil
Antidepressant
I
Meperidine + Mepergan Narcotic analgesic
Metoprolol + Lopress
I
Hydrochlorothiazide
i
Promethiazine
I
Mesoridazine Serentil Antipsychotic
Minocycline Minocin
I
l
Tranquilizer Minoxidol Rogaine
Methacycline Rondomycin, Antibiotic Nabilone Cesame
I
' Methazolamide Neptazane Antiglaucoma
I
J
Methdilazine
Tacaryl Antihistamine
Nadolol + Corzide
I
t
Antipruritic
Bendroflumethiazide
I Methotrexate Folex Antimetaboli te
Nalidixic Acid NegGrar
!
Antipsoriatic
Naprosyn Anaproll
" Methotrexate Antimetabolite
" NaproxE
Antipsoriatic
Nifedipine Adalat
" Mexate & Antimetabolite
Mexate-AQ Anti psoriatic
Methyclothiazide Aquatensen Antihypertensive
Thiazide Diuretic
" Procardi
" Enduron Antihypertensive
Thiazide diuretic
Methyclothiazide + Enduronyl Antihypertensive
Norfloxacin Noroxin Deserpidine
Thiazide diuretic
Methyclothiazide + Diutensen-R Antihypertensive
Nortriptyline Pamelor
Reserpine
Thiazide diuretic
Methyldopa + Aldoril Antihypertensive
Non-Steroidal Anti-Inflammatory 0
Hydrochlorothiazide
Thiazide diuretic
Methyldopa + Aldoclor Antihypertensive
Chlorothiazide
Thiazide diuretic
12
1
Therapeutic Therapeutic
me Class Generic Name Brand Name Class
Antihypertensive Metolazone Diulo Antihypertensive
Thiazide diuretic Thiazide diuretic
Antihypertensive
"
Mykrox Antihypertensive
Thiazide diuretic Thiazide diuretic
Anticholesterol
..
Zaroxolyn Antihypertensive
Antidepressant
Thiazide diuretic
Narcotic analgesic
Metoprolol + lopressor HCT Beta-adrenergic blocker
Antipsychotic
Minocycline Minocin Antibiotic
Tranquilizer Minoxidol Rogaine Hair growth stimulator
:in Antibiotic Nabilone Cesamet Antiemetic
Anti glaucoma
Antinausea
Antihista1J1ine
Nadolol + Corzide Antihypertensive
Antipruritic _
Bendroflumethiazide Beta-adrenergic blocker
Antimetabolite
Nal idixic Acid NegGram Antimicrobial
Anti psoriatic
Naprosyn Anaprox NSAID, antiarthritic
ate _Antimetabolite
"
Naproxen NSAID, antiarthritic
Antipsoriatic
Nifedipine Adalat Antianginal
Antimetabolite
Antihypertensive

Antipsoriatic
Calcium channel
n Antihypertensive
blocker
Thiazide Diuretic
" Procardia Antianginal
Antihypertensive
Antihypertensive
Thiazide diuretic
Calcium channel
blocker
Antihypertensive
Norfloxacin Noroxin Antibacterial
Thiazide diuretic
R Antihypertensive
Nortriptyline Pamelor Antidepressant (tricyclic)
Thiazide diuretic
Antihypertensive
Thiazide diuretic
Antihypertensive
Thiazide diuretic
13
i
Therapeutic
Generic Name Brand Name Class Generic Name Brand
Oxytetracycline Terramycin Antibiotic Quinethazone Hydro
Perphenazine Trilafon Antipsychotic
Tranquilizer Quinidine Gluconate
Quina!
Perphenazine + Etrafon Antidepressant (tricyclic)
Dura
Amitriptyline Tranquilizer Quinidine Sulfate Quind
" Triavil Antidepressant (tricyclic)
" Quino
Tranquilizer
Quinine Quina
Phenylbutazone Butazolidin NSAID, antiarthritic
Rauwolfia Serpentina + Rauzic
Phenylpropanolamine + Ornade Antihi stamine Bendroflumethiazide
Chlorp.he_niramine Spansule Decongestant
Reserpine + Diupn
Phenylpropanolamine + Triaminic TR Antihistamine Chlorothiazide
Pheniramine + Decongestant
Reserpine + Hydro
_!>.yrilamine
Hydrochlorothiazide
Phenytoin Oilantin Anticonvulsant
" Serpa!
Piroxicam Feldene NSAID, antiarthritic
Polythiazide Renese Antihypertensive Reserpine + S e r A ~
Thiazide diuretic Hydralazine +
Prazosin + Polythiazide Minizide Antihypertensive
Hydrochlorothiazide
Thiazide diuretic Selegiline Eldepr
Prochlorperazine Compazine Antinausea
Anti-vomiting Spironolactone + Aid act
Promethazine Phenergan Antihistamine
Hydrochlorothiazide
Propranolol + lnderide Beta-adrenergic blocker
Sulfacytine Reno<
Hydrochlorothiazide Thiazide diuretic Sulfadoxine + Fansic
Protriptyline Vivactil Antidepressant (tricyclic)
Pyrimethamine
Pyrazinamide Pyrazinamide Anti-infective
Sulfamethizole + Thiosr
Antituberculosis
Phenazopyridine
Sulfamethoxazole Ganta
~ .
14
15
Therapeutic
Quinethazone Hydromox Antihypertensive
Diuretic
Quinidine Gluconate Quinaglute Antiarrhythmic
Dura-Tabs
Quinidine Sulfate Quindex Extentabs Antiarrhythmi c
" Qui nora Antiarrhythmic
Quinine Quinamm Antiprotozoal
Rauwolfia Serpentina + Rauzide Antihypertensive
Bendroflumethiazide Thiazide diuretic
Reserpine + Diupres Antihypertensive
Chlorothiazide Thiazide diuretic
Reserpine + Hydro pres Antihypertensive
"
Serpasii-Esidrix Antihypertensive
Thiazide diuretic
Reserpine + Ser-Ap-Es Antihypertensive
Hydralazine + Thiazide diuretic
Hydrochlorothiazide
Selegiline Eldepryl Anti-Parkinsonism
MAO inhibitor
Spironolactone + Aldactazide Antihypertensive
Sulfacytine Renoquid Antibiotic
Sulfadoxine + Fansidar Antimalarial
Pyrimethamine Antiprotozoal
Sulfamethizole + Thiosulfii-A Urinary analgesic
Phenazopyridine Antibiotic
Sulfamethoxazole Gantanol Antibiotic
15

Class
Sulfamethoxazole +. Azo Cantanol Antibiotic
Phenazopyridine Urinary analgesic
Sulfapyridine (Generic only) Dermatitis herpetiformis
suppresant
Sulfasalazine Azulfidine Bowel anti-inflammatory
Sulfinpyrazone Anturane Antigout agent
Antihyperuricemic
Sulfasoxazole Cantrisin Antibiotic
Sulfasoxazole + Azo Gantrisin Antibiotic
Phenazopyridine Urinary analgesic
Sulfone Dapsone Antileprosy
Antimalarial
Sulindac Clinoril ( < 1%) NSAID, antiarthritic
Terfenadine Seldane Antihistamine
Tetracycline Achromycin Antibiotic
II
Sumycin Antibiotic
Thioridazine Mellaril Antipsychotic
Tranquilizer
Thiothixene Navane Antipsychotic
Tranquilizer
Timolol + Timolide Beta-adrenergic blocker
Tolazamide Tolinase Alltidiabeti c (oral)
Sulfonylurea
Tolbutamide Orinase Antidiabetic (oral)
Sulfonylurea
16
Therapeutic
Class
Tretinoin Retin-A Antiacne
(topical)
Triamterene Dyrenium Antihypertensive
Diuretic
Trifluoperazine Stelazine Antipsychotic
Tranquilizer
T riflupromazine Vesprin Antiemetic
Antipsychotic
Trimeprazine Temaril Antipsychotic
Tranquilizer
Trimethoprim Trimpex Antibiotic
Trimethoprim + Bactrim Antibiotic
Sulfamethoxazole
"
Septra Antibiotic
Trimipramine Surmontil Antidepressant (tricyclic)
Tripelennamine PBZ Antihistamine
Triprolidine Actidil Antihistamine
Triprolidine + Actifed Antihistamine
Pseudoephedrine Decongestant
Vinblastine Vel ban Antineoplastic
17
OTHER PHOTOSENSITIZING AGENTS
Classification or Use
Antifungals
Antimicrobials, antiseptics
Artificial sweeteners
Coal tar and coal t,ar for
psoriasis and chronic eczema and in
hair shampoos
Cosmetics and dyes
Deodorant and bacteriostatic
agents in soaps
18
Fentichlor
Jadit
Multifungin
Bithionol
Chlorhexidine
Hexachlorophene
Calcium cyclamate
Cyclamates
Sodium cyclohexyl
sulfamate
Anthracene
Many phenolic agents
Naphthalene
Phenanthrene
Pitch
Thiophene
Acridine
Eosin
Erythrocine
Fluorescein
Methylene blue
Methyl violet
Orange red
Paraphenylenediamine
Rose bengal
Toluidine blue
Trypaflavin
Trypan blue
Halogenated carbanilides
Halogenated phenols
Halogenated salicylanilides
Classification or Use Agent
Fluorescent brightening agent for cellulose, Blankophor
nylon, or wool fibers
Melanogenics (furocoumarins) Methoxypsoralens
Petroleum products
Psoralen
Perfumes and toilet articles (esssential oils) Ethereal oils
Musk Ambrette
Oil of Bergamot
Oil of Cedar
Oil of Citron
Oil of Lavender
Oil of Lemon
Oil of Lime
Oil of Rosemary
Oil of Sandalwood
Perfumes, flavoring, spices Rutaceae (plant)
Umbelliferae (plant)
Tattoos Cadmium sulfide
Sunscreens with a Reported Photosensitizing Ingredient
6-Acetoxy-2,4,-dimethyl-m-dioxane (preservative in sunscreens)
Benzophenones (Aramis, Clinique, and others)
Cinnamates (Aramis, Estee Lauder, and others)
Oxybenzone (Eclipse, PreSun, and others)
Paba esters (Eclipse, Block Out, Sea & Ski, and others)
Para-aminobenzoic acid (PABA-Pabagel, Pabanol, PreSun, and others)
19
BIBLIOGRAPHY
1. Physician's Desk Reference, 44th ed., Medical Economics,
Oradell, N.J., 1990.
2. Drug Interactions and Side Effects Index, Keyed to PDR, 44th ed.,
Medical Economics, Oradell, N.J., 1990.
3. Physician' s Desk Reference for Nonprescription Drugs, 11th ed.,
Medical Economics, Oradell, N.J., 1990.
4. long, J.W. The Essential Guide to Prescription Drugs, Harper &
Row Publishers, Inc., New York, 1990.
5. Griffith, H.W. Complete Guide To Prescription & Non-
Prescription Drugs, The Body Press, a Division of Price Stern
Sloan, Inc., los Angeles, CA, 1990.
6. Physician's 1990 Drug Handbook, Springhouse Corporation,
Springhouse, Pennsylvania, 1990.
7. USP-DI 1Oth ed., Drug Information for the Health Care
Professional, Volumes 1A and 1 B, by USPC, 12601 Twin brook
Parkway, Rockville, MD 20852, 1990.
8. The Merck Manual of Diagnosis and Therapy, Fifteenth Edition,
Merck Sharp & Dohme Research laboratories, Division of Merck
& Co., Inc., Rahway, N.J., 1987.
9. Ben-Hur, E. and Rosenthal, I. Photomedlclne, Volume Ill, Chapter
7,
11
Safety Measures In Optical Radiation Treatment, " pages 147-
189, CRC Press, Inc., Boca Raton, Florida, 1987.
11 U. S. Government Printing Office: 1991 292810 (40270)
20
I LL Docume n t De l i very
REG-14715483
DCUDJG
NLM -- W1 A126 (Gen) ; W1 A126 (Re f ); Film S02755; E-Journal w/ILL access
US FEDERAL TRADE COMMISSION ( FTC)
US FEDERAL TRADE COMMISSION LIBRARY
600 PENNSYLVANIA AVENUE NW
WASHINGTON, DC 20580
ATTN:
PHONE : 202- 326- 2395
FAX : 202- 326- 2732
E- MAIL : library@ftc.gov
SUBMITTED:
PRINTED:
REQUEST NO. :
SENT VIA:
DOCLINE NO. :
2009-06-18 10:05:43
2009-06-18 11:51:43
REG-14715483
DOCLINE
KYB-27277283
REG Copy Journal
TITLE:
PUBLISHER/PLACE :
VOLUME/ ISSUE/PAGES :
DATE :
TITLE OF ARTICLE :
ISSN:
OTHER NUMBERS/ LETTERS :
SOURCE :
MAX COST:
COPYRIGHT COMP .:
CALL NUMBER:
NOTES:
REQUESTER INFO:
DELIVERY:
REPLY :
ARCHIVES OF INTERNAL MEDICINE
American Medical Assn. Chicago
2008; 168 ( 15) : 1629-37 1629-37
2008
25- .... VITAMIN D LEVELS AND ...
0003-9926
Unique ID.: 0372440
KYB-27277283
LocatorPlus
$14.00
Guidelines
W1 A126 (Gen); W1 A126 (Ref); Film S02755;
E-Journal w/ILL access
CG
Evans James
E-mail: library@ftc.gov
Mail :
KEEP THIS RECEIPT TO RECONCILE WITH BILLING STATEMENT
For problems or questions, contact NLM at h ttp : //wwwcf . nlm.nih.
gov/ill/ill_web_form.cfm or phone 301-496-5511.
Include LIBID and request number .
NOTE: -THIS MATERIAL MAY BE PROTECTED BY COPYRIGHT LAW (TITLE 17, U.S. CODE )
NLM Collection Access Section, Bethesda, MD
:d
:d
ty
re
lb
l(j
ill
ll
" s
n
,
g
,,
--
25-Hydroxyvitamin D Levels and the Risk
of in the General Population
Micl1al L Melamed, MD, MHS; Erin D. Micltos, MD, MHS; Wendy Post, MD, MS; Brad Astor, PltD
Background: In patients l;lndergoing dialysis, therapy
with calcitriol or paricalcitol or other vitamin D agents
is associated \vith reduced mortality. Observational data
suggests that low 25-hydroxyvitamin D levels (2S(OH)D)
are associated with diabetes mellitus, hypertension, and
cancers. However, whether low serum 25(0H)D levels
are asSociated with mortali ty in the general population
is unknown.
. Methods: We tested the association of low 25( OH)D lev-
els with all-cause, cancer, and cardiovascular disease
( CVD) mortality in 13 331 nationally representative adults
20 years or older from the Third National Health and Nu-
trition Examination Survey (NHANES III) mor-
tality files. Participant vitamin D levels were-collected from
l988through 1994, and individuals were passively fol-
lowed for mortali ty through 2000.
deficiency (lowest quartile of 25(0H)D level, <17.8
nglmL (to convert to nanomoles per liter, multiply by
2.496)), while greater physical activity, vitamin D supple-
mentation, and nonwinter season were inversely associ-
ated. During a median 8. 7 years of follow-up, th ere were
1806 deaths, including 777 from CVD. In multivariate
models (adjusted for baseline demographics, season, and
traditional and novel CVD risk factors), compared with
the highest quartile, being in the lowest quartile (25 I OH) D
levels <17.8 nglmL) was associated with a 26% i n-
creased rate of all-cause mortality (mortality rate ratio,
1.26; 95% Cl, 1.08-1.46) and a population attributable
risk of3.1 %. The adj usted models of CVD and cancer mor-
tality revealed a higher risk'; which was not statistically.
significant.
Results: In cross-sectional analyses, in-
creasing age, female sex, nonwhite race/ethnicity, dia-
betes, current and higher body mass index were
all with higher odds of25(0H)D
Conclusion: The lowest quartile of 25( OH) D level ( < 17.8
nglmL) is independently associated \vith all-cause mor-
tality in the general population.
Arch Intent Med.
Author Affiliations:
Division or Nephrology,
Department or Medicine, and
Department of Epidemiology &
Population Health, Albert
Einstein College of Medicine,
Bronx, New York
(Or Melamed); Divisions of
Cardiology (Drs Michos and
Posr) and General Internal
Medicine (Dr Astor),
Department of Medicine,
johns Hopkins Uni\'ersity,
School of Medicine, Baltimore,
Maryland; and Department of
Epidemiology, johns Hopkins
Bloomberg School of Public
Heahh,'Baltimore (Drs Post
and Astor).
S
EVERAL STUDIES HAVE SUG-
gested that 25-hydroxyvita-
-- min (25(0H]p) deficiency is
an unrecognized contributor
to the development of cardio-
vascular disease (CVD), cancer, and mor-
tality. 1,25-Dihydroxyvitamin D affects the
renin-angiotensin system,
1
is associated with
cardiac myocyte hypertrophy/ and has anti-
inflammatory effects? all of which may in-
fluence CVD risk:' l n addition, 1,25-
dihydroxyvitamin D has antiproliferative
activity, which may infl uence cancer risk.
5
Therapy \vith calcitriol or paricalcitol or
other vitamin D agents (activated vitamin
D therapy) is associated \vith lower mor-
tality in those with end-stage renal dis-
ease.(>.10 In addition, low 25(0H)D levels
have been associated \vith multiple CVD risk
factors in the Third National Health and Nu-
trition Examination Survey (NHANES lll)
population
11
as well as with hyperten-
sion,12 congestive heart failure,B cancer/"
and diabetes mellitus. u
16
Low 25(0H)D lev-
els in incident hemodialysis patients have
recently been shown to be associated with
all-cause mortality.
17
A meta-analysis of 18
randomized clinical trials of vitamin D
supplementation in mostly older individu-
als found that randomization to vitamin D
supplementation was associated \vith lower
all-cause mortality.
18
Despite these sug-
associations, w.e. found no pub-
lished studies evaluating the relationship be-
tween 25(0H)D levels and mortality risk in
the general population.
The optimal level of25(0H)D has been
suggested to be 30 nglmL or greater (to con-
vert to nanomoles per liter, multiply by
2. 496),
19
.2 a level associated \vith maximal
suppression of parathyroid hormone andre-
duced fracture rates and postulated to be as-
sociated \vith better health outcomes. Ap-
proximinely 41% of men and 53% of women
in the United States, however, have levels
of25(0H)D below 28 nglmL.
21
ARCH INTERN MED/VOL 168 (NO. 15), AUG 11/25, 2008
1629
Material may be protected by copyright law (Title 17, U.S. Code)
We hypothesized that low serum 25(0H)D level is a
risk factor for cancer, CVD, and all-cause mortality among
adults in the United States. The NHANES Ill linked mor-
tality data set provides an excellent opportunity to test
the association of25(0H)D deficiency with mortality in
a large, multiracial sample of the general population.
METHODS
STUDY PARTICIPANTS
The Third National Health and Nutrition Examination Survey is
a natiom.,ide probability sample of noninstitutionalized civilian
persons conducted by the National Center for Health Statistics
(NCHS) of the Centers for Disease Control and Prevention. De-
tails regarding informed consent and statistical methods arc out-
lined elsewhere.
22
We restricted our analyses to adults 20 years
or older who had a physical examination and laboratory testing
at baseline (October 1988-0ctober 1994) and for whom vital sta-
tus information was known during the follow-up period (through
December 31, 2000). Non-Hispanic blacks, Mexican Ameri-
cans, and elderly persons were oversampled in NHANES Ill to
allow for more precise estimates in these groups. n To ensure com-
parable conditions at each NHANES survey site, northern states
were surveyed during the summer and southern states were sur-
veyed during the \vinter. Participants were excluded if they did
not have complete data on the study variables of interest. There-
maining l3 331 participants represem approximately 175 mil-
lion people in the United States. The institutional review boards
at the johns Hopkins I3loomberg School of Public Health, Balti-
more, Maryland, and the Albert Einstein College of Medicine,
Bronx, York, determined this analysis to be exempt.
STUDY VARIABLES
lmen\iew questions, physical examination, laboratory val-
ues, such as C-reactive protein, glucose, albumin, creatinine, and
lipids levels, were assessed in all study participants at baseline dur-
ing 1988 through 1994 and processed per standard protocol.
12
Laboratory measurements occurred during either morning,
afternoon, or evening Participants were asked to fast
12 hours before the morning examination or 6 hours for the af-
ternoon or evening examination. To minimize hemolysis of blood
samples, \viii affect serum calciu11_1 but not serum 2S(OH)D
levels, all phlebotomists were certified and completed training in
standardized laboratory procedures annually. Serum 2S(OH)D
was measured using the Diasorin radioimmunoassay (RIA) kit
(Diasorin, Stillwater, Minnesota) on frozen serum ( < -20C) be-
tween February 1994 and December 1995 (total coefficients of
variation from quality control samples, 13%-19%). The RIA kit
was calibrated using high-performance liquid chromatography-
purified 2S(OH)D every 6 months. Race/ethnicity was self-
identified. Hypertension was defined as systolic blood pressure
of 140 mm Hg or higher, diastolic blood pressure of90 mm Hg or
higher, and/or use of antihypertensive medications. A partici-
pant was considered to have diabetes mellitus if he or she re-
ported ever being told by a physician that he or she had diabetes
or diabetes" at a time other than during pregnancy, was
taking insulin or a "diabetes pill" at the time of the questionnaire,
or had a fasting blood glucose level greater than 126 mgldl (to
convert to millimoles per liter, multiply by 0.0555) or a nonfast-
ing blood glucose level greater than 200 mgldl. Smoking was clas-
sified as never, current, or forn1er smoker. Low socioeconomic
. status (SES) was defined as 200% of the poverty index or lower.
The use of cholesterol medication was based on the response to
the question, lower your blood cholesterol, are you now fol-
10\\-ing this advice to take prescribed medicine?"
Creatinine levels were calibrated to the Cleveland Clinic labo-
ratory standard by subtracting 0.23 mgldL (to convert to micro-
moles per liter, multiply by 88.4) from the NHANES Ill values,
and estimated glomerular filtration rate (eGFR) was calculated
based on serum creatinine level using the abbre.,iated Modifi-
cation of Diet in Renal Disease (MDRD) formula.H.zs We as-
signed an eGFR value of 200 mUmin/1. 73 m
2
for all indi\-iduals
with an eGFR higher than 200mUminll.73m
1
(n=94) because
these eGFR values were thought to be biologically implausible.
The main analyses were restricted to those participants ,.,ith an
eGFR higher than 15 mUmin/1.73 m
2
Urinary albumin to cre-
atinine ratio was based on a spot urine sample. Both albumin to
creatinine ratio and C-reactive protein values were log-
transformed to achieve approximate normality for statistical
analyses.
Physical activity level was based on the participant's meta-
bolic expenditures in the past month and categorized as low (:S3.5
METs [metabolic equivalent tasks!), moderate (3.6-14.9 METs),
or high (;=:ls METs) physical activity intensity.
26
No infonna-
tion was available about whether the phrsical activities were per-
formed indoors or outdoors. Use of vitamin D supplementation
was coded as positive for patients who reponed taking a supple-
ment or multivitamin containing more than 0 IU of vitamin D
(ergocalciferol or cholecalciferol). Those taking repletion doses
of vitamin D (50000 IU) were excluded from the analysis (n=3).
LINKAGE AND CAUSES OF DEATH
Mortality outcomes were collected on NHANES III participants
through December31, 2000, using probabilistic matching. \vith
up to I 2 identifying data elements, to National Death Index (NDI)
records and were made available in june 2005. A selected sample
of death certificates was r:eviewed manually to validate the pro-
cess. The underlying cause of death was coded according to the
lntemational Classification of Diseases, Ninth Revision, Clinical
(lCD-9-CM)
21
for deaths occurring between 1988
and 1998, and according to lntemational Statistical Classifi-
cation of Diseases, 1Oth Revision (lCD-I 0)
28
for deaths occurring
in 1999 and 2000. All deaths from 1988through 1998 coded un-
der ICD-9-CM guidelines were recoded into comparable groups
based on the ICD-lOunderl}ing cause of death.
27
.lS For this analy-
sis, CVD mortality included deaths coded as due to hyperten-
sive disease (codes ll0-113), ischemic heart disease (codes 120-
125), arrhythmia (codes 144-149), heart failure (code ISO),
cerebrovascular disease (codes 160-169), or atherosclerosis or other
diseases of the arteries (codes 170-178). Cancer mortality in-
cluded deaths coded as all malignant neoplasm deaths (codes
COO-C95) including malignant neoplasms of digestive organs
(codes C15-C26), respiratory and intrathoracic organs (codes
OO-C39), and genital organs (codes CSO-C63). A subanalysis
of cancers previously linked to low 25(0H)D levels included co-
lo rectal cancer (codes C18-C21), breast cancer (code C50), and
prostate cancer (code C61). Infectious deaths included deaths
from infectious and parasitic diseases (codes A00-899), menin-
gitis (codes GOO and G03), acute rheumatic fever (codes 100-
102), endocarditis (code 133), acute upper and lower respira-
tory tract infections including influenza and pneumonia (codes
]00-]06 andj10-j22), and urinary tract infections (code N39).
External causes of mortality included deaths from accidents, falls,
suicide, homicide, and complications of medical and surgical care
(codes V01-Y89).
STATISTICAL ANALYSES
Because of rJ1e complex sampling design ofNHANES Ill, weighted
analyses using the "survey" command in Stata 9.0 (StataCorp, Col-
lege Station, Texas) were used. The distributions of participant
characteristics were ftrst examined by unweighted serum 25(0H)D
ARCH INTERN l\IEO/VOL 168 (NO. 15). AUG 11125, 2008
1630

..
Table 1. Baseline Characteristics of 13331 Participants of NHAHES Ill by 25(0H)O Quartiles
25(0H)O Quartile, 11g/ml
> 32.1 24.4-32.1 17.8-24.3 < 17.8
Characteristic (n-33591 (na3242) (n-3344) (n3386) PValue
Age, Y 41.9 (0.5) 45.2 (0.6) 46.7 (0.6) 45.5 (0.6) <.001
Women,% 45.0(1.2) 50.5 (1 .0) 56.1 (1.4) 66.3 (1.4) <.001
Race/ettmicity, %
Non-Hispanic white 91.4 (0.9) 82.6 (1.2) 69.3 (1.8) 48.1 (2.5) <.001
Non-Hispanic black 2. 1 (0.3) 5.4 (0.5) 13.3 (0.9) 32.8 (1.9) <.001
Mexican American 2.6(0.3) 4.8 (0.5) 6.7 (0.6) 7.2 (0.7) <.001
Other 3.9 (0.9) 7.1 (0.9) 10.7 (1.3) 11.8 (1.3) <.001
Diabetes,% 3.8 (0.5) 6.6 (0.5) 8.2 (0.8) 10.9 (0.8) < .001
History of CVD; % 6.0 (0.5) 7.9 (0.8) 9.5(0.9) 8.2 (0.9) .002
Hypertension,% 19.3 (1.1) 23.8 (1.2) 27.8 (1.2) 28.9 (1.3) <.001
BMI 25.3 (0.1) 26.4 (0.2) 27.6 (0.2) 28.1 (0.2) <.001
Total cholesterol, mg/dl 202.2 (1.0) 202.8 (1.3) 207.5 (1.3) 203.5 (1.2) .04
HDLC, mg/dl 51.2 (0.5) 49.8 (0.5) 50.4 (0.5) 51.8 (0.4) .73
Use of cholesterol-lowering medication, o/o 2.4 (0.3) 2.7 (0.4) 3.3 (0.4) 3.7 (0.7) .02
Systolic BP. mm Hg 120.1 (0.6) 122.3 (0.5) 124.0 (0.5) 124.4 (0.6) <.001
Diastolic BP, mm Hg 73.6 (0.2) 74.2 (0.3) 74.7 (0.3) 75.0 (0.3) < .001
Smoking,%
Former 28.3 (1.2) 26.8 (1.1) 25.3 (1.2) 21.5 (1.3) .001
Current 29.0 (1.4) 26.0 (1.4) 26.9 (1.3) 32.6 (1.3) .26
Season, o/o
Winter, JanMar 10.6 (2.1) 15.3 (2.3) 20.6 (3.6) 27.3 (4.7) <.001
Spring, Apr-Jun 22.0 (4.1) 28.8 (4.6) 28.8 (4.3) 26.3 (4.3) .04
Summer, JuiSep 43.4 (5.5) 33.1 (5.6) 26.0 (4.2) 19.2 (3.5) <.001
Fall, Oct-Dec 24.0 (4.7) 22.8 (4.5) 24.7 (4.0) 27.2 (4.1) .34
eGFR <60 mUmin/1.73 m
1
,% 3.3 (0.4) 4.1 (0.4) 4.8 (0.5) 4.6 (0.6) <.001
Albumin _mg/g_of creatinine,%
...
6.1 (0.4) 7.4 (0.7) 9.9 (0.6) 12.6 (1.1) < .001
CRP >0.021 mg/L, 24.0 (1.5) 26.6 (1.6) 30.7 (1.6) 37.0 (1.7) <.001
--Serum albumin, g/dl 4.23 (0.02) 4.21 (0.03) 4.14 (0.02) 4.08 (0.02) <.001
LowSES, Ofo . 29.8 (1.5) 29.4 (1.4) 35.7 (2.0) 41.9 (1.6) <.001
Use of vitamin D supplementation, % 35.2 (1.3) 30.1 (1.2) 25.2 (1.2) 16.3 (1.3) <.001
Physical activity level, %
Moderate 55.6 (1.3) 56.4 (1.1) 53.7 (2 .. 0) 48.1 (1.4) .002
High 28.7 (1.5) 20.9 (1.0) 16.6 (1.1) 11.7 (1.3) <.001
Abbreviations: 25(0H)O, 25-hydroxyvitamin D; BMI, body mass index (calculated as weight In kilograms divided by height in meters squared); BP. blood
pressure; CRP, C-reactive protein; CVD, cardiovascular disease; eGFR, estimated glomerular filtration rate; HDL -C, high-den.sity lipoprotein cholesterol;
NHANES Ill, Third National Health and Nutrition Examination Survey; SES, socioeconomic status .
.-SI conversion factors: To convert 25(DH)O to nanomo!es per liter, multiply by 2.496; cholesterol to millimoles per liter, multiply by 0.0259; CRP to nanomoles
per liter, multiply by 9.524; and serum albumin to grams per liter, multiply by 10.
3
Data are give.n as mean (SE) value or percentage (SE) of participants unless otherwise specified.
level quartifc5 (Table 1 gives the quartile cutoff values). Multi-
variate logistic regression analyses were used to detem1ine inde-
pendent predictors of 25(0H)D deficiency, defined as being in
the lowest quartile (2S[OH) D levels < 17.8 nglml). A sensitiv-
ity analysis was performed using multivariate linear regression,
with log-transformed 2S(OH)D levels as the dependent vari-
able. Power calculations using PS power (http://biostat.mc
.vanderbilt.edu/r.vikilbin/vie\\/Main/PowerSampleSize) and s:1mple
size program
29
revealed a detect.able mortality rate ratio (MRR)
of greater than 1.17 or less than 0.84 for all-cause mort.ality (1.25
or 0.78, respectively, for CVD mortali ty) assuming 80% power
and an a level of .05.
Poisson regression models were used to examine the inde-
pendent associations between 25(0H)D levels and cancer, CVD,
other causes, and all-cause mortality. For the analyses of cause-
specific mortality, participants were censored at the time of death
from other causes. We explored the continuous, potent ially non-
linear, relationship of mortality risk associated with 25(0H) D
levels using fully adjusted restricted cubic spline models.
Because serum 25(0H)D levels vary by season,
30
all multi-
variate anai}'Ses were adjusted for season of examination. Inclu-
sion in the fi nal model was based on the variable of interest being
associated \vi th both 25(0H)D levels and mortality (P< .20) and
on a priori detemlination of confounders of the association be-
tween 25(0H)D levels and mortality. Co\oariates included in the
fi nal model were age, sex, race, season, hypertension, history of
CVD, diabetes mellitus, smoking, body mass index (BMI), high-
density lipoprotei n cholesterol, total chole.Sterol, the tiSe of cho-
lesterol-lowering medications, eGFR categories, serum albumin
level , log urinary albumin to creatinine ratio, log C-rc.active pro-
tein, physical activity level, vitamin D supplementation, and low
SES. No pairs of co variates \vith correlations of 0. 7 or greater were
included in the models to avoid issues of colinearity. Interac-
tions were tested by adding a product tenn for 2S(OH)D quar-
tile and each of the foiiO\ving covariates: age, sex, race, history of
CVD, diabetes status. h>'Pertension status, obesity status, physi-
cal activi ty level, smoking, and baseline eGFR lower than
60 mUmin/1. 73 m
2
Because non-Hispanic blacks have lower se-
rum 25(0H)D levels compared \vith non-Hispanic whites, the
highest quartile of 25(0H)D (the comparison group) only had
371 non-Hispanic blacks, compared \vith 216-+ in the lowest quar-
tile. Because this small comparison group may spuriously influ-
ARCH INTERN MED/VOL 168 (NO. 15), AUG 11125.2008 \VW\V. ARCHINTERNMED.COM
1631
Material may be protected by copyright law (Title 17, U. S. Code)
'
Table 2. Baseline Characteristics of 13 331 Participants of NHANES Ill by 25(0H)D QuartJies for Multlcategory Variables
25(0H)D Quartile, ngtml
> 32.1 24.4-32.1 17.8-24.3 <17.8
Characteristic (n-3359) (n 3242) (n=3344) (na3386)
Race/ethnicity, %
NooHispaoic white 43.5 (1.4) 28.5 (0.9) 18.5(0.8) 9.5 (0.6)
Non-Hispanic black 7.8 (0.8) 14.4 (1.0) 27.5 (0.9) 50.3 (2.1)
Mexican-American 2Q.3 (1.3) 27.1 (1.1) 29.2 (1.2) 23.4 (1.6)
Other 19.4 (2.7) 25.9 (1.9) 30.1 (2.4) 24.6 (2.3)
Smokino.%
Never 34.8 (1.5) 27.8 (0.9) 21.9 (0.9) 15.6 (0.9)
Former 39.9 (1.9) 27.4 (1.2) 20.1 (1.2) 12.6 (0.9)
Current 37.9 (1.4) 24.6 (1.2) 19.7 (1.0) 17.8 (1.1)
Season,%
Winter, Jan-Mar 23.7 (1.5) 24.8 (2.0) 26.0 (2.0) 25.5 (1.9)
Sprino. Apr-Jun 31.5 (1.9) 29.8 (1.1) 23.1 (0.8) 15.6 (1.2)
Summer, Jui-Sep 48.3 (1.8) 26.6 (1.4) 16.2 (06) 8.9 (0.7)
Fall, Oct-Dec 36.5 (1.7) 25.1 (1.4) 21.1 (8.8) 17.3 (1.4)
Physical activity, %
Low 23.9 (1.1) 25.1 (1.1) 25.5 (1.0) 25.5 (1.3)
Moderate 37.9 (1.3) 27.8 (0.9) 2Q.6 (0.9) 13.7 (0.8)
High 49.5 (22) 26.1 (1.0) 16.1 (1.2) 8.4 (0.8)
Abbreviations: 25(0H)O, 25-hydroxyvitamin 0; NHANES Ill, Third National Health and Nutrition Examination Survey.
3
Data are given as mean (SE) value or percentage (SE) of participants.
ence results, we elected to analyze non-Hispanic black-specific
quartiles for the subgroup analysis (the _c-uto(f values were > 24
nglmL (quartile 1), 18 to'S24 nglmL (quartile 21. 13.5 to Sl8
nglmL, and <13.5 nglmL (quartile 4]).
In addition, to test the robustness of the association, we per-
formed sensitivity analyses adjusting for serum calcium and phos-
phate levels, which are affected by vitam_in D status and are as-
sociated with mortality in patients \vith end-stage renal disease
11
and in tl1e general population.
32
Because diabetes mellitus and/or
hypertension may be in cai.lsal pathway between low 25(0H)D
levels and mortality, we developed models \vith and \vithout dia-
betes mellitus and hypertension. Because of potential nonlinear
associatiqns between continuous variables such as age, BMI, and
mortality, we also performed a sensitivity analysis in which all
variables were added to the model as categorical variables (quar-
tiles of the continuous variables). We also further tested the non-
linear association found in the spline by using the follo,ving cut-
offs for 25(0H)D levels: lower than 20 nglml, 20 to 29 nglmL,
30 to 39 nglmL, 40 to 49 nglmL, and 50 ng/ml or higher, \vith
30 to 39 nglmL as the reference group. We calculated the attrib-
utable risk percentage by dividing the elevation from baseline
rate (1.0) by the MRR for the lowest quartile of 25(0H)D level.
We calculated the population attributable risk percentage by mul-
tiplying the attributable risk percentage by the percentage of the
population who were in the lowest 25(0H)D quartile. For all
analyses, P < .05 for 2-tailed tests was considered statistically sig-
nificant. P values were not adjusted for multiple comparisons.
RESUlTS
FACTORS ASSOCIATED WITH
LOW 25(0H)D LEVELS
Older participants, women, and non-Hispanic blacks had
lower 2S(OH)D levels (Table 1 and Table 2 ). Across
decreasing quartiles of 25(0H)D level (highest quartile
to lowest), mean systolic and diastolic BP, mean BMI, and
percentages of patients with diabetes, with elevated al-
bumin to creatinine ratios, and with elevated C-reactive
protein levels increased and mean serum albumin levels
decreased. There were more participants with low SES
and fewer participants taking vitamin D supplements and
participating in high and moderate levels of physical ac-
tivity in the lowest 2S(OH)D quartile ( <17.8 nglmL).
In multivariate models, increasing age, female sex, non-
white race/ethnicity (particularly non-Hispanic black race/
ethnicity), diabetes, current smoking, and increasing BMI
were all independently associated with an increased odds
of being 2S(OH)D deficient (ie, lowest quartile) (Table 3 ).
Physical activity, vitamin D supplementation, and non-
\vinter season were associated with a decreased odds of
deficiency. In unadjusted analysis, low SES was associ-
ated with a higher risk of deficiency (OR, 1.60; 95% CI,
1.40-1.81), which after adjustment became a lower risk
of deficiency (OR, 0.73; 95% CI, 0.60-0.89). The primary
. confou.nders of this association were race and physical ac-
tivity; both associated \vith a greater than 15% in
the point estimate. When using log-transformed 2S(OH)D
levels in a linear regression, all of the associations given
in Table 3 remained the same, including the reversal of
the low SES point estimate \vith adjustment.
ASSOCIATIONS BETWEEN 25(0H)D LEVELS
AND ALL-CAUSE MORTALITY
During a median 8.7 years of follow-up (intraquartile range,
7.1-10.2 years), there were 1806deaths, of which 777 (43%)
wen! ascribed to CVD. 424 (23%) were ascribed to can-
cers, 105 (6%) were ascribed to infectious causes, and 92
(5%) were ascribed to external causes of death. Of those
who died of CVD, 590 (76%) died of atherosclerotic CVD,
145 (19%) of cerebrovascular disease, and 42 (5%) of con-
ARCH INTERN MED/VOL 168 (NO. 15), AUG 11125, 2008
1632
',
gestive heart failure. Those _who died had a mean age of
66.4 years and 46% were female, 58% were hypertensive,
19% had diabetes, and 32% had prior CVD.
In unadjusted analysis, having a 25(0H)D level in the
lowest quartile was associated with a 78% increased risk
of all-cause mortality (Table 4 ). The continuous, fully ad-
justed association between vitamin D levels and all-cause
mortality is shown graphically in Figure 1. After further
adjusting for known CVD risk factors including BMI
renal function, low SES, and markers of a healthy lifestyle
including physical activity and usc of vitamin D supple-
mentation, there was still a 26% higher rate of aU-cause mor-
tality for the lowest 25(0H)D quartile (MRR, 1.26; 95%
CI, 1.08-1.46) compared with the highest quartile. This ad-
. justed MRR translates to an attributable risk percentage of
20.6% and a population attrib"utable risk percentage of3.1%
associated with the lowest 25(0H)D quartile.
ASSOCIATIONS BETWEEN 25(0H)D LEVELS
AND CAUSE-SPECIFIC MORTALITY
For CVD mortality, the unadjusted analysis revealed that
having a 25(0H)D level in the lowest quartile was asso-
ciated with a 70% higher risk of mortality. This associa-
.tion.did not remain statistically significant in the fully ad-
justed model, though the magnitude of the point estimate_
was similar to that for all-cause mortality (MRR, 1.20; 95%
CI, 0.87-1.64) (Table 4). There was a nonsignificant as-
sociation between lowest 25(0H)p quartile and cancer:.
specific mortality in an unadjusted model (MRR. 1.31; 95%
CI, 0.96-1.81), whith disappeared after adjustment for age,
sex, race, and season (MRR, 1.05; 95% Cl, 0.74-1.47).
When testing only cancers previously associated with
25(0H)D levels (colorectal, breast, and prostate cancer
[n = 116)), the lowest quartile had an adjusted MRR of 1.38
(95% CI, 0.49-3.93) and the 17.8- to 24.4-nglmlquartile
had an MRR of 2.05 (95% Cl, 1.14-3.68).
SENSITIV1TY ANALYSES
Results did not change significantly when serum calcium
and phosphate levels were .ente_red in the fully adjusted
model (MRR for lowest 25(0H)D quartile, 1.27 (95% CI,
1.10-1.47) for all-cause mortality; 1.21 (95% Cl, 0.89-
1.65] for CVD mortality). Likewise, the associations re-
mained relatively unchanged in models assessing all-
cause mortality not including diabetes mellitus and
hypertension (Table 4) and, separately, in models not in-
(MRR,l.28; 95% Cl, 1.11-1.48) and not
including hypertension (MRR, 1.26; 95% CI, 1.08-1.45).
When continuous variables such as age were added to the
model as categorical variables, the association between the
lowest 25( 0 H) D quartile and mortality strengthened (M RR,
1.34; 95% Cl, 1.15-1.57). Results also did not differ when
analyzed by different 25(0H)D cutoff values. As shown
in Figure 2 , in women, having both low ( <20 nglmL)
and high (>50 nglmL) 25(0H)D levels was associated \vith
an increased rate of mortality.
SUBGROUP ANALYSES
The lo\vest 25(0H)D quartile was more strongly asso-
ciatedwith mortality among participants without a
Table 3. Independent Predictors of 25(0H)D Deficiency
(2510HJD l evels < 17.8 ng/ml) Among 13331 Adults
Older Tban 20 Years
OddsRaUo
(95% ConlldeiiC8
Cbaracterlstlcs lllleml)
/lqe, per 10 y 1.10 (1.04-1.17)
Female sex 2.26 (1.90-2.69)
Race/ethoicity
No!Hiispanic black 10.11 (8.1312.n)
Mexican American 2.45 (1.96-3.06)
Other 3.03 (2.18-4.20)
Diabetes 1.46(1.1&-1.84)
History of CVO 0.87 (0.64-1.18)
Hypertension 0.96 (O.n-1.21)
BMI, per1 U 1.04 (1.021.06)
Total cholesterol, per 10 mgldl 0.98 (0.97-1.00)
HDLC, per 10 mgldl 1.02 (0.971.06)
Use of cllolesteroHowerino medications 1.39 (0.85-2.29)
Former smoking 1.09 (0.911.31)
Current smoking 1.59 (1 .35-1.87)
Season
Spring, Apr-Jun 0.67 (0.49-0.92)
Summer, 0.31 (0.23-0.43)
Fall, OctDec 0.63 (0.45-0.86)
eGFR <60 mUmin/1.13 nr 1.05 (0.75-1.47)
Albuminuria > 30 mgtg of creatiniM 1.26 (0.96-1.64)
CRP >0.021 rngll 0.96 (0.78-1.20)
Serum albumin, per 1 o.93 (O.n-1.21)
lowSES 0.13 (0.60-0.89)
Use of vi1amin 0 supplementation 0.45 (0.36-0.56)
Modelllte physical activity 0.65 (0.53-0.80)
High physical activity 0.44 (0.32.0.61)
p
Value
.001
< .001
<.001
<.001
<.001
.002
.36
.74
< .001
.01
.44
.19
.33
<.001
.01
<.001
.005
.n
.09
.13
.58
.002
<.001
<.001
<.001
Abbreviations: 25(0H)D, 25-hydroxyvitamin 0; BMI, body mass index
(calculated as weight in kilograms divided by height in meters squared);
BP. blood pressure; CRP, Creactive protein; CVD, cardiovascular disease;
eGFR, estimated glomerular filtration rate; HDL-C, high-density lipoprotein
cholesterol; NHANES Ill, Third National Health and Nutrition Examination
Survey; SES, socioeconomic status.
Sl conversion factors: To convert 25(0H)D to nanomoles per liter. multiply by
2.496; cholesterol to millimoles per liter, multiply by 0.0259; CRP to nanomoles
per liter, multiply by 9.524; and serum albumin to grams per filer, multiply
by10.
a Multivariate model adjusted for all factors listed in the Table. Reference
categories are non-Hispanic white for race/ethnicity; nonsmokers for smoking
status: greater than 60 mUmin/1.73 m
1
for eGfR; low physical activity for
physical activity level; and winter season for season (Jan-Mar).
history of CVD (MRR, 1.54; 95% Cl, than
in lhosc with a history of CVD (MRR, 0.90; 95% Cl,
0.66-1.22) (P value for interaction, .006) (Table 5 ).
Stronger associations, although were
found for participants without hypertension (P value
for interaction, .09), participants without diabetes
(P value for interaction, .09), and for women (P value
for interaction, .06). No interaction was observed for
age, race, obesity, physical activity level, smoking, or
eGFR. .
In the general US population, we found that 25(0H)D de-
ficiency (lowest quartile, < 17.8 nglmL) was associated with
a 26% higher risk of all-cause mortality, independent of
baseline demographics, traditional and nontraditional CVD
ARCH INTERN MED/VOL 168 (NO. 15), AUG 11125, 2008 \V\VW.ARCHIKTERNMEO.Cm.t
1633
..
Table 4. Rale Ratios of All-Cause and CVD Mortality by 25(0H)D Quartlles In 13331 Partlclpanls of NHAHES m
2li(OH)D Ollartlll, 1g/llll
MRR > 32.1 24.4-32.1 17.824.3 < 17.8
All-ause MAR
Unadjusted 1 (Reference) 1.14 (0.94-1.39) 1.49 (1.24-1.78) 1.78 (1.44-2.21)
limiledb 1 (Reference) 0.92 (0.78-1.08) 1.11 (0.95-1.31) 1_.52 (1.31-1.77)
fuiiC
1 (Reference) 0.93 1.()6 (0.89-1.24) 1.26 (1.08-1.46)
Full without diabetes mellitus and hyperteoslon 1 (Reference) 0.94 (0.80-1.12) 1.06 (0.9G-1.26) 1.28 (1.11-1.48)
CVO MRR
Unadjusted 1 [Reference) 1.07 (0.82-1.41) 1.31 (1.01-1.71) 1. 70 (1 .222.37)
Umitedb 1 I Reference I 0.85 (0.67-1.09) 0.98 (0.76-1.27) 1.53 (1.12-2.08)
Fullc 1 (Reference) 0.83 (0.65-1.07) 0.88 1.20 (0.87-1.64)
Full without diabetes meHitus and hypertension 1 (Reference I 0.85 (0.66-1.09) 0.89 1.22 (0.9G-1.65)
Cancer MRR
Unadjusted 1 (Reference) 0.97 (0.65-1.45) 1.52 (1.1G-2.10) 1.31 (0.96-1.81)
limitedb
1 I Reference) 0.78 (0.53-1.16) 1.11 (0.821.51) 1.05 (0.74-1.47)
Full
6
1 (Reference) 0.80 (0.54-1.19) 1.08 (0.80-1.46) 0.91 (0.63-1.31)
Infectious disease MRR
Unadjusted 1 [Reference) 1.28 (0.65-2.53) 1.22 (0.67-2.21) 1.46 (0.73-2.91)
limitedb 1 (Reference) 1.06 (0.532.12) 0.90 (0.46-1 .76) 1.10 (0.47-2.56)
Fullc 1 [Reference) 1.01 (0.53-1.93) 0.87 (0.431.74) 0.84 (0.38-1.86)
External cause MRR
Unadjusted 1 [Reference) 0.88 (0.25-3.08) 1.59 (0.58-4.37) 1.17 (0.44-3.07)
limitedb 1 (Reference I 0.85 (0.2&2.83) 1.47 (0.54-3.99) 1.05 (0.45-2.48)
fulld 1 [Reference) 0.92 (0.27-3.11) 1.68 (0.63-4.47) 1.27 (0.56-2.87)
Abbreviations: 25(0H)O, 25hydroxyvitamin 0; CVO, cardiovascular disease; MRR, mortality rate ratio; NHANES Ill, Third National Health and Nutrition
Examination Survey.
Sl conversion factor. To convert 25(0H)O to nanQmoles per liter, multiply by 2.496.
aoata are given as MMR (95% confidence interval). _
bUmited model adjusted for age; sex, race, and season. .
cThe fully adjusted model includes age, sex, race, season, hypertension. history of cvo, diabetes, smoking, high-density lipoprotein cholesterol, total
cholesterol, use of cholesterol-lowering medications, estimated glomerular filtration rate categories, serum albumin level, log albumin to creatinine ratio, log
Creaclive protein level, body mass index, physical activity level, use of vitamin 0 supplementation. and low socioeconomic status.
d Fully adjusted cancer and external causes of death models include age, sex, race, season, cigareHe use, body mass index, log e-re active protein level, serum
albumin level. physical activity level, use of vitamin D supplementation. and l ow socioeconomic status.
:::-2.5
u

0 2.0

:::;: 1.0
..

o.s
..,
..,
<
0 10 20 30 40 50 60
Serum 25(0H)O l evel, ng/ml
Figure 1. Restricted cubic spline showing the fully adjusted associations
between serum 25-hydroxyvitamin 0 (25[0H)D) levels and all-cause
mortality in 13 331 participants of the Third National Health and Nutrition
Examination Survey. Knots are at 1 0.9, 20.5, 28.9, and 45.9 ng/ml. To
convert 25(0H)O to nanomoles per liter, multiply by 2.496. Cl indicates
confidence interval.
risk factors, and measures of a healthy lifestyle. The esti-
mated association with increased risk of CVD mortality
was similar, though not statistically significant. We did not
find an association with cancer mortality or other causes
of death. This is the first study, to our knowledge, to ex-

o female
Male
l
r
<20 20-29 30-39 4()-49 >50
25(0H)D level, nglml
Figure 2. Associations between 25-hydroxyvitamin 0 (25(0H)D) levels and
all-cause mortality in 13 331 participants of the Third National Health and
Nutrition Examination Survey, overall and by sex. To convert 25(0H)O to
nanomoles per liter, multiply by 2.496. Cl indicates confidence interval.
plore the association between 25(0H)D levels and mor-
tality in the general population.
Several lines of evidence suggest that vitamin D defi-
ciency may be a risk factor for cardiovascular, cancer, and
all-cause mortality. Ecological studies reveal that CVD
events are higher in the winter when vi tamin D levels are
ARCH INTERN MED/VOL168 (NO. 15), AUG I I nS, 2008 \V\V\V.ARCHINTERNMED.COM
163 ..
..
Table 5. The Adjusted AssociaUons of 25(0H)D levels in 13331 Participants of HHANES Ill and AIICauu Mortality
In Specific Participant Subgroups
25(0H)D Quartile, ngtmL
Subgroup (No. of RlstJNa. at Dntlls) > 32.1 24.4-32.1 17.8-24.3 < 11.1
Age, y
-
;:: 65 (2867/1243) 1 [Reference) 0.97 (0.79-1.19) 0.99 (0.821.20) 1.26 (1.03-1.54) J
< 65 (104641563) 1 [Reference] 0.81 (0.58-1.14) 1.13 (0.81-1.56) 1.28 (0.931.76)
Sex
Men (6277/1047) 1 (Reference] 0.82 (0.641.05) 0.94 (0.75-1.19) 1.04 (0.83-1.30) J
Women (70541759) 1 [Reference) 1.16 (0.87-1.55) 1.27 (0.97 -1.66) 1.51 (1.151.98)
Raoe/ettmiclty
Non-Hispanic white (5696/1 056) 1 {Reference) 0.92 (0.78-1.08) 1.06 (0.88-1.27) 1.22 (1.03-1.45)
Non-Hispanic blac;l(b (3597 /401) 1 [Reference) 0.80 (0.59-1.1 0) 1.03 (0.77-1.36) 1.05 (0.77-1.44)
Mexican American (3511/318) 1 [Reference] 1.28 (0.72-2.25) 1.40 (0.85-2.30) 1.46 (0.89-2.42)
Other (527 /31 I 1 (Reference] 0.65 (0.08-4.92) 0.78 (0.1 15.70) 2.50 (0.59-10.60)
CVD
Present (13961632) 1 [Referefl(ej 0.67 (0.49-0.92) 0.75 (0.53-1.04) 0.90 (0.66-1.22) J
Absent (1193511174) 1 (Reference] 1.10 (0.84-1.43) 1.22 (0.96-1.57) ' 1.54 ( 1.23-1.94)
Hypertension
Present ( 406411114) 1 (Reference] 0.87 (0.721.05) 1.00 (0.80-1.25) 1.12 (0.891.41) J
Absent (9267/692) 1 [Reference] 0.96 (0.661.38) 1.13 (0.861.50) 1.42 (1.00-2.01)
Diabetes
Present (1295/408) 1 (Reference I 0.77 (0.53-1.11) 1.17 (0.85-1.62) 1.10 (0.77-1.58) J
Absent (12036/1398) 1 (Reference) 0.97 (0.791.18) 1.03 (0.84-1.27) 1.36 (1.141.61)
BMI
<30 (996311434) 1 [Reference) 0.95 (0. 791.13) 1.01 (0.831.23) 1.22 (1.01-1.48) J
;:: 30 (33681372) 1 (Reference) 0.81 (0.56-1.17) 1.09 (0.74-1.62) 1.38 (0.92-1.08)
eGFR. mUmin/1.73 m
2
. < 60 (7431417) 1 [Reference) . 1.13 (0.851.50) 1.08 (0.781.49) . 1.28 (0.961.70) J
;:: 60 (1258811389). 1 [Reference J 0.87 (0.71-1.07) 1.03 (0.851.25) 1.20 (1.001.44)
Physical activity level :
1.13 (0.91:1.40) - Low (43591819) 1 (Reference) 0.96 (0.71-1.29) 1.35 (1.03-1.76) J
. Moderate (6772/896) 1 [Reference] 0.92 (0.75-1.12) 1.04 1.25 (0.97-1 .61)
High (2200191) 1 [Reference) 1.26 (0.682.33) 1.21 (0.66-2.23)' 0.95 (0.29-3.12)
Smoking status
Nonsmokers (6542/697) 1 [Reference) 0.98 (0.761.25) 1.15 (0.83-1 .57)
1.37 (0.98-1 .93) J
Former smokers (33221681) 1 (Reference) 0.97 (0.74-1.28) 0.99 (0.76-1.29) 1.18 (O.BS-1.63)
Current smokers. (3467/428) 1 I Reference I 0.83 (0.561.23) 1.09 (0.79-1.51) 1.16 (0.821.63)
Abbreviations: 25(0H)D, 25-hydroxyvitamin D; BMI, body mass index (calculated as weight in kilograms divided by height in meters squared);
CVO, cardiovascular disease; eGFR, estimated glomerular filtration rate; NHANES Ill, Third National Health and Nutrition Examination Survey.
Sl conversion factor: To convert 25(0H)D to nanomoles per liter, multiply by 2.496.
PValue far
tateradlon
.14
.06
.74
.87
.27
.006
.09
.09
.68
.35
.49
.58
1
Models were adjusted for age, sex, race, season, hypertension, history of CVD, diabetes, smoking, high-density lipoprotein cholesterol, total cholesterol, use
of cholesterol-lowering medications, eGFR categories, serum albumin level, log albumin to creatinine ratio, tog Creactive protein level, body mass Index, physical
activity level, use of vitamin 0 supplementation, and tow socioeconomic status. except for the variable in the subgroup.
bAfrican American-specific quartiles were the following: quartile 1, higher than 24 ng/ml; quartile 2, 18 to 24 ng/ml or lower; quartile 3, 13.5 to lower than
18 ng/ml; and quartile 4,1ower than 13.5 ng/ml.
lowerl
3
and that cancer survival is better if the cancer is
diagnosed in the summer when vitamin D levels are

In addition, observational data show that the

use of activated vitamin D therapy in patients with end-
stage renal disease is associated with decreased mortal-
ily.6-10 Vitamin D receptor knockout mice experience car-
diac myocyte hypertrophy.
1
Activated vitamin D has
anti proliferative properties.
5
In mice, vitamin Dis an in-
hibitor of the renin-angiotensin system. Other inhibi-
tors of the renin-angiotensin system, such as angiotensin-
converting enzyme inhibitors , reduce mortality and
morbidity in multiple disease states.
3
s
_l ow 25(0H)D levels are also associated \Vith hyperten-
sion, diabetes mellitus. insulin resistance, and an elevated
BMI, all of which are risk factors for CVD and all-cause mor-
tality. low 25(0H)D levels,u but not vitamin D intake,
36
are associated with incident hypertension. There are 2 small
clinical trials that suggest that vitamin D supplementa-
tion37,.3S reduced systolic BP. The association of 25(0H)D
deficiency \vith obesity,
39
i0 glucose intolerance,
15
1
M
1
and
the metabolic syndromeH is another potential mechanism
for increased CVD risk. We found that diabetes mellitus
and increased BMI were independently associated with in-
creased odds of being 25(0H)D deficient. We adjusted for
hypertension, cUabetes mellitus, and BMI in our analyses
and still found a statistically significant association \vith all-
cause mortality. Analyses with and without hypertension
and diabetes mellitus in the model did not reveal a differ-
ence in the risk estimates associated with low 25(0H)D lev-
els. However, this analysis did not account for incident dia-
betes and hypertension developing after the NHANES
survey period.
The association of low 25(0H)D levels with mortal-
ity was strongest in those without CVD, without hyper-
ARCH INTERN 1\IED/VOL 168 (NO. 15), AUG 11125, 2008
1635
tension, and without diabetes mellitus, arguing against
low vitamin D levels being only a marker of poor gen-
eral health. If diabetes mellitus and hypertension are in
the causal pathway between low 25(0H)D levels and mor-
tality (ie, low 25(0H]D leads to hypertension, which leads
to higher mortality), it follows that the effect is more pro-
nounced in those without preexisting diabetes mellitus,
hypertension, and CVD. The fact that the associations
were stronger in those without CVD at baseline sug-
gests that if a causal relationship exists, 25(0H)D defi-
ciency may play a role before CVD is established.
It is unclear why the association between 25(0H)D
levels and mortality was more pronounced among women.
ll may be that there is a hormone interaction between
estrogen levels and 25(0H)D. <;ompared with men,
women tend to develop atherosclerosis later in life,
suggesting that, as in the subgroup without CVD, low
25(0H)D levels play a role in the development of ath-
erosclerosis. This area deserves further research and may
help reveal the mechanisms behind the associat ions seen,
if in fact they are causal.
A recently published randomized clinical trial of cal-
cium and vitamin D
3
supplementation ( 400 IU/d) in gen-
erally healthy postmenopausal women did not show a
lower risk of cardiovascular events in those randomized
to vitami n D and calcium supplementation.
43
However,
a meta-analysis of 18 randomized clinical trials showed
that participants randomized to vitamin D supplemen-
tation experienced fewer deaths compared with those ran-
domized placebo.
18
In that analysis, as in ours, they
were noi' able to establish the specific cause of death re-
sponsible for the lower mortalily. .
Several authors have commented that the optimal lev-
els o'f 25(0H)D should be greater thi.m 30 nglmL.
19
20
In
our observational study, we found that there was a lower
risk of mortality at levels 'of 30 to 49 nglmL, but that at
levels greater than 50 nglmL there was again a higher risk
of mortality in women. This is similar to findings about
antioxidant vitamins and vitamin E, which show that too
much may be harmfu!.++
45
Several associations between vitamin D deficiency and
clinical factors need further explanatiori'.ln adjusted analy-
sis, low SES was associated with a lower risk of being
25(0H)D deficient. Most of the confounding was due to
the very strong association between non-Hispanic black
race and low 25(0H)D status. The mechanism for the pro-
tective effect of low SES is unknown. However, one may
conjecture that participants with low SES may spend more
time outdoors and therefore have more sun exposure.
Our study is limited in that it is an observational study,
and therefore causality cannot be inferred. We adjusted
for all known traditional CVD risk factors and nontra-
ditional risk factors, including markers of a healthy life-
s tyl e, such as the use of vitamin D supplementation. Re-
sidual confounding, however, may still exist. Specifically,
the excess mortality observed in the lowest quartile may
be a reflection of the participant's overall poor condi-
tion or an unmeasured confounder, which we could not
fully adjust for in our analysis. We were not able to dis-
tinguish the specific causes of mortality that accounted
for the elevated all-cause mortality risk associated with
25(0H)D deficiency in ourpopulation, perhaps owing
to limitations of power or to the use of potentially im-
precise death certificate data. Northern states were only
sampled in the summer in NHANES III, and therefore
the full extent of 25(0H)D deficiency in the population
is probably underestimated. This probably attenuates the
association seen in our results. Finally, one must view
the subgroup analyses with caution because of issues of
multiple comparisons. .
In conclusion, the lowest 2S(OH)D quartile ( < 17.8 ngl
ml) is associated with a higher risk of all-cause mortality
in the general US population. Further observational stud-
ies are needed to confirm these findings and establish the
mechanisms underlying these observations. If con-
firmed, randomized clinical trials will be needed to deter-
mine whether vitamin D supplementation at higher doses
could have any potential benefit in reducing future mor-
tality risk in those with 2S(OH)D deficiency.
Accepted for Publication: j anuary 15, 2008.
Correspondence: Michal L Melamed, MD, MHS, Division
of Nephrology, Department of Medicine, Albert Ein
stein College of Medicine, 1300 Morris Park Ave, Ull-
mann 615, Bronx, NY 10461 (mmelamed@aecom.yu
. edu).
Author Contributions: Drs Melamed and Michos had full
access to all the data in the study, take responsibility for
the integrity of the data and the accuracy of the data analy-
sis, and contributed equally to this work. Study concept
and design: Melamed and Michos. Acquisition of data:
Melamed and Michos. Analysis and interpretation of data:
Melamed, Michos, Post, and Astor. Drafting of t11e manu-
script: Melamed and Michos. Critical revision of the manu-
script for important intellectual content: Melamed, Michos,
Post, and Astor. Statistical analysis: Melamed, Michos,
and Astor. Obtained funding: Melamed. Study
sion: Michos, Post, and Astor.
Financial Disclosure: Dr Michos has received consult-
ing fees from Abbott Pharmaceuticals.
Funding/Support: Dr Melamed and this analysis were sup-
ported by grant F32-DK069017 and K23-DK078774 and
Dr Michos was supported by grant T32-HL007024 from
the National Institutes of Health. Dr Michos is also sup:
ported by the PJ Schafer Cardiovascular Research Fund.
Dr Post is supported, in part, by the Paul Beeson Physi-
cian Faculty Scholars in Aging Program.
Role of the Sponsor: The funding organizations had no
role in the design and conduct of the study; the collec-
tion, management, analysis, and interpretation 9f data; or
the preparation: review, or approval of the manuscript.
Disclaimer: All analyses, interpretations, and conclu-
sions are made by the authors and do not represent views
of the National Center for Health Statistics (NCHS).
Additional Information: The NCHS is the source of the
data used in this analysis.
Additional Contributions: Christopher Rogers, PhD, of
the NCHS aided in the analysis.
1. U vc. Kong J, Wei M, ChenZf. Uu SO. Cao LP. 1,250ihydroxyvitamin 0(3) is
a negative endocrine regulator of the renin-angiotensin system. J Ctin Invest.
2002;110(2):229238.
ARCH INTERN MED/VOL 168 (NO. 15), AUG 11125, 2008 W\V\V.ARCHINTERNMED.COM
1636
s.
6.
7.
8.
9.
10.
11.
12 .
13.
14.
15.
16.
17.
18.
19.
20.
21.
22. I
'
23.
{
I
f
I
..
2. Xiang W, Kong J, Chen S, et al. Cardiac hypertrophy In vitamin D receptor knock
out mice: role of the systemic and cardiac renin-angiotensin systems. AmJ Physiol
Endocrinol Metab. 2005;288(1):E12H132.
3. Mathieu C. Adorini L The coming of age of 1 ,25dihydroxyvitamin D(3) analogs
as immunomodulatory agents. Trends Mol Med. 2002;8(4):174179.
4. Levin A, Li YC. Vitamin D and its analogues: do they protect against cardiovas
cular disease in patients with kidney disease? Kidney Int. 2005;68(5):1973
1981.
5. Banerjee P, Chatterjee M. Antiprotiferative role of vitamin D and its analogs-a
brief overview. Mol CetiBiochem. 2003;253(12):247254.
6. Teng M, Wolf M, Ofsthun MN, et at. Activated injectable vitamin D and hemodialy
sis survival: a historical cohort stud:t. JAm Soc Nephrol. 2005; 16(4):11151125.
7. Melamed Ml, Eustace JA, Ptanlinga l, et al. Changes In serum calcium, phos
phate. and PTH and the risk of death in incident dialysis patients: a longitudinal
study. Kidney Int. 2006;70(2):351357.
8. Tentori F. Hunt we. Stidley CA. et al. Mortality risk among hemodialysis pa
tients receiving different vitamin D analogs. Kidney Int. 2006;70(10):1858
1865.
9. KalantarZadeh K, Kuwae N, Regidor DL. et al. Survival predictability of time
varying Indicators of bone disease In maintenance hemodialysis patients. Kid:
ney Int. 2006;70(4):771-780.
10. Shoji T, Shinohara K, Kimoto E. et al. l ower risk for cardiovascular mortality in
oral1alphahydroxy vitamin 03 users in a haemodialysls population. Nephrol
Dial Transplant. 2004;19(1):179184.
11. Martins D. Wolf M, Pan D, et al. Prevalence of cardiovascular risk factors and
the serum levels of 25hydroxyvitamin Din the United States: data from the Third
National Health and Nutrition Examination Survey. Arch Intern Med. 2007;167
(11):11591165.
12. Forman JP, Giovannucci E, Holmes MD, et al. Plasma 25hydroxyvitamin D lev
els and risk of Incident hypertension. Hypertension. 2007;49(5):10631069.
13. Zittermann A. Schteithoff SS, Tenderich G. Berthold HK, Karter R, Stehle P.
Low vitamin D status: a contributing factor in the pathogenesis of congestive
heart failure? JAm Col/ Cardiol. 2003;41(1): 105112.
14. Giovannucci E, Liu Y, Rimm EB, et al. Prospective study of predictors of vitamin
D status and cancer incidence and mortality in men. J Nat! Caricer lnst. 2006;
. 98(7):451:459. ____ .... .
15. Pittas AG, B, ll T, et al. Vitamin D and calcium intake in relation
to type 2 diabetes in women. Diabetes Care. 2006;29(3):650656.
16. Chonchol M, Scragg R. 25Hydroxyvitamin D. insulin resistance. and kidneyfunc
lion in the Third National Health and Nutrition Examination Survey. Kidney Int.
2007;71(2):134139.
17. Wolf M. Shah A, Gutierrez 0, et al. Vitamin D levels and early mortality among
incident hemodialysis patients. Kidney Int. 2007;72(8):10041013.
18. Autier P. Gandini S,.Vitamin D. Supplementation and total mortality: a meta
analysis of randomized controlled trials. Arch Intern Med. 2007;167(16):1730
1737. . . .
19. Jhomas MK, lloyd-Jones OM, Thadhani Rl, et al. Hypovitaminosis Din medical
inpatients. N Eng/ J Med. 1998;338(12):777783.
20. BischoffFerrari HA. Giovannucci E. Willett WC, Dietrich T, Dawson-Hughes B.
_. Estimation of optimal serum concentrations of 25hydroxyvitamin D for multiple
health outcomes [published corrections appear in Am J Clin Nutr. 2006;84(5):
1253 and Am J Clin Nutr. 2007;86(3):809]. Am J Clin Nutr. 2006;84(1):1828.
21. Zadshir A. Tareen N, Pan D. Norris K, Martins D. The prevalence of hypovita
minosis 0 among US adults: data from the NHANES Ill. Ethn Dis. 2005;15
(4)(supp15):S597 S51 01. .
22. Plan and operation of the Third National Health and Nutrition Examination Sur-
vey, 198894: series 1: programs and collection procedures. Vital Health Stat f.
July 1994;(32):1-407.
23. Enati TM. Massey JT. Waksberg J. ChuA, Maurer KR. Sample design: Third Na
tiona! Health and Nutrition Examination Survey. Vital Health Stat2. September
1992;(113):135.
24. Coresh J, Astor BC, McQuillan G, et at. Calibration and random variation of the
serum creatinine assay as critical elements of using equations to estimate glo
merular filtration rate. Am J Kidney Dis. 2002;39(5):92()-929.
25. Levey AS. Coresh J, Balk E. et at. National Kidney Foundation practice guidelines
for chronic kidney disease: evaluation. classi fication, and stratification. Ann In
tern Med. 2003;139(2):137147.
26. Zhu S, StOnge MP. Heshka S, Heymsfield SB. lifestyle behaviors associated
with tower risk of having the metabolic syndrome. Metal!olism. 2004;53(11):
15031511.
27. World Health Organization. International Classification of Diseases, Ninth Revi
sion, Clinical Modification {ICD9CM). Geneva, Switzerland: World Health Or
ganization; 1991.
28. World Health Organization. lnternationaiStatistica/CiassificationofDiseases. 10th
Revision {ICD10). Geneva. Switzerland: World Health Organization: 1992.
29. Dupont W, Plummer WO. PS power and sample size program available for free
on the Internet (letter] . Control Clin Trials. 1997;18:274.
30. looker AC. Dawson-Hughes B. Calvo MS. Gunter EW, Sahyoun NR. Serum 25
hydroxyvitamin D status of adolescents and adults in two seasonal subpopula
lions from NHANES Ill. Bone. 2002;30(5):771777.
31. Block GA. Klassen PS,LazarusJM, Ofsthun N,lowrle EG. ChertowGM. Mineral
metabolism, mortality, and morbidity In maintenance hemodialysis. JAm Soc
Nephrol. 2004;15(8):22082218.
32. Ohlngra R. Sullivan LM. Fox CS, et al. Relations of serum phosphorus and cal
cium levels to the incidence of cardiovascular disease in the community. Arch
Intern Med. 2007;167(9):879885.
33. Zinermann A, Schleithoff SS, Koerfer R. Putting cardiovascular disease and vi
tamin D insufficiency into perspective. Br J Nutr. 2005;94(4):483-492.
34. Urn HS. Roychoudhuri R. Peto J, Schwartz G. Baade P, Motter H. Cancer sur
vivalls dependent on season of diagnosis and sunlight exposure. lniJ Cancer.
2006;119(7):15301536. .. -.
35. Yusuf S, Sleight P, Pogue J, Bosch J, Davies R. Dagenais G; the Heart Out
comes Prevention Evaluation Study Investigators. Effects of an angiotensin
converting-enzyme inhibitor, ramlpril. on cardiovascular events in highrisk patients.
N Eng/ J Med. 2000;342(3):145153. .
36. Forman JP, Bischoff-Ferrari HA, Willett WC, Stampfer MJ, Curhan GC. Vitamin
D intake and risk of incident hypertension: results from three large prospective
cohort studies. Hypertension. 2005;46(4):676-682.
37. KrauseR, Buhring M, W, Holick MF, Sharma AM. Ultraviolet Band
blood pressure. Lince/. 1998;352(9129):709710.
38. Pfeifer M, Begerovi B, Minne HW, Nachtigall D, Hansen C. Effects of a short
term vitamin D(3) and calcium supplementation on blood pressure and para
thyroid hormone levels in elderly women. J Clin Endocrinol Meta b. 2001 ;86
(4):16331637. .
39. Worts man J. Matsuoka LV, Chen TC,lu z. Holick MF. Decreased bioavallability
of vitamin Din obesity. Am J Clin Nutr. 2000;72(3):690693.
40. Arunabh S, Pollack S, Yeh J, Aloia JF. Body fat content and 25-hydroxyvitamin
D levels in healthy women. J Clin Endoctinol Metab. 2ooj;88(1):157161.
41 . Hypp6nen E, Power C. Vitamin 0 status and glucose homeostasis in the 1958
British birth cohort: the rote of obesity. Diabetes care. 2006;29(10):22442246.
42. Ford ES, Ajani UA. McGuire LC. liu S. Concentrations of serum vitamin D and
the metabolic syndrome among U.S. adults. Diabetes care. 2005;28(5):1228
1230.
43. Hsia J, Heiss G. Ren H. et al. Calcium/vitamin D supplementation and cardia
. vascular events. Circulation. 2007;115(7):846-854. .
44. Miller ER Ill, PastorBarrius9 R, Dalal D, Riemersma RA, Appel LJ, Guallar E.
Meta-analysis: highdosage vitamin E supplementation may increase allcause
mortality. Ann Intern Med. 2005;142(1):3746.
45. Bjelakovic G, Nikolova 0, Gtuud Ll, Simonetti RG. Gluud C. Mortality in random
ized trials of antioxidant supplements for primary and secondary prevention: sys
tematic review and meta-analysis. JAMA. 2007;297(8):842857.
ARCH INTERN MEO/VOL 168 (1'>0. 15), AUG 11/25, 2008
1637

BRIEF REPORTS
I
Induction of withdrawal-like symptoms in a
small randomized, controlled trial of opioid
blockade in frequent tanners
Mandeep Kaur, MD," Anthony Liguori, PhD,b Wei Lang, PhD,c Stephen R. Rapp, PhD,c,d
Alan B. Fleischer )r, MD," and Steven R. Feldman, MD, PhDa,c,e
Winston-Salem, North Carolina
Frequent tanning has reinJorcing propert.ies. We tested whether opioid antagonism blocks potential
reinforcing effect of indoor tanning in 8 frequent tanners and 8 infrequent tanner control subjects. Opioid
blockade reduced ultraviolet preference in frequent tanners. Four of 8 frequent tanners, but no infrequent
tanners, exhibited withdrawal-like symptoms with naltrexone administration. A limitation of this study is
its small size. (] Am Acad Dermatol 2006;54: 709-11.)
W
hen given a blinded choice between
ultraviolet (UV)- and non-UV-emitting
tanning beds, frequent tanners over-
whelmingly prefer UV-emitting beds.
1
A question-
naire testing Diagnostic and Statistical lvlanua! of
Mental Disorders, Fourth Edition's addiction criteria
suggests that frequent tanning has features of an
addictive behavior
2
Inducti on of cutaneous endor-
phins by UV light occurs in vitro and coul d play a
role in driving t N exposure behavior.
3
If cutaneous
endorphins are involved in a reinJorcing effect of
UV exposure, then endorphin blockade would be
expected to block the effect.
1
An open trial using 50 mg of the opioicl. antagonist
naltrexone was performed to test whether endor-
phin blockade reduces the reinforcing effect of UV
in frequent tanners. The study was stopped when
2 of 3 frequent tanners developed symptoms con-
sistent with physiological withdrawal (eg, nausea,
From the Center for Dermatology Research, Departments of
Dermatology, Physiology/Pharmacology,b Public Health Sci-
ences,' Psychiatry and Behavioral Medicine,d and Pathology,
Wake Forest University School of Medicine.
Funding sources: The Center for Dermatology Research is sup-
ported by a grant from Galderma Laboratories, L.P. This study
was supported by a grant from the National Institutes of
Health (DA 14014-01) and by a pilot grant f rom the Wake Forest
University Women's Healt h Center of Excellence.
Conflicts of interest: None identified.
Reprint requests: St even R. Feldman, MD, PhD, Department of
Dermatology, Wake Forest University School of Medicine,
Medical Center Boulevard, Winston-Salem, NC 27157-1071.
E-mail: sfeldman@wfubmc.edu.
0190-9622/ $32.00
2006 by the American Academy of Dermatology, Inc.
doi:1 0.1 016/j.jaad.2005.1 1.1 059
jitteriness, shaking).
4
This finding is unexpected but
is consistent with the hypothesis that frequent tann-
ing is driven by an opioid-dependent mechanism.
The possibility that the withdrawal-like symptoms
were simply nonspecific nausea related to naltrexone
administration seems unlikely because naltrexone is
generally well tolerated in non-opioid-dependent
inclivicluals.
5
6
\Y/e performed a randomized, con-
t.rollecl trial of naltrexone exposure in frequent
and infrequent tanners to test whether opioid block-
ade abrogates frequent tanners' preference for UV
versus non-UV stimuli and to determine if opioid
blockade induces withdrawal symptoms in frequent
tanners.
METHODS
Eight frequent tanners and 8 infrequent tanners
between the ages of 18 and 34 years were recruited.
Frequent tanning was defined as tanning 8-15 times
a month, more than is required to maintain a tan.
Infrequent tanners were defined as having used
tanning beds previously, but no more than 12 times
in any given year. We included infrequent tanners
rather than non-tanners so as not to expose indoor
tanning-naive individuals to conditions that we
felt could potentially have addictive properties. The
study was approved by the Institutional Review
Board. Minimal erythema dose (MED) phototesting
was performed to ensure that the amount of UV light
exposure deli vered in this study did not cause
sunburn. We excluded individuals with Fitzpatrick
skin type I to reduce potential side effects.
We performed a double-blind, placebo-controlled
trial of UV and non-UV exposure in conjunction with
placebo-controlled administration of an escalating
709
710 Brief reports J AM ACAD D ERMATOL
APRIL 2006
Subjects I
Frequent tanners (n=8) Infrequent tanners (n=8)
5mg naltrexone or placebo 5mg naltrexone or placebo
l n=8 ! n=8
l n=6 l n=8
Fig 1. Flow diagram of patient allocation. Eight frequent tanners and 8 infrequent tanners were
given escalating doses of naltrexone. At each dose level, subjects were randomized to receive
either naltrexone or placebo; they were then given the other at the next visit. All 8 infrequent
tanners completed the trial without adverse events. Two frequent tanners discontinued the
study due to adverse events at the 15 mg naltrexone dose, and 2 others had adverse events
associated with administration of 15 mg of naltrexone but continued in the study.
dose of naltrexone (Fig 1). Naltrexone is a relatively
pure narcotic antagonist thal effectively blocks both
central and peripheral opioid receptors. This block-
age results in withdrawal symptoms in opioid-
dependent individuals, but is generally well tolerated
in non-opioid-dependent people. Only 5o/o or fewer
of normal, healthy individuals report adverse effects
when given naltrexone
5
6
Block randomization was
used.
Subjects agreed to not tan in UV beds or to tan
outdoors for the duration of the study (2 weeks). We
queried additional exposure to natural sunlight and
treated it as a covariate in our analyses. Participants
were given either placebo or naltrexone and, after a
1- hour rest period, were exposed in both UV and
non-UV tanning beds in random order. Frequent
tanners were exposed to the UV bed for a period up
to 10 minutes on each visit, and the infrequent
tanners were exposed to the UV bed for a period
of time that was based on the results of previous
MED phototesting. After completing each session,
participants completed a survey which indicated
their preference for a particular tanning bed. A 9-
point rating scale that compared the two beds
directly was used to assess the tanning bed prefer-
ence (1 = greatl y preferred first bed; 5 = no prefer-
ence; 9 =greatly preferred second bed). Participant
responses were analyzed using SAS version 8 (SAS
Institute, Cary, NC). Adverse events, when they
occurred, were recorded following each tanning
session. Because of the small size of this study,
significance testing was not conducted.
RESULTS
With placebo and with a 5-mg dose of naltrexone,
frequent tanners preferred the UV stimulus; how-
ever, frequem tanners showed a reduced preference
for UV exposure with naltrexone closes of 15 and
25 mg (Fig 2). Placebo-treated infrequent tanners
exhibited less preference for the UV stimulus as
compared with frequent tanners (Fig 2).
There were no adverse events reported by either
frequent or infrequent tanners with exposure to
placebo, and there were no reported adverse events
among the infrequent tanners at any administered
naltrexone dose. The frequent tanners did not expe-
rience any adverse side effects with a 5-mg close of
naltrexone. However, at a naltrexone dose of 15 mg,
4 of the 8 frequent tanners reported adverse events
(nausea and/or jitteriness), causing 2 of these 4 pa-
tients to remove themselves from the study. There
) AM ACAD DRMATOL
Brief rep011s 711
9=Preference
T r ~ fornon-UV
-'----------'----------------+----------''---1-a_nn_e_r ______ __, 1 =Preference
for UV
Fig 2. Preference for UV vs non-UV exposure in frequent and infrequent t.;mners given placebo
or increasing doses of naltrexone. Mean preference scores less than 5 indicate preference
for UV. With placebo and with 5 mg naltrexone, frequent tanners preferred the UV stimulus;
however, frequent tanners showed a reduced preference for UV exposure with nal trexone
doses of 15 and 25 mg.
were no adverse events reported in any of the re-
maining participants at a dose of 25 mg of naltrexone.
DISCUSSION
Some people will go to a tanning bed for just a few
visits w obtain a tan before an event, while others
may ran intermittently in order to maintain a ran.
There are other individuals-whom we define as
frequent tanners--who tan several times a week,
more than is required to maintain a tan. It has been
suggested that tanning is an obsession for frequent
tanners (BBC, "Young 'tanorexics' risking cancer";
broadcast May 24, 2005). These individuals may be
at greatest risk of adverse effects of UV.
The character of the adverse events observed in
frequent tanner participants is consistent ,,.vitb symp-
toms of opiate withdrawal induced by the opioid
blockade. We recognize that this was onl y a small
study and that a larger study confirming these find-
ings would be valuable. Nevertheless, the finding
that 4 of 8 frequent tanners developed withdrawal-
like symptoms with an opioid antagonist--while
such symptoms were not observed with placebo or
with infrequent tanners receiving naltrexone-lends
support to the hypothesis that the reinforcing effects
of UV exposure may be at least in part mediated by
opioids. Further study confirming a high prevalence
of withdrawal symptoms among frequent tanners
treated with opioid blockade would further support
that UV exposure may have an addictive quality .in
frequent tanners.
REFERENCES
1. Feldman SR, liguori A, Kucenic M, Rapp SR, Fleischer AB,
Lang W, et al. UV exposure is a reinforcing stimulus in frequent
indoor tanners. J Am A cad Dermatol 2004;5 1:45-51.
2. Scerra C. Addicted to the sun? Dermatology Times, September
1, 2003. Available at: http://www.dermatologytimes.com/
dermatologytimes/article/articleDetail.jsp?id=69605. Accessed
January 18, 2006.
3. Levins PC, Carr DB, Fisher JE, Momtaz K, Parrish JA. Plasma
beta-endorphin and beta-lipoprotein response to ultraviolet
radiation [letter). Lancet 1983;2:166.
4. Kaur M, Liguori A, Fleischer AB, Feldman S. Side effects of
naltrexone observed in 2 of 3 frequent tanners: Could fre-
quent tanners have chronically high opioid levels? J Am Acad
Dermatol 2005;52:916.
5. Perez-Reyes M, Wall ME. A comparative study of the oral,
intravenous, and subcutaneous administration of H-Naltrexone
to normal male volunteers. Naltrexone: Research Monograph
28. In: Willette RE, Barnett G, editors. Bethesda (MD): National
Institute on Drug Abuse; 1980. pp. 93-101. Available at: ht tp:/ /
www.drugabuse.gov/pdf/monographs/28.pdf. Accessed Janu-
ary 18, 2006.
6. Gonzalez JP, Brogden RN. Naltrexone: a review of its pharma-
codynamic and pharmacokinetic properties and therapeutic
efficacy in the management of opioid dependence. Drugs 1988;
35: 192-213.
Ultraviolet Radiation Related
Exposures
Introduction
Uluavioler radiation (UVR) is electromagnetic radiation found
between X-rays and li ght in the electromagnetic specrrum. It is
emitted by the su n and artificial devices, including sunbeds or
sunl amps. UVR can be divided into UVA, UVB, and UVC
components.
Solar radiation and exposure to sunlamps or sunbeds were first
listed in the Ninth Report on Carcinogens (2000) and broad-spectrum
UVR and its components ultraviolet A radiation (UVA), ultraviolet B
radiation (UVB), and ultraviolet C radiation (UVC) were first listing
in the Tenth Report on Carcinogens (2002). Much of the evidence for
listing the various UVR related-exposures applies to more than one
rype of UVR, rhus the profiles for these listings are discussed together.
Evidence for the carcinogeniciry of broad-spectrum UVR comes from
studies on solar radiation and exposure to sunlamps or sunbeds.
Similarly, studies ro evaluate the carcinogenicity of solar radiation in
animals and to determine the mechanism(s) by which it causes cancer
(mechanistic studies) involve exposure to broad-spectrum UVR or its
UVA, UVB, or UVC components. Use of sunlamps or sunbeds
el1[ails exposure to ultraviolet radiation. Evidence for the
carcinogeniciry of the UVR-related exposures is discussed separately
and follows this introduction. However, most of the information on
additional information relevant ro carcinogenicity, properties, use,
production, exposure, and regulations is common to all listings for
exposures related to UVR and therefore has been combined into one
sect ion following the carcinogeniciry discussions. The listings for
exposures related to UVR are as follows:
Solar radiation is known to be a human carcinogen
Exposure to sunlamps or sunbeds is known to be a human
carcinogen
Broad-spectrum UVR is known to be a human carcinogen
Ultraviolet A radiation is reasonably anticipated to be a human
carcinogen
Ultraviolet B radiation is reasonably anticipated to be a human
ca1cinogm
Ulrravioler C radiation is reasonably anticipated to be a human
carcinogen
Solar Radiation
Known ro be a human carcinogen
First Listed in rhe Ninth Report on Carcinogens (2000)
Carcinogenicity
Solar radiation is known to be a human carcinogen based on sufficient
evidence of carcinogeniciry from studies in humans, which indicate a
causal relationship benveen exposure to solar radiation and skin cancer
(both cutaneous malignant melanoma and non-melanocytic skin
cancer). Some studies suggest rhar solar radiation also may be
associated with melanoma of rhe eye and non-Hodgkin's lymphoma
(IARC 1992).
Exposure to Sunlamps or Sunbeds
First Lis red in the Ninth Report on Carcinogens (2000)
SUBSTANCE PROFILES
Carcinogenicity
Exposure to sunlamps or sunbeds is known to be a human carcinogen,
based on sufficient evidence of carcinogenicit y from studies in
humans, which indicate a causal relationship between exposure ro
sunlamps or sunbeds and human cancer. Sunlamps and sunbeds emit
primarily UVA and UVB radiat ion. Epidemiological studies have
shown chat exposure to sunlamps or sunbeds increases the risk of
malignant melanoma (Swerdlow eta!. 1988, Walter et al. 1990,
Autier eta!. 1994, Wescerdal1l et aL 1994, Chen eta!. 1998, Walter et
aL 1999, Wesrerdahl et aL 2000). The longer the exposure, the greater
the risk, especially in people exposed before the age of 30 or people
who have been sunburned. Malignant melanoma of the eye also is
associated with use of sunlamps (lARC 1992).
Broad-Spectrum UVR
First Listed in the Tenth Report on Carcinogem (2002)
Carcinogenicity
Broad-spectrum UVR is known to be a human carcinogen based on
sufficient evidence of carcinogenicity from studies in humans that
show a causal relationship between exposure and human cancer.
Epidemiology studies have shown that exposure ro broad-spectrum
UVR (solar radiation) causes skin cancer (both melanocyric and non-
melanocyric). Studies of humans exposed to solar radiation, artificial
devices emitt ing broad-spectrum UVR, or devices emitting
predominantly UVA or UVB all contribute to these conclusions.
Evidence that the broad-spectrum UVR component of solar radiation
is carcinogenic comes from (1) studies of human cancer associated
with exposure co devices rhat emit artificial broad-spectrum UVR, (2)
the face that tumors develop at the same sites both in humans exposed
ro sunlight and in animals exposed ro broad-spectrum UVR from
artificial sources, and (3) mechanistic studies in which human tissue
was exposed to artificial sources of broad-spectrum UVR.
Broad-spectrum UVR is absorbed by DNA and causes direct and
indirect DNA damage with rhe potential to result in mutations, as
demonstrated by mechanistic studies using human rissue. Mutations
found in the p53 rumor suppressor gene of human skin cancer are
specific for broad spectrum UVR-induced damage.
The findings in humans are supported by evidence in experimental
animals. Exposure ro broad-spectrum UVR caused ski n rumors
(papilloma and squamous cell carcinoma) and eye rumors (spindle-cell
sarcoma) in albino rats and skin tumors (fibrosarcoma and/or squamous-
cell carcinoma) in mice, hamsters, and opossums (lARC 1992).
UVA
Reasonably anticipated ro be a human carcinogen
First Listed in the Tenth Report on Carcinogem (2002)
Carcinogenicity
UVA is teasonably anticipated to be a human carcinogen based on
limited evidence from studies in humans and sufficient evidence from
studies in experimental animals. Epidemiological studies on rhe effects
of sunlight or artificial broad-spectrum UVR cannot identify effects
due specifically to UVA, UVB, or UVC exposure. However,
information about the speci fic effects of UV A, UVB, and UVC
expos ure can be inferred by comparing the results of human
epidemiology studies of broad-spectrum UVR exposure with the
results of studies on the effects of specific broad-spectrum UVR
componentS in experimental animals and human tissues.
REPORT ON CARCINOGENS, ELEVENTH EDITION
SUBSTANCE PROFILES
In studies where most of the UVR exposure was to UV A (i.e.,
exposure to solar radiation or UVA-emitting sunbeds), there was an
increased risk of skin cancer. Westerdahl et al. (2000) studied
exposure to sunbeds emitting mainly UVA (with 0. 1 o/o to 2.1 o/o UVB)
and found an increased risk of melanoma. The available data from
experimental animals show that exposure to UV A caused skin tumors
in mice (squamous-cell carcinoma and papilloma) and fish
(melanoma) (IARC 1992).
UVB
Reasonably anticipated to be a human carcinogen
First Listed in the Tenth Report on Carcinogens (2002)
Carcinogenicity
UVB is teasonably anticipated to be a human catcinogen based on
limited evidence from studies in humans and sufficient evidence from
studies in experimental animals. Mechanistic studies with human
tissue have demonstrated that the UVB component in solar radiation
is absorbed by DNA, resulting in DNA damage rhat leads ro the
characteristic p53 gene mutations observed in human skin cancer.
However, epidemiologic studies linking these exposures to skin cancer
are limited because they lack information on the specific wavelengths
of UVR ro which the individuals were exposed. Although increased
skin cancer is clearl y associated wit h exposure to UVB (as a
component of solar radiation or from sunlamps used before the early
1970s), t he people in these studies also were exposed t o other
components of broad-spectrum UVR. Therefore, the studies could
not distinguish between the effects of UVB and other components of
UVR. Sunlamps used in the early 1970s produced significant amounts
of UVB (22 to 40%); one smdy found that exposure w UVB-
emiHing sunlamps increased the risk of malignant melanoma of the
skin (Chen et al. 1998). In experimental animals, prolonged exposure
to devices emirting primarily UVB caused skin t umors in rats
(papilloma), mice (squamous-cell carcinoma, fi brosarcoma, papilloma,
and keratoacanthoma), guinea pigs (fibroma and rrichofolliculoma),
and opossums (melanocytic hyperplasia and melanoma) (lARC 1992) .
uvc
Reasonably anticipated to be a human carci nogen
First Listed in the Tenth Report on Carcinogens (2002)
Carcinogenicity
UVC is reasonably anticipated to be a human carcinogen based on limited
evidence from mechanistic studies with human t issue and sufficiem
evidence from studies in experi mental ani mals. H uman studies,
including those with cultured human cells, have shown that exposure to
UVC causes DNA damage. UVC is absorbed by DNA and causes
damage similar to that caused by UVB. However, no epidemiological
studies have adequately evaluated UVC carcinogenicity in humans.
UVC is absorbed by the ozone layer and does not contribute to solar
exposure. In studies of exposure to artificial devices emitting UVC, the
devices also emitted other components ofUVR. Exposure to high doses
of radiation from devices emitti ng primarily UVC caused skin rumors
in rats (keratoacanthoma-like tumors) and mice (squamous-cell
carcinoma and fi brosarcoma) (IARC 1992).
Ultraviolet Radiation Related
Exposures
Additional Information Relevant to Carcinogenicity
Broad-spectrum UVR causes skin cancer via DNA damage,
suppression of the immune system, tumor promotion, and mutations
in the p53 tumor suppressor gene. Broad-spectrum UVR causes
mutations in cultured human cel ls; the type of damage depends on the
specific wavelength of UVR and whether the affected cel ls can repair
the damage without er ror. DNA absorbs broad-spectrum UVR
(mainly UVB and UVC), and this reaction yields products that can
cause mutations (discussed below under "Properties"). UVB causes the
following four major DNA base modifications (changes to DNA's
structure) in people: cyclobutane-type pyrimidine dimers, (6-4)
phocoproducts, the corresponding Dewar isomers, and thymine
glycols. Both UVA and UVB induce 8-hydroxydeoxyguanosine
production from guanosine by the action of singlet oxygen (Griffiths
eta/ 1998).
UVA, UVB, and UVC as individual components of broad-
spectrum UVR cause genotoxic damage in several in vitro test systems,
including bacteria, yeast, rodent cells, and human cells. Moreover,
exposure to each of the three components of broad-spectrum UVR
causes DNA damage in humans. UVA's biological effects are indirect
and largely the result of energy transferred through reactive oxygen
intermediates (free radicals), whereas UVB and UVC are absorbed by
DNA and directly damage DNA through base modifications. Based
on t he number of studies showi ng genet ic damage, UVC is the
strongest genotoxin of the three components of broad-spectrum UVR,
and UV A is the weakest.
More than 90% of human squamous-cell carcinomas contain
mutations of the p53 tumor suppressor gene. These mutations were
found in 74% of sun-exposed normal human skin and only 5% of
unexposed skin, indicating a strong association with sun exposure.
Observed p53 gene mutations were most frequently C to T or CC to
TT transitions at pyrimidine- pyrimidine sequences. These specific
p53 mutations now are considered a signature of broad-spectrum
UVR carcinogenesis (Brash et at. 1991, Ziegler et al 1993, Griffiths et
at. 1998, Wikonkal and Brash 1999).
Exposure to solar radiation and broad-spectrum UVR alters
immune function in humans and experimental animals (lARC 1992).
Evidence that immunosuppression is related to ski n cancer comes
from the following observat ions: ( 1) immunosuppressed organ
transplant recipients showed a marked increase in skin cancer ,
particularly squamous-cell carcinoma, (2) broad-spectrum UVR
decreased the abili ty t o mount a delayed- type hypersensit ivity
response, and (3) mice exposed to low levels of broad-spectrum UVR
fai led to reject highly immunogenic tumor cell lines (Quinn 1997).
Exposure of human skin grafts on mice to UVB radiation after
pretreatment with the carcinogen dimerhylbenz(a)anthracene causes
human ski n tumors (squamous-cell carci noma, act inic keratoses,
melanocytic hyper plasia, and mel anoma) (At ill asoy et al 1997).
Exposure of human skin grafts on mice t o UVB alone causes
precancerous lesions (melanocytic hyperplasia).
Properties
Solar radiation includes most of the electromagnetic spectrum. Of the
bands withi n the optical radiation spectrum, UVR is the strongest and
most damaging to living things (IARC 1992) . Broad-spectrum UVR
includes wavelengths of light ranging from 100 to 400 nm. UVR is
divided into wavelength ranges identified as UVA (315 to 400 nm),
UVB (280 to 315 nm), and UVC (100 to 280 nm) . Of the solar UV
energy reaching the equator, 95% is UVA and 5% is UVB. No
measurable UVC from solar radiation reaches the earth's surface,
because t he shorrest UV wavelengths are completely absorbed by
ozone, molecular oxygen, and water vapor in the upper atmosphere
(Farmer and Naylor 1996).
Molecules that absorb UVR and visible light (photoreactive
molecules) contain segments that react with light (called
chromophores), in which photons of light excite electrons from the
ground state to higher-energy states. These molecules then generally
re-emir light on rerurning to lower-energy or ground states (Dyer
1965) . The various molecules sensitive ro UVR differ in rhe
wavelengths ofUVR that rhey absorb and the light that they emit.
Photochemical and photobiological interactions occur when photons
react with a phororeacrive molecule, forming eirher an altered molecule or
two separate molecules (Phillips 1983, Smirh 1989). For such a reaction
to occur, the photons must have enough energy to alter a photoreactive
chemical bond (i.e., to break rhe original bond or form new bonds).
The photobiological reactions related co skin cancer risk due to
UVR exposure are the reactions with the main chromophores of the
skin's outer layer- urocanic acid, DNA, r.ryprophan, tyrosine, and the
melanins. The products resulting from UVR's reaction with DNA
(DNA phoroproducrs) include pyrimidine di mers, pyrimidine-
pyrimidone (6-4) photoproducts, thymine glycols, and DNA exhibiting
cytosine and purine damage and other damage, such as DNA strand
breaks and cross-links and DNA-protein cross-links. The various DNA
photoproducts differ in their mutagenic potential (IARC 1992).
UVR-induced DNA phoroproducrs cause a variety of cellular
responses char cont ribute ro ski n cancer. U nrepaired DNA
photoproducts may result in the release of cytokines that contribute ro
tumor promotion, tumor progression, immunosuppression, and the
induction oflatent viruses (IARC 1992, Yarosh and Kripke 1996).
l..NB is considered to be the major cause of skin cancer, despite che
met that it does nor penetrate the skin as deeply as UV A or react with
the outer skin layer as vigorously as UVC. Its high reactivity with
macromolecules, coupled wirh the depth ro which it penetrates skin,
makes UVB the most potent portion of the UV spectrum for both
short-term and long-term biological effects. UVA, while possibly nor as
dangerous, also cau.ses biological damage {Farmer and Naylor 1996).
Use
Broad-spectrum UVR has many uses as a natural source of energy and
is important in various biological processes. Solar radiation is required
for l ife. Plants must have sunli ght ro grow and to produce
carbohydrates and oxygen. Broad-spectrum UVR from solar radiation
helps produce vitamin D in human skin cells. Vitami n D metabolites
promote the absorption of calci um by the intestinal tract; therefore, it
is essential for rhe growth and development of healthy bones. Brief
exposure to sunlight on a regular basis is sufficient to produce all of
the vitamin D most people need. This vitamin also can be obtained
from dietary sources. Artificial sources of broad-spectrum UVR have
many uses, including tanning, medical diagnosis and treatment, and
promotion of polymerization reactions (e.g., cur ing of protect ive
coatings). Sunbeds use anificially produced UVR to enable individuals
to develop a suntan for cosmetic reasons. Originally, sunbeds were
built with mercury arc lamps, which emitted large quantities of UVB
and UVC. Now, sunbeds and solaria emit mostly UVA (IARC 1992).
Broad-specrnun UVR has both diagnostic and therapeutic uses in
medicine and dentistry. More than 30 disorders now can be created
through UVA exposure combined with compounds called psoralens
(PUV A therapy). Psoriasis and eczema are rhe skin diseases most
frequently treated with PUVA therapy. PUVA can also be used with
UVB exposure to treat psoriasis patients who are not good candidates
for systemic therapy with methotrexate or etretinare (Morrison 1992) .
In addition, broad-spectrum UVR and, more commonly, UVB are
used with coal-tar creams to treat psoriasis (Reid 1996). l..NB also
may be used to convert 7-dehydrocholescerol (provitamin D) to
vitami n D in the skin of vitamin D-deficienc patients.
SUBSTANCE PROFILES
UVA may be a component of the phorocherapy ro treat neonatal
jaundice or hyperbilirubinemia. Typically an infant is irradiated with
visible light for several hours a day, for up to one week; however, the
lamps also may emir UVR, and one commercial neonatal
phototherapy unit was found to emit UVA and shorter wavelengths of
UVR (IARC 1992). UV A has been found to react with melatonin, a
hormone thar helps co regulate sleep-wake cycles. Although the
phocoproducts of melatoni n have nor been identified, melaronin has
been predicted to be moderately phototoxic (Kim et aL 1999).
Broad-spectrum UVR has many industrial applications. One of its
major industrial uses is in photopolymerization, including curing of
protective coatings and inks. Broad-spectrum UVR is used to simulate
weathering of various materials, such as polymers. UVR (usually UVC
at 260 to 265 nm) is used to sterilize and disinfect tools and materials.
O cher uses include UV photography, UV lasers, and in dental
examinations to derecr early dental caries, dental plaque, and calculus
(IARC 1992).
Sources
In the broadest sense, broad-spectrum UVR is formed when
something is heated or when elecrrons that have been raised to an
excited state return to a lower energy level. Broad-spectrum UVR is
naturally emitted by the sun. An estimated two-thirds of the energy
emitted by the sun penetrates the atmosphere. Broad-spectrum UVR
constitutes approximately 5% of rhe solar radiation that reaches the
earth's surface (IARC 1992).
Six artificial sources of broad-spectrum UVR have been identified:
incandescent lights, gas discharge lamps, arc lamps, fluorescent lamps,
metal halide lamps, and elecrrodeless lamps. Incandescent sources
provide visible radiation in a continuous spectrum. Gas discharge lamps
produce visible radiation when an electrical currem is passed through a
gas. The type of gas present in the lamp determines the emission
wavelengths; low gas pressures produce narrow bands, whereas higher
pressures produce broad bands. Arc lamps are intense sources of broad-
spectrum UVR and often are used to simulate solar radiati on.
Fluorescem lamps emit radiation from a low-pressure mercury discharge,
which produces a strong emission at 254 nm; this radiation excites the
phosphor-coated lamp to produce fluorescence. Various emission spectra
can be obtained by altering the makeup and thickness of the phosphor
and the glass envelope. In metal halide lamps, metal halide sal ts are
added ro a mercury-vapor discharge lamp, thus creating extra emission
lines. Elecrrodeless lamps use magnetrons to generate microwave energy,
which then is absorbed by the discharge tube (IARC 1992).
Low-pressure mercmy vapor lamps, sLmlamps, and black-light lamps
are considered to be low-intensity UVR sources. High-intensity UVR
sources include high-pressure mercury vapor lamps, high-pressure xenon
arc lamps, xenon-mercury arc lamps, plasma rorches, and welding arcs.
Sunlamps and sunbeds emit broad-spectrum UVR. Sunbeds now
chiefly emit UVA; however, before the mid 1970s, t hey more
commonly emitted UVB and UVC (IARC 1992). Three different
UVA phosphors have been used in sunlan1ps sold in rhe United Stares
since the late 1970s, producing emission spectra that peak at 340,
350, or 366 nm. Two modern sunlamps evaluated by the U.S. Food
and Drug Administration emi tted 99.0% and 95.7% UVA; the
remaining radiation was UVB. A new high-pressure UVA sunbed
wich eighteen 1600-warr filtered arc lamps emitted 99.9% UVA. An
older type of sunlamp, used prior ro the lace 1970s (UVB/FS type),
emitted 48.7% UVA (Miller eta!. 1998) .
Exposure
The greatest source of human exposure to broad-spectrum UVR is
solar radiation; however, the exposure varies with geographical
location. Informacion on global broad-spectrum UVR levels has been
compiled from data gathered for epidemiological studies of skin cancer
SUBSTANCE PROFILES
and other health effects, such as premature aging of the ski n, cataracts,
and suppression of the immune response. Despite the large number of
measurements, estimating human exposure is complex. The UVR
wavelengths to which an individual is exposed vary considerably with
latitude, alriwde, time of day, and season. People also vary in their
length of outdoor exposure, the pans of the body they expose, and the
shapes of rheir bodies. Nevertheless, many studies have estimated
exposure to broad-spectrum UVR. Few studies, however, were able to
distinguish between UV A, UVB, and UVC exposure (I ARC 1992).
Various factors influence terrescrial levels of UVA (i.e., levels
found at the earth's surface) . UVA levels decrease with increasing
distance from t he equator and increase with increasing altitude.
Terrestrial UVA levels also are decreased by stratospheric ozone,
which varies with latitude and season. When there is less ozone, more
UVA reaches the earth's surface. Time of day also influences UVA
levels. Clouds reduce the amount of UVA reaching ground level. Air
pollurion, including tropospheric ozone, can decrease UV A exposure,
especially in urban areas. Surface reflection also contributes to
personal exposures to UV A and can result in exposure to body parts
that othe1wise would be shaded from the sun (IARC 1992).
Terrestrial UVB levels are affected by the same factors as terrestrial
UV A levels; however, since UVB is absorbed more by stratospheric
ozone than is UV A, differences in latitude and altitude affect UVB
exposure more than UVA exposure. Seasonal changes affect UVB
levels, mostly in temperate regions. Generally, cloud cover scarrers less
than 10% of the UVB under a dear sky; however, very heavy cloud
cover virtually eliminates UVB, even in the summer. Surface reflection
also contributes to human UVB exposure (IARC 1992).
Commonly used fluorescent sunlamps deliver 0. 3 to 1.2 times the
annual UV A dose from the sun to a rypical tanner exposed for 20
sessions at 2 minimal erythemal doses (MED) per session (Miller eta!.
1998). (The MED is the lowest UVR exposure sufficient ro produce
well-defined reddening of the skin with in 24 hours of exposure.) A
frequent tanner (100 sessions at 4 MED/session) receives 1.2 to 4.7
times the annual solar UV A dose, while the newer high- pressure
sunlamps deliver 12 times the annual solar VV A dose to the frequent
tanner.
Approximately 25 million people in the United States use sunbeds
each year, and one to nvo million people visit tanning facilities as
often as 100 times per year (Sikes 1998, Swerdlow and Weinsrock
1998). Teenagers and young adults are prominent among users. A
1995 U.S. survey found that of commercial tanning salon patrons,
8% were 16 to 19 years old, 42% were 20 to 29 years old, and 71%
were female (Swerdlow and Weinstock 1998).
Anyone working outside (such as agricultural, construction, and
road work laborers) is exposed to solar radiation on the job. For a
group of more than 800 outdoor workers in the United Stares at 39
N latitude (the latitude of Philadelphia), personal annual exposure of
the face was estimated at 30 ro 200 MED (Rosenthal eta!. 1991).
However, this estimate may be low because Rosenthal and colleagues
assumed facial exposure to be only 5% ro I Oo/o of ambient exposure,
whereas other data suggested that it could be as high as 20%. Based
on rhis higher estimate, the annual facial exposure doses for these
outdoor workers would be 80 to 500 MED (IARC 1992).
Occupational exposure to artificial broad-spectrum UVR occurs in
indusrria1 photo processes, principally UV curing of polymer inks,
coatings, and circuit board phororesisrs; sterilization and disinfection;
quality assurance in the food industry; medical and dental practices;
and welding (IARC 1992). UV lasers, such as those used in cornea
shaping and coronary angioplasty, are another potential source of
occupational exposure, with relative risks that could be comparable to
risks for individuals in outdoor professions (Sterenborg eta!. 1991).
Electric arc welders are the largest occupational group with exposure
to arrificial broad-spectrum UVR. It is estimated that more than
500,000 welders in the Un ited States have been occupationally
exposed to broad-specuum UVR. Occupational exposure to artificial
broad-spectrum UVR depends on both the source of exposure and the
protective methods used to decrease exposure. Some artificial broad-
spectrum UVR sources (such as germicidal lamps in some uses) are
self-contained and present no risk to workers. Other occupational
uses, such as use of UVR in laborarories, UV photography, and UV
lasers, inevitably lead to broad-spectrum UVR exposure, wh.ich may
include intense short-term exposures (!ARC 1992).
Regulations
FDA
Performance standards for sunlamps and other devices that emit ultraviolet radiation
have been developed
User instructions and warning labels must accompany sunlamps and other devices
that emit ultraviolet radiation
Guidelines
ACGIH
Threshold limit values tTLVsl have been developed for over 60 diHerent wavelengths
(ranging from 180 to 400 nm) in the ultraviolet spectrum. In addition to these TLVs,
specific protections for the eye to exposures from UV radiation in the 315 to 400
nm spectral range also have been developed
1
NIOSH
Comprehensive recommendations for standards have been developed that include
various exposure limits, labeling and warning sign requirements, and numerous
work practice requirements
1
1
See Introduction for information on where to obtain additional detail on regulations
and recommendations.
REFERENCES
Alillasoy. E. S .. R. Elenilsas. E. R. Sauler, P. W. Soballe and M. Herl yn. 1997. UVB induction of epithelial
tumors in human skin using a RAG-I mouse xenograft model. J Invest Dermatol 109(61: 7049.
Autier. P .. J. F. Oore. f. Lejeune, K. f. Koelmel. 0. Geffeler. P. Hille. eta/. 1994. Cutaneous malignant
melanoma and exposure to sunlamps or sunbeds: an EORTC multicenter case-control study in
Belgium. France and Germany. EORTC Melanoma Cooperative Group.lnt J Cancer 58(6): 80913.
Brash, 0. E., J. A. Rudolph. J. A. Sirnon. A. Lin. G. J. McKenna. H. P. Baden. A. J. Halperin and J. Ponten.
1991. A role for sunlight in skin cancer: UVinduced p53 mutations in squamous cell carcrnoma. Pror.
Nail Acad Sci US A 88(221: 10124-8.
Chen. Y. T., R. Dubrow. T. Zheng, R. L. Barnhill. J. fine and M. Berwick. 1998. Sunlamp use and the risk of
cutaneous malignant melanoma. a population-based case-control study in Connecticut, USA. 101 J
Epidemiol 2715): 758-65.
Oyer, J. R. 1965. Introducti on. In Applications of Absorption Spectroscopy of Organic Compounds.
Englewood Cliffs, NJ: PrenticeHall. Inc. p. 321.
Farmer. K. C. and M. F. Naylor. 1996. Sun exposure. sunscreens. and skin cancer prevention: a year-round
concern. Ann Pharmacother 30(61: 66273.
Griffiths, H. R .. P. Misuy, K. E. Herlle11 and J. lunec. 1998. Molecular and cellular effects of ultraviolet
lightinduced genotoxicity. Crit Rev Clin l ab Sci 35(31: 189-237.
IARC. 1992. Solar and Ultraviolet Radiation. IARC MonOl]raphs on tile Evaluation of Carcinogenic Risk of
Chemicals to Humans, vof. 55. Lyon. France: International Agency for Research on Cancer. 316 pp.
Ki m. Y. 0 . H. J. Chung. S. T. Chung. J. H. Kim. J. H. Park. K. S. Kil and 0. H. Cho. 1999. Phototoxicity of
melatonin. Arch Pharm Res 22(2): 14350.
Miller, S. A., S. L. Hamilton, U. G. Wester and W. H. Cyr. 1998. An analysis of WA emissions from sun
lamps and the potential importance for melanoma. Photochem Photobiol 68(1): 6370.
Morrison, W. l. 1992. Phototherapy and pllotochemotherapy. Adv Oermatol 7: 255270.
Phillips, R. 1983. Sources and Ajlplications of Ultraviolet Radiati on. London: Academi c Press.
Quinn. A. G. 1997. Ultraviolet radiation and skin carcinogenesis. Br J Hosp Med 58l61: 2614.
Reid. C. 0. 1996. Chemical photosensi tivity: another reason to be careful in the sun. FDA Consumer
Magazine. May 1996. hnp://www.fda.gov/ fdac/features/496_sun.htmf.
Rosenthal. F. S .. S. K. West. B. Munoz. E. A. Emmett. P. T. Strickland and H. R. Taylor. 1991. Ocular and
facial skin exposure to ultraviolet radiation in sunlight: a personal exposure mO<Iel witll applic<llion
to a woflter populatron. Health Phys 61(1): 77-86.
Sikes. R. G. 1998. The history of suntanning: a love/hale affair. J Aesthetic Sci 1(2): 67.
Smith. K. C., ed. 1989. The Science of Photobiology. 2nd ed. New York. Plenum. p. 4753.
Sterenborg, H. J., F. R. de Gruijl. G. Kelfkens and J. C. van der Leun. 1991. Evaluation of skin cancer risk
resulting from long term occupational exposure to radiation from ultraviolet lasers in the range from
190 to 400 nm. Photochem Photobiol 54151: 77580.
Swerdlow, A. J., J. S. English, R. M. MacKie. C. J. O'Doherty. J. A. Hunter, J. Clark and D. J. Hole. 1988.
Fluorescent lights. ultraviolet lamps, and risk of cutaneous melanoma. Bmj 297(66491: 647-50.
Swerdlow. A. J. and M. A. Weinstock. 1998. Do tanning l amps cause melanoma? An epidemiologic
assessment. J Am Acad Dermatof 38(1 ): 8998.
Walter. S. 0 .. W. D. King and l. D. Marrett. 1999. Association of cutaneous malignant melanoma with
intermittent exposure to ultraviolet radiation: results of a case-tonuol study in Ontario, Canada. lnt J
Epidemioi2Bi31: 41 8-27.
Walter. S. 0 .. L. 0. Marrett. l. From. C. Hertzman. H. S. Shannon and P. Roy. 1990. The association of cuta-
neous malignant melanoma with the use of sunbeds and sunlamps. Am J Epidemiol 131121: 23243.
Westerdahl, J .. C. lngvar, A. Masback, N. Jonsson and H. Olsson. 2000. Risk of cutaneous malignant
melanoma in relation to use of sunbeds: further evidence for UVA carcinogenicity. Br J Cancer 82191:
1593-9.
Westerdahl. J., H. Olsson. A. Masback, C. lngvar, N. Jonsson. l. Brandt, P. E. Jonsson and T. Moiler. 1994. Use
of sunbeds or sunlamps and malignant melanoma in southern Sweden. Am J Epidemiol 140181: 691-9.
Wikonkal. N. M. and 0. E. Brash. 1999. Ultraviolet radiation induced signature mutations in photocarcino
genesis. J lnvestig Oermatol Symp Proc 4{1 ): 6-10.
Yarosh. D. B. and M. l. Kripke. 1996. DNA repair and cytokines in antimutagenesis and anticarcinogenesis.
Mutat Res 35011 ): 25560.
Ziegler, A., 0. J. leffell, S. Kunala. H. W. Sharma, M. Gailani, J. A Simon, et at. 1993. Mutation hotspots due
to sunlight in the p53gene of non melanoma skin cancers. Proc Natl Acad Sci USA 90(91: 4216-20.
SUBSTANCE PROFILES
FINAL
Report on Carcinogens
Background Document for
Broad-Spectrum
Ultraviolet (UV) Radiation
and
UV A, and UVB, and UV C
December 13-14,2000
Meeting of the
NTP Board of Scientific Counselors
Report on Carcinogens Subcommittee
Prepared for the:
U.S. Department of Health and Human Services
National Toxicology Program
Research Triangle Park, NC 27709
Prepared by:
Technology Planning and Management Corporation
Canterbury Hall, Suite 310
4815 Emperor Blvd
Durham, NC 27703
Contract Number N01-ES-85421
Dec. 2000
RoC Background Document for Ultraviolet Radiation
Criteria for Listing Agents, Substances or Mixtures in the Report on Carcinogens
National Toxicology Program
Known to be Human Carcinogens:
There is sufficient evidence of carcinogenicity from studies in humans, which
indicates a causal relationship between exposure to the agent, substance or
mixture and human cancer.
Reasonably Anticipated to be Human Carcinogens:
There is limited evidence of carcinogenicity from studies in humans which
indicates that causal interpretation is credible but that alternative explanations
such as chance, bias or confounding factors could not adequately be excluded; or
There is sufficient evidence of carcinogenicity from studies in experimental
animals which indicates there is an increased incidence of malignant and/or a
combination of malignant and benign tumors: (1) in multiple species, or at
multiple tissue sites, or (2) by multiple routes of exposure, or (3) to an unusual
degree with regard to incidence, site or type of tumor or age at onset; or
There is less than sufficient evidence of carcinogenicity in humans or laboratory
animals, however; the agent, substance or mixture belongs to a well defined,
structurally-related class of substances whose members are listed in a previous
Report on Carcinogens as either a known to be human carcinogen, or reasonably
anticipated to be human carcinogen or there is convincing relevant information
that the agent acts through mechanisms indicating it would likely cause cancer in
humans.
Conclusions regarding carcinogenicity in humans or experimental animals are based on
scientific judgment, with consideration given to all relevant information. Relevant
information includes, but is not limited to dose response, route of exposure, chemical
structure, metabolism, pharmacokinetics, sensitive sub populations, genetic effects, or
other data relating to mechanism of action or factors that may be unique to a given
substance. For example, there may be substances for which there is evidence of
carcinogenicity in laboratory animals but there are compelling data indicating that the
agent acts through mechanisms which do not operate in humans and would therefore not
reasonably be anticipated to cause cancer in humans.
Dec. 2000
ii
Dec. 2000
Summary Statement
Broad-Spectrum Ultraviolet (UV) Radiation and UVA, and UVB, and UVC
Carcinogenicity
Broad-spectrum ultraviolet radiation (UVR) is known to be a human carcinogen based on
sufficient evidence of carcinogenicity from studies in humans. Epidemiology studies
clearly demonstrate that exposure to broad spectrum UVR increases both melanocytic
and non-melanocytic skin cancer. Studies of humans exposed to solar radiation, artificial
devices emitting broad-spectrum UVR, and devices emitting predominantly ultraviolet A
radiation (UV A) or ultraviolet B radiation (UVB) all contribute to this conclusion.
Exposure to solar radiation is associated with an increased risk of malignant melanoma of
the skin, non-melanoma skin cancer, malignant melanoma of the eye, and cancer of the
lip (IARC 1992, NTP 2000). Evidence for the role of the UVR component of solar
radiation in carcinogenicity comes from studies of human cancers associated with
exposure to artificial UVR-emitting devices, tumor site-concordance between humans
exposed to sunlight and animals exposed to UVR from artificial sources and human
mechanistic studies using artificial sources ofUVR. Exposure to sunlamps or sunbeds
has been associated with malignant melanoma of the skin (Autier et al. 1994, Swerdlow
et al. 1988, Walter et al. 1990, 1999, Westerdahl et al. 1994, 2000, Chen et al. 1998).
Mechanistic studies using human tissue demonstrate that UVR is absorbed by DNA and
causes direct and indirect DNA damage with mutagenic potential. Mutations found in the
p53 tumor suppressor gene of human skin cancer are specific for UVR-induced damage
(see below).
The findings in humans are supported by evidence in experimental animals. Exposure to
broad spectrum UVR induced skin tumors (papilloma and squamous cell carcinoma) and
eye tumors (spindle cell sarcoma) in albino rats and skin tumors (fibrosarcoma and/or
squamous cell carcinoma) in mice, hamsters and opossum.
The epidemiological literature does not provide a basis for subdividing the effects of
sunlight or artificial UVR into components attributable specifically to UV A, UVB, or
ultraviolet C radiation (UVC). However, information regarding the specific effects of
UV A, UVB, and UVC can be inferred from the results of human epidemiology studies of
mixed UVR exposure together with the results of studies on the effects of specific UVR
components in experimental animals and human tissues.
UV A is reasonably anticipated to be a human carcinogen based on limited evidence
from studies in humans and evidence from studies in experimental animals. Studies in
which UV A has contributed substantially to human exposure (solar radiation and UV A
emitting sunbeds) have demonstrated an excess of skin cancer. Westerdahl et al. (2000)
reported an association of melanoma with exposure to sunbeds when the majority of the
exposure was considered to be from sunbeds emitting mainly UV A (source reported to
emit 0.1% to 2.1% UVB). The fmding in humans is supported by evidence in
experimental animals. UVA exposure induced skin tumors in mice (squamous cell
carcinoma and papilloma) and fish (melanoma).
111
Dec. 2000
UVB is reasonably anticipated to be a human carcinogen based on limited evidence
from studies in humans and evidence from studies in experimental animals. Mechanistic
studies in humans have demonstrated that the UVB component in solar radiation is
responsible for the mutagenic photoproducts that lead to the signature p53 mutations
observed in human skin cancer. However, epidemiologic studies are limited by lack of
information identifying exposure wavelength specificity. Although exposure to UVB, as
a component of solar radiation or from sunlamps used before the early 1970s, is clearly
associated with excess skin cancer, these human exposures are not solely to UVB but are
confounded by exposures to other components of the UVR spectrum. ln one study,
exposure to sunlamps used in the early 1970s, which produced significant amounts of
UVB (22% to 40%), was associated with cutaneous malignant melanoma (CMM) (Chen
et al. 1998). The finding in humans is supported by evidence in experimental animals.
Prolonged exposure to devices emitting primarily UVB caused the development of skin
tumors in rats (papilloma), mice (squamous cell carcinoma, fibrosarcoma, papilloma,
keratoacanthoma), guinea pigs (fibroma and trichofolliculoma), and opossums
(melanocytic hyperplasia and melanoma).
UVC is reasonably anticipated to be a human carcinogen based on limited evidence
from human mechanistic studies and evidence from studies in experimental animals.
Studies of human tissue have demonstrated that both in vivo and in vitro exposure to
UVC causes DNA damage. UVC is absorbed by DNA and induces mutagenic
photoproducts similar to the types of damage caused by UVB. However, there are no
epidemiologic studies adequate for evaluation ofUVC carcinogenicity in humans. UVC
is absorbed by the ozone layer and does not contribute to solar exposure, and studies
using artificial devices emitting UVC are not specific for UVC radiation. Exposure of
experimental animals to high doses of radiation from devices emitting primarily UVC
caused skin tumors in rats (keratoacanthoma-like skin tumors) and mice (squamous cell
carcinoma and fibrosarcoma).
Other Information Relating to Carcinogenesis or Possible Mechanisms of
Carcinogenesis
Broad-spectrum UVR causes skin cancers via mechanisms that include DNA damage,
immunosuppression, tumor promotion, and mutations in the p53 tumor suppressor gene.
Broad-spectrum UVR induces mutations in cultured human cells, the type of damage
depends upon the specific wavelength applied and the competence of an affected cell to
repair the damage without error. DNA is a major cellular chromophore absorbing UVR
(mainly UVB and UVC) and responds to irradiation by yielding free radical reactive
intermediates and various photoproducts with mutagenic potential. UVB photons cause
the following four major DNA base modifications in humans: (i) cyclobutane-type
pyrimidine dimers, (ii) (6-4) photoproducts, (iii) the corresponding Dewar isomers, and
(iv) thymine glycols. Both UV A and UVB induced 8- hydroxydeoxyguanosine produced
from guanosine by the action of singlet oxygen.
UV A, UVB, and UVC as individual components of UVR are genotoxic in prokaryotes,
lower eukaryotes, non-human mammalian cells, and human cells. Moreover, in vivo
exposure from all three components ofUVR results in DNA damage in humans. UV A's
iv
Dec. 2000
biological effects are indirect and largely the result of energy transferred through active
oxygen intermediates, whereas UVB and UVC photons are absorbed by DNA and direct
damage occurs through DNA base modifications. Based on the number of positive
genotoxic studies, UVC is the most potent and UV A is the least potent genotoxin of the
components of broad spectrum UVR
More than 90% of human squamous-cell carcinomas contain mutations of the p53 tumor
suppressor gene. These mutations were found in 74% of sun-exposed normal human skin,
compared with 5% ofunexposed skin, indicating a strong association with sun exposure.
Observed p53 gene mutations were most frequently C to T or CC to TT transitions at
pyrimidine-pyrimidine sequences. These specific 53 mutations are now considered a
signature ofUVR carcinogenesis.
Exposure to solar radiation and UVR has been found to alter immune function in humans
and experimental animals. Evidence that immunosuppression is related to skin cancer
incidence comes from the following observations that: (i) immunosuppressed organ
transplant recipients showed a marked increase in skin cancer, particularly squamous-cell
carcinoma, (ii) UVR decreased the ability to mount a delayed type hypersensitivity
response, and (iii) mice exposed to low levels of UVR failed to reject highly
immunogenic tumor cell lines.
Human skin grafts on mice also yielded human skin tumors (squamous cell carcinomas,
actinic keratoses, melanocytic hyperplasia and melanoma) following irradiation with
UVB after pretreatment with the carcinogen dimethylbenz(a)anthracene. Precancerous
lesions (melanocytic hyperplasia) were found in human skin grafts on mice treated with
UVB alone.
v
Dec. 2000
vi
Dec. 2000
Table of' Contents
Criteria for Listing Agents, Substances of Mixtures in the Report on Carcinogens ... ... ...... ....... ... .. i
Summary Statement ... .... .... ..... ..... .......... .... ..... .... ... ..... .... ... ... ..... .... .... ..... ..... .......... .... ..... .... ... ..... .. . iii
1 Introduction ...... .... .... .... ..... .. ..... .... ...... .. .... .... .. ...... .... ... ..... .... .... .... ..... .. ..... .... ..... ... .... ..... ........ .... 1
1.1 Identification of UVR by type ... .. ........ .. .... .... .... .... ...... .. .... .. .... ... .. ....... .. ... .. .. .............. 2
1.2 Physical properties .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . 2
1.3 Photochemical and photobiological activities .. .. .. .... ..... .. ...... .... .... .. ...... .... .... .. ...... .. ... 3
2 Human Exposure ... ........ .. .... .......... ...... .... .. ............ ... ..... ........ ........ .. .... .......... ...... .... .. ............ ... . 5
2.1 Use ..... ... ...... ....... .... .... .... .... .... ... ... ... ...... ..... .... .... ... ...... ....... .... .... ... ..... .... ... ... ... ...... .. ... . 5
2.1.1 Cosmetic use .... .. ... .. .. .. .. ...... .. .. .. ....... ... .. ...... .. .. .. .. .. ...... .. ...... .. .. .. .. .. ...... .. .. .. .. 5
2.1.2 Medical and dental applications .... .... .. ...... .... .... ...... .. ...... .. ...... .. .. .. .. .. .. .... .. .. 5
2.1.3 Industrial applications .. .. .. .. .... .. .. .. .. .... .... .... .. .. .. ...... .. .. .. ...... .. .. .. .. .. .. .. .. .... .. .. . 6
2.2 Production .. ... .... ..... .. .... .. .. .. .... .. .. ... ....... .... .. .. .. .. .. .. .. .. .. ... .... ..... .. .... .. .. .. .... .. .. ... ....... .... .. 6
2.3 Analysis ... ..... ..... ... .... ...... .. .... ..... ..... .. ...... ... .... ... ....... ..... ..... ... .... ...... .. .... ..... ..... .. ...... .. .. 6
2.3 .1 Spectroradiotnetry ...... .. .... ...... .... .... .... ... .... .... .... .... .... ...... .... .... ... ..... .... .... .. .. 6
2.3.2 Wavelength-independent (thermal) detectors ....... .... .............. .. ...... .. ..... .... . 6
2.3.3 Wavelength-dependent detectors .... ....... ... .... .......... .. ...... ... ...... .. ..... ... ...... .. . 7
2.4 Environn1ental occurrence ... .... .... .. .. ... ... .. .. ... ..... .. ...... ... .... .... .... .... .. .. .. .. ...... .. ....... .. ..... 7
2.5 Environmental exposure .... ... .. ... ... .. .... .... .. .... .. .. .... .. ....... ... .. .. .... .... .. .... .. .. .. ..... .. ... ... .. ... 7
2.5.1 Solar u-vR ... .. .. .. ..... .... .... .... ..... .. .. .. .. .. .. .. .. .. .. .. .. .. ...... .. .. .. .. .. .......... ... .. .. .... .. .. . 7
2.5.2 Artificial sources .. .. .. .. .. ...... .. ... .. ... .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. ...... .. .. .. .. .. ...... .. ..... 10
2.6 Occupational exposure .. .. ...... ...... .. .. .. .. ...... .. .. .. .. .... ... .. .. .. .. ......... ... .. .. .. .. .. .. ........... .... . 11
2.6.1 Solar UVR ..... .. .. .. .. .. ... .. ...... .... ... .. .. .. .. .. .. .. .. .. .. .. .. ....... .. .. .. .. ... .. .. .. .. ... .. .. .... .. . 11
2.6.2 Artificial UVR .. ...... ... .. ... .. ... .... ....... .. ...... ... ...... .. ............. .. ...... ... .. ... .. ... .... .. 11
2.7 Biological indices of exposure .. .... .. ........ .. .... .... ...... ....... ....... .... .... .. ...... .................. . 11
2.8 Regulations .... .... .... ... ...... .... .... ... ..... ........ ... .... .... ..... .... ... .... .... ... ...... .... .... ... ..... ........ .. 11
3 Human Cancer Studies .. .. .. .. .. .. .... .. .. .. .. .. .. .. .. .. .. .. .. .. .. ... ... .. .. .. .. .. .. .. .. .. .. .. .. .. .... .. .. .. .. .. .. .. .. .. .. .. .. .. .. 17
3.1 Solar radiation .. .... .... ...... .... .... .... .. ...... .... .. ...... .. ...... .... .... .... .... .. ...... .... .... .... ... .... .... ... 18
3.1.1 Evaluations by the IARC (1992) and the NTP (2000) ........... .. ....... .. ...... .. 18
3 .1.2 Recent epidemiologic studies .... ...... .. ..... ... .... ......... .... ..... .. ..... ..... ... .... ..... .. 19
3.2 UVR frotn artificial sources ..... .. .... ... .. ... .. ..... .... .. ....... .. ..... ... ............... .. ..... .. ...... .. .... 20
3 .2.1 Cosmetically related UVR exposure ...... .. ...... .. ..... .................. .... .. ........... . 20
3.2.2 Medically related UVR exposure .. .. .. ... .... .. .. .. .. ...... .. .......... .. ... .. .. .. .. .. ...... .. 25
3.2.3 Occupationally related UVR exposure .. ..... ... .. .. .. .. .. .. ... .. .. .. .. .. ... .. ... .. ...... .. 27
3.3 DNArepair .... ... ...... .. .. ... .. .. .. .. .. ... ... .. .. .. .. ... .. .. .. .. .. .. .. .. .. .. ....... .. ....... .. ... ... .. .. .. .. .. .. ... .. .. . 28
3.4 Discussion .. .. .. .. .. .. .. .. .. .. .. .. .. .... .. .. .. .. .. .. ... ... .. .. .. .. .. ... ... .. .. .. .. .. .. .. .. .. .. .. .. .. .... .. .. .. .. .. .. ... ... 28
3.4.1 UVA .. .. .. .. .. .. .. ..... .... ... .. .. .. ... .... ... ... .... .. ...... .... ... ... ...... .. ...... .. .. .. .... .. .. ... ... ... .. 28
3.4.2 UVB .. ... ...... .... .. .. .. .. ... .. .. .. .. .. .. .. .. .. ... .. ... .. .. .. ...... .. .. ... ..... ... .. ... .. .. ... .. .. .. .. .. .. .. . 29
3.4.3 uvc .. ............................................. ............................................ ............... 30
vii
Dec. 2000 RoC Background Document for Ultraviolet Radiation
3.5 Sununary .. ....... ........ ........ .. ...... .. ..... ... .... ... ..... ........ .......... ........ ........ .. ...... .. ..... ... .... ... 30
4 Studies of Cancer in Experimental Animals ... ... ........ ..... ..... ... .... .... .... ... ....... ... ... ..... ... ... .... ... . .47
4.1 Broad-spectrum UVR ..... ....... .. ....... ...... ... ..... .... ...... ....... .. ..... ........ .... ...... ... ..... .. ....... . 47
4 . l .1 Rats ... .. ... ... .. .. .. .. .... .. .. .... .. .. ..... ... .... .. .. .... .. .. .... .. .. .. ... ... .. .. .. .. .... .. .. .... .. .. ..... ... 4 7
4.1 .2 Mice .. .... .. .. .. .. .. .. .... .... .... ........ .... .. .. .... .... ........ .... .... .. .. .. .. .. .. .. .. .... .... ........ .... 47
4.1.3 Hamsters ...... .... .. .... .... .. ........... ... .. .. .... .. .. .. .. ...... .. .... .... .. .. .. ..... ........ .... .... ..... 48
4. 1.4 Guinea pigs ......... ... .... ... .. .. .. .. .. .. .... .... .. .. ........ .. .. .... .. .. .... .. .. .. .. .. .. .... ... ... ... ... 48
4. 1.5 Other species ...... .. .... .... .... .. .. .. .. ... ... .. ... ... .. ... ...... .. .. .. .. .. .. .... .. .. .. .. .. .. .. ... .. ..... 48
4. 1.6 Action spectra .. ... ... ..... .... ... ... .. ... .. .. ... .. .. ........ .... .... .. .. .... .... .. .. .. .. .... .. ...... ... . 49
4.2 Primarily UV A ..... ..... ... ...... .. ..... .... ... ..... .... .... ....... .. .... ... ... .... ..... ... ...... .. ..... .... ... ..... ... 49
4.2. 1 Mice .... ... ... ..... .... ... ..... .... .. ...... .... .... ... ...... ... .... ..... ... ... ..... .... ... ..... .... .. ....... .. 49
4.2.2 Other species .... .... ..... ... ...... ... ...... ....... .... ... .... ..... ... ...... .. ...... .. ...... .... ...... .... 52
4.3 Primarily UVB ... .. ...... .. ...... .. ..... .. ...... .. ...... .... .... .... ...... .. ..... .. ...... .. ...... .. ..... .. ...... .. ..... 53
4.3. 1 Rats ... ....... ....... .. ...... .... ........... ... ...... ... ..... ........ .. ....... ....... .. ...... .... ........... ... 53
4.3.2 Mice .... .. ..... .... ..... ....... .... .. ...... ..... .. ..... .... .... .... .. ... .. ..... .... ..... ....... .... .. ...... ... 53
4.3.3 .. .... .... .. .. .. .... .. .. .. ...... .. .. .... .... .... .. .. .. .. .. .. .... ... .... .... .... .... .... ... ..... .... . 54
4.3.4 Guinea pigs ..... .. ..... .. .... ..... ... ... .. ... .... .. ... .. .. .. .... ... ... ...... .. .. .. ........ ...... .. ...... .. 54
4.3.5 Other species .... .. .. .... .... .. .. .. ... ... .. .. ... .... ... ........ ... .... .... .... .... .. .. .. .. .... .. ..... ... .. 55
4.4 Primarily UVC ... .. .... .... ... ..... ........ .... .. .. ........ .... .... .... .. .. ...... .. .... .... ... ..... ... ..... .... .. .. .... 56
4.4.1 Rats ... .... .. .. .. ... ... .. .. .. .. .... .. .. .... .. .. .. .. .. .. .. .. .. .. .. .. .. .. .... .. .. .. ... ... .. .. .. .. .... .. .. .... .. .. 56
4.4.2 Mice ... ... ..... .... ..... .. ..... ..... ... .... .... .... .... .... .... .... .... ... ..... .... ..... .. ..... ..... ... .... ... 56
4.5 Cancer development in human-mouse chimera models ... ..... .. ..... ........ ..... .... .... .... ... 57
4.6 Sum1nary .... .. ....... .... ... .. ....... ....... .. ..... .... ...... ...... .... ..... .. ....... .... ... .. ....... ....... .... ... .... ... 57
5 Genotoxicity .. .. .. .. ........ .... .. .. .. .. .. .. .... .. .. .. .. .... .. ...... .. ... ... .... .. .. ........ .... .. .. .. .. .. .. .... .. .. .. .. .... .. ....... ... 59
5.1 Methods for identifying and quantifying UVR-induced DNA lesions .... .... ..... .... ... 59
5.2 UVR-induced DNA photoproducts ... .... .... ... ..... .... .. .. .. .. ..... .. ...... .. ..... .... ..... ... ..... .... .. 59
5.2. 1 UYA-induced indirect DNA damage ..... .. ..... ... ...... ... ..... ........ .. ....... ........ .. 60
5.2.2 UVB-induced direct DNA damage .... ... .. ... ... ... ... ...... ..... ..... ....... ..... .. ..... ... 60
5.2.3 Cellular mechanisms for minimizing UVR-induced DNA damage ........ . 60
5.2.4 Cellular responses to UVR-induced DNA damage .. ...... ... ...... .. ..... ... .. .. .. .. 61
5.3 Prokaryotic systems .. .... .. .. .. .. .. .. .. .. .. .. .. .. .. .. .... .. .. .... .. .. .. .. .... .. .. .. .. .... .. .. .. .. .. .. .. .. .. .. .. .. .. .. 61
5.3 . 1 Induction of mutation in Salmonella typhimurium .... .... .... ... .... .... ..... ... .... 61
5.3.2 Induction of mutation in Saccharomyces cerevisiae .. .. .. .. .. .. .. ..... .... .... .... .. 64
5.4 Plants and lower eukaryotic systems ...... .. .. .. .... .. .. ...... .. .... .... .... .... .. .. .. .. .. .. .. .. .. ... .. ... .. 64
5.5 Mam1nalian systems ...... .... ..... ... ... ..... .... ... ..... .... .... .... .... .. ..... .... ..... .... ..... ... ... ..... .... ... 64
5.5. 1 Nonhuman mammalian in vitro assays ..... .... ...... .. .... .... ..... .. ....... .... .. ...... .. 64
5.5.2 Human in vitro assays ..... .. ..... .... ..... ... ...... .. ... ..... ... .... ...... ... .... ... ...... .. ..... ... 65
5.5.3 Nonhuman mammalian in vivo assays .... .. .... ... ...... ... .... ...... .... ...... ... ...... ... 66
5.5.4 Human in vivo assays .... .... .... ............ .... ..... ... ............ .... ..... .. ..... .... .... .... ... 66
5.5.5 Other in vitro and in vivo end points ..... ..... .... ......... ........ ....... .. ....... .. ...... . 68
viii
Dec. 2000 RoC Background Document for Ultraviolet Radiation
5.5.6 Molecular epidemiological studies of DNA repair capacity ... .................. 68
5.6 Summary .... ..... ... ..... .. ... ... .... ... .. .. .... ..... ... .... .. ... .... ... .... ..... ... ..... .. ... ... .... .... ... .... ..... ... .. 69
5.6.1 UVA .. .. .. .. .. .. .. .. .. .. .. .. .. .... .. .... .. .. .. ... .... .... .... .... .... .... .... .... .. .. .. .. .. .. .. .. .. .. .... .. .. . 69
5.6.2 UVB .. .. ... ... .. .. .... .... .. ... ... .. ...... .. .. ... .... ... .... ........ .... ...... ... .. .... .. .. .. .. .. .. .. ... ... ... 69
5.6.3 uvc .. .... .. .............. .. .. .. .. .. .. .. .. .. .. .......... .. .............. .. ...... ... .. ...... .. .. .. .. .. .. .. .. ... 70
6 Other Relevant Data .............................................................................................................. .. 73
6.1 Absorption and transmission of UVR in biological tissues .. ... .. .. .. .. .. .. .. .. .. .. .. .... ... ... 73
6.2 Mechanisms of UV-induced skin cancer .. .. .... .. .. .. ...... .. ......... ........ .... .. .... .. .. .. ...... .. .. 74
6.2.1 DNA damage .. .. .. .. .. .. .. .. ... ... ... .. ... .. ... .... .... ... .. ... ... ... .. .. .... .. .. .. .. .. .. .... .. ...... .. .. 75
6.2.2 DNA repair ........ .. ... .... ..... ................ .. ....................... ................................. 77
6.2.3 Mutations ..... ............ .. ............. .... ......... ........................ ............ .. .............. . 78
6.2.4 Tumor suppressor gene expression and mutation .. ....... ..... ...... ............. .. .. 78
6.2.5 Immunosuppression ........ ................ .. ..... ........ ..... ... ... ........ .. ............ ........ .. 79
6.3 h1itiation and promotion ......................... ......... ......................................................... 80
6.4 Summary .. ....... ... ...... .. ...... .. .... ... ............. .... ... .. ................ ... ...... .. ...... .. .... ... ............. .. 80
7 References .......... ... ........ ........ ...... ................................... ................ ................ .. ...... ................. 83
Appendix A: IARC Monograph of Evaluation of Carcinogenic Risks to Humans. Solar
and Ultraviolet Radiation. Vol. 55. Lyon, France. World Health Organization. 1992.
pp. A-1 - A-279 .. ........ .. .. .. .. .. .. .... .. ..... .. .. ... ..... ..... ... .... .... .... .. ...... .. .. .. .... .... .... .... .... .. .. .. .. .. ...... .. .. 99
Appendix B: Profile for Solar Radiation and Exposure to Sunlamps and Sunbeds. Report
on Carcinogens, Ninth Edition (2000) ... ... .... .. .. .... ... .... .... .... .. ... ... ..... .... ... ..... .... ..... ... .... .. .. .... 101
List of Tables
Table 1-1 Optical radiation spectrum ......... ...... ................................... ...... .. ................. ................ ... 1
Table 2-1. Percentage of daily UV A radiation received during two periods on a clear
su1nn1er day .... .... .... ..... .. ...... .. ..... ... ..... .... ..... .. ...... ... ..... ... .... .... ..... .. ...... .. ..... ... ..... .... ..... .. ...... ... ..... .. .. 8
Table 2-2. Typical values for ambient daily and annual UVB radiation expressed as
minimal erythema dose ..... .... .. .. .. .. .. .. .. .. .... .. .. .... .. .. .. .. .... .. .. .. .. .... ........ .... .. .. .. .. .. .. .. .. .... .. .. .... .. .. .. ... ... .. 8
Table 2-3. Percentage of daily UVB radiation received during two periods on a clear
sumn1er day ... .... .... .... .. .......... .... ........ ........ .. .. ...... ...... .... .... .... .... .. .......... .... ........ ........ .. .. ...... .... .. .. .. .. 9
Table 2-4. Representative terrain reflectance factors for horizontal surfaces measured
with a UVB radiometer at 12:00 PM at various U.S. locations ......... ......... ...... .... .. ...... .. .. .. ...... .. .. 10
Table 2-5. FDA regulations .. .. ... .. ... .. ... ... .. ... ... ... .. .... .. .. ... ... .... .... .. ...... .. .. .. .. .. .. ..... ... .. .. .. .. .. .. .. .... ... ... 12
Table 2-6. OSHA Regulations .. ..... ...... .. ... .. .. .... .. .. .. .. .. ... .. .. ......... .. .. .. .. .... ... .. .. .. .. .... .... .... ........ .. .. ... 15
Table 3-1 . Epidemiologic studies of the relationship between cutaneous malignant
melanoma and exposure to sunlamps or sunbeds (listed in chronological order by
publication date) .. ........ ........ .......................................... ...... ........ ........ .......................................... 31
Table 3-2. Recent epidemiologic studies of the relationship between cancer and medically
related UV exposure ... ... ..... .. ...... ... ..... .. ...... .. ...... .... ..... .. ...... ... ..... .. ...... .. ..... ..... .... ... ............. ... ..... .. 42
ix
Dec. 2000
Table 3-3. Recent epidemiologic studies of the relationship between cancer and
occupational UV exposure .... ... .. .. .... ..... .... ..... .. .... ... .... ... .. ... ...... .... ..... .... ... .... ...... .... .. ... ...... ... .. .... .. 44
Table 4-1 . Tumor incidences in female C3H/ Tif mice exposed to UV A tanning sources
with differing UVB emission levels .... .. .... .... .... .... ...... ... .... ... ...... .... .... .... .. .... ...... .... .... .... .... ... ....... 51
Table 4-2. Tumor incidences in female C3H/Tif mice exposed to broad-spectrum UVR
and/or UV A ... .. ... .. .. .... .. ... ... ...... .. ... .. .. .. .. .. .. .. .. .. .. .. .. ..... ... .. .. .. .. .. .. .. .. .. .. .. .. ...... .. .. .. ...... .. .. .. .. .. .. .. .. .. .. .. 52
Table 4-3. Incidences of melanoma in hybrid fish (Xiphophorus) exposed to various
wavelengths ofUVR .. .. .. .. .... .... .... ... ... .. .... ... ... .. .. ... .. ... ... .... .... .... ..... .... ... .. ... .... ... .... .... ..... .. .... ... .. .... 53
Table 4-4. Dose-response to (mainly) UVB in SHHl albino hairless mice ........... .. .. .. .. .. .. ... ...... . 54
Table 5-l . Genetic and related effects ofUVR exposure reviewed in IARC (1992) .. .. .. .... .. .. .. .. . 61
Table 5-2. Genetic and related effects ofUVA, UVB, and UVC exposure reviewed in
IARC ( 1992) .. ....... ... .. .. .. .. ..... .. .. ... .. ... .... ....... .. ... ... .... .. .. .. ....... ... .. .. .. .. ..... .. .. ... .. ... .... ....... .. .. .. .. ... ... .. .. 71
Table 6-1 Characteristics ofUVR .... ...... .. ...... .. .. .. .. .. ...... .. .. .... .. ....... .. .. .. ..... .... .. .. ....... .. .... .. .......... .. 74
List of Figures
Figure 1-1. Electromagnetic spectrum ...... ... .... ... ..... .... .... ...... .... .. ...... .. .. ... ... .... ....... .... ... .... ... ...... .... 3
Figure 4-1. Dose-effect relationship for the induction of < 1-mm skin tumors in hairless
mice by exposure to UVB over a wide range of daily doses; tm = median induction time .......... . 55
X
Dec. 2000
1 Introduction
Ultraviolet radiation (UVR) was nominated for listing in the Report on Carcinogens by
Dr. Hiroshi Yamasaki, of the International Agency for Research on Cancer (IARC), on
the basis of the !ARC's classification ofUVR as carcinogenic to humans (Group 1)
(IARC 1992). In 1997, the National Toxicology Program (NTP) reviewed the effects of
solar radiation, which includes most of the electromagnetic spectrum, and exposure to
sunlamps and sunbeds, which provide exposure to radiation primarily in the ultraviolet A
(UV A) and ultraviolet B (UVB) portions of the spectrum (NTP 1997). The NTP
recommended that solar radiation and exposure to sunlamps and sunbeds be listed in the
Ninth Report on Carcinogens (RoC), where they are listed as known to be human
carcinogens, based on studies in humans that (1) clearly indicate a causal relationship
between exposure to solar radiation and cutaneous malignant melanoma and
nonmelanocytic skin cancer and (2) have shown that exposure to sunlamps or sunbeds is
associated with cutaneous malignant melanoma (NTP 2000). Malignant melanoma of the
eye also is associated with use of sunlamps. In contrast, there is little support for
association of exposure to sunlamps or sunbeds with nomnelanocytic skin cancer (IARC
1992). The 1997 NTP review recommended that broad-spectrum UVR, including UVA,
UVB, and ultraviolet C (UVC), be reviewed for possible separate listings in the Tenth
RoC.
The sun is the major source of UVR. UVR is a small portion of the solar spectrum
outside the visible range. The bandwidths within the optical radiation spectrum are listed
in Table 1-1.
Table 1-1 Optical radiation spectrum
Region Wavelength range
uv 100 to 400 nm
uvc 100 to 280 nm
UVB" 280 to 315 nm
UVA
3
315 to 400 nm
Visible 400 to 780 nm
Infrared (IR) 780 nm to 1 mm
IRA 780 run to 1 .4 j..Lm
IRB 1.4 to 3.0 j..Lm
IRC 3.0 j..Lm to 1 mm
Source: Adapted from ACGIH 1996
"Photobiological designations of the Commission lntemationale de l'Eclairage (International Commission
on Illumination), cited in lARC 1992.
Dec. 2000
Various conventions are used to classify the optical radiation spectrum into separate
bands (e.g., on the basis of transmission and absorption properties). These spectral-band
categories are used to identify approximate wavelengths; they do not designate fine
dividing lines below which an effect is present and above which it does not occur.
1.1 Identification of UVR by type
UVR contains wavelengths from 100 to 400 run and is classified as follows: UV A, 315 to
400 nm; UVB, 280 to 315 nm; and UVC, l 00 to 280 nm. This nomenclature is not
always rigorously followed, as different researchers use slight variations in these ranges.
The relative position of UVR in the electromagnetic spectrum is shown in Figure 1-1.
1.2 Physical properties
The atmosphere does not absorb UV A, which is the most abundant of the three UVR
bands and accounts for 95% of the UV energy reaching the earth's surface at the equator.
UVB normally is absorbed by the ozone layer; it constitutes 5% of solar UVR and is the
most biologically critical part of solar UVR (Farmer and Naylor 1996, cited in NTP
2000). Naturally occurring UVC, the shortest UV wavelength produced by the sun, is the
type ofUVR most harmful to the genome; however, it is totally absorbed by the earth's
atmosphere (Daya-Grosjean et al. 1995, cited in NTP 2000).
2
Dec. 2000
Adapted from NASA 2000
Figure 1-1. Electromagnetic spectrum
1.3 Photochemical and photobiological activities
Photochemical and photobiological interactions occur when a photon reacts with a
molecule of matter, producing either a photochemically altered species or two dissociated
molecules (Phillips 1983, Smith 1989, both cited in IARC 1992). For this reaction to be
effective, the amount of photon energy must be sufficient to alter molecular bonds.
Photon energy typically is expressed in electronvolts (the photon energy of light of
wavelength 300 nm = 4.1 e V) (WHO 1979, cited in fARC 1992). The number of altered
molecules produced relative to the number of absorbed photons is referred to as
3
Dec. 2000
"quantum yield" (Phillips 1983, cited in IARC 1992). The efficiency of a photochemical
interaction per incident quantum and the photobiological effects per unit radiant exposure
vary widely with wavelength (Jagger 1985, cited in IARC 1992).
4
Dec. 2000
2 Human Exposure
2.1 Use
UVR has many uses as a natural source of energy and is important in various biological
processes. Artificial sources ofUVR are used for tanning, medical diagnosis and
treatment, and promoting polymerization reactions. Exposure to UVR usually is
expressed as a dose rate in watts per square meter (the power striking a unit surface area
of an irradiated object). The commonly used unit of effective dose is the minimal
erythema dose (MED), which is defined as the lowest radiant exposure to UVR sufficient
to produce erythema of the skin with sharp margins within 24 hours of exposure. Though
imprecise, MEDs are useful, because they are related to the biological consequences of
the exposure (IARC 1992).
2. 1. 1 Cosmetic use
Tanning beds use artificial light to allow individuals to develop "suntan" for cosmetic
reasons. Originally, tanning beds were built with mercury arc lamps, which emitted large
quantities ofUVB and UVC. Now, sunbeds and solaria emit mostly UV A (IARC 1992).
Table 2-5 summarizes the characteristics of various light sources used for tanning.
Radiation emission (%) Contribution to tanning (%)
Lamp UVA UVB uvc UVA UVB uvc
Mercury arc sunlamp 40 40 20 0 35 65
Simulated sunlight lamp 95 5 0 20 80 0
Type I UV A lamp 99 1 0 60 40 0
Type II UV A lamp > 99.9 < 0.1 0 > 90 < 10 0
Optically filtered high-pressure lamp 100 0 0 100 0 0
Summer UV sunlight 95 5 0 20 80 0
Source: IARC 1992
2. 1.2 Medical and dental applications
UVR has both diagnostic and therapeutic uses in medicine and dentistry. More than 30
disorders can now be treated through UVA exposure with psoralens (PUVA). Psoriasis
and eczema are the skin diseases most frequently treated with PUV A therapy. PUV A can
also be used with UVB exposure to treat psoriasis patients who are not good candidates
for systemic therapy with methotrexate or etretinate (Morison 1992). UVR (most
commonly UVB) and coal-tar creams also are used to treat psoriasis (FDA 1996). In
addition, UVB is used to convert ?-dehydrocholesterol (provitamin D3) to vitamin Din
the skin of vitamin D-deficient patients.
UV A has been used to treat neonatal jaundice or hyperbilirubinemia. Although treatment
usually involves irradiating the infant with visible light for several hours a day, for up to
one week, one commercial neonatal phototherapy unit was found also to emit UV A and
radiation at wavelengths down to 265 run (in the UVC range) (!ARC 1992). UVA has
been found to alter the molecular structure of melatonin, a honnone that helps regulate
5
Dec. 2000
sleep-wake cycles, to unidentified photoproducts; moderate phototoxicity of melatonin
has been predicted (Kim et al. 1999). UVR also has been used to detect various dental
disorders, such as early dental caries, dental plaque, and calculus (IARC 1992).
2.1.3 Industrial applications
UVR has many industrial applications. One of the major industrial uses involves
photopolymerization, which includes curing of protective coatings and inks. UVR also is
used to simulate weathering of various materials, such as polymers. It is used to sterilize
and disinfect, usually in the range of 260 to 265 nm (UVC). Other uses include UV
photography and use ofUV lasers. UVR is a byproduct of electric-arc welding (IARC
1992).
2.2 Production
In the broadest sense, UVR is formed when a body is heated (through incandescence) or
when electrons that have been raised to an excited state return to a lower energy level.
UVR is naturally emitted from the sun. Around two-thirds of the energy emitted by the
sun penetrates the atmosphere. UVR comprises approximately 5% of the solar radiation
that reaches the earth's surface. Artificial sources ofUVR include tungsten/halogen, gas
discharge, arc, fluorescent, metal halide, and electrodeless lamps (IARC 1992).
2.3 Analysis
UVR can be measured with chemical or physical detectors, often in conjunction with a
monochromator or band-pass filter for wavelength selection. Chemical detectors include
photographic emulsions, actinometric solutions, and UV -sensitive plastic films. Physical
detectors include radiometric devices and photoelectric devices (IARC 1992).
2. 3. 1 Spectroradiometry
Spectroradiometry is generally considered the best way to characterize a source of UVR
and is based on measurement of its spectral power distribution (radiated power as a
function of wavelength). Spectral measurements are used to calculate biologically
weighted radiometric quantities. A spectroradiometer consists of three parts. (1) Input
optics collect the incident radiation and conduct it to (2) the entrance slit of a
monochromator, which disperses the radiation with one or two dispersion devices
(diffraction grating or prism). The monochromator then guides the radiation to the exit
slit by way of mirrors, where it enters (3) the radiation detector, normally a photodiode,
or a photomultiplier tube for higher sensitivity. The accuracy of UVR measurements is
affected by various parameters, including wavelength calibration, bandwidth, stray
radiation, polarization, angular dependence, linearity, and calibration sources. Double
monochromators are used to provide accurate UVR readings.
2.3.2 Wavelength-independent (thermal) detectors
Thermal detectors usually are used to measure the total radiant power of a source, rather
than just the UV component. Thermal detectors operate on the principle that UVR
absorbed by a receiving element will cause a temperature rise in the element. This rise is
measured, usually with a thermopile or pyroelectric detector. Thermopiles must have a
window made of fused silica for measuring UVR at wavelengths as low as 250 nm.
6
Dec. 2000
Pyroelectric detectors rely on voltage generated by temperature changes in a lithium
tantalate crystal.
2.3.3 Wavelength-dependent detectors
The accuracy of wavelength-dependent detectors varies depending upon the types of
detectors and filters used. The most common is the Robertson-Berger meter, which
incorporates optical filters, a phosphor, and a vacuum phototube or photovoltaic cell. The
meter measures wavelengths < 330 nm in the global spectrum. The spectral response rises
sharply with decreasing wavelength.
Detectors incorporating a photodiode or vacuum photocell in conjunction with optical
filters and suitable input optics (such as a quartz hemispherical detector) have been used
to match a number of different action spectra. The American Conference of
Governmental Industrial Hygienists (ACGIH) uses one of these detectors, the
International Light Model 730 UV Radiometer, to evaluate the health hazards of
exposure to UVR.
A complementary approach to evaluating UVR is the use of photosensitive films. By
relating the degree of deterioration of the films, usually measured as changes in their
optical properties, the user can determine the dose of incident UVR. The most widely
used photosensitive film is polymer polysulfone.
It is difficult to achieve a prescribed UVR spectral dose with wavelength-dependent
detectors. Accurate results require detectors that are calibrated against the appropriate
source spectrum with a spectroradiometer. If this is not done, dosimetric errors will arise.
Measuring UVB radiation also is difficult, as only 0.3% of the sun's total radiant energy
is UVB.
2.4 Environmental occurrence
Solar radiation is scattered by various components of the atmosphere, and about two-
thirds of it penetrates to the earth's surface. UVC exists in the extraterrestrial solar
spectrum, but is completely filtered out by the ozone layer. Most UVB is absorbed by
ozone in the stratosphere, and only a small fraction (around 5%) of the total radiation
penetrating to the earth's surface is UVB (IARC 1992).
2.5 Environmental exposure
2.5.1 Solar UVR
Information on global UVR levels has been compiled from data gathered for
epidemiological studies of skin cancer and other health effects, such as premature aging
of the skin, cataracts, and suppression of the immune response. Despite the large number
of measurements, estimating human exposure is complex. UVR spectral irradiance varies
considerably with latitude, altitude, time of day, and season. People also vary in their
length of outdoor exposure and parts of the body exposed. In addition, individual
exposure geometry complicates efforts to estimate human exposure. Although UVR
levels were estimated for many studies, few were able to differentiate among UV A,
UVB, and UVC (IARC 1992).
7
Dec. 2000
2.5.1.1 UVA
Various factors influence terrestrial levels ofUV A. UV A levels decrease with increasing
distance from the equator and increase with increasing altitude (decreasing with distance
below sea level). Terrestrial UV A levels also are decreased by stratospheric ozone, which
varies with latitude and season. When there is less ozone, more UV A wi ll reach the
earth's surface. Time of day also influences daily UV A levels (IARC 1992). Table 2- 1
shows the proportion of UV A radiation received during two periods on a summer day at
three latitudes (altitude not specified).
Table 2-1. Percentage of daily UV A radiation received during two periods on a clear
summer day
UVA (% of daily total)
Latitude (
0
N) 11 :00 AM - 1 :00 PM 9:00 AM- 3:00 PM
20 27 73
40 25 68
60 21 60
Source: IARC 1992
Clouds reduce the amount of UV A reaching ground level. Air pollution, including
tropospheric ozone, can decrease UV A exposure, especially in urban areas (IARC 1992).
Surface reflection also contributes to personal exposures to UV A.
2.5.1.2 UVB
Terrestrial UVB levels are affected by the same factors that influence terrestrial UV A
levels. However, because UVB is absorbed more by stratospheric ozone than is UV A,
differences in latitude and altitude affect UVB exposure more than UV A exposure.
Seasonal changes affect UVB levels, mostly in temperate regions. Table 2-2 gives UVB
exposure levels for various latitudes and seasons (altitude not specified).
Table 2-2. Typical values for ambient daily and annual UVB radiation expressed as
minimal erythema dose
Diurnal UVB (MED)
Latitude (
0
N) Winter Spring/ Autumn Summer Annual
20, Hawaii 14 20 25 6,000
30, Florida 5 12 15 4,000
40, New Jersey 2 7 12 2,500
50, Washington 0.4 3 10 1,500
Source: IARC 1992
8
Dec. 2000
Time of day at a given latitude also affects UVB levels, as shown in Table 2-3 (altitude
not specified).
Table 2-3. Percentage of daily UVB radiation received during two periods on a clear
summer day
UVB (% of daily total)
Latitude (
0
N) 11 :00 AM - 1 :00 PM 9:00 AM - 3:00 PM
20 30 78
40 28 75
60 26 69
Source: IARC 1992
Variation in stratospheric ozone with latitude and season affects UVB levels. Air
pollution decreases UVB exposure, and clouds also affect UVB levels. Generally, cloud
cover scatters less than 10% of the UVB under a clear sky. However, very heavy cloud
cover virtually eliminates UVB, even in the summer. Surface reflection contributes to
human UVB exposure. Exposure due to reflection is important, as body parts normally
shaded are exposed to reflected radiation (IARC 1992). Table 2-4 summarizes reflectance
for various types of terrain.
9
Dec. 2000
Table 2-4. Representative terrain reflectance factors for horizontal surfaces
measured with a UVB radiometer at 12:00 PM at various U.S. locations
Material Reflectance(%)
Lawn grass, summer, Maryland, California, and Utah 2.0-3.7
Lawn grass, winter, Maryland 3.0- 5.0
Wild grasslands, Vail Mountain, Colorado 0.8- 1.6
Lawn grass, Vail, Colorado 1.0-1.6
Flower garden, pansies 1.6
Soil, clay and humus 4.0-6.0
Sidewalk, light concrete 10-12
Sidewalk, aged concrete 7.0- 8.2
Asphalt roadway, freshly laid (black) 4.0- 5.0
Asphalt roadway, two years old (gray) 5.0-8.9
House paint, white, metal oxide 22
Boat dock, weathered wood 6.4
Aluminum, dull, weathered 13
Boat deck, wood, urethane coating 6.6
Boat deck, white fiberglass 9.1
Boat canvas, weathered, plasticized 6.1
Chesapeake bay, Maryland, open water 3.3
Atlantic Ocean, New Jersey coastline 8.0
Sea surf, white foam 25-30
Atlantic beach sand, wet barely submerged 7.1
Atlantic beach sand, dry, li ght 15- 18
Snow, fresh 88
Snow, two days old 50
Source: IARC 1992
2.5.1.3 uvc
No data on environmental exposure to UVC were found in the published literature.
2.5.2 Artificial sources
Six artificial sources of UVR have been identified. (1) Incandescent sources provide
optical radiation that appears as a continuous spectrum. A "color temperature" usually
describes incandescent sources. UVR emission occurs when the color temperature
exceeds 2,500K (2,227C). (2) Gas discharge lamps produce optical radiation by passing
an electrical current through a gas. The type of gas present in the lamp determines
emission wavelengths. At low pressures, fine lines are produced, while higher pressures
create broad bands. Low-pressure discharge lamps filled with mercury, argon, xenon,
krypton, or neon are used to create specific bands for spectral calibrations. (3) Arc lamps
are intense sources ofUVR. They are operated under extreme pressures and have color
10
Dec. 2000
temperatures of 6,000K (5,727C). Arc lamps often are used to simulate solar radiation.
(4) Fluorescent lamps create radiation from a low-pressure mercury discharge, which
produces a strong emission at 254 nm. This in turn excites the phosphor-coated lamp to
produce fluorescence. Various emission spectra can be obtained by alteration of the
composition and thickness of the phosphor and the glass envelope. (5) Metal halide lamps
add metal to a mercury discharge lamp, allowing for lines in addition to the mercury
emission spectrum. ( 6) Electrode less lamps use magnetrons to generate microwave
energy, which then is absorbed by the discharge tube (IARC 1992).
2.6 Occupational exposure
2. 6. 1 Solar UVR
Occupational exposure to solar UVR occurs for anyone working outside. For a group of
more than 800 outdoor workers in the United States at 40 N latitude, personal annual
facial exposure doses were estimated at 30 to 200 MED (Rosenthal et at. 1991, cited by
!ARC 1992). This unusually low estimate may be due to the fact that Rosenthal assumed
facial exposure to be about 5% to 10% of ambient exposure. Other data suggest that
facial exposure is around 30% of ambient exposure. By the latter estimate, the annual
facial exposure doses for these outdoor workers would be 80 to 500 MED.
2.6.2 Artificial UVR
Electric arc welders are the largest occupational group with exposure to artificial UVR. It
has been estimated that over half a million welders in the United States have been
occupationally exposed to UVR. Levels of effective UV irradiance (relative to the action
spectrum of the ACGIH) around electric arc welding equipment at 1m with an arc
current of 400 A ranged from 1 to 50 W/ m
2
, and the unweighted UVA irradiance ranged
from 3 to 70 W/m
2
, depending upon the type of welding and the metal being welded.
Other occupational exposures to artificial UVR are low, ranging from 10 W/m
2
(offices
and discotheques) to 20 W/m
2
(sunbed shop with 20 or more tanning appliances).
Occupational exposure to artificial UVR depends upon both the source and the protective
methods used to decrease exposure. Some artificial UVR sources are self-contained, such
as germicidal lamps in some uses, and present no risk to workers. Other occupational
uses, such as use ofUVR in laboratories, UV photography, and UV lasers, inevitably lead
to UVR exposure where short-term and intense exposures may occur (IARC 1992).
2.7 Biological indices of exposure
The common biological indices of exposure to UVR are erythema and photokeratitis.
Erythemas, or "sunburns," are used as a simple indicator of the biological consequences
of UVR exposure. One study determined the action spectra for DNA photodamage in
different human epidermal layers in situ. Overall, the action spectrum for erythema is 280
to 340 nm (UVB and part ofUVA) (Young et al. 1998).
2.8 Regulations
The U.S. Food and Drug Administration (FDA) regulates UVR, establishing safe uses for
irradiation in the production, processing, and handling of food. The FDA also sets forth
labeling requirements for drugs containing coal tars for use with UVR. The FDA
II
Dec. 2000
regulates various devices that emit UVR, such as sunlamps, sunbeds, medical lamps, and
purifiers. The Occupational Safety and Health Administration (OSHA) regulates UVR
exposure among welders and cutters; regulations cover safety precautions, guidelines,
and treatment. Table 2-5 summarizes FDA regulations that affect UVR, and Table 2-6
summarizes OSHA regulations that affect UVR.
Table 2-5. FDA regulations
Regulatory action Effect of regulation and other comments
21 CFR - Specific Labels on dietary food low in fat may identify one or
Requirements for Health Claims. Promulgated: 58 FR more of the following risk factors for development of
2801, 0 l/06/93. Health claims: dietary lipids and cancer. cancer: family history of a specific type of cancer,
cigarette smoking, alcohol consumption, overweight and
obesity, ultraviolet or ionizing radiation, exposure to
cancer-causing chemicals, and dietary factors.
21 CFR 179- PART 179- IRRADIATION IN THE Subparts A through C govern the radiation, radiation
PRODUCTION, PROCESSING AND HANDLING OF sources, and packing materials for irradiated foods in the
FOOD. Promulgated: 42 FR 14635,03115/77. U.S. production, processing, and handling of food.
Codes: 21 U.S.C. 321, 342, 343, 348, 373, 374.
21 CFR 179.39-Ultraviolet radiation for the processing Ultraviolet radiation for the processing and treatment of
and treatment of food. Promulgated: 61 FR 42383, food may be safely used under the following conditions:
08/ 15/96. ( 1) The radiation sources consist of ultraviolet emission
tubes designed to emit wavelengths within the range of
2200-3000 il units with 90% of the emission being the
wavelength 2537 il units. (2) The ultraviolet radiation is
used or intended for use as follows: surface
microorganism control for food and food products and
the sterilization of potable water used in food
production.
21 CFR 358-PART 358-MISCELLANEOUS For labeling of products containing coal tar identified in
EXTERNAL DRUG PRODUCTS FOR OVER-THE- 358.71 O(c) for the control of psoriasis, under the heading
COUNTER HUMAN USE. Promulgated: 55 FR 33255, "Indications," the labeling of the product will state: "Do
08/ 14/90. U.S. Codes: 21 U.S.C. 321,351,352,353, not use this product with other forms of psoriasis therapy
355, 360, 371. Labeling of drug products for the control such as ultraviolet radiation or prescription drugs unless
of dandruff, seborrheic dermatitis, or psoriasis. directed to do so by a doctor."
21 CFR 872.6010ff.--Miscellaneous Devices. An ultraviolet activator for polyme1ization is a device
Promulgated: 52 FR 30097,08/ 12/87. U.S. Codes: 21 that produces ultraviolet radiation intended to
U.S.C. 351, 360, 360c, 360e, 360j, 371. Ultraviolet polymerize (set) resinous dental pit and fissure sealants
activator for polymerization. or restorative materials by transmission of light through
a rod. It is classified as a Class II product.
21 CFR 878-PART 878-GENERAL AND PLASTIC This part sets forth the classification of general and
SURGERY DEVICES. Promulgated: 53 FR 23872, plastic surgery devices intended for human use that are
06/24/88. U.S. Codes: 21 U.S.C. 351 , 360, 360c, 360e, in commercial distribution.
360j, 3601,371.
12
Dec. 2000
21 CFR 878.4630---Ultraviolet lamp for dennatologic An ultraviolet lamp for dermatologic disorders is a
disorders. Promulgated: 53 FR 23872, 06/24/88. U.S. device (including a fixture) intended to provide
Codes: 21 U.S.C. 351, 360, 360c, 360e, 360j, 3601, 371. ultraviolet radiation of the body to photoactivate a drug
in the treatment of a dermatologic disorder if the
labeling of the drug intended for use with the device
bears adequate directions for the device's use with that
drug. It is classified as a Class 11 product.
21 CFR 878.4635- Ultraviolet lamp for tanning. An ultraviolet lamp for tanning is a device that is a lamp
Promulgated: 55 FR 48440, 11/20/90. U.S. Codes: 21 (including a fixture) intended to provide ultraviolet
U.S.C. 351, 360, 360c, 360e, 360j, 3601,371. radiation to tan the skin. This device is classified as a
Class I product and therefore is exempt from the
premarket notification procedures in subpart E of part
807 of this chapter.
21 CFR 880-PART 880-GENERAL HOSPITAL This part sets forth the classification of general hospital
AND PERSONAL USE DEVICES. Promulgated: 45 FR and personal use devices intended for human use that are
69682-69737, 10/21/80. U.S. Codes: 21 U.S.C. 351, in commercial distribution.
360, 360c, 360e, 360j, 371.
21 CFR 880.6500-Medical ultraviolet air purifier. A medical ultraviolet air purifier is a device intended for
Promulgated: 45 FR 69682-69737, 10/21/80. U.S. medical purposes that is used to destroy bacteria in the
Codes: 21 U.S.C. 351, 360, 360c, 360e, 360j , 371. air by exposure to ultraviolet radiation. This device is
classified as a Class II product (perfonnance standards).
21 CFR 880.671 0---Medical ultraviolet water purifier. Identification. A medical ultraviolet water purifier is a
Promulgated: 45 FR69682-69737, 10/21/80. U.S. device intended for medical purposes that is used to
Codes: 21 U.S.C. 351 , 360, 360c, 360e, 360j, 371. destroy bacteria in water by exposure to ultraviolet
radiation. This device is classified as a Class II product
(performance standards).
21 CFR 1000-PART 1000-GENERAL. Promulgated: Examples of electronic products that may emit
38 FR 28624, 10/ 15173. U.S. Codes: 21 U.S.C. 360hh- ultraviolet radiation are biochemical and medical
360ss. Examples of electronic products subject to the analyzers, tanning and therapeutic lamps, sanitizing and
Radiation Control for Health and Safety Act of 1968. sterilizing devices, black-light sources, and welding
equipment.
13
Dec. 2000
21 CFR 1040- PART 1040- PERFORMANCE Sunlamp products and ultraviolet lamps manufactured
STANDARDS FOR LIGHT-EMITTING PRODUCTS. on or after May 7, 1980, but before September 8, 1986,
Promulgated: 50 FR 36550, 09/06/85. U.S. Codes: 21 are subject to the provisions of this section. Sunlamp
U.S.C. 351,352,360, 360e-360j, 371, 381; 42 U.S.C. product means any electronic product designed to
263b-263n. Sunlamp products and ultraviolet lamps incorporate one or more ultraviolet lamps and intended
intended for use in sunlamp products. for irradiation of any part of the living human body, by
ultraviolet radiation with wavelengths in air between 200
and 400 nm, to induce skin tanning. Timer systems,
control for termination of radiation emission, protective
eyewear requirements, and labeling requirements are
described. A warning statement with the words
"DANGER-Ultraviolet radiation. Follow instructions.
Avoid overexposure. As with natural sunlight,
overexposure can cause eye and skin inj ury and allergic
reactions. Repeated exposure may cause premature
aging of the skin and skin cancer. WEAR
PROTECTIVE EYE WEAR; F AlLURE TO MAY
RESULT IN SEVERE BURNS OR LONG-TERM
INJURY TO THE EYES. Medications or cosmetics may
increase your sensitivity to the ultraviolet radiation.
Consult physician before using sunlamp if you are using
medications or have a history of skin problems or
believe yourself especially sensitive to sunlight. If you
do not tan in the sun, you are unlikely to tan from the
use of this product" must be placed on each sunlamp
product. Each ultraviolet lamp shall have a label which
contains the words "Sunlamp-DANGER-Ultraviolet
radiation. Follow instructions."
Source: The regulattons tn thts table have been updated through the 1999 Code of Federal Regulations 21
CFR, 1 April 1999.
14
Dec. 2000
Table 2-6. OSHA Regulations
29 CFR 1910.250-SUBPART Q-Welding, Cutting Where the work permits, the welder should be enclosed
and Brazing. Promulgated: 55 FR 13696, 04/ 11190. U.S. in an individual booth painted with a finish of low
Codes: 29 U.S.C. 653, 655, 657. reflectivity, such as zinc oxide (an important factor for
absorbing UVR) and lamp black, or shall be enclosed
with noncombustible screens similarly painted. Booths
and screens shall permit circulation of air at floor level.
Workers or other persons adjacent to the welding areas
shall be protected from UVR by noncombustible or
flameproof screens or shields or shall be required to
wear appropriate goggles.
29 CFR 1926.350-SUBPART J-Welding and Since the inert-gas metal-arc welding process involves
Cutting. Promulgated: 58 FR35179, 06/30/93. U.S. the production of ultraviolet radiation of intensities of 5
Codes: 29 U.S.C. 653, 655, 657, 40 U.S.C. 333. Inert- to 30 times those produced during shielded metal-arc
gas metal-arc welding. welding, employees shall not be permitted to engage in
or be exposed to the process until the following special
precautions have been taken: (1) The use of chlorinated
solvents shall be kept at least 200 feet, unless shielded,
from the exposed arc, and surfaces prepared with
chlorinated solvents shall be thoroughly dry before
welding is permitted on such surfaces. (2) Employees in
the area not protected from the arc by screening shall be
protected by filter lenses. When two or more welders are
exposed to each other's arc, filter lens goggles of a
suitable type shall be worn under welding helmets. Hand
shields to protect the welder against flashes and radiant
energy shall be used when either the helmet is lifted or
the shield is removed. (3) Welders and other employees
who are exposed to radiation shall be suitably protected
so that the skin is covered completely to prevent bums
and other damage by ultraviolet rays. Welding helmets
and hand shields shall be free of leaks and openings, and
free of highly reflective surfaces.
Source: The regulattons tn thts table have been updated through the 1999 Code of Federal Regulattons 29
CFR, I July 1999.
15
Dec. 2000
16
Dec. 2000
3 Human Cancer Studies
Humans can be exposed to UVR from natural (solar) and artificial sources (see Sections
1 and 2). The IARC (1992) and the National Toxicology Program (NTP 2000)
reviewed the evidence for human carcinogenicity of solar radiation and exposure to
sunlamps or sunbeds. Both reports concluded there was sufficient evidence in humans
that solar radiation was carcinogenic, causing malignant melanoma of the skin and non-
melanoma skin cancer. Solar radiation is classified by the IARC ( 1992) as carcinogenic
to humans (Group 1) and is listed in the Ninth RoC (2000) as known to be a human
carcinogen.
The 1992 IARC review also considered artificial sources ofUVR. The IARC Working
Group characterized the human evidence concerning the carcinogenicity of artificial
sources of UVR as limited, and classified exposures associated with the use of
sunlamps and tanning beds as probably carcinogenic to humans (Group 2A). The NTP
(2000) review concluded that there was sufficient evidence from human studies to list
exposure to sunlamps or sunbeds as known to be a human carcinogen, based on
epidemiological studies evaluated by the IARC and studies published after the 1992
IARC review. The NTP (2000) conclusions about the carcinogenicity of solar radiation
and exposure to sunlamps and sunbeds were based on the NTP background document
( 1997) prepared to evaluate these exposures.
The purpose of this section is to review evidence in humans regarding the potential
carcinogenicity of broad-spectrum UVR and its components (UV A, UVB, and UVC).
The most extensive literature comes from studies on sunlight and cancer; however,
these studies are not specific for UVR. Evidence for the role of the UVR component of
solar radiation in carcinogenicity comes from studies with artificial sources of UVR,
twnor-site concordance between humans exposed to solar radiation and animals
exposed to UVR from artificial sources (see Section 4), and human mechanistic studies
using artificial sources ofUVR (see Sections 5 and 6). Epidemiologic studies
evaluating exposure to artificial sources of UVR are valuable for assessing the effects
ofUVR itself and the role ofthe UVR component in solar radiation. Human
epidemiologic evidence on the carcinogenicity of specific components of the UVR
spectrum, including UV A, UVB, and UVC, is limited. The IARC Working Group
noted that none of the studies reviewed had assessed the emission spectra of artificial
UV sources, and little additional information from human studies has been produced
since the 1992 IARC evaluation. This section summarizes the 1992 IARC review, the
1997 NTP review, and post-1992 reviews of the extensive literature on solar radiation,
and reviews human studies evaluating carcinogenic effects of exposure to UVR from
artificial sources (including broad-spectrum UVR and specific UVR components),
concentrating on exposure to sunlamps or sunbeds.
17
Dec. 2000
3.1 Solar radiation
3. 1. 1 Evaluations by the /ARC (1992) and the NTP (2000)
The IARC (1992) evaluation provides extensive information on the evidence for the
carcinogenicity of solar radiation in humans. The studies reviewed examined malignant
melanoma of the skin, non-melanoma skin cancer, malignant melanoma of the eye, and
cancer of the lip, with the majority of the evidence pertaining to the first two cancers.
The results of descriptive epidemiologic studies suggest that exposure to sunlight
increases the risk of nonmelanocytic cancer. Nonmelanocytic tumors occur
predominantly on regions of the body exposed to sunlight. Evidence to suggest that
these cancers are associated with the UVR component of sunlight comes from latitude
studies. There is a strong inverse relationship between latitude of residence and cancer
incidence or mortality and, conversely, a positive relationship between measured or
estimated ambient UVR and cancer incidence or mortality. Three case-control studies
found a significantly increased risk of cancer of the lip associated with outdoor work (a
proxy for UVR exposure).
The analytic epidemiologic literature on the relationship between malignant melanoma
of the skin and exposure to sunlight is extensive. Population-based case-control studies
in western Australia, Queensland, western Canada, and Denmark showed consistent
positive associations of malignant melanoma with residence in sunny environments
throughout life, in early life, and for short periods in early adult life, and with measures
of cumulative sun damage, such as microtopographical changes or history of keratosis
or nonmelanocytic skin cancer. Most studies showed positive associations with
measures of intermittent sun exposure, but associations with total (lifetime) sun
exposure or occupational sun exposure were inconsistent.
Only one study reviewed by the IARC referred to a specific component of the UVR
spectrum. A cross-sectional study of Maryland fishermen included estimates of annual
and lifetime exposure to UVB obtained through a combination of self-reported history
and measurements with film dosimeters (Vitasa et al. 1990, cited in IARC 1992). After
adjustment for age, eye color, childhood freckling, and skin reaction to sunlight,
squamous-cell carcinoma was associated with cumulative UVB exposure above the
75th percentile (odds ratio [OR] = 2.53, 95% CI = 1.18 to 5.01 ), but basal-cell
carcinoma was not associated with exposure to UVB. Basal-cell carcinoma is more
strongly associated with nonoccupational than occupational sun exposure and with
intermittent than total exposure (English et al. 1997). No other study providing
information about the association of specific UV wavelengths with skin cancer was
identified.
The relationship between solar radiation and non-Hodgkin's lymphoma is less clear.
The NTP background document on solar radiation and exposure to sunlamps or
sunbeds evaluated four studies (Bentham and Aase 1996, Newton et al. 1996, Hartge et
al. 1996, McMichael and Giles 1996) that provided limited support for an association
of solar radiation with non-Hodgkin's lymphoma. Two of these studies evaluated the
relationship of cancer with levels of solar UVB. In a U.S. study, Hartge et al. (1996)
reported that state annual average estimated solar UVB levels (adjusted for latitude,
18
Dec. 2000
altitude, and cloud cover) were positively correlated with state mortality rates for
melanoma and non-melanoma skin cancer in white males, but negatively correlated
with mortality rates for non-Hodgkin's lymphoma (P < 0.0001 for all coefficients). In
contrast, in a worldwide study, McMichael and Giles (1996) reported that the
incidences of non-Hodgkin's lymphoma and cutaneous malignant melanoma in white
Caucasoid populations (from 49 registries in 19 countries) were positively correlated
with estimated average annual UVB exposure (as MED, based on latitude and adjusted
for cloud cover). The correlation coefficients were 0.50 in males and 0.51 in females
for non-Hodgkin's lymphoma and 0.75 in males and 0.67 in females for melanoma (P <
0.001 for all coefficients). Neither of these studies was specific for UVB radiation,
because they were based on estimates ofUVB levels as a portion of total solar UVR,
which also includes a UV A component.
3. 1.2 Recent epidemiologic studies
Epidemiologic studies of sun exposure and skin cancer published after the 1992 IARC
evaluation were reviewed by Elwood ( 1996) and Armstrong and Kricker ( 1996).
Elwood (1996) provided a comprehensive review of studies on melanoma and sun
exposure published through 1995, including eight case-control studies published after
the IARC review, and Armstrong and Kricker (I 996) reviewed studies of malignant
melanoma and non-melanoma skin cancer. These reviews reinforced the !ARC's
fundamental conclusions, but presented no new information relating specifically to
UVAorUVB.
The importance of the conditions of sunlight exposure with respect to melanoma has
been further evaluated in recent studies. Elwood and Jopson ( 1997) reported an overall
analysis of 35 case-control studies that evaluated the relationship between cutaneous
malignant melanoma and sun exposure (intermittent, occupational, and total) and age-
specific history of sunburn. Overall, risk was significantly increased by intermittent
exposure (OR = 1.71, 95% CI = 1.54 to 1.90) and significantly reduced by high
occupational exposure (OR= 0.86, 95% CI = 0.77 to 0.96); a small excess risk
associated with total exposure was marginally significant (OR = 1.18, 95% CI = 1.02 to
1.38). The estimates of risk with respect to sun exposure showed considerable
heterogeneity (P < 0.001). For intermittent exposure, 21 of23 studies with relevant
exposure infonnation found a positive association with melanoma, which was
statistically significant in 16 studies. Sunburn at all ages or as an adult significantly
increased the risk of melanoma (OR = 1.91, 95% CI = 1.6 to 2.17), as did sunburn in
adolescence or in childhood. The authors suggested that the association with sunburn
also reflected the effect of intetmittent exposure.
Recent studies evaluating the relationship between sunlight and non-Hodgkin's
lymphoma provided little additional information bearing on the conclusions of the
Ninth RoC (2000). Adami et al. (1999) conducted a population-based cohort study in
Sweden, which assessed UVR exposure by occupation (using job titles obtained from
the census) and latitude (based on classification of each individual's home and work
addresses). Data for incidences ofnon-Hodgkin's lymphoma, chronic lymphocytic
leukemia, malignant melanoma, and squamous-cell carcinoma were obtained from the
Swedish Cancer Registry. Adami et al. (1996) reported a positive association between
19
Dec. 2000
latitude of residence and sex-specific age-adjusted relative risks of non-Hodgkin's
lymphoma but did not find an association with occupation, where job title and industry
served as a surrogate for exposure (indoor versus outdoor occupations). In a population-
based case-control study in the United States, Freedman et al. (1997) reported an
inverse association between non-Hodgkin's lymphoma mortality and sunlight exposure,
as assessed from occupational and residential information on death certificates. Two
separate case-report studies reported positive associations with residential and
occupational surrogates for sunlight exposure, for skin cancer mortality in one study
and for melanoma in the other (Freedman eta/. 1997).
3.2 UVR from artificial sources
Humans are exposed to artificial sources ofUVR for cosmetic purposes (sunlamps or
sunbeds), for medical treatment (PUV A and UVB treatment of psoriasis), and through
occupational exposure (e.g., fluorescent lights or welding) (see Section 2). In most of
these studies, with the possible exception of medical exposure, exposure was to broad-
spectrum UVR, or the type ofUVR was unknown (see Tables 3-1, 3-2, and 3-3).
3.2. 1 Cosmetically related UVR exposure
As mentioned above, the most extensive epidemiological evidence for evaluation of the
relationship between human cancer and exposure to artificial UVR comes from studies
where the exposure was to sunlamps or sunbeds. This section reevaluates the literature on
cutaneous malignant melanoma and exposure to sunlamps or sunbeds, because of the
importance of these human studies in evaluation of the carcinogenicity of UVR radiation,
their relevance in elucidating the role of UVR in the carcinogenicity of solar radiation
and to address a recent epidemiologic review and assessment of exposure to tanning
lamps and malignant melanoma that was published since the 1997 NTP background
document (Swerdlow and Weinstock 1998).
The IARC (1992) classified exposure to sunlamps or sunbeds as probably carcinogenic
to humans. Two case-control studies published between the 1992 TARC review and the
1997 NTP assessment (Autier et al. 1994, Westerdahl et al. 1994) provided evidence
that exposure to sunlamps or sunbeds increased the risk of melanoma. The Ninth RoC
listed exposure to sunlamps or sunbeds as known to be a human carcinogen (NTP
2000), based on these two studies and the studies reviewed by the IARC (1992). Since
the 1997 NTP assessment, a review article and three additional studies have been
published. Swerdlow and Weinstock (1998) reviewed 19 case-control studies
evaluating the relationship of exposure to sunlamps and sunbeds with cutaneous
malignant melanoma, including the nine studies reported in the 1997 NTP background
document. The authors concluded that "although several investigations have found a
positive relation between tanning lamp use and melanoma, in some instances including
dose-response or duration-response effects, the methodologic limitations preclude any
firm conclusions regarding a causative relation".
Since Swerdlow and Weinstock's review, there have been three additional publications
evaluating the relationship of exposure to sunlamps or sunbeds to melanoma; one study
provided positive evidence (Westerdahl et al. 2000) and another provided limited
20
Dec. 2000
evidence (Chen et at. 1998). In addition, Walter et al. (1999) reanalyzed the case-
control study (Walter et al 1990) discussed in the NTP background document (1997),
providing further support for an elevated risk of melanoma with sunlamp or sun bed
exposure. The following sections evaluate the case-control studies on exposure to
sunlamps or sunbeds and cutaneous malignant melanoma and address the methodologic
concerns raised by Swerdlow and Weinstock (1998).
3.2. 1. 1 Epidemiologic studies of melanoma and sunlamp or sunbed exposure
The 22 publications evaluating the association between exposure to sunlamps or
sunbeds and malignant melanoma (19 reviewed by Swerdlow and Weinstock and three
more recent) relate to 21 case-control studies, because two of these publications
analyzed the same population (Walter et al. 1990, 1999); these reports were considered
as one study for the purpose of this evaluation. Two other case-control studies cited by
Swerdlow and Weinstock were not evaluated, because one study (Autier et al. 1991)
was descriptive rather than analytical, and exposure in the second study (Dubin et al.
1989) was not specific for sunlamp or sunbed use, but was characterized only as
medical and occupational. The remaining 19 case-control studies were reviewed.
Because these studies varied greatly in quality, including power to detect an effect,
characterization of exposure, and analysis of the effect, they did not contribute equal
information to the assessment of causality. The power of some studies was limited by
small numbers of exposed cases or because cases were accrued at an earlier time period
and so were inadequate to detect exposures that occurred in the 1980s (when tanning
salons became more popular). Some studies included "ever-use" of sunlamps or
sunbeds as part of larger studies focusing primarily on other risk factors for melanoma,
and provided little information about frequency or duration of exposure, age at
exposure, location of exposure, or body sites exposed. Also, several studies did not
report a risk estimate or reported little subgroup analysis with respect to such factors as
exposure, histologic type of cancer, or patient characteristics. Stratified analyses can
increase the sensitivity to detect an effect and provide other pertinent information
concerning sensitive subgroups.
Studies lacking sufficient power, detailed exposure assessment, or detailed analyses
were difficult to evaluate and provided little information about cancer effects due to
exposure to sunlamps or sunbeds. On the other hand, a few studies provided relatively
detailed exposure assessment and analyses. Thus, in an effort to evaluate causality, the
case-control studies were grouped into four tiers with respect to the quality of the
information concerning the exposure to sunlamps or sunbeds and its relationship to
cancer. Some studies differ in the ranking criteria according to analysis, exposure or
power; priority generally was given to the quality of exposure information. The case-
control studies are summarized in Table 3-1.
3.2. 1.2 Criteria for the four tiers and ranking of the studies
Tier 1. Exposure assessment: limited information; exposure was reported only as ever-
use. Analyses: a quantitative risk estimate was not calculated or reported; percentages of
21
Dec. 2000
exposed cases and controls were not reported, so risk estimates could not be calculated.
Power: limited by small numbers of exposed cases.
Studies: Klepp and Magnus 1979, Holly et al. 1987, Beitner et al. 1990.
Tier 2. Exposure assessment: limited information; exposure was reported only as ever-
use. Analyses: no detailed analyses, but information was provided (e.g., percentages of
exposed cases and controls) allowing a risk estimate to be calculated. Power: limited by
small numbers of exposed cases.
Studies: Adam et al. 1981, Gallagher et al. 1986, Holman et al. 1986, Zanetti et al. 1988,
MacKie et al. 1989, Dunn-Lane et al. 1993, Garbe et al. 1993.
(Note: Gallagher et al. reported that they had queried more detailed information on
frequency of exposure; however, they did not described the frequency of use, the number
of individuals exposed, or a risk estimate, thus this study was grouped in Tier 2.)
Tier 3. Exposure assessment: some infom1ation with respect to duration or frequency.
Analyses: more information with respect to risk calculation; some subgroup analysis.
Power: larger sample sizes; higher percentages of exposed individuals, but duration or
lifetime usage was low, so the numbers of highly exposed cases were small.
Studies: Elwood et al. 1986, Osterlind et al. 1988, Holly et al. 1995.
Tier 4: Exposure assessment: detailed information with respect to duration, frequency,
and other factors, such as age when exposure occurred or location of exposure. Analyses:
detailed subgroup analyses with respect to exposure characteristics, patient
characteristics, or histologic type of melanoma. Power: larger study populations, higher
percentages of individuals exposed to sunlamps or sunbeds, and/or longer durations of
usage. Exposure to sunlamps or sunbeds generally was the major focus of these studies.
Studies: Swerdlow et al. 1988, Walter et al. 1990 (reanalyzed in Walter et al. 1999),
Autier et al. 1994, Westerdahl et al. 1994, Chen et al. 1998, Westerdahl et al. 2000.
3.2.1.3 Evaluation of the evidence for association of malignant melanoma with exposure to
sunlamps or sunbeds
The three Tier 1 studies (Klepp and Magnus 1979, Holly et al. 1987, Beitner et al.
1990) and five ofthe seven Tier 2 studies (Gallagher et al. 1986, Holman et al. 1986,
Zanetti et al. 1988, Dunn-Lane et al. 1993, Garbe et al. 1993) found no association
between malignant melanoma and exposure to sunlamps or sunbeds. The other two Tier
2 studies (Adam et al. 1981, MacKie eta/. 1989) reported that a larger percentage of
cases than controls had used sunbed or sunlamps. Because the studies in Tiers 1 and 2
were limited in their ability to detect an effect or did not report information needed in
order to evaluate an effect, they provided little or no information for assessing
causality.
None of the studies in Tier 3 found a positive association between exposure to
sunlamps and malignant melanoma (Elwood et al. 1986, Osterlind et al. 1988, Holly et
al. 1995). Elwood et al. (1986) reported information on the duration of exposure;
22
Dec. 2000
however, this study was limited in power by the relatively small number of exposed
cases (15) resulting from the small number of malignant melanoma cases (83), and a
low level of exposure (average duration was 2.3 hours). Both Oster lind et al. ( 1988)
and Holly et al. (1995) evaluated malignant melanoma risk in relation to number of
sunlamp or sunbed uses. Holly et al. (1995) found no association between melanoma
and either low or high categories of sunlamp exposure but did not define the exposure
levels in each category, making it difficult to compare the exposures with those in other
studies. Osterlind et al. (1998) found no relationship between melanoma and number of
sunlamp uses, but did not report an OR for each exposure category. Exposures to
sunlamps were for both medical and cosmetic reasons. Melanoma risk in this study also
was not related to sunbed usage. The Tier 3 studies contributed some information to the
evaluation of causality.
The studies in Tier 4 provided the most information concerning causality, because they
contained detailed exposure assessments and analyses. Moreover, most of these studies
were better able to detect an effect, because of adequate study populations (mostly
> 400 cases), a higher proportion of exposed cases(> 20%), and a higher level of
lifetime exposure (total number of uses). Five of the six studies reported increased risk
of malignant melanoma associated with exposure to sunlamps or sun beds (Swerdlow et
al. 1988, Walter et al. 1990, 1999, Autier et al. 1994, Westerdahl et al. 1994, 2000).
The sixth study (Chen et al. 1998) provided limited support, because elevated risks
were observed only after subgroup analysis (e.g., stratification by the number of types
oflamps used and location and timing of exposure), but not for ever-use of sunlamps or
sunbeds (crude OR= 1.3; adjusted OR= 1.1).
In the Tier 4 studies, odds ratios for ever-use of sunlamps or sunbeds ranged from 1.1
to 2.9. Higher odds ratios were found for the higher exposure strata and in subgroup
analyses by patient characteristics (younger patients), exposure characteristics (younger
age of exposure), body site of cancer (mostly trunk and legs), and histologic type of
melanoma (superficial spreading and lentigo maligna). Four of the five studies that
tested for an exposure-response relationship reported a positive association (Swerdlow
et al. 1998, Walter et al. 1990, Westerdahl et al. 1994, 2000), though Westerdahl et al.
(2000) reported an exposure-response relationship only up to a total of 250 uses. Chen
et al. (1998) reported no relationship between the total number of sunlamp uses and
melanoma risk.
Few ofthese studies provided information on the types of sunlamps or sunbeds used.
This factor is important, because exposure in the 1970s was more likely to take place at
home with devices that emitted greater amounts ofUVB and UVC radiation, whereas
exposure in the 1980s increasingly occurred in commercial salons using devices that
emitted mainly UV A.
Chen et al. (1998) was the only study that obtained information concerning the type of
sunbed or sunlamp used (e.g., desktop models, floor models, beds, or walk-in booths).
This infonnation was obtained by showing subjects pictures of various types of
sunlamps and sunbeds. The study found a nonsignificant elevated risk of malignant
melanoma associated with the use of desktop sunlamps and heavyweight floor-model
23
Dec. 2000
sunbeds and a statistically significant tripled risk associated with use of more than two
types of sunlamps, compared with no use of sunlamps. Increased risk of melanoma also
was associated with first use of sunlamps before 1971 and with sunlamp use at home.
However, the study had insufficient power to detect an association between melanoma
and use of sunlamps in the late 1970s and 1980s, because of insufficient follow-up time
for cases accrued between 1987 and 1989. Walter et al. (1990) also found a greater risk
for exposures that occurred at home than at commercial sites.
In contrast, Westerdahl et al. (2000) reported a greater risk associated with commercial
than with home use of sunbeds. This was the first population-based case-control study
to accrue cases in the late 1990s; thus, it had greater power to detect the effects of
exposure in the 1980s. Most (80%) of the exposure to sun beds in this study took place
in the 1980s, probably from predominantly UV A-emitting sunbeds. This contrasts with
the exposures reported by Chen et al. (1998), in which only 59 of the subjects (25% of
the exposed subjects) had used sunlamps in a commercial setting after 1970, and in
which the follow-up for exposures that occurred in the 1980s was shorter.
3.2. 1.4 Methodologic concerns
Swerdlow and Weinstock (1998) discussed seven biases or methodologic limitations
present in many of case-control studies listed in Table 3-1; however, many of these
limitations were not specific for exposure to sunlamps or sunbeds, but are inherent to
most retrospective case-control studies. Three of the limitations concerned exposure
assessment: inadequate information on the types of sunlamps used (discussed above),
inadequate classification according to the level of exposure, and misclassification of
exposure through inclusion ofboth medical and cosmetic exposure. The fourth
limitation related to the limited power to detect an association because only a small
proportion of subjects had ever used sunlamps or sun beds or had used tanning devices
at an exposure level sufficient for an effect to be detected. Both exposure
misclassification and limited power would diminish the strength of an association with
melanoma. These issues were addressed by ranking of the studies in the four tiers
described above. The studies in Tier 4, which largely overcame these limitations,
showed positive associations of melanoma with exposure.
The other three biases, confounding due to sun exposure, recall bias, and publication
bias, may induce an artifactual association. Regarding confounding, several studies
(Autier et al. 1994, Swerdlow et al. 1988, Westerdahl et al. 1994) reported an
association between exposure to sunlamps or sunbeds and increased risk of melanoma
after adjusting for recreational sun exposure or indicators of sun exposure (raised nevi
and number of sunburns) (Westerdahl et al. 2000). However, the control of recreational
sun exposure may not be appropriate in this situation, because UVR presumably is the
relevant exposure underlying both exposures, solar radiation and sunlamps or sunbeds;
thus, the two exposures may have an additive effect on the risk of melanoma. Thus,
controlling for sun exposure may lead to an underestimation of the effect of exposure to
sunlamps or sunbeds. All studies reporting a positive association between sunlamp or
sunbed exposure and malignant melanoma adjusted for phenotypic indicators of sun
sensitivity.
24
Dec. 2000
Several studies (Autier et al. 1994, Walter et al. 1990, Westerdahl et al. 2000, 1994)
used measures to control for recall bias. Autier et al. (1994) focused on recall bias in
the training of the interviewers; neither interviewers nor subjects were informed of the
study's objective. Westerdahl et al. (1994) used a questionnaire with many variables
and stated that at the time of the interview ( 1988 to 1990), the population was unaware
of the relationship between sunlamps or sunbeds and malignant melanoma. Westerdahl
et al. (2000) used identical procedures of data collection for cases and controls and
collected information from melanoma patients shortly after diagnosis. Walter et al.
( 1990) reported that rates of sunbed use were similar in patients interviewed before and
after the diagnosis of melanoma, suggesting that recall bias was not important. The fact
that studies with negative results and methodological limitations (small sample sizes
and low exposure) were published suggests that publication bias probably was not a
major factor.
3.2.2 Medically related UVR exposure
As discussed in Section 2, UVR has been used to treat psoriasis, alone or in
combination with chemical agents; e.g., PUV A (UV A plus methoxsalen), UVB, or
UVB plus coal tar. Most human studies evaluating health effects of medically related
exposure to UVR have been compromised by exposure of subjects to another potential
carcinogen; coal tar, for instance, is a known to be human carcinogen (NTP 2000).
3.2.2.1 /ARC and NTP evaluations
Methoxsalen (methoxypsoralen) with UV A therapy (PUVA) is known to be a human
carcinogen based on sufficient evidence in humans (IARC 1982, 1987, NTP 2000).
Squamous-cell carcinoma was reported in patients treated with PUV A therapy. UVB
therapy, either alone or in combination with other treatments, has not previously been
reviewed for carcinogenic risk by either the IARC or the NTP.
3.2.2.2 Recent epidemiologic studies
The studies with PUV A provided only limited infonnation concerning the
carcinogenicity ofUVR exposure, because of the co-exposure with psoralens, which
may be photocarcinogens (see Section 5 for discussion of the genotoxicity ofPUVA
therapy). A wealth of literature has been published on PUV A treatment and cancer. In a
review of the literature published after 1992, Studniberg and Weller (1993) concluded
that a long-term multicenter prospective study following psoriasis patients treated with
PUV A (Stem et al. 1979, 1984, Stem and Lange 1988) provided evidence that PUV A
was an independent carcinogen in humans, capable of initiating and promoting the
formation of squamous-cell carcinoma. These fmdings were supported by several long-
tenn retrospective studies (Forman et al. 1989, Lindelof et al. 1991, Bruynzeel et al.
1991 ). At the time of the review, the relationship of basal-cell carcinoma to PUV A
alone was not well established.
Since this review, Stern et al. (1997, 1998) reported the results of a 15-year follow-up
of the PUVA cohort with respect to both non-melanoma skin cancer and melanoma.
Risk of basal-cell carcinoma was elevated only in psoriasis patients exposed to high
levels of PUV A (Stern et a/. 1998). An excess risk of malignant melanoma, relative to
25
Dec. 2000
the age- and sex-specific rates for the U.S. population, also was reported (Stern et al.
1997). This risk did not become evident until the period from 1991 to 1996, suggesting
that a long follow-up time was needed to detect melanoma. The risk of melanoma was
higher among patients receiving at least 250 PUV A treatments. This study was
criticized by Whitmore and Morison (1997) for (1) inaccurate statistics, as the use of
cancer statistics from the National Cancer Institute's Surveillance, Epidemiology, and
End Results data may underestimate the true incidence of melanoma, (2) confounding
variables, as the cohort study lacked a control group of patients with psoriasis who
never received PUV A, and (3) surveillance bias, as cohort members were aware that
they were being followed for adverse effects ofPUV A therapy.
A Swedish prospective study that followed a cohort ofPUV A-treated patients did not
find an increased risk of malignant melanoma (Lindelof et al. 1999). However, the
treatment regimen was different in this study; one-fifth of the cohort received PUV A
bath therapy, in which the UV A dose is 15 to 20 times lower than in oral therapy.
Moreover, both the mean and cumulative UV A doses for PUV A treatment generally are
much lower in Europe than in the United States (Studniberg and Weller 1993).
Pasker-de Jong et at. (1999) conducted a systematic review of nine human studies
evaluating the relationship between UVB psoriasis treatment and non-melanoma skin
cancer. All studies followed cohorts of psoriasis patients, some of whom had received
UVB treatment. Three studies evaluating the effects of UVB therapy without coal tar
found no excess of cancer in UVB-exposed individuals (Larko and Swanback 1982,
Bhate et al. 1993, Maier et al. 1996). Two studies evaluated the effect of exposure to
UVB and coal tar in the same PUV A cohort used in Stern et al. ( 1997, 1998) discussed
above. Elevated risks of genital SCC (RR = 4.6 [Stern 1990]) and non-melanoma skin
cancer (OR = 2.4, 95% Cl = 2.2, 10.0 [Stem et at. 1980]) were reported in patients
exposed to over 300 treatments with UVB and/or over 90 months of treatment with coal
tar compared with members of the PUV A cohort without high exposure to UVB or coal
tar. However, a later follow-up of the cohort no longer found a significant association
between non-melanoma skin cancer and long-term exposure to UVB or coal tar after
controlling for PUVA exposure and other confounders (Stern and Laird 1994).
Pittelkow et al. (1981) also did not find an increase in the cumulative incidence of non-
melanoma skin cancer in psoriasis patients treated with UVB and coal tar, compared
with the age-specific incidence of non-melanoma skin cancer for that geographical
area.
A cohort study (Hannuksela-Svahn et al. 2000) published after the Pasker-de Jong et al.
(1999) review, studied a population of psoriasis patients diagnosed between 1973 and
1984 and treated with different UVR therapies (30 cases and 13 7 controls for
squamous-cell carcinoma and 19 cases and 110 controls for non-Hodgkin 's lymphoma).
The mean length of follow-up was 14 years. Because increased incidences of
squamous-cell carcinoma (30), non-Hodgkin's lymphoma (19), and laryngeal cancer
(11) were observed for the cohort as a whole, a nested case-control study was used to
evaluate the role of prior exposures to different psoriasis treatments. An elevated but
nonsignificant risk of squamous-cell carcinoma (RR = 1.6, 95% CI = 0.4 to 6.4) from
prior UVB treahnent was reported. Risk ofnon-Hodgkin's lymphoma was not
26
Dec. 2000
increased by any treatment, including UVB, and results for laryngeal cancer were not
reported.
3.2.3 Occupationally related UVR exposure
3.2.3. 1 /ARC evaluation
The IARC commented that epidemiological studies evaluating effects of exposure to
artificial UVR had not measured actual doses of UVR nor considered the emission
spectrum, and that subjects were exposed to sources of varying intensity and emission
spectra. The lARC reviewed eight case-control studies evaluating the relationship
between fluorescent lighting and melanoma. Most ofthese studies provided limited
information. Two studies reported an increased risk of melanoma from exposure to
fluorescent lamps (Beral et al. 1982, Elwood et al. 1986), but the measurement of
exposure was crude in one of the studies (Beral et al. 1982) and the effects were
inconsistent depending on the method of ascertainment of information in the other
study (Elwood et al. 1986). Exposure to UVR from arc welding and other occupational
sources was not associated with malignant melanoma. However, exposure to arc
welding torches increased the risk for melanoma of the eye (OR = 8.3, 90% CI = 2.5 to
27.10) in a Canadian study (Siemiatycki et al. 1991), though not in an U.S. study
(Seddon et al. 1990).
3.2.3.2 Recent epidemiologic studies
Studies evaluating the effects of occupational UVR exposure and cancer published
since the lARC evaluation include three analytic studies and one case report. The case
report was of five cases of non-melanoma skin cancer in welders, reported from the
Skin Cancer Clinic in Bedford, England (Currie and Monk 2000).
Bajdik et al. ( 1996) evaluated the risk of non-melanoma skin cancer from nonsolar
radiation in a population-based case-control study of226 basal-cell carcinoma and 180
squamous-cell carcinoma cases and 406 age-matched controls. Subjects were asked
about job history and exposure to fluorescent lighting, sunlamps, welding torches,
mercury-vapor lamps, ultraviolet or black lights, printing or photocopying lights, UV
lamp treatments, or horticultural growth-inducing lights. Slightly elevated but
nonsignificant risks of basal-cell carcinoma were observed for exposure to sunlamps,
mercury-vapor lamps, and horticultural growth-inducing lights, and similar
nonsignificant elevated risks of squamous-cell carcinoma were observed for exposure
to sunlamps and welding torches. However, the authors noted that the statistical power
was low because of the limited number of exposed individuals (except for exposure to
fluorescent lighting or welding torches).
Holly et al. (1996) reported that welding exposure was a risk factor for uveal
(intraocular) melanoma (OR = 2.2, 95% CI = 1.3 to 3.5) in a population-based case-
control study (221 patients and 447 controls) in the western United States. Other
occupational groups that were also exposed to UVR also had an increased risk of uveal
melanoma (OR = 3.0, 95% CI = 1.2 to 7.8) for sailors, ship officers or fisherman and
(OR= 1.2, 95% CI = 0.74 to 1.9) for agricultural occupations. For these occupations,
the source of UVR exposure was sunlight.
27
Dec. 2000
The relationship between fluorescent light exposure and cutaneous malignant
melanoma was evaluated in a population-based case-control study (583 cases and 608
controls) in Ontario, Canada (Walter et al. 1992). In males, significantly increased risk
of melanoma was associated with cumulative years of occupational exposure (with an
exposure-response relationship) and with various indices of exposure to domestic
fluorescent light. In females, results were inconsistent. The observed increased risk
remained after adjustment for other major risk factors, including time spent outdoors
for occupation.
3.3 DNA repair
Xeroderma pigmentosum is a rare autosomal recessive genetic disease characterized by
an excision repair defect, as observed in cultured skin fibroblasts damaged by UVR.
Patients display cellular and clinical hypersensitivity to UVR and have a> 200-fold
excess of sunlight-related skin cancer (IARC 1992, Cleaver and Kraemer 1989, cited in
Wei et al. 1994). Xeroderma pigmentosum is a rare disease, resulti11g in exceptionally
low DNA repair capacity.
DNA repair capacity may also vary in the general population and thus may be a
hereditary susceptibility factor for skin cancer. Wei et al. (1994, 1995) provided evidence
that DNA repair capacity may be the underlying cause of sunlight-induced basal-cell
carcinoma resulting from certain known risk factors (susceptible skin type, poor tanning
ability, history of multiple sunburns, frequent sunbathing, exposure to chemicals, or
multiple medical irradiations) (see Section 5 for discussion of DNA repair assays).
3.4 Discussion
The studies reviewed by the IARC (1992) and the substantial number of studies
published since provide strong evidence that exposure to solar radiation causes malignant
melanoma and basal- and squamous-cell carcinoma of the skin. Terrestrial sunlight is a
mixture of UVR, visible, and infrared light, so it can be deduced that one or more of these
components is carcinogenic. Studies using artificial sources of UVR, mainly sunlamps
and sun beds, suggest that UVR is the carcinogenic component of solar radiation. Positive
associations between exposure and skin cancer have been reported both for early models
of sunlamps emitting high percentages of UVB and for later models of sun beds emitting
mainlyUVA.
The epidemiological literature, while extensive, does not provide a basis for subdividing
the effects of solar radiation or UVR from artificial sources into components attributable
specifically to UV A, UVB, or UVC. However, some information with respect to the
specific effects of UV A, UVB, and UVC can be inferred from the results of studies in
which the predominant exposure was to a specific UVR component.
3.4.1 UVA
Evidence for carcinogenic effects of UV A exposure comes from studies on solar
radiation and melanoma, sunscreen usage, sunlamps and sunbeds, and PUV A treatment.
It has been suggested that UV A is important in the development of melanoma. Solar
radiation contains varying amounts ofUV A and UVB, depending on latitude. In
28
Dec. 2000
descriptive epidemiological studies of worldwide incidence of cutaneous malignant
melanoma, cancer incidence correlated better with latitude changes in UVA intensity than
latitude changes in UVB intensity; correlations of latitude with melanoma incidence and
correlations of latitude with UV A intensity had similar slopes (Moan et al. 1999).
Several, but not all, studies showed sunscreen use to be a risk factor for melanoma,
possibly as a result of longer exposure to sunlight (because of protection from sunburn)
or inadequate blocking ofUVA radiation (early sunscreens blocked mainly UVB
radiation) (Gasparro 2000). Westerdahl et al. (2000) reported an association between
malignant melanoma and exposure to sunbeds, the majority of which probably emitted
mainly UV A (0.1% to 2.1% UVB). PUV A therapy is a known human carcinogen. Most
studies showed an association between PUV A therapy and non-melanoma skin cancer,
and a recent study reported an association with melanoma (Stem et al. 1997). However,
these studies are compromised by co-exposure to psoralens and the use of psoriasis
patients as study populations.
3.4.2 UVB
Individuals are exposed to UVB radiation from the sun and from artificial sources, such
as sunlamps and sunbeds, medical therapies, fluorescent lighting, and welding. The
strongest evidence for UVB carcinogenicity comes from the importance of the UVB
component to the association of cancer with solar radiation and exposure to sunlamps and
sunbeds. There is a strong inverse relationship between latitude and both the incidence of
nonmelanocytic skin cancer and measured or estimated ambient UVR. The yearly
average intensity of all wavelengths in sunlight increases with decreasing latitude;
however, the greatest increase is in UVB exposure, because the stratospheric ozone layer
is thicker at higher latitudes and absorbs much more UVB than UV A. In fact, several
studies have used estimated solar UVB as an indicator of exposure to solar radiation (e.g.,
Hartge et al. 1996, McMichael and Giles 1996).
In one study, exposure to sunlamps used in the early 1970s, which produced significant
amounts ofUVB (22% to 40%), was associated with malignant melanoma (Chen et al.
1998). Other studies using artificial sources ofUVB radiation gave mainly negative or
inconsistent results or were limited by confounding with exposure to other potential
carcinogens. UVB therapy does not appear to be a risk factor for psoriasis patients.
Fluorescent lighting devices generate light by emitting UV radiation, which strikes a
phosphor on the interior lining of the tube. The glass tubes absorb most of the radiation
below 290 nm, but longer wavelengths, particularly above 297 nm, are more readily
transmitted. There is some evidence that fluorescent lighting may increase the risk of skin
cancer (Walter et al. 1992); however, results of earlier studies were inconsistent (IARC
1992). Other occupational exposures also appear to involve mainly UVB-emitting
devices. Welding used to join metal components produces ultraviolet light (250 to 297
nm). Some evidence suggests that welding may increase the risk ofuveal melanoma;
however, confinnatory studies are needed, and the effects of welding fumes are unknown
(Holly et at. 1996).
29
Dec. 2000
3.4.3 uvc
The effects ofUVC are harder to evaluate. Solar UVC is filtered by the ozone layer, and
few studies have examined the effects of exposure to artificial sources ofUVC. Desktop
sunlamps used before the 1970s may have emitted UVC (see Section 2), as may welding
torches.
3.5 Summary
Epidemiologic studies have clearly demonstrated that exposure to broad-spectrum UVR
increases both melanocytic and nonmelanocytic cancer. Studies of solar radiation,
artificial devices emitting broad-spectrum UVR, and devices emitting predominantly
UV A or UVB all have contributed to this conclusion. Both UV A and UVB components
of solar radiation appear to be important, and they may contribute differently to risks of
different types of cancer. Some evidence suggests that UV A or UVB alone also may
increase the risk of skin cancer or melanoma of the eye, but it is not conclusive. Little
information from human studies was available to evaluate UVC.
30
Dec. 2000
Table 3-1. Epidemiologic studies of the relationship between cutaneous malignant melanoma and exposure to sunlamps or
sunbeds (listed in chronological order by publication date)
Reference
Study location
Exposure
Years cases Study
Percent exposed Comments
accrued design Population (case/controls) Effects Ranking tier
Klepp and hospital-based cases: 89 melanoma exposure to UV lamps no difference between cases sunlamps not a major focus;
Magnus 1979 case-control
controls: 227 hospital
(not clear whether and controls no subgroup analysis
Norway controls with malignant
sunbeds or sunlamps)
poor exposure assessment,
1974-1975
lymphoma, testicular
assessed by
little exposure information,
cancer, or bone or soft-
questionnaire
and sources of exposure not
tissue sarcoma use of artificial light very clear
The study was restricted
rare
limited power due to rare use
to 78 cases and 131 % exposed not given of lamps and small sample
controls from Oslo and size
surrounding areas because
possible selection bias
of differences in
because controls were cancer
geographical distribution
patients
between cases and
controls.
Tier 1
Adam et a/. 1981 case-control cases: 169 women with sunlamp used assessed use of sunlamps was low, but sunlamps not a major focus;
England
malignant melanoma by postal questionnaire; significantly higher in cases no subgroup analysis
controls: 503 women
other information than controls (P < 0.05)
little exposure information
1971- 1976
matched by age and
assessed from medical
calculated (not reported) crude
records
limited power due to small
marital status randomly OR = 2.9
sample size and low
selected from general 8/3
exposure rate
practitioners
Tier2
response to questionnaire:
Ill cases and 342
controls
31
Dec. 2000
Reference
Study location
Exposure
Years cases Study
Gallagher eta/. population- cases: 595 newly exposure to sunlamps no association between sunlamps not a major focus;
1986 based case- confirmed including frequency and sunlamp use and melanoma subgroup analysis by sex and
Canada
control
controls: 595 age- and
duration; assessed by (X
2
, NS) body site, but nos. used for
sex-matched controls
interview with a
no association by gender or
risk estimates not reported,
1979-1981
from insurance
standardized
body site
and OR not calculated
subscribers
questionnaire
excluded lentigo maligna
exposure characterized cases
as moderate and
Tier2
relatively limited
% exposed not given
Holman eta/. population- cases: 511 preinvasive or exposure to sun and crude OR= l.l (0.6-1.8) sunlamps not a major focus;
1986 based case- invasive melanoma sunlamps assessed by no subgroup analysis because
Australia
control
controls: 511 sex- and
structured questionnaire of small no. of exposed
1980- 1982
age-matched controls
administered by nurse subjects
from electoral rolls or
interviewers
Tier2
student rolls of public 9 overall
schools
Elwood et al. hospital-based cases: 83 malignant home exposure to no association with risk sunlamps not a major focus;
1986 case-control melanoma identified from f1uorescent lighting and
calculated (not reported) crude
no subgroup analysis
England
pathology services of2 the use of sunlamps
OR = 1.3 little information on
1981-1984
hospitals
average exposure = 2.3 h assessment of association; no
controls: 83 age-, sex-,
18/ 14
risk estimate given
and residence-matched
hospital controls (in or
sample size and short
out)
duration of exposure
lentigo maligna melanoma
excluded
Tier 3
32
Dec. 2000
Reference
Study location
Exposure
Years cases Study
Holly et al. 1987 case-control cases: 121 consecutive exposure to use of melanoma patients similar to sunlamp use not a major
u.s.
melanoma patients at tanning salon assessed controls with respect to use of focus; no subgroup analysis;
clinic by questionnaire tanning salons no information on risk
1984-1985
estimate (none given)
controls: 139 sex- and %exposed not given
age-matched patients at very little exposure
same clinic (not information.
dermatology)
lentigo maligna melanoma
excluded
small sample size
Tier 1
Zanetti et at. 1988 population- cases: 208 histologically exposure to UV A lamps crude OR = 1.5 sunlamp use not a major
Italy
based case- confirmed malignant assessed by
adjusted OR = 0.9 (0.4-2.0)
focus; no subgroup analysis
control melanoma from the questionnaire
adjusted for age, hair color,
1984- 1986
regional tumor registry
7/ 5 skin reaction to the sun,
controls: 416 from sunburn in childhood, and
National Social Service education
Registry
Tier 2
33
Dec. 2000
Reference
Study location
Exposure
Years cases Study
Osterlind et al. population- cases: 474 melanoma exposure to sunlamps sunbeds: crude OR= 0.7 (0.5- risk estimate for sunbeds
1988 based case-
and sunbeds, including 1.0) given, but analysis by dose
Denmark
control
age-matched randomly
no. ofuses (< or > 10),
sunlamps: use not related to
not stated
selected from population
assessed with structured
risk and risk not related to no. sunlamp use evaluated by
1982-1985
registrar
questionnaire at home
of uses dose, but information on
interview
analysis and risk estimates
sunbeds: 14/ 18 not given
50% used < 10 times somewhat limited power for
sunlamps: 45/42
sunbeds due to low
percentage of individuals
exposed in the higher
exposure group (-7o/>-8%
used sun beds > 10 times)
excluded lentigo maligna
melanoma
Tier 3
Swerdlow et al. hospital-based cases: 180 malignant exposure to UV lamps ever use OR = 2.9 (1.3- 6.4) sunlamps a major focus
1988 case-control melanoma from university and sunbeds, including
exposure response for detailed exposure
Scotland
stratum depts. of dem1atology and ever use, duration, age at
increasing duration (P < 0.05) information and subtype
1979-1984
plastic surgery first exposure, and when
greater risk for first use before
analyses
exposure occurred ( 5 yr.
controls: 197 hospital in-
before presentation)
age 30 (OR= 3.8) adj. for nevi, skin type, hair
and out-patients with
various nonmalignant
assessed by interview greater risk for use > 5 years
and eye color, and sun
diseases, stratum-matched 21 /8
before presentation (OR= 9.1)
exposure
for age, sex, and city elevated risk for cancer on
cells small after stratification
where treated legs and trunk
by duration or exposure
characteristics
elevated risk for superficial
Tier4
spreading melanoma and
nodular melanoma
34
Dec. 2000
Reference
Study location
Exposure
Years cases Study
MacKie et a!. hospital-based cases: 280 (181 women exposure to artificial artificial sources ofUV: sunlamps not a major focus;
1989 case-control and 99 men) identified sources of UV (classed
M: cntde OR= 2.6 (0.9-
no subgroup analysis
Scotland
from registry as modem sunbeds or
7.3), conditional regression
1989-1987
older sunlamps);
adj. OR= 1.3 (0.2- 7.9) method repotted but details
sex- matched hospital
exposure to sunbeds 1 or
F: crude OR= 1.5
relating to matching not
patients with non-
2 times/wk for at least 12
described
dennatological illness
wk sun beds:
artificial UV sources: calculated ORs (by 2 x 2
no. of exposed cases
12/3
table) from nos. of exposed
adjusted ORs for nevi (total,
cases and controls
sunbed: atypical), freckling tendency,
M: crude OR = 8.6
skin type, severe sunburn,
M: 8/ 1
F: crude OR = 3.8
F: 10/3
tropical residence
Tier 2
Beitner eta!. 1990 population- cases: 523 incident exposure in solariums no increased risk with frequent sunlamps not a major focus;
Sweden
based case- malignant melanoma assessed by exposure to solariums no subgroup analysis
control
questionnaire
no information given with
1978- 1983
sex-matched controls % exposed not given respect to risk estimate or no.
selected from population of individuals exposed
registry
poor exposure assessment
Tier 1
35
Dec. 2000
Reference
Study location
Exposure
Years cases Study
Walter et al. 1990 population- cases: 583 histologically exposure, including ever ever use (crude OR): detailed exposure
Ontario, Canada
based case- confirmed use, year began, months
M: 1.9 (1.2-3.0)
information and subgroup
control
controls: 608 randomly
of use, no. uses/wk,
F: 1.5 (0.2-2.1)
analysis
1984-1986
length of exposure,
selected from tax rolls,
adj. did not change OR for
adj. for age, nevus density,
matched for sex, age, and
location, and parts of
skin color, skin reaction to
body exposed, assessed
either sex
municipality sun, and socioeconomic
by interview with exposure-response for
status; adjusted analyses gave
questionnaire cumulative min. ofuse
same effect as unadjusted
M: 24/ 14
(P < 0.01)
analysis
F: 28/21
slightly greater risk for face,
recreational sun exposure a
head, or neck than trunk; little
possible confounder
risk for legs; greater risk for
Tier 4
trunk in M than F
ORs for histol. type:
lentigo maligna and
Hutchison's melanotic
freckle: M = 2.4,
F = 3. 1
superficial spreading and in
situ: M = 1.90,
F = 1.4
nodular: M = 1.7,
F = 1.4
greater risk for home use
greater risk for first use before
age30
greater risk for 5 yr. since last
use
36
Dec. 2000
Reference
Study location
Exposure
Years cases Study
Dunn-Lane 1993 hospital-based cases: 100 consecutive exposure to sunlamps 17% cases and 15% controls sunbeds not a major focus;
Ireland
case-control patients at 7 hospitals and sunbeds assessed by used sunbeds; duration of use some indication that duration
controls: I 00 sex- and
standard pre-coded similar was considered, but risk
1985-1986
age-matched orthopedic
questionnaire
calculated (not reported) clUde
estimates not given and
hospital controls with 17/ 15 OR= 1.2
details on duration of use not
limb injuries excluding
described
sports injuries little exposure information
sample size
Tier 2
Garbe et at. 1993 case-control cases: 1,079 melanoma exposure to sunbeds adj. OR= 1.5 (0.9- 2.4) for sunlamps not a major focus;
Germany
patients from Central assessed by 885 cases and 705 controls no subgroup analysis
Malignant Melanoma questionnaire and with known information
little exposure information
1983-1990
Registry interview
controls: 778 outpatients 817
low percentages with
exposure (7. 7, 7 .1)
from dermatology clinics
adj. for no. of nevi, hair
excluding patients with
previous UV treatment for
color, skin type, age, and
skin disorders, skin
participating center
cancer, or nevi Tier2
37
Dec. 2000
Reference
Study location
Exposure
Years cases Study
Autier et al. 1994 population- cases: 420 consecutive exposure assessed with ever use: crude OR = 1.0 (0.7- detailed exposure
Europe
based case- malignant melanoma respect to ever use, 1.3) information and subgroup
control patients location of exposure,
for tanning purposes: crude
analysis; adj. for age, sex,
1991
type of machine, hair color, and no. of
controls: 447 controls in OR for sunlamps
the same municipality,
duration of exposure
= 1.8 (1.0-3.3)
holidays wk spent in sunny
randomly chosen by a
session, year first used,
OR for 10+ h exposure for
resort; overall OR given only
quota sampling method,
no. of sessions, reason as crude
with no skin cancer
for use by interview and
tanning purposes:
insufficient follow-up for
history
questionnaire first exposure before 1980:
exposure occurring after
26/27
2.1 (0.8-5.34)
1980
experience of sunburn:
Tier 4
7.4 (1.7- 32.3)
Westerdahl eta/. population- cases: 400 patients from exposure to sunbeds or adj . OR: detailed exposure
1994 based case- South Swedish Health sunlamps, including ever
ever use: 1.3 (0.9-1.8)
information (with respect to
Sweden
control Care Region use and how often,
>10 uses: 1.8 (1.0-3.2)
dose); subgroup analysis
1988- 1990
controls: 640 randomly
assessed by
exposure response (P < 0.06)
adj. for history of sunburn,
selected from population
comprehensive
hair color, raised nevi, and
registry matched by sex,
questionnaire greater risk age < 30 yr:
history of frequent summer
age, and parish 30125 ever use: OR = 2.7 (0.7-9.8) sunbathing
use > 10 times: OR= 7.7 (1-
small cell numbers after
64)
stratifying by no. of uses and
for use> 10 times vs. none: age
greater risk for trunk (OR= Tier4
4.2) than head or extremities
(OR= 1.1)
38
Dec. 2000
Reference
Study location
Exposure
Years cases Study
Holly et al. 1995 population- cases: 452 women with ever use sunlamps OR for exposure category: some subgroup analysis on
u.s.
based case- melanoma (medical or cosmetic);
lower: 0.9
no. of uses but no definition
control
controls: 930 age-matched
how many times in lite
higher: 1.1
of lower and higher
1981-1986
(excluding last 3 yr) categories; subgroup analysis
women in the same
no difference or elevation in
counties identified by 37/38
risk due different histologic
calculated for histologic type
random-digit dialing
types of melanoma
both medical and cosmetic
exposure included
Tier 3
Chen el a/. 1998 population- cases: 624 newly sunlamp use assessed by OR for ever use: sunlamps major focus of
u.s.
based case- diagnosed malignant nurse-interviewers with a
crude = 1.3 (1.0-1.7)
study; detailed exposure
control melanoma structured questionnaire
adj. = 1.13 (0.8- 1.5)
assessment, including
1115/87
and classified by type, attempt to define type of
no relationship between risk
age-matched community
year first used, and lamp used; detailed analysis
location, as well as
and total no. of uses
controls selected by adj. for phenotype index (hair
random-digit dialing
information on potential age at first use< 25 yr:
and eye color, skin type or
confounders adj . OR = 1.4 (0.9-2.1)
tanning ability) and
23/19 no signif. increased risk for recreational sun exposure
any type of sunlamp
insufficient follow-up time
used > 2 types of lamp: for later exposures
adj. OR= 3.5 (1.3- 9)
Tier4
adj. OR for location:
home: 1.4 (1-2)
commercial 0.8 (0.5-1.3)
used before 1970:
adj. OR = 1.3 (0.8-2.1)
39
Dec. 2000
Reference
Study location
Exposure
Years cases Study
Walter et al. 1999 population- cases: 583 newly exposure to sunlamp use, OR for ever use: good exposure assessment;
Ontario, Canada
based case- diagnosed (1984-1986), including year, duration,
clUde = 1..6 (1.2-2.2)
detailed analysis
control histologically confirmed location, and part of
adj. = 1.5 (1.2-2.1) adj. for sex, age, skin sun
1984-1986
body exposed, assessed
controls: 608 selected
no difference in risk by body
response
reanalysis of
from property tax
by in-person interview
Walter et al. 1990 with a structured
location potential confounder is
assessment and chosen to
recreational sun exposure
match the case
questionnaire risk for lentigo maligna
distributions with respect 26/ 18
highest: OR= 2.8 (1.4- 5.3); Tier4
to age, sex, and
risk signif. for superficial
municipality of residence
spreading and in situ: OR=
1.5 (1.1-2.0); elevated for all
types
no difference in risk by skin
reaction, but signif elevated
only for burners (larger
sample size)
no difference by age at
diagnosis
40
Dec. 2000
Reference
Study locati on
Exposure
Years cases Study
Westcrdahl eta/. population- cases: 571 malignant exposure to sunbcds, OR for regular use: detailed exposure
2000 based case- melanoma from the including ever use,
crude = 1.6 (1.1-2.4)
information and analysis
Sweden
control population-based regular use, exposure
adj. = 1.8 (1.2-2.7) adj. for hair color, skin type,
1995- 1997
Regional Tumor Registry time, no. of times/wk,
use at age < 35: OR = 2.3
rai sed nevi, no. of sun
in South Swedish Health no. of wklyr, location,
Care Region season, age at first and
use at age> 35: OR= 1.6
exposures
controls: 913 selected by
last use, assessed by F: OR= 2. 1
not adj . for recreational sun
random sampli ng and
comprehensive M: OR = 1.3
exposure; however,
questionnaire
controll ing for nevi may take
matched by sex, age, and darker hai r: OR = 2.3
this into account, since they
parish 44/4 1 light hair: OR = 1.5
are related to both sun
cotmnercial: OR = 2.2 exposure and skin type
home: OR = 1.5
Tier4
risk greater for use in winter;
small sample size in summer,
OR < 1
greater risk indi viduals aged
< 36
ri sk greatest for lesions of
extremities, then trunk; no risk
for face; in M, no risk for
upper extremities
exposure response up to 250
total uses or IS uses/yr, after
which the ORs decreased
test for trend, times/yr, 0.06,
total uses, 0.26
41
Dec. 2000
Table 3-2. Recent epidemiologic studies of the relationship between cancer and medically related UV exposure
UVR
Reference treatment Adjustments
Study Type of Study Potential
location cancer design Population Exposure Effects confounders
Stem et al. PUVA cohort 1,380 psoriasis patients psoralen 0.4- squamous-cell carcinoma: overall analysis not
1998
basal-cell and multicenter
enrolled in 16-university 0.6 mg!kg orally,
overall RR= 17.6 (15.6-
adj. for other
U.S. squamous-
prospective
center study followed by UV A,
19.8); dose-related;
therapies;
cell from 1/ 1/75 to 10/ 1/76
usual dose 8-15 J/cm
2
substantial risk at all doses
multivariate
carcinoma
followed-up until
4 PUVAdose
high dose adj. for other
analysis for
categories, based on
different dose
9/ 1/97 therapies: RR = 8.6 (4.9-
groups adj. for
no. oftreatments:
15.2)
65% males
< 100, 100-159 160-
therapies as well as
1,042 basal-cell 336, > 337
basal-cell carcinoma: age, sex, area of
1,422 squamous-cell interview,
overall RR = 4.1 (3. 7- 4.6);
residence, and
anatomic site
mean age 44
documentation of
dose-related; substantial risk
PUV A therapy and
only at highest dose no psoriasis
PUV A Follow-up Study
other treatments for high dose adj. for other
controls;
psoriasis therapies: RR=4.7 (3.1- 7.3)
surveillance bias
reference group: PUVA < 100
treatments (low dose)
Stern et al. PUVA cohort follow-up until 2/29/96 two exposure groups II melanoma SEER incidence
1997
melanoma PUVA controls: U.S. population
based on no. of
RR = 2.3 (1.1-4. 1)
rates used for
u.s. Follow-up (SEER)
treatments: expected
1975- 1990: 4 melanoma
Study low< 250 no psoriasis
(above) high> 250
RR = 1.1 (0.3- 2.9)
controls;
1990-1996: 7 melanoma surveillance bias
RR = 5.4 (2.2- 11 .0)
42
Dec. 2000
UVR
Reference treatment Adjustments
Study Type of Study Potential
location cancer design Population Exposure Effects confounders
Lindelof et al. PUVA cohort 4,799 patients treated information obtained entire cohort, RR: other therapies
1999
melanoma multicenter
with PUV A at I I centers from patient' s records
melanoma: no psonas1s
Sweden 64% psoriasis patients
at each center.
controls
prospective M: 1.1 (0.5- 2.2, n = 8)
mean follow-up:
77% received oral
F: 1.1 (0.4-2.3, n = 7)
surveillance bias
PUVA.
M 15.9, F 16.2
squamous-cell carcinoma:
subcohort of I ,867
UV A dose varied by
M: 5.6 (4.4-7. 1, n = 68)
followed 15- 21 yr
disease; average dose
for psoriasis patients
F: 3.6 (2.1-5.8, n = 17)
400-600 J/cm
2
subcohort, RR:
45 patients received > no excess melanoma
400 treatments
squamous-cell carcinoma:
537 patients received
M: 8.1 (6.1- 10.6)
> 1,000 J/cm
2
F: 6.4 (3.3-11.2)
Hannuksela- UVB cohort 5,687 psoriasis patients exposure assessed RR for UVB treatment: other therapies
Svahn et al.
squamous- nested case-
from 1973- 1984 from patients' files
squamous-cell carcinoma: small number of
2000
cell control PUV A Finnish Cancer percent exposed 1.6 (0.4-6.4) cases in nested
Finland carcinoma Registry (cases/controls):
non-Hodgkin's lymphoma:
case-control study
non- follow-up until squamous-cell 0.1 (0.0-0.8)
Hodgkin's 1/31/ 1995 carcinoma: 70/46
lymphoma
nested study: 67 cases, non-Hodgkin's
199 age- and sex- lymphoma: 16/47
matched controls chosen
laryngeal cancer:
from cohort using
55/38
density sampling
principle
43
Dec. 2000
Table 3-3. Recent epidemiologic studies of the relationship between cancer and occupational UV exposure
Exposure Adjustments
UVexposure
Percent exposed Potential
(cases/ confounders or
Reference Type of cancer Study design Population controls) Effects limitations
Walter et al. fluorescent li ghting population-based same population interview: various occupational exp.: adj. for
1992
melanoma
case-control used for sunbeds exposures to solar and
M: OR = 1.47
socioeconomic
Sweden
(Walter et al. nonsolar UVR,
(0.98-2.14) for exp.
status, sun exp. ;
1999) residential and occup.
I 0 yr ago, dose-
most results not
use of fluorescent lamps,
related for yr of
altered by adj. for
potential confounders
cumulative
risk factors (history
occup. use vali dated with exposure
of sunburn,
employers, residential
socioeconomic
use validated by mail
F: OR = 1.06 (0.76-
status, occupational
1.48) for exp. I 0 yr
sun exp.)
surveys
ago, no dose-
occup. exposed 10 yr response
fluorescent lighting
ago:
domestic exp. :
is ubiquitous
M: 77/70
consistent risk in M
retrospective
F: 56/56
assessment
but not F for
various indices of
exposure
44
Dec. 2000
Exposure Adjustments
UVexposure
Percent exposed Potential
(cases/ confounders or
Reference Type of cancer Study design Population controls) Effects limitations
Bajdik et al. nonsolar UVR population-based cases (1983-1984), Interview: job history, no increased risk of small no. of exposed
1996
basal-cell and
case-control 180 squamous-cell outdoor exposure, from nonsolar UVR individuals except
Alberta, Canada squamous-cell 226 basal-cell from
fluorescent lighting, exposures (e.g., for fluorescent lights
carcinoma Alberta Cancer
other measures ofUV fluorescent lights, and welding
Registry
exposure, confounders welding torches, UV
insufficient follow-
406 aged-match
welding:
lamps)
up
controls from squamous-cell exposure
Alberta health carcinoma: 3l/26 misclassitlcation
insurance plan basal-cell carcinoma:
subscriber list 28/26
other exposures: < I 0
Holly et al. 1996 cccupational population-based 221 cases 1978- interview: potential welding: adj. for age, no. of
Western U.S. uveal melanoma
case-control 1987 confounders and occup.
adj. OR= 2.2 (1.3-
nevi, eye color, and
(intraocular)
447 controls from
history
3.5), no dose
skin response
population within 18/ 11 relationship not adj. for solar
5-yr age group radiation exp.
white males
45
Dec. 2000
46
Dec. 2000
4 Studies of Cancer in Experimental Animals
The IARC reviewed carcinogenicity studies of UVR in rats, mice, and hamsters. The
animals were tested with broad-spectrum UVR or with discrete UV A, UVB, or UVC or a
combination ofthese for carcinogenic effects on the skin and eye (IARC 1992;
Appendix A).
4.1 Broad-spectrum UVR
4.1.1 Rats
The carcinogenic potential ofUVR was recognized from the observation that daily
irradiation of six albino rats with broad-spectrum UVR from a mercury-vapor lamp at a
distance of 18 inches ( 46 em) for one minute, three times a week, resulted in the
formation of skin tumors (papillomas) in one rat (Findlay 1930, cited in IARC 1992). Six
hundred rats were exposed to solar radiation for an average of five hours a day (exposure
around solar noon in the summer was avoided). About 60% of the rats died from
sunstroke. Of the 235 surviving rats, 70% developed tumors on the ears, eyes, nose, tail ,
neck, or paws. Squamous-cell carcinoma and spindle-cell sarcoma were the predominant
tumor types. In complementary experiments, rats exposed to filtered sunlight did not
develop tumors, but all 150 rats exposed to quartz mercury lamps developed tumors
(types and sites unspecified) (Roffo 1934, cited in TARC 1992). Subsequent studies in
which 2,000 white rats were exposed to sunlight yielded similar results (Roffe 1939,
cited in IARC 1992). The IARC Working Group concluded that these studies provided
adequate evidence of carcinogenicity in rats for UVR from sunlight.
In other studies, tumors (papillomas, squamous-cell carcinomas, and occasionally basal-
cell carcinomas) were detected in rats (strain not specified) that were almost continuously
exposed to broad-spectrum UVR from a quartz mercury lamp for 11 months (Putschar
and Holz 1930, cited in IARC 1992). Squamous-cell carcinomas and, rarely, spindle-cell
carcinomas and sarcomas, round-cell carcinomas, and basal-cell carcinomas of the skin
were seen in 20 rats (strain unspecified) exposed for up to 10 months to broad-spectrum
UVR from a mercury-vapor burner at a distance of75 em (Hueper 1942, cited in IARC
1992). Two of seven white rats exposed to UVR from a solar lamp, for two hours a day,
six days a week, for a year or more, developed spindle-cell sarcomas of the eye
(Huldschinsky 1933, cited in IARC 1992). Freeman and Knox (1964, cited in IARC
1992) exposed 66 pigmented and 12 unpigmented rats to UVR from mercury lamps, five
days a week, for one year. The doses per session corresponded to approximately 1 MED.
A total of 98 eye tumors developed. About two-thirds of the tumors were fibrosarcomas,
and the rest were hemangioendotheliomas.
4.1.2 Mice
Daily irradiation of mice with broad-spectrum UVR from a mercury-vapor lamp at a
distance of 18 inches (46 em) for one minute, three t imes a week, resulted in the
formation of skin papillomas within eight months (Findlay 1930, cited in IARC 1992).
An unspecified number of mice exposed to sunlight developed squamous-cell carcinomas
and spindle-cell sarcomas ofthe ear, eyes, paws, tail, and nose (Roffe 1939, cited in
47
Dec. 2000
IARC 1992). The IARC Working Group concluded that these studies provided adequate
evidence of carcinogenicity in mice for UVR from sunlight.
A strain mice were exposed to broad-spectrum UVR at weekly doses of3.6 to 43 x 10
7
ergs/cm
2
(40 to 430 kJ/m
2
); 5% developed skin and eye tumors (spindle-cell sarcomas or
fibrosarcomas, mostly in the cornea) and hemangioendotheliomas (Blum and Lippincott
1942, Lippincott and Blum 1943, Grady et al. 1943, all cited in IARC 1992). Of more
than 600 A strain mice exposed to daily doses of broad-spectrum UVR at 0.32 to 8.6 x
I 0
7
ergs/cm
2
(3 to 86 kJ/ m
2
) from unfiltered medium-pressure mercury arc lamps, over
90% developed skin tumors, mainly on the ears, the only site for which quantitative data
were provided (Blum 1959, cited in IARC 1992).
More recent studies, mostly in mice, provide additional evidence for the carcinogenicity
ofbroad-spectrum UVR (Emmett 1973, Urbach et al. 1974, Epstein 1978, 1985, Kripke
and Sass 1978, WHO 1979, van der Leun 1984, Forbes et al. 1982, Staberg et al. 1983,
Young et al. 1990, Menzies et al. 1991, all cited in IARC 1992).
4.1.3 Hamsters
Hemangioendotheliomas and fibrosarcomas developed in 14 eyes in a group of 19
hamsters (nine pigmented, 10 unpigmented) that were exposed to broad-spectrum UVR
from mercury lamps at 50 em from the skin, five days a week, for one year (Freeman and
Knox 1964, cited in IARC 1992).
4.1.4 Guinea pigs
No tumors were found in the eyes of 17 guinea pigs that were exposed to broad-spectrum
UVR from mercury lamps at 50 em from the skin, five days a week, for one year
(Freeman and Knox 1964, cited in IARC 1992).
4.1.5 Other species
Several researchers have reported skin and eye tumors in domestic animals (cows, goats,
sheep, cats, dogs, horses, and swine) following exposure to sunlight (Emmett 1973, Dorn
et al. 1971, Madewell et al. 1981, Nikula et al. 1992, all cited in IARC 1992).
Monodelphis domestica, a South American opossum that is unusually prone to
photoreactivation, developed actinic keratoses and skin tumors (mostly fibrosarcomas
and squamous-cell carcinomas) following exposure to broad-spectrum UVR from a
Westinghouse FS-40 sunlamp (280 to 400 nm) (Ley 1985, Ley et al. 1987, both cited in
IARC 1992). In a later study, 40 opossums (19 male and 21 female) were e x ~ o s e to
broad-spectrum UVR (FS-40 sunlamps, 280 to 400 nm) at a dose of 250 Ji m , three times
weekly, for 70 weeks, and 29 control opossums (14 male and 15 female) were exposed to
fluorescent lamps emitting primarily visible light (Kusewitt et al. 1991). Both groups of
animals had their backs shaved and were housed under red lights to prevent
photoreactivation. The UVR-exposed opossums developed a variety of hyperplastic and
neoplastic skin lesions on the backs and on a single ear; 20 developed skin tumors (50%),
and 13 (32.5%) had more than one tumor. Tumors included 25 papillomas,
four keratoacanthomas, seven carcinomas in situ, three microinvasive squamous-cell
48
Dec. 2000
carcinomas, two invasive squamous-cell carcinomas, one basal-cell tumor, 10 dennal
spindle-cell tumors, two benign melanomas, and one malignant melanoma. No skin
tumors were observed in the control animals.
4.1.6 Action spectra
The action spectrum for tumor induction in SKHl albino hairless mice was studied from
a database containing information for approximately 1,100 mice treated with 14 different
broadband UVR sources with spectral ranges from mainly 254 nm (from a Philips TUV
germicidal lamp) to > 400 nm (from a Philips Xe3.0 fluorescent lamp) (de Gruijl et al.
1993). UVB at 293 nm was most effective in inducing tumors. However, because of a
lack of data, the action spectrum for longer-wavelength UV A (340 to 400 nm) was much
less well defined. A follow-up study showed that radiation from a custom-made Philips
365-nm source was carcinogenic in hairless mice but was a factor of 10
4
less effective
than UVB at 293 nm. UV A radiation at 365 nm induced the same types of skin tumors as
UVB exposure (mainly squamous-cell carcinomas and precursor lesions) (de Laat et al.
1997).
4.2 Primarily UV A
Numerous experiments have been performed to assess the carcinogenicity of UV A
(reviewed in IARC 1992). A large percentage of these studies, conducted primarily in
hairless mice, did not detect tumors. The IARC Working Group noted that the doses of
radiation (generally in the daily dose range of 160 kJ/m
2
) may have been too small, or
exposure periods may have been too short. In other experiments, tumors clearly were
induced by radiation purported to have been UV A, but the IARC Working Group noted
that efforts to eliminate all UVB were likely insufficient. The studies reviewed below
were considered to have controlled for the presence ofUVB (IARC 1992).
4.2.1 Mice
Groups of 24 male and female SKH1 albino hairless mice were exposed to UV A from a
bank of Philips TL40W/09 fluorescent tubes filtered through a 10-mm glass plate that
strongly absorbed UVB. Animals were exposed 12 hours a day, seven days a week, for
approximately one year. The daily dose was 220 kJ/m
2
. Most animals developed
scratching lesions before they developed skin tumors. All animals had skin tumors, with a
median time to appearance of265 days. Larger lesions were examined microscopically
(selection criteria not disclosed). Of the lesions examined, 60% were classified as
squamous-cell carcinomas, 20% as benign tumors, and 20% as mild cellular and nuclear
atypia. These lesions were similar to those observed in a parallel experiment with UVB,
but the tumor latency period in the UV A-exposed animals was longer (van Weelden et al.
1986, 1988, cited in IARC 1992). However, residual UVB radiation was not believed to
be responsible for the effect because more than 100,000 times the actual amount of
residual UVB present would have been required to induce the observed tumor rate.
Groups of 48 male and female SKHl albino hairless mice were exposed to UV A
(> 340 nm) at 220 kJ/m
2
, for two hours per day, seven days per week, for up to 400 days.
Radiation was generated from mercury metal iodide lamps and passed through liquid
filters. UVB was effectively eliminated from the radiation. Most of the animals
developed skin tumors, and 31 exhibited tumors before any observed scratching. The
49
Dec. 2000
largest tumors (15/20) were examined microscopically and were classified as squamous-
cell carcinomas (Sterenborg and van der Leun 1990, cited in IARC 1992).
In several studies, mice were exposed to UVR sources from which UVB was excluded so
vigorously that shorter-wavelength UV A (315 to 340 run) also was excluded; most of the
animals developed squamous-cell carcinomas. In these experiments, exposure was
mainly to wavelengths in the region of340 to 400 nm (van Weelden et al. 1988, 1990,
Sterenborg and van der Leun 1990, all cited in IARC 1992).
In one of these studies, when female SK.H1 mice were exposed to filtered UVR (340 to
400 nm) at daily doses of360 and 600 kJ/m
2
, 19/44 mice surviving at 18 weeks had skin
tumors (mostly papillomas). At week 100, 22 surviving mice had 40 tumors, many of
which were considered clinically to be squamous-cell carcinomas (it was not clear
whether microscopic examination was used in classifying tumors) (Kligman et al. 1990,
1992, both cited in IARC 1992).
The carcinogenicity of short-wavelength UV A (315 to 340 nm) was investigated in a
study using fluorescent tubes with peak emission near 330 nm and filtering UVB with
glass. Groups of24 male and female SK.Hl mice were exposed to average daily doses of
20 or 56 kJ/m
2
, seven days per week, for 650 days. All mice in the high-dose group had
multiple tumors, initially classified as mainly papillomas, but later as predominantly
squamous-cell carcinomas. In the lower-dose group, three mice had skin tumors, all of
which were papillomas (Keltkens et al. 1991, cited in IARC 1992).
Bech-Thomsen (1997) investigated the carcinogenic effects of various UV A and UVR
sources and their interactions in a series of studies with female C3H/Tif lightly
pigmented hairless mice. In the first study (Bech-Thomsen et al. 1988), 200 mice were
exposed to UVA (341 to 400 nm) from a filtered source at 150 to 200 kJ/m
2
, six days a
week, for four weeks. No skin tumors were observed during the 57-week observation
period (total dose= 4,050 kJ/m
2
). Among mice exposed to broad-spectrum UVR (UV A,
UVB, and< 1% UVC) for 13 and 26 weeks, 35% and 88%, respectively, developed
tumors by 57 weeks. Exposure to UV A (for four weeks at 4,200 KJ/m
2
) before exposure
to broad-spectrum UVR (for 13 or 26 weeks) significantly delayed tumor development.
In subsequent studies, exposure of female C3H/Tif mice to UV A, alone or before or after
exposure to broad-spectrum UVR, increased the incidence of tumors. In one study (Bech-
Thomsen et al. 1991 ), mice were divided into 14 groups of 20 animals each, and three
UVA sources, emitting varying amounts ofUVB, were used either alone or before
irradiation with simulated solar (broad-spectrum) UVR. All UV A exposures were for
20 minutes a day, five days a week, for 13 to 98 weeks. Exposure to broad-spectrum
UVR was for 10 minutes a day, four days a week, for the lifetimes of the animals. One
control group was not exposed to any UVR source, and one control group was exposed to
broad-spectmm UVR only. All three UV A sources induced skin tumors. Of the 260
irradiated mice, 232 developed tumors; 230 developed multiple tumors that later fused by
growth. Pre-irradiation with UV A sources with relatively high UVB outputs enhanced the
carcinogenic effect of broad-spectrum UVR. The carcinogenic potential ofUV A sources
was directly related to their emission below 320 run. In a follow-up study, Bech-Thomsen
50
Dec. 2000
et al. (1992) administered UV A radiation alone or after 12 weeks of exposure to broad-
spectrum (simulated solar) UVR. This study demonstrated that a UVA source with a low
carcinogenic potential could significantly increase the carcinogenic effect ofbroad-
spectrum UVR.
Bech-Thomsen and Wulf (1993) also investigated whether the carcinogenic potential of
UV A sources could be estimated from the International Commission on Illumination
(CIE) human erythema action spectrum, which is used worldwide to assess the risk from
UVR-emitting appliances used in the home. Two groups of 40 C3H/Tif mice were
exposed to broad-spectrum UVR (with a UVB output of 16.7%) for 84 days.
Subsequently, each group was exposed to one of two commercial UV A sources with
different levels ofUVB emissions (2.2% and 6.9%). After pre-irradiation with identical
broad-spectrum UVR, exposure to the same erythemogenic dose from the differing UVA
sources resulted in simi lar times to tumor development. An inverse relationship between
the daily exposure dose and the tumor induction time was noted, whether the UV A was
administered alone or after broad-spectrum UVR exposure. These researchers concluded
that the CIE erythema action spectrum could be used to compare the carcinogenic
potential of different UVR sources. The results of the Bech-Thomsen studies are
summarized in Tables 4-1 and 4-2.
Table 4-1. Tumor incidences in female C3H/Tif mice exposed to UV A tanning
sources with differing UVB emission levels
Daily dose (kJ/m
2
)
Duration Tumor
UVA
3
UVBb (weeks) incidence Reference
121 8.9 41 20/20 Bech-Thomsen and Wulf 1993,
Bech-Thomsen et al. 1991
81 6 59 21/21 Bech-Thomsen and Wulf 1993
162 3.6 86 13/20 Bech-Thomsen and Wulf 1993
21 0.5 88 1/22 Bech-Thomsen and Wul f 1993
199 0.4 97 l/20 Bech-Thomsen and Wulf 1993
82 6.1 47 20/20 Bech-Thomsen et al. I 992
163 3.7 74 13/20 Bech-Thomsen et al. 1992
199 0.4 85 1/20 Bech-Thomsen et al. 1992
281-320 nm; b 3 2 1 ~ nm.
51
Dec. 2000
Table 4-2. Tumor incidences in female C3H/Tif mice exposed to broad-spectrum
UVR and/or UV A
Total dose (kJ/m
2
)
UVA
3
UVBb
4,050 0
710 230
4,760 230
1,410 460
5,670 480
4,260 20
Source: Bech-Thomsen et al. 1988
321-400 nm; b28l-320 nm.
4.2.2 Other species
4.2.2. 1 Opossums
Duration Tumor
Exposure regimen (weeks) incidence
UVA4wk 57 0/24
UVR 13 wk 54 8/23
UVA 4 wk, UVR 13 wk 57 0/25
UVR26 wk 29 22/25
UVR 3 d, UV A 3.5 wk, 38 23/25
UVR26wk
UVR 3d, UVA 3.5 wk 57 0/25
M. domestica developed non-melanoma skin tumors or melanocytic hyperplasia (a
melanoma precursor lesion) following exposure to UV A (Ley 1997). Thirty dorsally
shaved M. domestica were exposed three times a week for 81 weeks to 25,000 J/m
2
of
UVA radiation from filtered F40BLB fluorescent lamps (black lights). The incidences of
non-melanoma skin tumors and melanocytic hyperplasia were 4% and 22%, respectively,
in the exposed animals. These data suggest that the action spectra for the induction of
melanoma and non-melanoma skin tumors are different.
4.2.2.2 Fish
Heavily pigmented backcross hybrids of the genus Xiphophorus (cross between platyfish
and swordtails) are very sensitive to melanoma induction by UVR. Groups of six-day-old
fish were irradiated with narrow-wavelength bands at 302, 313, 365, 405, and 436 nm
and scored for melanomas four months later. Two groups of controls were used because
the researchers realized that the initial control group was exposed to some ambient UV A
and visible radiation. This could explain the high incidence of melanoma in the first
control group. The second control group was kept under subdued yellow light for two
months and had a much lower incidence of melanoma. The action spectrum (sensitivity
per incident photon as a function of wavelength) for melanoma induction showed
appreciable sensitivity at 365, 405, and probably 436 nm (Setlow et al. 1993). The tumor
incidence for each wavelength is shown in Table 4-3.
52
Dec. 2000
Table 4-3. Incidences of melanoma in hybrid fish (Xiphophorus) exposed to various
wavelengths of UVR
No. of exposure No. of fish with melanoma
Wavelength (nm) levels No. of fish (%)
Control"
- 124 30 (24.2)
302 4 123 37 (30.1)
313 4 124 46 (37.1)
365 6 85 38 (44.7)
Controlb - 20 I (5.0)
405 4 61 18 (29.5)
436 2 21 5 (23.8)
Source: Setlow el al. 1993
controls were in ambient light in shaded greenhouse for the following irradiations: 313 nm, 7 of 9 at
302 nm, 7 of9 at 365 nm, and 2 of 5 at 405 nm.
bControls were in covered tanks for 2 months for the following irradiations: 436 nm and for 3 of 5 at
405 nm.
4.3 Primarily UVB
4.3.1 Rats
Skin-tumor induction was studied in a group of 40 (shaved) female NMR rats, eight to 10
weeks old at the initiation of the experiment. Animals were irradiated for 60 weeks
(duration and frequency of exposures were not specified) at a distance of 3 7.5 em from a
commercial sunlamp emitting mainly UVB. Weekly doses of radiation were described as
being 5.4 to 10.8 x 10
4
J/m
2
A total of25 skin tumors, most of which were papillomas of
the ears, developed in 16/40 animals (Stenback 1975, cited in IARC 1992).
4.3.2 Mice
Several studies have clearly indicated in albino mice a dose-response to UVB in the
development of skin tumors. Forbes et al. (1981, cited in !ARC 1992) demonstrated a
dose-response relationship in the time to onset of skin tumors in SKHl albino hairless
mice exposed to UVB. Groups of24 male and female mice, six to eight weeks old, were
exposed to sunlamps emitting mainly UVB (< 1% below 280 nm; two-thirds from 280 to
320 nm, and one-third above 320 nm). Animals were irradiated five days per week, for up
to 45 weeks. Although the duration of daily exposures was not stated, the dail y dose of
radiation was computed. Time to onset of skin tumors is summarized in Table 4-4.
53
Dec. 2000
Table 4-4. Dose-response to (mainly) UVB in SHHl albino hairless mice.
Weeks to 50%
Daily dose (J/m
2
) tumor incidence Week terminated
420 38.6 45
587 33.3 45
822 29.2 45
1,152 20.0 36
1,613 17.6 36
2,259 12.9 25
Source: Forbes et al. 1981, cited in TARC 1992
All animals eventually developed at least one skin tumor, with an inverse-dose-related
latency for the appearance of skin tumors in 50% of the animals in the exposure groups.
Tumors > 4 mm in diameter tended to be squamous-cell carcinomas, and tumors 1 to
4 mm formed a continuum from carcinoma in situ to squamous-cell carcinoma. Tumors
< 1 mm in diameter were epidermal hyperplasia and squamous metaplasia, tending
toward carcinoma in situ. Fibrosarcomas accounted for less than 1% of the tumors.
In a similar experiment, six groups of22 to 44 male and female SKH1 albino hairless
mice were exposed to mainly UVB at daily doses ranging from 57 to 1,900 J/m
2
(de Gruijl et al. 1983, cited in IARC 1992). Although the highest dose tested was not
sufficient to induce erythema, most animals in the study developed skin tumors. There
was a clear dose response in the time required for 50% of the animals to develop skin
tumors (Figure 4-1). Squamous-cell carcinomas developed in 71% ofthe mice in the
lowest dose group, while only two skin tumors were observed in 24 nonirradiated control
m1ce.
4.3.3 Hamsters
Stenback (1975, cited in IARC 1992) irradiated 40 shaved female Syrian golden
hamsters, eight to 10 weeks of age at the initiation of the experiment, with mainly UVB.
Weekly doses of radiation were 5.4 to 10.8 x 10
4
J/m
2
. A total of30 skin tumors were
observed in 14/40 animals, of which 22 were papillomas (14 animals), four were
keratoacanthomas (three animals), one was a squamous-cell carcinoma of the skin, and
three were papillomas of the ear (all in one animal).
4.3.4 Guinea pigs
Stenback (1975, cited in IARC 1992) irradiated shaved guinea pigs with mainly UVB.
Weekly doses of radiation were 5.4 to 10.8 x 10
4
J/m
2
. Only 2/25 animals had skin
tumors (a fibroma in one animal and a trichofolliculoma in the other).
54
Dec. 2000
1000
800
500
.......
..
300
>-
tl
"0

100
e
.....
80
50
3 5 1 0 30 50 80 1 00
Dose (as % of maximal dose)
Source: de Gruijl eta!. 1983, cited in IARC 1992
Figure 4-1. Dose-effect relationship for the induction of< 1-mm skin tumors in
hairless mice by exposure to UVB over a wide range of daily doses; tm = median
induction time
4.3.5 Other species
4.3.5. 1 Opossums
M. domestica developed actinic keratoses, fibrosarcomas, and squamous-cell carcinomas
following exposure to a UVR sunlamp (280 to 400 nm). In another study, opossums were
shaved and exposed three times per week for 70 weeks to 250 J/m
2
of mainly UVB
radiation with relative emissions of0.04, 0.27, 0.69, 1.0, or 0.09 at wavelengths of280,
290,300,313, or 360 mm, respectively. Melanomas were observed in 5/13 exposed
animals and melanocytic hyperplasia in 8/ 13 exposed animals (Ley et al. 1989, cited in
IARC 1992). In a subsequent study (Ley 1997), 30 dorsally shaved M. domestica were
exposed three times a week for 81 weeks to 250 J/m
2
of UV radiation from FS-40
sunlamps (approximately I 50 J/m
2
ofUVB radiation). The incidences of non-melanoma
skin tumors and melanocytic hyperplasia in exposed animals were 71% and 31%,
respectively. Although the incidence of non-melanoma skin tumors was significantly
higher than observed in opossums exposed to UV A, the incidence of melanocytic
hyperplasia was similar to that in UV A-exposed animals (see Section 4.1.5).
4.3.5.2 Fish
Melanocytic neoplasms were induced in a group of 460 hybrid fish (Xiphophorus),
following exposure to mainly UVB from FS-40 sunlamps. The sunlamps were filtered
with acetate sheets transmitting> 290 nm or> 304 nm at various doses ( 150 or 300 J/m
2
per day for > 290 nm; 850 or 1,700 J/m
2
per day for > 304 nm) for 1 to 20 consecutive
days. Melanocytic tumors were found in 19% to 40% of the exposed fish. Of l 03 controls
55
Dec. 2000
from the two parent strains, 13% and 2% developed these tumors (Setlow et al. 1989,
cited in IARC 1992).
4.4 Primarily UV C
No studies were found in which animals were exposed solely to UVC. In the studies
reviewed below, the source ofUVC was a low-pressure mercury discharge germicidal
lamp, which emitted 90% to 95% of its radiation at 254 nm, but also emitted significant
amounts of UVB, W A, and visible light.
4.4.1 Rats
Nine groups of six to 12 male CD-1 rats, 28 days of age, were shaved and exposed to
varying doses ofUVC from a germicidal lamp (Strickland et al. 1979, cited in IARC
1992). The dose range was 0.08 to 26.0 x 10
4
11m
2
. Exposure duration was not specified.
Survival ranged from 75% to 92% in the experimental groups. Keratoacanthoma-like skin
tumors developed at a yield that was approximately proportional to radiation throughout
the dose range of0.65 to 26.0 x 10
4
11m
2
. No tumors were observed at or below 0.32 x
10
4
11m
2
.
4.4.2 Mice
A group of 40 female C3HIHeNCriBr mice was exposed to radiation from germicidal
lamps at a weekly dose rate of3 x 10
4
11m
2
. The duration of the experiment was not
specified. Three animals died without tumors at experimental weeks 9, 43, and 63. All
other animals had tumors, with 97% of the mice affected by 52 weeks. The median time
to tumor onset was 43 weeks, and the mean number of tumors per tumor-bearing animal
was 2.9. Microscopic examination revealed that of the 83 lesions initially considered to
be tumors, 66 were squamous-cell carcinomas, 10 were proliferative squamous-cell
lesions, and six were invasive fibrosarcomas (Lill1983, cited in !ARC 1992). The !ARC
Working Group noted that the 4% UVB content of the radiation source provided a
weekly dose of 1,170 11m
2
, which could not be excluded as a contributing factor in the
induction of skin tumors (IARC 1992).
Groups of 24 male and female S.KH1 albino hairless mice, 6 to 10 weeks of age, were
exposed to UVC from germicidal lamps seven days per week, for 75 minutes per day, at a
dose of230, 1,460, or 7,000 11m
2
. The highest dose applied was 60% lower during the
initial seven days of the experiment. A total of 65 squamous-cell carcinomas of the skin
were found. The numbers of animals with tumors were not reported, but the investigators
noted that both the numbers of animals with tumors and the numbers oftumors per
mouse were strongly dose-related (Sterenborg and van der Leun 1988, cited in IARC
1992). By comparing tumor incidences and onset times in their own UVC experiment to
those from experiments with UYB, Sterenborg and van der Leun (1988, cited in !ARC
1992) concluded that the UVB emitted from the germicidal lamp was insufficient to
cause the tumors observed in their experiment. They estimated that the UVB present
would require at least 850 days of exposure to induce skin tumors at the rate at which
they had observed tumors after 161 days of exposure to the UVC. Further, they noted a
qualitative difference between WC- and UVB-induced skin tumors in mice, in that
UVC-induced tumors were scattered more widely over the skin than were tumors
56
Dec. 2000
associated with UVB. Also, the dose-response curve was steeper in UVB-exposed mice
than in mice exposed to the germicidal lamp radiation. The IARC Working Group noted
that the observations given to exclude UVB as a causative factor in skin tumorigenesis
did not rule out a possible interaction between the two types of radiation (IARC 1992).
4.5 Cancer development in human-mouse chimera models
Several researchers investigated UV -induced skin cancers .in human skin grafted to mice.
Atillasoy et al. (1997) grafted white human skin onto 158 recombinase activating gene-1
(RAG-1) knockout mice. Mice were divided into four groups: control, a single
administration of dimethyl(a)benzanthracene (DMBA), exposure to UVB (290 to 320
nm) at 500 J/ m
2
three times per week, or a combination ofDMBA and UVB. Mice were
examined three times a week, and all surviving mice were euthanized and autopsied after
a median observation period of 10 months (range 3 to 16 months). About half of the
grafts exposed to UVB (alone or with DMBA) developed milia, compared with 3% of
DMBA-exposed grafts and none of the controls. Actinic keratoses were observed in 9%
of the grafts exposed to UVB alone and 19% of the grafts exposed to DMBA plus UVB.
Invasive squamous-cell carcinomas developed in 10% of the grafts exposed to DMBA
plus UVB. None of the controls developed actinic keratoses or squamous-cell
carcinomas. Melanocytic hyperplasia was found in 68% of the grafts exposed to UVB
only and 77% of the grafts exposed to both UVB and DMBA. One human nodular-type
malignant melanoma developed in a graft exposed to both DMBA and UVB (Atillasoy et
al. 1998).
In a follow-up study (Sauter et al. 1998), 25 RAG-I mice with human skin grafts
received a single administration ofDMBA followed by three weekly exposures to UVB
(500 J/m
2
) for at least five months. Cysts, hyperplasia, precancers, or invasive cancers
were seen in 24 of 25 exposed grafts, compared with none of the controls. Two
squamous-cell carcinomas were observed. Of grafts exposed for seven or more months,
83% (15/ 18) developed squamous precancer or squamous-cell carcinoma of human
origin, and 44% (8/ 18) developed melanocytic hyperplasia or melanoma. Direct
correlations between p53 tumor suppressor gene expression and cell proliferation and the
degree of histologic change were observed for both squamous epithelial and melanocytic
cells.
Hun1an skin was transplanted to severe combined immunodeficient mice and exposed to
UVB (280 to 360 nm) at daily doses of at 7.3 x 10
5
to I .8 x I 0
6
J/m
2
for two years
(Nomura et al. 1997). Actinic keratoses developed in 77.8% (14/ 18) and squamous-cell
carcinoma in 16.7% (3/18) of grafts exposed to UVB. None ofthe 15 control grafts
developed actinic keratoses or squamous-cell carcinomas. The same p53 mutation at
codon 242 (C TGC to C CGC) was observed in actinic keratoses and squamous-cell
carcinomas, and double or triple mutations were observed in all skin cancers and three of
eight actinic keratoses.
4.6 Summary
Broad-spectrum UVR was carcinogenic to albino rats, inducing skin tumors (papilloma,
squamous-cell carcinoma, spindle-cell sarcoma and carcinosarcoma, and basal-cell
57
Dec. 2000
carcinoma) and eye tumors (spindle-cell sarcoma and squamous-cell carcinoma). Broad-
spectrum UVR induced skin or eye tumors (spindle-cell sarcoma or fibrosarcoma, mostly
in the cornea) and hemangioendothelioma in mice and hamsters and caused skin tumors
(mostly fibrosarcoma and squamous-cell carcinoma) in opossums. Broad-spectrum UVR
also has been implicated in tumor development in domestic animals (cows, goats, sheep,
cats, dogs, horses, and swine).
UVA induced skin tumors in mice (squamous-cell carcinoma and papilloma), opossums
(melanocytic hyperplasia) and fish (melanoma). Prolonged UVB exposure caused skin
twnors in rats (papilloma), mice (squamous-cell carcinoma, fibrosarcoma, papilloma, and
keratoacanthoma), guinea pigs (fibroma and trichofolliculoma), opossums (melanocytic
hyperplasia and melanoma), and fish (melanocytic neoplasms). Exposure of experimental
animals to high doses of UVC caused skin tumors in rats (keratoacanthoma-like skin
twnors) and mice (squamous-cell carcinoma and fibrosarcoma). Human skin grafts on
mice also yielded skin tumors (squamous-cell carcinoma, actinic keratosis, melanocytic
hyperplasia, and melanoma) following irradiation with UVB alone or after exposure to
DMBA.
58
Dec. 2000
5 Genotoxicity
The IARC conducted an extensive review of the literature through 1991 on the
genotoxicity of solar and ultraviolet radiation, to develop a better understanding of
exposure to UVR, the intermediate biological responses, and their consequences, with
emphasis on carcinogenesis (IARC 1992).
This section discusses pertinent genotoxicity information from the IARC review and
from recent genotoxicity studies, focusing on UVR, including UV A, UVB, and UVC. It
is important to recognize that many exogenously supplied photosensitizers, including
some pharmaceuticals, can affect the biological response to UVR. In some cases,
interactions with photosensitizers have therapeutic application; for example, UV A may
be used in combination with furocoumarins to treat skin diseases or tumors (IARC 1992,
Muller et al. 1998). However, UVR interactions with exogenous chemical agents are
considered outside the scope of this document and are not addressed.
5.1 Methods for identifying and quantifying UVR-induced DNA lesions
Griffiths et al. (1998) reviewed the measurement and significance of DNA lesions
induced by UVR. UVR-induced DNA lesions and methods for identifying and
quantifying them may be categorized as follows:
Single- and double-strand DNA breaks. UVR causes strand breakage interfering with
inter- and intra-strand stabilization and inevitably resulting in some degree of a-helical
unwinding. Several assays rely on this phenomenon and do not require DNA extraction,
but are based on fluorescence labeling of DNA. Examples of such assays are the
fluorescence-activated DNA unwinding assay, DNA sedimentation analysis, and the
single-cell gel electrophoresis (or comet) assay (Griffiths et al. 1998).
Specific DNA sequences containing damage. UVR elicits antigenicity by altering DNA
sequences through denaturation. Polyclonal and monoclonal antibodies have been raised
against specific lesions that begin with thymine dimers. These antibodies have been used
to recognize sequence-specific damage both on fixed section slides and in fluorescence-
activated cell-sorter-type flow cytometry systems. For instance, Herbert et al. (1994,
cited in Griffiths et al. 1998) developed a polyclonal antibody specific for a cyclobutane
thymidine dimer with an adjacent 3' or 5' thymidine.
Specific DNA base lesions. Franklin and Haseltine (1984, cited in Griffiths et al. 1998)
developed and demonstrated a high-perfonnance liquid chromatographic (HPLC) assay
that can separate and quantitate cyclobutane-type pyrimidine dimers and (6-4)
photoproducts.
5.2 UVR-induced DNA photoproducts
It is well documented that UVR induces mutations in both prokaryotic and eukaryotic
cells, and any cell that is UV -irradiated will likely sustain DNA damage. UV A, UVB,
and UVC have induced mutations in bacterial systems and cultured mammalian cells. In
addition, UV A has induced mutations in yeast, and UVC has induced mutations in plants
59
Dec. 2000
and amphibians (IARC 1992). The type of damage induced depends upon the specific
wavelength(s) applied and the competency of an affected cell to repair the damage
without error. DNA is a major cellular chromophore absorbing UVR (mainly UVB); it
responds to irradiation by yielding single-electron reactive intermediates and, depending
on exposing wavelengths and energy produced, various identifiable photoproducts. All
photoproducts are expected to have mutagenic potential; however, their specificity and
potency vary (IARC 1992).
5.2. 1 UVA-induced indirect DNA damage
Over 90% of the UV radiation reaching the surface of the earth is in the form ofUV A.
Upon absorption ofUV A by cells and subsequent generation of activated oxygen, the
energy is transferred to DNA. DNA poorly absorbs UV A; therefore, the induced
genotoxic damage is due to absorption of photons by other endogenous chromophores
(IARC 1992). Examples of endogenous chromophores within mammalian cells are
riboflavin, porphyrins, quinones, and reduced nicotinamide cofactors (Griffiths et al.
1998).
UVA-excited endogenous photosensitizers produce a much lower level ofbase loss than
does UVB (Cadet et al. 1992). The major DNA base lesions induced are 8-
hydroxydeoxyguanosine (8-0HdG), produced from guanosine by the action of singlet
oxygen; hydroxyhydroperoxides, indirectly generated from the radical cation of thymine
under aerobic conditions; and pyrimidine photoproducts (however, their induction
requires a six-fold greater energy input than UVC-induced lesions at a similar frequency).
UVA does not induce formation of (6-4) photoproducts (Griffiths et al. 1998).
5.2.2 UVB-induced direct DNA damage
UVB photons directly cause the following major DNA base modifications: cyclobutane-
type pyrimidine dimers, (6-4) photoproducts, the corresponding Dewar isomers, and
thymine glycols. The pyrimidine dimers are five to 1 0 times more abundant than the
other DNA base modifications. Depending upon the conditions of exposure, these
pyrimidine dimers occur as cytosine-cytosine, thymine-thymine, or mixed dimers. The
absorption spectra for cytosine and thymidine match the action spectrum for dimer
formation and, in (6-4) photoproduct induction, the cytosines 5' upstream of adjacent
pyrimidines present perfect targets for such DNA damage (Griffiths et al. 1998).
UVB also is responsible for induction ofDNA strand breaks. The incidence of DNA
strand breaks increases as a function of increasing wavelength. Single-base lesions,
mainly ring-saturated thymines known as thymine glycols, are also observed. Along with
these, 8-0HdG adducts are induced over the dose range of 4 to 750 mJ/cm
2
(Stewart et
al. 1996, cited in Griffiths et al. 1998). UVB exposure also causes DNA-protein
cross! inks, mostly affecting cysteine residues. At equivalent doses, UVB induces DNA-
protein crosslinks at about one-tenth the frequency that UV A does (Griffiths et al. 1998).
5.2.3 Cellular mechanisms for minimizing UVR-induced DNA damage
Healthy eukaryotic cells can minimize UVR-induced DNA damage by several defense
mechanisms, which interact to protect cells against toxic effects of UVR. These
60
Dec. 2000
mechanisms include production of antioxidant enzymes, production of detoxification
enzymes, and repair ofUVR-induced DNA lesions by means of direct reversal, base
excision repair, nucleotide excision repair, transcription repair coupling, and
mitochondrial repair of UV -induced lesions (Griffiths et al. 1998).
5.2.4 Cellular responses to UVR-induced DNA damage
Transcriptional activation of mammalian "early response genes" (e.g., c-fos and c-jun) is
induced within minutes ofUVR exposure. Early and secondary response genes also
include genes mediating protein binding to DNA damage sites, cell proliferation control
genes (e.g., growth arrest and DNA damage genes), genes coding for enzymes involved
in signal transduction (e.g., protein kinase C) or for antioxidants (e.g., heme oxygenase),
and the p53 tumor suppressor gene (Griffiths et al. 1998).
UVR-induced photoproducts have genotoxic consequences that vary depending on the
particular exposure circumstances. In the survey below, genotoxic effects are classified
according to the test system in which they were assessed. Data presented in IARC ( 1992)
are summarized in Table 5-l. Studies that were not reviewed in IARC ( 1992) are
discussed in the following text.
5.3 Prokaryotic systems
5.3. 1 Induction of mutation in Salmonella typhimurium
UVC exposure unambiguously increased the frequencies of reverse gene mutations in
several S. typhimurium tester strains, including repair-defective strains hisG46 and
hisG428 (Cebula et al. 1995) and recA-uvrB (Hartmann et al. 1996, cited in Griffiths et
al. 1998).
Table 5-1. Genetic and related effects of UVR exposure reviewed in IARC (1992)
Results (no. positive/no. studies)
Test system End point UVA UVB uvc UVR
8
Prokaryotic
Salmonella typhimurium mutation 1/1 1/1
Escherichia coli mutation 8/8 1 /1 6/6
Escherichia coli DNA damage 5/5
Bacillus subtilis mutation 1/1
Lower eukaryotic
Saccharomyces DNA damage or 3/3 212 1/1
cerevisiae pyrimidine dimers
Saccharomyces aneuploidy 1/ 1
cerevisiae
Saccharomyces mutation 2/2
cerevisiae
61
Dec. 2000
8
Plant
Wheat mutation 1/ 1
Unspecified plant cells DNA damage 1/ 1
Nicotiana tabacum unscheduled DNA 1/ 1
synthesis
Chlamydomonas pyrimidine dimers 1/ 1
reinhardtii
Chlamydomonas mutation 1/ 1
reinhardtii
Tradescantia chromosomal 1/ 1 1/ 1
aberrations
Nonmammalian eukaryotic
Drosophila DNA damage 1/ 1
melanogaster
ICR 2A frog cells DNA damage 1/ 1 2/2
ICR 2A frog cells SCE, 1/ 1 2/2 1/ 1
chromosomal
aberrati ons
A8W243 Xenopus frog chromosomal 1/ 1
cells aberrations
Fish (in vitro) DNA damage 1/ 1
Chick embryo fibroblasts SCE, 2/2
chromosomal
aberrations
Nonhuman mammalian in vitro
Chinese hamster ovary DNA damage 2/2 1/ 1 1/ 1
cel ls
Chinese hamster ovary SCE, 2/2 1/ 1 2/2
cells chromosomal
aberrations
Chinese hamster ovary mutation 3/3 2/2 3/3 2/2
cells
Chinese hamster chromosomal 2/2
fibroblasts aberrations
Chinese hamster V79 DNA damage 2/2 1/ 1 2/2
lung cells
Chinese hamster V79 mutation 2/2 4/4 4/4 3/3
lung cells
62
Dec. 2000
8
Chinese hamster V79 SCE, 3/3
lung cells chromosomal
aberrations
Chinese hamster CHEF- chromosomal 1/ 1
125 cells aberrations
Syrian hamster embryo cell transformation 1/2 1/ 1 3/3
cells
C3H 1 OT1 /2 mouse cells DNA damage 1/ 1
L5178Y mouse mutation 1/ 1 1/ 1 1/1 1/ 1
lymphoma cells
Mouse splenocytes micronuclei 1 / 1
New Zealand black chromosomal 1/ 1
mouse fetal fibroblasts aberrations
Mouse epidermal cells, cell transformation 717 6/6
embryo cells, fibroblasts,
fibrosarcoma cells
Human in vitro
Fibroblasts DNA damage or 4/4 4/4 8/8 4/4
pyrimidine dimers
Fibroblasts mutation 1/ 1 1/ 1 5/5 1 / 1
Fibroblasts micronuclei 1/1 1/ 1
Fibroblasts SCE, 9/9 2/2
chromosomal
aberrations
Fibroblasts cell transformation 1/ 1 3/3
Keratinocytes and DNA damage 1/ 1 1/1
melanocytes
Epithelial P3 cells DNA damage 1/ 1
Epi thelial cells mutation 1/ 1 1/ 1 1/ 1
Teratoma or DNA damage 3/3 3/3 3/3
teratocarcinoma cells
Lymphoblastoid cells mutation 0/ 1 0/ 1 1/ 1
Hel a cells DNA damage 1 /1
Hela cells mutations 1/ 1
Melanoma cells SCE 1/ 1
Melanoma cells mutation 1/ 1
Melanoma cells micronuclei 1/ 1
63
Dec. 2000
8
Lymphocytes mutation 2/2
Lymphocytes SCE, 4/4
chromosomal
aberrations
Nonhuman mammalian in vivo
Mouse skin DNA damage or 3/3 1/1 1/1
pyrimidine dimers
Mouse skin fibroblasts cell transformation 1 /1
Marsupial corneal cells DNA damage 2/2
Human in vivo
Epidermis or skin cells DNA damage or 1/ 1 1/ 1 2/2
pyrimidine dimers
Cornea unscheduled DNA 1/1
synthesis
Fibroblasts DNA damage 2/2 2/2
Includes solar, simulated solar, and sunlamp irradiation.
5.3.2 Induction of mutation in Saccharomyces cerevisiae
UVB and natural sunlight exposure increased the frequencies of pyrimidine dimer
fonnation (Armstrong and Kunz 1992), single-base-pair substitution (Kunz and
Annstrong 1998), and gene mutation (Annstrong and Kunz 1990) in S. cerevisiae.
Natural sunlight and UVB induced similar G-C to A-T transitions; however, natural
sunlight induced a higher percentage of G-C to T -A or C-G transversions. Dipyrimidine
adducts likely were responsible for the transitions and are now recognized as a signature
of sun exposure (Sarasin 1999). These data suggest that one type of DNA damage leads
to most of the mutations associated with UVB exposure, whereas two different types of
DNA damage may be involved in sunlight mutagenesis (Kunz and Annstrong 1998).
5.4 Plants and lower eukaryotic systems
No additional genotoxicity studies in plant or eukaryotic systems were identified in the
current literature.
5.5 Mammalian systems
5.5. 1 Nonhuman mammalian in vitro assays
Oxidative damage in DNA is caused by UVB irradiation and results in the formation of a
DNA adduct, 8-0HdG. Studies demonstrated a decrease in antioxidant enzyme defenses
in SKHI hairless albino mice after UVB radiation, implicating antioxidant status in
protection against oxidative damage (Cameron and Pence 1992). A further study by this
group examined mechanisms ofUVB-induced DNA damage and subsequent modulation
64
Dec. 2000
by the antioxidants vitamin C (ascorbic acid), selenite, or Trolox (a water-soluble vitamin
E analog). BALB/c MK-2 mouse keratinocytes were exposed to UVB at a dose range of
4 to 750 mJ/cm
2
. Adducts were measured via HPLC coupled with electrochemical and
UV absorbency detection. Preincubation ofthe cells for two days with 0.4 or 0.8 J..tg/ml of
ascorbic acid, 10 or 20 J..lg/ml of Trolox, and 5 or 12.5 J..l.M selenite significantly
decreased the number of 8-0HdG adducts per 10
5
deoxyguanines induced by UVB at
500 mJ/cm
2
The results further elucidated mechanisms through which UVB altered DNA
exposed ex vivo in cultured mouse skin cells and indicated that antioxidant nutrients
might protect skin cells against UVB damage (Stewrut et al. 1996, cited in Griffiths et al.
1998).
5.5.2 Human in vitro assays
Murata-Kamiya et al. (1995, cited in Griffiths et al. 1998) demonstrated that oxygen free
radicals caused DNA base and sugru modifications and DNA strand breaks. They showed
that a known mutagen, glyoxal, was produced by exposure of DNA to an oxygen-radical-
forming system (5 mM ferrous sulfate-ethylenediaminetetraacetic acid, +37 C, 60 min).
Glyoxal was produced with a 17-times-higher efficiency than 8-0hdG, with adduct
formation at guanine sites. The authors predicted that this type of exposure of DNA to an
oxygen-radical-forming system, with following glyoxal and guanine adduct formation,
constituted one of the major types of UV A-induced DNA damage.
Mizuno et al. (1991, cited in Griffiths et al. 1998) conducted a study using a thymine
dimer-specific monoclonal antibody (TDM-1), which was produced against mouse and
human DNA after exposures of cells ex vivo to 313-nm UVB in the presence of
acetophenone. When UVB-irradiated DNA was incubated with photolyase from E. coli
and visible light, TDM-1 binding and the presence ofthymine dimers were reduced. It
was shown that TDM-1 binding to UVB-irradiated DNA was inhibited by photolyase, but
not by 64M-1 antibody specific for (6-4) photoproducts. The authors concluded that the
TDM-1 antibody had affinity for cyclobutane-type DNA thymine dimers. They
measured, by competitive assessments with the two antibodies, the amount of each type
of DNA damage in DNA extracted from UVB-irradiated mammalian cells. Repair
experiments indicated that (6-4) photoproducts were excised from UVB-irradiated
cellular DNA more rapidly than thymine dimers. Excision rates of both photoproducts
were lower in mouse (NIH3T3) cells than in human fibroblasts.
Immunocytochemical methods were used to measure cyclobutane pyrimidine dirners, (6-
4) photoproducts, and Dewar isomers in normal human mononuclear cells following ex
vivo irradiation by natural sunlight or a UVB sunlamp (Clingen et al. 1995, cited in
Griffiths et al. 1998). The induced photoproducts were detected following a 30- to 60-
minute sunlight exposure, or with sunlamp irradiation as low as 50 to 100 J/m
2
. A dose-
dependent increase in the binding of monoclonal antibodies specific for pyrimidine
dimers, (6-4) photoproducts, and Dewar isomers was observed. The relative ratio of
Dewar isomers to (6-4) photoproducts was much greater after exposure to natural
sunlight than after exposure to broad-spectrum UVB. Use of the (6-4) monoclonal
antibody indicated that binding sites increased slightly after a one-hour exposure to
natural sunlight and remained relatively constant with further exposure. The authors
65
Dec. 2000
hypothesized that following irradiation with natural sunlight, most (6-4) photoproducts
were converted into Dewar isomers, and that this conversion was likely caused by the
UVA component. They concluded that the (6-4) photoproducts probably did not
contribute directly to sunlight-induced carcinogenesis.
Human skin explants were studied with a e
2
P]-HPLC method for recognizing and
measuring lesions (cyclobutane dimers, [6-4] photoproducts, and Dewar isomers)
induced in DNA after exposure to UV A, UVB, or UVC (Bykov and Hemminki 1996).
The experimental method was sensitive enough to detect the lesions at a UVB radiation
dose of 10 J/m
2
. Dewar isomers were detected only at a high doses ofUVB. The
compounds were identified by their photochemical reactivity and by spiking with
prepared standards. Treatment with nuclease Pl was used to identify the 5'-terminal
nucleotide. UV A caused no detectable adducts .
5.5.3 Nonhuman mammalian in vivo assays
Formation of8-0HdG adducts was evaluated in the epidermis of hairless mice after
repeated exposure to UVB (Hattori et al. 1996, cited in Griffiths et al. 1998). Exposure of
hairless mice to UVB at a dose of either 3.4 kJ/m
2
(2 MED) or 16.8 kJ/m
2
(10 MED),
three times a week, for two weeks, induced a 2.5- or 6.1 -fold increase, respectively, in the
levels of 8-0HdG in DNA. An immunohistochemical method, using a monoclonal
antibody specific for 8-0HdG, showed stronger and more extensive staining in the nuclei
of UVB-irradiated epidem1al cells than in those of nonirradiated cells. Western blots
probed with antibodies against 4-hydroxy-2-nonenal-modified proteins confirmed the
involvement of reactive oxygen species in the epidermal damage induced by chronic
UVB exposure. The authors suggested that three pathways might regulate the fonnation
of 8-0HdG after UVB exposure: photodynamic action, lipid peroxidation, and
inflammation. They concluded that 8-0HdG might be active in sunlight-induced skin
carcinogenesis.
5.5.4 Human in vivo assays
5.5.4.1 DNA damage and repair
DNA synthesis, measured by eHJ thymidine incorporation after lymphocyte activation,
was studied in circulating leukocytes from patients with widespread psoriasis who were
being treated with PUV A (oral 8-methoxypsoralen and high-intensity UV A) (Kraemer
and Weinstein, 1977, cited in IARC 1992). Of 13 psoriasis patients treated with PUV A,
seven demonstrated a significant (P < 0.05) reduction in lymphocyte incorporation of
eHJthymidine immediately after UV A treatment, compared with incorporation before
UV A treatment. In addition to its therapeutic effects on epidermal cells, PUV A treatment
affected circulating blood cells in some psoriasis patients. However, in 10 control
subjects who received UV A alone, lymphocytes were capable of normal activation and
DNA synthetic activity. This study raised the possibility of genotoxic effects in
circulating lymphocytes. Strauss et al. (1979, 1980) observed induction of presumed
mutations at the HPRT locus in lymphocytes in UV A-exposed patients, but not in the
absence of psoralens. In another study, patients treated with PUV A, but not UV A alone,
showed evidence of local and systemic impairment of the delayed cellular
hypersensitivity component of the immune response, providing evidence for a possible
66
Dec. 2000
mechanism oftumor promotion in the skin (Bridges and Strauss 1980, Strauss 1982).
Human studies evaluating PUV A treatment and cancer risk are reviewed in Section 3.
Irradiation of human buttock skin with 300-nm UVR in situ induced thymine dimers and
(6-4) photoproducts (Chadwick et al. 1995, cited in Griffiths et al. 1998). Irradiation of
human buttock skin with UVC (260 nm) immediately followed by OVA (320 nm)
induced the Dewar isomers of the (6-4) lesions. All three lesions were detected in
methanol-fixed paraffin sections through the use of specific monoclonal antibodies. The
lesions were analyzed in an automated image analysis system, and the level of inununo-
diaminobenzidine-peroxidase was measured in individual epidermal-cell nuclei . Staining
patterns indicated a decrease with depth of about 2.5% per cell layer. Following
irradiation with a shorter wavelength (260 nm), staining decreased rapidly with depth
(39% per cel1layer). The results showed effective penetration and damage induction by
UVB in human skin after in vivo exposure.
Hori et al. (1992, cited in Griffiths et al. 1998) studied DNA extracted from a variety of
human skin tumors and control tissues, including femoral skin and white blood cells, with
an immunoblotting method using antibodies against UV-irradiated calf thymus DNA.
The antibodies used were reactive to cyclobutane-type pyrimidine dimers.
Immunoprecipitates were observed for facial actinic keratosis and keratosis-derived
squamous-cell carcinoma specimens. Through the use of photoreactivation enzyme plus
visible light, both immunoprecipitates were found to be specific for cyclobutane-type
pyrimidine dimers. Immunofluorescence studies of actinic keratosis tissue showed that
unremoved photodamage in DNA remained in the nucleus of actinic keratosis cells. The
authors suggested that the tumor cells might be deficient in an enzyme required for
repairing cyclobutane-type pyrimidine dimer damage.
Clingen et al. (1995, cited in Griffiths et at. 1998) used specific monoclonal antibodies in
situ and a computer-assisted image analysis system to determine the relative induction of
cyclobutane dimers, (6-4) photoproducts, and Dewar isomers in human mononuclear
cells and fibroblasts following irradiation with UVC, broad-spectrum UVB, and narrow-
spectrum UVB. DNA lesions were produced in different proportions, with broad-
spectrum UVB inducing a greater combined yield of (6-4) photoproducts and Dewar
isomers per cyclobutane dimer than UVC or narrow-spectrum UVB. Relative induction
ratios of (6-4) photoproducts versus cyclobutane dimers were 0.15, 0.21, and 0.10
following irradiation with UVC or broad- or narrow-spectrum UVB, respectively.
5.5.4.2 Tumor suppressor and ras gene mutations
Brash et al. ( 1991, cited in IARC 1992) reported five C to T, four C to A, and three CC to
TT mutations at various codons of the p53 tumor suppressor gene in 24 invasive
squamous-cell carcinomas taken from sun-exposed skin; about 90% of squamous-cell
carcinomas examined in this study contained p53 mutations. CC to TT transitions have
not been found in any internal tumors, suggesting that sun exposure plays a role in p53
mutations. Pierceall et al. ( 1991, cited in IARC 1992) reported one C to T transition and
one C to A transversion in 10 squamous-cell carcinomas examined. Ouhtit et al. (1997)
investigated the frequency ofp53 mutations in normal skin from Japanese patients. More
mutations were found in skin samples taken from sites chronically exposed to the sun
67
Dec. 2000
than from covered sites. A recent study showed that 50% of mutations of the PTCH
tumor suppressor gene found in basal-cell carcinomas were UVR-specific (Quinn 1997).
Melanomas from 37 patients with varying sun exposure were examined for N-ras
mutations (van 't Veer et al. 1989, cited in IARC 1992). N-ras mutations were found in
tumors from seven patients who were continuously exposed to the sun. All mutations
were base substitutions at TT or CC sites that are potential targets for UV photoproducts.
In other studies, N-ras, Ki-ras, and Ha-ras base substitution mutations were found in
melanomas, basal-cell carcinomas, and squamous-cell carcinomas (Sekiya et al. 1984,
Corominas et al. 1989, Keijzer et al. 1989, Shukla et al. 1989, van der Schroeff et al.
1990, all cited in IARC 1992).
5.5.5 Other in vitro and in vivo end points
DNA 8-hydroxy-2'-deoxyguanosine is a mutation-prone (G-C toT-A transversion) DNA
base-modified product generated by reactive oxygen species or photodynamic action. G-
C toT-A transversions were observed in the p53 and ras genes ofUVB-induced skin
tumors from mice and in squamous- and basal-cell carcinomas from human skin exposed
to sunlight (Hattori et al. 1996, cited in Griffiths et al. 1998).
5.5.6 Molecular epidemiological studies of DNA repair capacity
Wei et al. (1994) evaluated the relationship between DNA repair capacity and basal-cell
carcinoma in 88 cases and 135 controls. Cases were Caucasian patients with
histopathologically confirmed primary basal-cell carcinoma recruited from physician
practices in the Baltimore area between 1987 and 1990. Controls were patients from the
same physician practices who were cancer-free and were frequency-matched to cases by
age. Cancer patients and controls provided a blood sample and completed a self-
administered questionnaire that collected infonnation with respect to demographics,
family history, and potential confounders for basal-cell carcinoma. Lymphocytes were
isolated from the blood. DNA repair was assessed with the host-cell reactivation assay,
which measures the ability of lymphocytes from the participants to repair damaged DNA.
Plasmids containing UVR-irradiated (0, 350, or 700 J/m
2
) chloramphenicol acetyl
transferase (CAT) reporter genes were transfected into lymphocytes, and the ratio of
CAT gene expression of irradiated plasmids to that of non-irradiated plasmids was
calculated as the percentage of residual repair activity at a given UVR dose. The mean
DNA repair capacity of all basal-cell carcinoma patients was 5% lower than that of
controls, a difference of borderline significance. However, among subjects with red hair
and skin type I, DNA repair capacity was significantly lower in cancer patients than in
controls. Moreover, among subjects who reported frequent sunbathing, poor tanning
ability, a history of multiple sunburns, exposure to chemicals, or multiple medical
irradiations, the basal-cell carcinoma patients had significantly lower DNA repair
capacity than the controls (P < 0.05), which suggested that DNA repair might be a
susceptibility factor and the underlying molecular mechanism of sunlight-induced skin
carcinogenesis in the general population.
Hallet al. (1994) used the host-cell reactivation assay to evaluate the relationship
between DNA repair capacity and basal- or squamous-cell carcinoma in a population-
68
Dec. 2000
based case-control study. The study participants were residents of Australia between the
ages of 40 and 64 who were listed on the electoral roll. They were invited to attend a skin
cancer screening clinic, where they were examined by a dermatologist and interviewed.
Cases were 87 individuals who had one or more skin cancers diagnosed at the survey or
in the preceding year. Controls (86) were chosen by random sampling of the remaining
survey attendees and matched by age and sex. DNA repair capacity was greater in
subjects with skin cancer than in controls, but the difference was not statistically
significant; for each 350-J/m
2
increment in UV dose to the plasmids, repair capacity was
greater by a factor of 1.07 (95% CI = 0.94 to 1.26) in subjects with basal-cell carcinoma
and by a factor of 1.04 (95% CI = 0.85 to 1.26) in subj ects with squamous-cell
carcmoma.
5.6 Summary
The !ARC (1992) summarized genetic and related effects ofUVR according to type
(predominant wavelengths), test system, result (positive, negative, or conditional), and
study reference (see Appendix A, Tables 32- 35). Table 5-2 (updated from IARC 1992)
summarizes genetic and related effects according to test system, UV irradiation type, and
result.
5.6.1 UVA
UV A (315 to 400 run) was genotoxic in prokaryotic and lower eukaryotic systems. Its
biological effects are indirect and largely the result of energy transferred through active
oxygen intermediates. In mammalian cell ex vivo exposure systems, UV A induced gene
mutation, cytogenetic damage, and other fom1s of DNA damage. Few data are available
on DNA damage in human skin and circulating blood from UV A in vivo exposures. The
lARC (1992) cited twelve studies in prokaryotic systems; results were positive in nine for
gene mutation and three for DNA damage. Of ten cited nonhuman mammalian in vitro
studies, results were positive in two for DNA damage, six for gene mutation, and two for
cytogenetic damage. Of 11 cited human in vitro studies, results were positive in eight for
DNA damage and three for gene mutation. The one human in vivo study gave positive
results. UV A radiation can induce cellular and viral gene expression. Based on the
published literature, UVA (without exogenous photosensitizers) is a less potent genotoxic
agent than UVB or UVC.
5.6.2 UVB
UVB (280 to 315 nm) was genotoxic in prokaryotic, lower eukaryotic, and plant systems.
UVB photons are absorbed by DNA, and direct damage occurs through DNA base
modifications. In mammalian cell ex vivo exposure systems, UVB induced gene
mutation, cytogenetic damage, and other fonns of DNA damage. In a number of studies,
UVB caused DNA damage and gene mutation in human skin and circulating blood after
in vivo exposure. IARC (1992) cited three studies in prokaryotic systems; two showed
gene mutation, and one showed cytogenetic damage. Of 12 cited nonhuman mammalian
in vitro studies, results were positive in three for DNA damage, seven for gene mutation,
and two for cytogenetic damage. Of 11 cited human in vitro studies, results for gene
mutation were positive in two studies and negative in one study; results were positive in
two studies for cytogenetic damage and eight studies for DNA damage. Five animal in
69
Dec. 2000
vivo studies were cited, all with positive results. The two cited human in vivo studies both
demonstrated DNA damage. UVB radiation can induce cellular and viral gene
expression. Based on the published literature, UVB is a more potent genotoxic agent than
UV A, but less potent than UVC.
5.6.3 uvc
UVC (100 to 280 nm) was genotoxic in prokaryotic, fungal, plant, and insect test
systems. UVC photons are absorbed by DNA, and direct damage occurs through high-
energy reactions. In mammalian cell ex vivo exposure systems, UVC induced gene
mutation, cytogenetic damage, and other forms of DNA damage. In the few in vivo
studies reviewed, UVC caused DNA damage and gene mutation in animal and human
blood and skin. The IARC (1992) cited twenty-three studies in prokaryotic and lower
eukaryotic systems; positive results were found in nine for gene mutation, two for
cytogenetic damage, and 12 for DNA damage. Of 24 cited mammalian in vitro studies,
two showed DNA damage, eight showed gene mutation, and 14 showed cytogenetic
damage. Of 39 cited human in vitro studies, positive results were found in 14 for DNA
damage, 11 for gene damage, and 14 for cytogenetic damage. The one cited animal in
vivo study showed positive results for DNA damage, as did the two cited human in vivo
studies. UVC radiation can induce cellular and viral gene expression. Based on the
published literature, UVC is a more potent genotoxic agent than UV A or UVB.
70
Dec. 2000
Table 5-2. Genetic and related effects of UV A, UVB, and UVC exposure reviewed in IARC (1992)
Nonmammalian systems
Prokary
otic
Lower
eukaryotic Plant In vitro
Animal cell
D G D G ADG CD
UVA
UVB
uvc
+
Source: Adapted from IARC 1992
Mammalian systems
Human cell Animal
In vivo
Human
A
A- aneuploidy; C - chromosomal aberrations; D - DNA damage; G- gene mutation; M - micronuclei; S - sister chromatid exchange; T - cell transformation
Consensus oflARC Working Group:+ = positive for the specific end point and level of biological complexity.
+ 1 =positive, but only one valid study was available to the Working Group.
71
Dec. 2000
72
Dec. 2000
6 Other Relevant Data
6.1 Absorption and transmission of UVR in biological tissues
UVR may be transmitted, reflected, scattered, or absorbed by chromophores (biological
molecules that absorb radiant energy) in tissue, such as the skin. Absorption ofUVR
depends on the wavelength of the UVR and the properties of the target chromophores.
Absorption ofUVR by a tissue chromophore is a prerequisite for any photochemical or
photobiological effect; however, absorption does not necessarily have a biological
consequence (IARC 1992, Gould et al. 1995). A molecule's absorption spectrum (the
range of wavelengths in which it absorbs UVR) differs from its action spectrum (the
range of wavelengths in which biological responses are produced), but this range is the
same in most instances (Gould et al. 1995). Measured transmission ofUVR was maximal
in the cornea at 380 nm, 80% in the aqueous humor at 400 nm, 90% in the lens at
320 run, and 80% in the vitreous humor at 350 1m1 (Boettner and Wolter 1962, cited in
IARC 1992). UV transmission at 300 to 400 nm in nonnal human lenses decreases with
age (Lerman 1988, cited in IARC 1992).
Skin epidermis (the outer layer of the skin) can be divided into two regions based on
function: an outermost, nonliving part called stratum corneum and an inner region of
living cells (IARC 1992). In the skin, UVR is absorbed by the chromophores. The main
chromophores present in the skin are melanin, DNA (Amax 260 nm at pH 4.5), urocanic
acid (Amax 277 nm at pH 4.5), and the aromatic amino acids tryptophan (Amax 280 nm at
pH 7) and tyrosine (Amax 275 nm at pH 7) (Morrison 1985, cited in IARC 1992). Urocanic
acid, the deamination product of histidine, exists in two isomeric forms; the trans isomer
is converted to cis upon UVR exposure. The amino acids tryptophan and tyrosine absorb
UVR through the epidermis. Melanins are produced by melanocytes and are transferred
to keratinocytes; they absorb broadly over the UVR spectrum (IARC 1992).
The depth to which UVR penetrates the human skin also is wavelength dependent. The
atmosphere filters out UVC, the shortest wavelength produced by sunlight and the most
potentially harmful to the genome, before it reaches the earth's surface. Therefore, UVC
plays only a minimal role in biological photochemical reactions. UVC produced by
artificial sources and reaching the skin can penetrate only the epidermis. UVB has the
potential to penetrate the epidermis and upper layer of the dennis, or papillary dermis.
Although UVB makes up only 5% of the UV photons reaching the earth's surface, it is
the most biologically important component of sunlight. UV A, with the longest
wavelength, reaches the deeper layer of the dermis, or reticular dennis (Table 6-1)
(Gould et al. 1995, Farmer and Naylor 1996).
73
Dec. 2000
Table 6-1 Characteristics of UVR
Type of UV
8
Percent of solar radiati on Wavelength Depth of skin
reaching earth's surface (nm) penetrance
UVA 6.3 3 2 ~ papillary, reticular
dermis
UVB 1.7 290-320 epidermis,
papillary dermis
uvc 0 200-290 epidennis
Source: Gould et al. 1995
wavelength classifications ofUV A, UVB, and UVC are slightly different from the CIE designations.
6.2 Mechanisms of UV -induced skin cancer
The frequency of skin cancer, including melanoma, has increased dramatically over the
past 40 years and currently accounts for about 40% of all cancer in the United States
(Whittaker 1996, Quinn 1997, Gilchrest et al. 1999). Although the reasons for this rapid
increase in skin cancer are not completely understood, increased exposure to solar
radiation and altered patterns of sun exposure are strongly implicated (Stary et al. 1997,
Gilchrest et al. 1999). Aging is an imp01tant risk factor for skin cancer; however, even
when age is excluded, UVR emerges as a primary factor in the three major types of skin
cancer (squamous-cell carcinoma, basal-cell carcinoma, and melanoma) (Gilchrest et al.
1999).
Risk factors for non-melanoma skin cancer and melanoma are different (Ablett et al.
1998, Gilchrest et al. 1999). Squamous-cell and basal-cell carcinomas are most often
found on areas ofthe body receiving maximum sun exposure (i.e., face, forearms, and
backs of hands). In these cases, total cumulative UVR exposure is an important risk
factor. Melanoma appears to be related to intense, intermittent UVR exposure. Several
recent reviews and studies support UVR exposure as an important risk factor for
melanoma (see Section 3 for a discussion ofhuman studies on intermittent sun exposure
and melanoma). Riinger (1999) suggested that UV A may play an important role in the
pathogenesis of malignant melanoma. UV A induces melanoma in the platyfish-swordtail
hybrid fish model and melanoma hyperplasia in the opossum Monodelphis domestica
(see Section 4). Atillasoy et al. (1998) reported that chronic UVB irradiation, with or
without an initiating carcinogen, could induce melanoma (see Section 4).
Human-mouse chimera models, in which human skin is grafted onto SCID/RAG mice,
have been used to study etiological factors important in the genesis of human tumors
(Satyamoorthy et al. 1999). UVB in combination with DMBA induced precancerous
lesions and invasive squamous-cell carcinoma and melanoma (see Section 4.5).
UVR produces both direct and indirect damage to DNA that may alter gene expression
and lead to mutations in protooncogenes and tumor suppressor genes. Ifunrepaired, these
lesions can result in cancer. Other factors (e.g., immunological responses, antioxidant
defenses, and genetic predisposition) also are important considerations (Streilein et al.
1994, Sarasin 1999). The evidence for DNA damage, DNA repair, and
74
Dec. 2000
immunosuppression as important mechanisms in UVR carcinogenesis is reviewed in the
following sections.
6.2. 1 DNA damage
UVR damage to biological systems occurs via phototoxic reactions that are either direct
or mediated by photosensitizers in the target tissues (Cadet eta/. 1997). In the skin, the
effects of UVR are mediated by photosensitization reactions characterized by structural
and functional changes in keratinocytes, melanocytes, Langerhans cells, and fibroblasts
(Pathak 1996). The mechanism of UVR-induced DNA damage differs distinctly with
wavelength (Cadet et aL 1997, lto and Kawanishi 1997) (Figure 6-1). Damage to DNA
by UVA proceeds indirectly via photosensitizers (non-DNA molecules) in
photosensitization reactions, because DNA does not readily absorb UV A. In contrast,
wavelengths shorter than 320 run (UVB, UVC) directly photoactivate the DNA molecule
to generate mainly pyrimidine photoproducts. Direct and indirect mechanisms of DNA
damage are discussed below.
Wavelength
(nm)
visible j
400-
UVA
320-
UVB
280
-1
uvc
Chromophore
Process
Damage
Photosensitizer (P)
_........ .,. DNA strand breaks
po + 0 )'r 0
2
. - .,. Fenton
/ reaction -- ---. oxidized bases
3p Type
1
... Electron abstraction ... oxidized bases (8-oxoGua)
' Q2
Type ~ 10
2
(singlet oxygen) __ .,. 8-oxoGua
p
hv
1p
electron transfer
_ .,. 8-oxoGua
DNA bases
___ hv __ .,. photohydration ... cytosine hydrates
photoaddition
dimerization
...
Pyr(6-4)Pyo + Dewar
Pyr<>Pyr
Thy-Ade
250-
hv
_ .,. DNA -
pyrimidine dimer (6-4)
photoproduct
P= photosensitizer, hu =radiation, Pyr = pyridine, Pyo = pyrimidone, Thy = thymidine, Ade = adenine,
Gua = guanine.
Source: Cadet et a!. 1997, lto and Kawanishi 1997
Figure 6-1. Mechanisms ofUV-induced DNA damage
75
Dec. 2000
6.2. 1. 1 Direct mechanisms
DNA is the primary cellular chromophore for UVB (Griffiths et al. 1998). The direct
excitation of DNA bases by the UVB component ofUVR gives rise, predominantly
through oxygen-independent reactions, to three base modifications: cyclobutane-type
pyrimidine dimers, pyrimidine (6-4) pyrimidone photoproducts, and related Dewar
isomers (Cadet et al. 1997, Griffiths et al. 1998, Sarasin 1999). These mutagenic
photoproducts result in C to T or CC to TT transitions that are always at pyrimidine-
pyrimidine sequences and are now considered a signature of sun exposure (Sarasin 1999).
In addition, UVB may cause some DNA strand breaks and single-base lesions. The yield
of strand breaks increases with increasing wavelength, and single-base lesions are
primarily thymine glycols (Griffiths et al. 1998).
Although UVB is more effective than UV A in generating direct DNA damage, UV A
does induce some direct damage. Young et al. (1998) demonstrated that both UV A (320
to 360 nm) and UVB (300 nm) readily induced thymine dimers in both melanocytes and
keratinocytes from human skin that was not exposed to sunlight. Furthermore, these data
showed that thymine dimer levels in melanocytes were comparable to those observed in
keratinocytes.
Like UVB, UVC causes direct excitation of DNA bases, through oxygen-independent
reactions, leading mainly to formation of dimeric pyrimidine photo lesions and relatively
minor yields of DNA photoproducts that include the thymine-adenine photo-adducts, the
"cytosine photohydrates" (Herrlich et al. 1994, Cadet and Vigny 1990, cited in Cadet et
al. 1997), and a few purine decomposition products (Cadet et al. 1997).
Formation of 5,6-dihydroxydihydrothymine-type lesions (thymine glycols) in DNA
following UVC irradiation also have been observed. It has been suggested that this
photoproduct arises from the action ofUVR-produced hydroxyl radicals (Hariharan and
Cerruti 1976, 1977, cited in IARC 1992). Thymine compounds irradiated with UVC in
the frozen state rapidly lose their absorption (Beukers et al. 1958, cited in IARC 1992); a
dimer of thymine (two molecules linked by a cyclobutane ring involving the 5 and 6
carbon atoms) was shown to be responsible for the loss of absorption (Beukers and
Berends 1960, Wulff and Fraenkel 1961, cited in IARC 1992). Continued irradiation
leads to a wavelength-dependent equilibrium between dimer fonnation and dimer
splitting to reform the monomer. Dimer formation is favored where the ratio of the dimer
to monomer absorbency is relatively small (at wavelengths> 260 nm), whereas
monomerization is favored at shorter wavelengths (around 240 nm), where the ratio is
larger (Johns et al. 1962, cited in IARC 1992).
Pigmented mouse melanocytes, melan-b (brown) and melan-a (black), were more
resistant than melan-c (albino) melanocytes to being killed by UVC or UVA, but were
less resistant to being killed by UVB or UV A + UVB. In both the melanocytes and
mouse melanoma cells, more pyrimidine dimer DNA damage was observed in pigmented
cells than in nonpigmented cells. These results indicate that pigment does not protect
against direct DNA damage in the form of pyrimidine dimers, nor does it necessarily
protect against cell death (Hill et al. 1997).
76
Dec. 2000
6.2. 1.2 Indirect mechanisms
In vitro experiments have firmly established that UV A is genotoxic by indirect
mechanisms. Endogenous chromophores (photosensitizers) for UV A include riboflavin,
porphyrins, quinones, tryptophan, and reduced nicotinamide cofactors (NADH and
NADPH) (Ito and Kawanishi 1997, Cadet et al. 1997, Griffiths et al. 1998). The effects
of exogenous photosensitizers, such as psoralens, porphyrins, coal tar, some antibiotics,
and some nonsteroidal anti-inflammatory agents (Gould et al. 1995), are outside the
scope of this document and are not discussed.
Following absorption ofUV A, chromophores generate reactive oxygen species and
radicals that can damage DNA (Griffiths et al. 1998). There are two competitive
photosensitized reactions: type I reactions do not require oxygen and produce a radical
intennediate via an electron transfer, whereas type II reactions require oxygen and
produce singlet oxygen eo
2
) (Ito and Kawanishi 1997).
Griffiths et al. (1998) reviewed indirect mechanisms ofUV A-induced DNA damage.
UV A interactions with photosensitizers in the target tissues promote the formation of
three base lesions, as well as base loss (at a much lower level). One base lesion is 8-
0HdG, the formation of which from guanosine appears to be mediated by singlet oxygen
and is reported to be induced by UV A in mammalian cells at 10 times the rate of DNA
strand breaks. Another base lesion is isomeric hydroxyhydroperoxides, produced through
indirect generation of the radical cation of thymine in the presence of oxygen. The third
base lesion is pyrimidine photoproducts; however, UV A generates this type of lesion
much less efficiently than does UVB.
For both type I and type II mechanisms, 8-0HdG appears to be the major oxidation
product of guanine in DNA (Ito and Kawanishi 1997). Peak et al. (1990, cited in Ito and
Kawanishi 1997) reported the formation of H
2
0
2
in human cells exposed to UV A.
Neither 02- nor H20 2 can cause DNA damage in aqueous solution. However, in the
presence of metal ions, highly reactive species, such as the hydroxyl radical (OH) and
metal-oxygen complexes, can be generated via metal-catalyzed reactions. Hydroxyl
radicals generated from the Fenton reaction of iron with H
2
0
2
may react with any of the
bases and sugar moieties of DNA (Cadet et al. 1997, Ito and Kawanishi 1997).
6.2.2 DNA repair
Yarosh and Kripke (1996, cited in NTP 1997) found that UV -induced DNA
photoproducts produced a variety of cellular responses contributing to skin cancer.
Unrepaired DNA photoproducts cause the release of cytokines that contribute to tumor
promotion, tumor progression, immunosuppression, and the induction of latent viruses.
DNA repair enzymes are an important gene protection mechanism, because they can
repair DNA photoproducts and block the carcinogenic responses triggered by cytokines.
See Sections 3 and 5 for discussion of xeroderma pigmentosum patients and the role of
DNA repair capacity in skin cancer.
77
Dec. 2000
6.2.3 Mutations
The photoproducts formed from UVR exposure as a result of DNA damage have varying
mutagenic potentials. Cyclobutane-type thymine dimers, the major UVR photoproducts,
are only weakly mutagenic (Banerjee et al. 1988, 1990, both cited in IARC 1992),
whereas the less common (6-4) thymine-thymine photoproduct is highly mutagenic
(LeClerc et al. 1991, cited in IARC 1992). UVR-induced cyclobutane dimer formation is
directly involved in UVR carcinogenesis. Such dimers prevent gene transcription.
Malignant transformation of the cell may result when the affected gene is a growth-
regulating gene, such as an oncogene or tumor suppressor gene. DNA repair mechanisms
include excision repair and photoreactivation. In the latter, the photoreactivating enzyme
repairs UVR-induced cyclobutane dimers and (6-4) photoproducts; the enzyme is
activated by UV A and visible light. Thus, photoreactivation repair of cyclobutane dimers
effectively reduced the incidence ofUVR-induced tumors in the opossum M. domestica
(Ley et al. 1991, cited in Grabbe and Granstein 1994).
The mutagenicity also varies with the type ofUVR. Peak et al. (1987, cited in Robert et
al. 1996) found that the frequency of single-strand breaks per genome per lethal event
was higher upon exposure of a human teratoma cell line to UV A than to UVB and/or
UVC. This is consistent with the finding that UV A induces a greater proportion of
rearrangements than UVB, 39% vs. 24%, possibly as a result of repair of single-strand
breaks (Robert et al. 1996).
6.2.4 Tumor suppressor gene expression and mutation
Loss of p53 function is an important factor in multistep carcinogenesis. Burren et al.
(1998) exposed human skin to sunlight and analyzed the skin for p53 expression and
pyrimidine dimers. The exposed human skin showed increased levels of pyrimidine
dimers and p53 protein expression. These effects varied according to the dose and
wavelength ofUVR. At equivalent biological doses, p53 expression was twice as high
after exposure to simulated solar radiation than after exposure to UV A. At lower doses of
UV A, expression of p53 was limited to the basal-cell keratinocytes; however, at higher
doses, all layers of the epidermis were affected. The researchers found that even sub-
erythemal doses of simulated solar radiation induced both pyrimidine dimers and p53
expression in human skin in situ (Burren et al. 1998).
Berg et al. (1996, cited in Griffiths et al. 1998) unequivocally demonstrated that
constitutive p53 tumor suppression gene product alterations are an early event in the
induction of skin cancer and are causally linked to UVB exposure. Sequencing data from
a large number of skin tumors showed that p53 was mutated in over 90% of squamous-
cell carcinomas (Brash et al. 1991, Ziegler et al, 1993, Wikonkal et al. 1997, cited in
Wikonkal and Brash 1999). These p53 mutations were found in 74% of sun-exposed
normal skin, compared with 5% in unexposed skin, indicating a strong association with
sun exposure. The majority of the mutations were C to T transitions occurring at
dipyrimidine sites, with single C to T transitions occurring in 70% of the cases and
tandem CC to TT in 10% of the cases, suggesting a causal relationship between
pyrimidine photoproducts and UVB carcinogenesis. The p53 tumor suppression gene
product is involved in cell-cycle regulation and is responsible for initiating cell apoptosis.
78
Dec. 2000
Lack of p53 tumor suppressor gene product results in failure to arrest the cell cycle in G
1
phase or to initiate the apoptotic pathway of cell death. Attempts by cells to replicate the
damaged genome will result in accumulated mutations that will, in turn, contribute to
genomic instability and reduced efficiency of DNA repair, leading to carcinogenesis
(Hanawalt 1996). Although detection of p53 mutations in skin tumor cells suggests that
p53 mutations are involved in some malignant melanomas, the role of p53 mutations in
melanoma may not be as large as their roles in skin basal-cell carcinoma or squamous-
cell carcinoma (Griffiths et al. 1998).
Sarasin (1999) reported that the PTCH tumor suppressor gene might have a role in skin
cancer development. This gene is involved in signal transduction related to cell
development and differentiation. Point mutations in PTCH were found in patients with
Godin's syndrome (nevoid basal-cell carcinoma syndrome), who have a high incidence
ofbasal-cell carcinomas; in 30% to 60% ofbasal-cell carcinomas from DNA-repair-
proficient individuals; and in 50% to 80% of basal-cell carcinomas from xeroderma
pigmentosum patients.
6.2.5 Immunosuppression
Exposure to solar radiation and UVR has altered immune function in experimental
animals and humans (IARC 1992). Studies of patients with DNA repair disorders such as
xeroderma pigmentosum, cockayne syndrome, and sun-sensitive trichothiodystrophy
have shown that DNA repair defects and elevated levels of sunlight-induced mutations in
the skin are insufficient to explain the high incidence of skin cancer in xeroderma
pigmentosum patients. Therefore, UVR-induced mutations in critical genes may be
necessary but not sufficient for skin cancer (Bridges 1998). Immunosuppression has been
suggested as a possibly important tumor-controlling mechanism (Quinn 1997, Bridges
1998, Sarasin 1999).
A study of mice with a defective XP A gene showed the full XP phenotype. These mice
were hypersensitive to UVB and showed several immunological defects similar to those
seen in human xeroderma pigmentosum patients (Bridges 1998). Quinn (1997) noted
several other findings indicating that immunosuppression is related to skin cancer
incidence: ( 1) immunosuppressed organ transplant recipients showed a marked increase
in skin cancer, particularly squamous-cell carcinoma, (2) UVR decreased the ability to
mount a delayed-type hypersensitivity response, and (3) mice exposed to low levels of
UVR failed to reject highly immunogenic tumor cell lines.
UVB increases tumor necrosis factor, which may suppress the function of a neoplastic
population of clonal T -cells in the skin, in a process mediated by urocanic acid and
serving as an immune upregulator. Urocanic acid, one of the main chromophores present
in the skin, exists in two isomeric forms, trans and cis. UVB converts trans-urocanic acid
into cis-uracanic acid, which is reported to be immunosuppressive (Streilein 1993,
Streilein et al. 1994, Hemnann et al. 1995). cis-Urocanic acid is thought to exert its
immunosuppressive action by causing a local accumulation of tumor necrosis factor-a
(Streilein et al. 1994), in turn preventing normal induction of contact hypersensitivity in
the skin (Streilein 1993, Cadet et al. 1997). Pre-irradiation of mice with low doses of
UVB (100 to 700 J/m
2
of fluorescent sunlamp radiation daily for four hours) suppressed
79
Dec. 2000
the development of contact hypersensitivity to sensitizing chemicals (such as 2,4-
dinitrofluorobenzene) subsequently applied to the irradiated skin (Toews et al. 1980,
Elmets et al. 1983, cited in IARC 1992). Local suppression of contact hypersensitivity by
UVB radiation also was observed in hamsters (Streilein and Bergstresser 1981, cited in
!ARC 1992).
UVB radiation decreases the alloactivating and antigen-presenting capacity of
Langerhans cells and increases interleukin-2 and interleukin-6 production by human
keratinocytes (Herrmann et al. 1995). In UV-irradiated skin cells, cell markers for
Langerhans cells are diminished. In concert with and because of the resultant abrogation
of the antigen-presenting function of Langerhans cells in these skin cells, suppressor T-
cell activation and tolerance to antigen results in immunosuppression. Such
immunosuppression has resulted in the growth of immunogenic neoplasms in mice and
may facilitate the growth of human neoplasms (Baadsgaard 1991 ).
6.3 Initiation and promotion
The evidence indicates that UVR is a complete carcinogen; that is, it both initiates and
promotes carcinogenesis (Matsui and DeLeo 1991, IARC 1992, Soballe et al. 1996,
Wikonkal and Brash 1999). The carcinogenic effects of UVR have been attributed largely
to UVB, which has been reported to be at least 5,000 times more effective as a complete
carcinogen than UV A (Forbes 1985, cited in Matsui and DeLeo 1991). However, in some
animal studies, UVA administered alone hasinduced skin cancer (see Section 4.2).
Matsui and DeLeo (1991) reviewed the evidence that UV A acts as a classic promoter and
discussed possible mechanisms. UV A was shown to promote squamous-cell carcinoma in
albino hairless mice. A constant dose ofUV A was least effective in inducing cancer, and
a regimen ofUVA plus UVB was most effective. Other studies indicated that UVA (320
to 400 nm) induced responses in vivo and in cultured mammalian cells similar to
treatment with the phorbol ester tumor promoter 12-0-tetradecanoylphorbol-13-acetate.
Current evidence indicates that UVA's promotional effects are through modulation of
protein kinase C, whereas UVB and UVC do not affect protein kinase C activity. UV A
may also promote carcinogenesis through mechanisms involving reactive oxygen species
(de Laat and de Gmijl 1996).
6.4 Summary
UV A is the most abundant component of UVR that reaches the surface of the earth.
Although UVB is partially filtered out by the atmosphere, it is the most biologically
significant component of solar UVR reaching the earth's surface, because it is absorbed
by biologically critical targets in the skin, such as DNA. UVR may be transmitted,
reflected, scattered, or absorbed by tissue chromophores in a wavelength- and
chromophore-dependent manner. UVB and UVC induce damage to biological systems
directly, whereas UVA-induced damage is indirect, mediated via endogenous
photosensitizers in the target tissues in photodynamic or nonphotodynamic phototoxic
reactions. These reactions result in damage to DNA (base mutations and dimerizations,
strand breaks, and DNA-protein crosslinks for UV A; base dimerizations and strand
breaks for UVB; and base dimerizations and glycol formation, strand breaks, and
elevation of gene transcription for UVC). UVB causes skin cancer via mechanisms that
80
Dec. 2000
include DNA damage, immunosuppression, tumor promotion, and mutations in the p53
gene. There is some evidence that UV A, under certain conditions, may act as a complete
carcinogen; however, there is more evidence that UV A acts as a tumor promoter. UVC
radiation is filtered by the earth's atmosphere and does not occur in sunlight. UVC is
known to cause direct damage to DNA, as does UVB; therefore, its potential role in
human carcinogenicity would result from exposure to artificial sources ofUVR, such as
germicidal lamps, rather than sunlight.
81
Dec. 2000
82
Dec. 2000
7 References
1. Ablett,E., J.Pedley, P.A.Dannoy, R.A.Sturm P.G.Parsons. (1998). UVB-specific
regulation of gene expression in human melanocytic cells: cell cycle effects and
implication in the generation of melanoma. Mutat Res 422:31-41.
2. ACGIH. 1996. Threshold limit values for chemical substances and physical
agents and biological exposure indicies. American Conference of Governmental
lndustrial Hygienists, Cincinnati, OH.
3. Adam,S.A., J.K.Sheaves, N.H. Wright, G.Mosser, R.W.Harris, and M.P.Vessey.
(1981 ). A case-control study of the possible association between oral
contraceptives and malignant melanoma. Br J Cancer 44:45-50.
4. Adami,J., G.Gridley, O.Nyren, M.Dosemeci, M.Linet, B,Glimelius, A.Ekbom,
S.H.Zahm. (1999). Sunlight and non-Hodgkin's lymphoma: a population-based
cohort study in Sweden. Int J Cancer 80:641-645.
5. Armstrong,B.K. and A.Kricker. ( 1996). Epidemiology of sun exposure and skin
cancer. Cancer Surv 26:133-153.
6. Armstrong,J.D. and B.A.Kunz. ( 1992). Photoreactivation implicates cyclobutane
di.mers as the major promutagenic UVB lesions in yeast. Mutat Res 268:83-94.
7. Armstrong,J.D. and B.A.Kunz. (1990). Site and strand specificity ofUVB
mutagenesis in the SUP4-o gene of yeast. Proc Natl Acad Sci USA 87:9005-9009.
8. Atillasoy,E.S., R.Elenitsas, E.R.Sauter, P.W.Soballe, M.Herlyn. (1997). UVB
induction of epithelial tumors in human skin using a RAG-1 mouse xenograft
model. J Invest Dermatol 109:704-9.
9. Atillasoy,E.S., J.T.Seykora, P.W.Soballe, R.Elenitsas, M.Nesbit, D.E.Elder,
K.T.Montone, E.Sauter, and M.Herlyn. (1998). UVB induces atypical
melanocytic lesions and melanoma in human skin. Am J Pathol 152: 1179-86.
10. Autier,P., M.Joarlette, F.Lejeune, D.Lienard, J.Andre and G.Achten. (1991).
Cutaneous malignant melanoma and exposure to sunlamps and sunbeds: a
descriptive study in Belguim. Melanoma Res 1:69-74.
11. Autier,P., J.F.Dore, F.Lejeune, K.F.Koelmel, O.Geffeler, P.Hille, J.P.Cesarini,
D.Lienard, A.Liabeuf, M.Joarlette, P.Chemal, K.Hakim, A.Koeln, and
U.R.K.leeberg. (1994). Cutaneous malignant melanoma and exposure to sunlamps
or sunbeds: an EORTC multicenter case-control study in Belgium, France and
Germany. EORTC Melanoma Cooperative Group. lnt J Cancer 58:809-813.
12. Baadsgaard,O. (1991). In vivo ultraviolet irradiation ofhuman skin results in
profound perturbation of the immune system. Relevance to ultraviolet-induced
skin cancer [published erratum appears in Arch Dermatol1991 Sep;127(9):1368].
Arch Dermatoll27:99-109.
13. Bajdik,C.D., R.P.Gallagher, G.Astrakianakis, G.B.Hill, S.Fincham, and
D.I.McLean. (1996). Non-solar ultraviolet radiation and the risk of basal and
squamous cell skin cancer. Br J Cancer 73:1612-1614.
83
Dec. 2000
14. Banerjee,S.K., R.B.Christensen, C.W.Lawrence, and J.E.LeClerc. (1988).
Frequency and spectrum of mutations produced by a single cis-syn thymine
cyclobutane dimer in a single-stranded vector. Pro. Nat. Aca. Sci USA 85:8141-
8145.
15. Banerjee,S.K., A.Borden, R.B.Christensen, J.E.LeClerc, and C.W.Lawrence.
(1990). SOS-dependent replication past a single trans-syn T-T cyclobutane dimer
gives a different mutation spectrum and increased error rate compared to
replication past this lesion in uninduced cells. J Bacterial. 172:2105-2112.
16. Bech T.N., H.C.Wulf, T.Poulsen, and K.Lundgren. (1988). Pretreatment with
long-wave ultraviolet light inhibits ultraviolet-induced skin tumor development in
hairless mice. Arch Dermatol124: 1215-1218.
17. Bech-Thomsen,N. ( 1997). Aspects of exposure to UV A tanning sources.
Carcinogenic effect in mice and melanogenic effect in man and mice. Dan Med
Bull 44:242-256.
18. Bech-Thomsen,N. and H.C.Wulf. (1993). Carcinogenic potential of fluorescent
UV tanning sources can be estimated using the CIE erythema action spectrum. lnt
J Radiat Bioi 64:445-450.
19. .Bech-Thomsen,N., H.C.Wulf, T.Poulsen, F.G.Christensen, and K.Lundgren.
(1991 ). Photocarcinogenesis in hairless mice induced by ultraviolet A tanning
devices with or without subsequent solar-simulated ultraviolet irradiation.
Photodermatol Photoimmunol Photomed 8:139-145.
20. Bech-Thomsen,N., T.Poulsen, F.G.Christensen, K.Lundgren, and H.C.Wulf.
( 1992). UV A tanning devices interact with solar-simulated UV radiation in skin
tumor development in hairless mice. Arch Dermatol Res 284:353-357.
21. Beitner,H., S.E.Norell, U.Ringborg, G.Wennersten, and B.Mattson. (1990).
Malignant melanoma: aetiological importance of individual pigmentation and sun
exposure. Br J Dermatol 122:48-51.
22. Bentham, G. and A.Aase. (1996). Incidence of malignant melanoma of the skin in
Norway, 1955-1989: associations with solar ultraviolet radiation, income and
holidays abroad. Tnt J Epidemiol 25: 1132-1138.
23. Beral,V, S.Evans, H.Shaw, and G.Milton. (1982). Malignant melanoma and
exposure to fluorescent lighting at work. Lancet 2(8293):290-3.
24. Berg,R.J., H.J.van Kranen, H.G.Rebel, A.de Vries, W.A.van Vloten, C.F.Van
Kreijl, J.C.van der Leun, and F.R.de Gruijl. (1996). Early p53 alterations in
mouse skin carcinogenesis by UVB radiation: immunohistochemical detection of
mutant p53 protein in clusters of preneoplastic epidermal cells. Proc Natl A cad
Sci U.S.A. 93:274-278.
25. Beukers,R., J.Ylstra, and W . .Berends. (1958). The effect of ultraviolet light on
some components of nucleic acids. II. In rapidly frozen solutions. Reel Trav chlim
Pays-Bas 77:729-732.
84
Dec. 2000
26. Beukers,R. and W.Berends. (1960). Isolation and identification of the irradiation
product of thymidine. Biochim Biophys Acta 41:550-551.
27. Bhate,S.M., G.R.Sharpe, J.M.Marks, S.Shuster, and W.M.Ross. (1993).
Prevalence of skin and other cancers in patients with psoriasis. Clin Exp Dermatol
18:401-4.
28. Blum,H.F. (1959) Carcinogenesis by Ultraviolet Light, Princeton, NJ:Princeton
University Press.
29. Blum,H.F. and S.W.Lippincott. (1942). Carcinogenic effectiveness of ultraviolet
radiation ofwavelength 2537 d. J Nat/ Cancer lnst 3:211-216.
30. Boettner,E.A. and J.R.Wolter. (1962). Transmission of the ocular media. Invest
Ophthalmol1:776-783.
31. Brash,D.E., J.A.Rudolph, J.A.Simon, A.Lin, G.J.McKenna, H.P.Baden,
A.J.Halperin, and J.Ponten. (1991). A role for sunlight in skin cancer: UV-
induced p53 mutations in squamous cell carcinoma. Proc Natl Acad Sci USA
88:10124-10128.
32. Bridges,B.A. (1998). UV-induced mutations and skin cancer: how important is
the link? Mutat Res 422:23-30.
33. Bridges,B. and G.Strauss. (1980). Possible hazards of photochemotherapy for
psoriasis. Nature 283:523-524.
34. Bruynzeel,I., W.Bergrnan, H.M.Hartevelt, C.C.Kenter, V.van d, A.A.Schothorst,
and D.Summond. (1991 ). "High single-dose" European PUVA regimen also
causes an excess of non-melanoma skin cancer. Br J Dermatol124:49-55.
35. Burren,R., C.Scaletta, E.Frenk, R.G.Panizzon, L.A.Applegate. (1998). Sunlight
and carcinogenesis: expression of p53 and pyrimidine dimers in human skin
following UV A I, UV A I + li and solar simulating radiations. Int J Cancer
76:201-6.
36. Bykov,V.J. and K.Hemminki. (1996). Assay of different photoproducts after
UVA, Band C irradiation of DNA and human skin explants. Carcinogenesis
17:1949-1955.
37. Cadet,J. and P.Vigny. (1990). The photochemistry of nucleic acids. In:
Bioorganic Photochem.istty, H. Morrison, ed. (New York: Wiley and Sons), pp 1-
272.
38. Cadet,J., C.Anselmino, T.Douki, and L.Voituriez. (1992). Photochemistry of
nucleic acids in cells. J Photochem Photobiol B 15:277-298.
39. Cadet,J., M.Berger, T.Douki, B.Morin, S. Raoul, J.L.Ravanat, and S.Spinelli.
(1997). Effects ofUV and visible radiation on DNA-final base damage. Bioi
Chern 378:1275-1286.
40. Cameron,G.S. and B.C.Pence. (1992). Effects of multiple applications of tumor
promoters and ultraviolet radiation on epidennal proliferation and antioxidant
status. J Invest Dermatol 99:189-192.
85
Dec. 2000
41. Cebula,T.A., E.N.Henrikson, P.E.Hartman, and W.H.Biggley. (1995). Reversion
profiles of cool white fluorescent light compared with far ultraviolet light:
homologies and differences. Photochem Photobiol6l :353-359.
42. Chadwick,C.A., C.S.Potten, O.Nikaido, T.Matsunaga, C.Proby, and A.R.Young.
(1995). The detection of cyclobutane thymine dimers, (6-4) photolesions and the
Dewar photoisomers in sections ofUV-irradiated human skin using specific
antibodies, and the demonstration of depth penetration effects. J Photochem
Photobiol B 28:163-170.
43. Chen,Y.T., R.Dubrow, T.Zheng, R.L.Bamhill, J.Fine, and M.Berwick. (1998).
Sunlamp use and the risk of cutaneous malignant melanoma: a population- based
case-control study in Connecticut, USA. Int J Epidemiol27:758-765.
44. Cleaver,J.E. and K.Kraemer. ( 1989). Xeroderma pigmentosum. In The Metabolic
Basis of Inherited Diseases II, C.R.Scriver, A.L.Beaudet, W.S.Sly, and D. Valle,
eds. McGraw-Hill, New York. pp 2949-2971.
45. Clingen,P.H., C.F.Arlett, J.Cole, A.P.Waugh, J.E.Lowe, S.A.Harcourt,
N.Hermanova, L.Roza, T.Mori, and O.Nikaido. (1995). Correlation ofUVC and
UVB cytotoxicity with the induction of specific photoproducts in T-lymphocytes
and fibroblasts from normal human donors. Photochem Photobiol 61: 163-170.
46. Corominas M, H.Kamino, J.Leon, and A.Pellicer. (1989). Oncogene activation in
human benign tumors of the skin (keratoacanthomas): is H-ras involved in
differentiation as well as proliferation? Proc Nat/ Acad Sci USA 86:6372-6376.
47. Currie C.L. and B.E. Monk. (2000). Welding and non-melanoma skin cancer. Clin
Exp Dermatol25(1):28-9.
48. Daya-Grosjean,L., N.Dumaz, and A.Sarasin. (1995). The specificity ofp53
mutation spectra in sunlight induced human cancers. J Photochem Photobiol B
28:115-124.
49. de Gruijl,F.R. , H.J.C.M.Sterenborg, P.D.Forbes, R.E.Davies, C.Cole, G.Kelfkens,
H.van Weelden, H.Slaper, and J.C.van der Leun. (1993). Wavelength Dependence
of Skin Cancer Induction by Ultraviolet Irradiation of Albino Hairless Mice.
Cancer Research 53:53-60.
50. de Gruij l,F.R., J.B.Van Der Meer, and J.C.van der Leun. (1983). Dose-time
dependency of tumor formation by chronic UV exposure. Photochem Photobiol
37:53-62.
51. de Laat,J.M., and F.R.de Gruij l. (1996). The role ofUV A in the aetiology of non-
melanoma skin cancer. Cancer Surv 26:173-91.
52. de Laat,A., J.C.van der Leun, and F.R.de Gruijl. (1997). Carcinogenesis induced
by UV A (365-nm) radiation: the dose-time dependence of tumor formation in
hairless mice. Carcinogenesis 18(5): 1013-1020.
53. Dom,C.R., D.O.Taylor, and R.Schneider. (1971). Sunlight exposure and risk of
developing cutaneous and oral squamous cell carcinomas in white cats. J Natl
Cancer lnst 46:1073-8.
86
Dec. 2000
54. Dubin,N., M.Moseson, and B.S.Pasternack. (1989). Sun exposure and malignant
melanoma among susceptible individuals. Environ Health Perspect 81: 139-151.
55. Dunn-Lane,J., B.Herity, M.J.Moriarty, and R.Conroy. (1993). A case control
study of malignant melanoma. Irish Med J 86:57-59.
56. Elmets,C.A., P.R.Bergstresser, R.E.Tigelaar, P.J.Wood, and J.W.Streilein. (1983).
Analysis of the mechanism of unresponsiveness produced by haptens painted on
skin exposed to low dose ultraviolet radiation. J Exp Med 158:781-794.
57. Elwood,J.M. (1996). Melanoma and sun exposure. Semin Oncol23:650-666.
58. Elwood,J.M. and.l.Jopson. (1997). Melanoma and sun exposure: an overview of
published studies. lnt J Cancer 73:198-203.
59. Elwood J.M., C.Williamson, and P.J.Stapleton. (1986). Malignant melanoma in
relation to moles, pigmentation, and exposure to fluorescent and other lighting
sources. Br J Cancer 53:65-74.
60. Emmett,E.A. (1973). Ultraviolet radiation as a cause of skin tumors. CRC Crit
Rev Toxicol2:211-255.
61. English,D.R., B.K.Atmstrong, A.Kricker, C.Fleming. (1997). Sunlight and
cancer. Cancer Causes Control8:271-283.
62. Epstein,J.H. (1978). Photocarcinogenesis: a review. Natl Cancer lnst Monogr 13-
25.
63. Epstein,J.H. (1985). Animal models for studying photocarcinogenesis. In: Models
in Dermatology, H.Maibach and N.Lowe, eds. , Vol.2, (Basel, Karger), pp 303-
312.
64. Fanner,K.C. and M.F.Naylor. (1996). Sun exposure, sunscreens, and skin cancer
prevention: a year-round concern. Ann Pharmacother 30:662-673.
65. FDA. (1996). http://www.fda.gov/fdac/features/496_sun.html.
66. Findlay,G.M. ( 1930). Cutaneous papillomata in the rat following exposure to
ultra-violet light. Lancet i: 1229-1231.
67. Forbes,P.D., H.F.Blum, and R.E.Davies. (1981). Photocarcinogenesis in hairless
mice: Dose-response and the influence of dose-delivery. Photochem Photobiol
34:361-365.
68. Forbes,P.D., R.E.Davies, F.Urbach, D.Berger, and C.Cole. (1982). Simulated
stratospheric ozone depletion and increased ultraviolet radiation: effects on
photocarcinogenesis in hairless mice. Cancer Res 42(7):2796-803.
69. Forbes P.D. (1985). UV-A and photocarcinogenesis. Photochem Photobiol42:7S.
70. Forman,A.B., H.H.Roenigk Jr, W.A.Caro, and M.L.Magid. (1989). Long-term
follow-up of skin cancer in the PUV A-48 cooperative study. Arch Derm.atol
125:515-9.
87
Dec. 2000
71. Franklin,W.A. and W.A.Haseltine. (1984). Removal ofUV light-induced
pyrimidine-pyrimidone(6-4) products from Escherichia coli DNA requires the
uvrA, uvrB, and urvC gene products. Proc Nat/ Acad Sci U.S.A. 81:3821-3824.
72. Freedman,D.M., S.H.Zahm, and M.Dosemeci. (1997). Residential and
occupational exposure to sunlight and mortality from non-Hodgkin's
lymphoma:composite (threefold) case-control study. BMJ 314:1451-1455.
73. Freeman,R.G. and J.M.Knox. (1964). Ultraviolet-induced corneal tumors in
different species and strains of animals. J Invest Dermatol 90: 814-817.
74. Gallagher,R.P., J.M.Elwood, and G.B.Hill. (1986). Risk factors for cutaneous
malignant melanoma: the western Canada melanoma study. Recent Results
Cancer Res 102:38-55.
75. Garbe,C., J.Weiss, S.K.riiger, E.Garbe, P.Biittner, J.Bertz, H.Hoffmeister,
I.Guggenmoos-Holzmann, E.G.Jung, and C.E.Orfanos. (1993). The German
melanoma registry and environmental risk factors implied. Recent Results Cancer
Res 128:69-89.
76. Gasparro,F.P. (2000). Sunscreens, skin photobiology, and skin cancer: the need
for UV A protection and evaluation of efficacy. Environ Health Perspect 108
(Suppl 1):71-8.
77. Gilchrest,B.A., M.S.Eller, A.C.Geller, and M.Yaar. (1999) The pathogenesis of
melanoma induced by ultraviolet radiation. N Eng! J Med 340:1341-8.
78. Gould,J.W., M.G.Mercurio, and C.A.Elmets. (1995). Cutaneous photosensitivity
diseases induced by exogenous agents. JAm A cad Dermatol33 :551-573.
79. Grabbe,S., and R. D.Granstein. (1994). Mechanisms of ultraviolet radiation
carcinogenesis. Chern. lmmunol. 58:291-313.
80. Grady,H.G., H.F.Blum, and J.S.Kirby-Smith. (1943). Types of tumor induced by
ultraviolet radiation and factors influencing their relative incidence. J Nat/ Cancer
Inst 3:371-378.
81. Griffiths,H.R., P.Mistry, K.E.Herbert, and J.Lunec. (1998). Molecular and cellular
effects of ultraviolet light-induced genotoxicity. Crit Rev Clin Lab Sci 35:189-
237.
82. Hall,J., D.R.English, M.Artuso, B.K.Armstrong and M.Winter. (1994). DNA
repair capacity as a risk factor for non-melanocytic skin cancer- a molecular
epidemiological study. lnt J Cancer 58:179-184.
83. Hanawalt,P.C. (1996). Role of transcription-coupled DNA repair in susceptibility
to environmental carcinogenesis. Environ Health Perspect 104(Suppl3):547-551.
84. Hannuksela-Svahn,A., E.Pukkala, E.Laara, K.Poikolainen, and J.Karvonen.
(2000). Psoriasis, its treatment, and cancer in a cohort of Finnish patients. J Invest
Dermatol 114:587-590.
88
Dec. 2000
85. Hariharan,P.V. and P.A.Cerutti. (1976). Excision of ultraviolet and gamma ray
products of the 5,6-dihydroxydihydrothymine type by nuclear preperations of
xeroderma pigmentosum cells. Biochim Biophys Acta 447:375-378.
86. Hariharan,P.V. and P.A.Cerutti. (1977). Formation of products ofthe 5,6-
dihydroxydihydrothymine type by ultraviolet light in HeLa cells. Biochemistry
16:2391-2395.
87. Hartge,P., S.S.Devesa, D.Grauman, T.R.Fears, and J.F.J.Fraumeni. (1996). Non-
Hodgkin's lymphoma and sunlight. J Natl Cancer Inst 88:298-300.
88. Hartrnann,A., H.Blaszyk, and J.S.Cunninghamn. (1996). Overexpression and
mutations ofp53 in metastatic malignant melanoma. lnt J Cancer 67:313-317.
89. Hattori,Y., C.Nishigori, T.Tanaka, K.Uchida, O.Nikaido, T.Osawa, H.Hiai,
S.Imamura, and S.Toyokuni. (1996). 8-hydroxy-2'-deoxyguanosine is increased in
epidermal cells of hairless mice after chronic ultraviolet B exposure [published
erratum appears in J Invest Dermatol 1997 Feb; 108(2):237]. J Invest Dermatol
107:733-737.
90. Herbert,K.E., N.Mistry, H.R.Griffiths, and J.Lunec. (1994). Immunochemical
detection of sequence-specific modifications to DNA induced by UV light.
Carcinogenesis 15:2517-2521.
91. Herrlich,P., C.Sachsenrnaier, A.Radler-Pohl, S.Gebel, C.Blattner, and
H.J.Rahmsdorf. (1994). The mammalian UV response: mechanism of DNA
damage induced gene expression. Adv Enzyme Regul34:381-395.
92. Herrmmm,J.J., H.H.J.Roenigk, and H.Honigsmann. (1995). Ultraviolet radiation
for treatment of cutaneous T-celllymphoma. Hematol Oncol Clin North Am
9:1077-1088.
93. Hill,H.Z., W.Li , P.Xin, and D.L.Mitchell (1997). Melanin: A two edged sword?
Pigment Cell Res 10:158-161.
94. Holly,E.A., J.W.Kelly, S.N.Shpall and S.Chiu. (1987). Number ofmelanocytic
nevi as a major risk factor for malignant melanoma. JAm Acad Dermatol17:459-
468.
95. Holly,E.A., D.A.Aston, R.D.Cress, D.K.Ahn, and J.Kristiansen. (1995).
Cutaneous melanoma in women. Am J Epidemiol141:923-933.
96. Holly,E.A., D.A.Aston, D.K.Ahn, and A.H.Smith. (1996). Intraocular melanoma
linked to occupations and chemical exposures. Epidemiology 7:55-61.
97. Holman,C.D.J. , B.K.Armstrong, P.J.Heenan, J.B.Blackwell, F.J.Cumrning,
D.R.English, S.Holland, G.R.H.Kelsall, L.R.Matz, LL.Rouse, A.Singh, R.E.J.Ten
Seldam, J.D.Watt, and Z.Xu. (1986). The causes of malignant melanoma: results
from the west Australian lions melanoma research project. Recent Results Cancer
Res 102:18-37.
98. Hori,M. , M.U.Udono, H.Yoshida, Y.Urata, K.Koike, and M.Thara. (1992).
Immunochemical detection of unrepaired cyclobutane-type pyrimidine dimers of
DNAs extracted from human skin tumours. Arch Dermatol Res 284:283-289.
89
Dec. 2000
99. Hueper,W.C. (1942). Morphological aspects of experimental actinic and arsenic
carcinomas in the skin of rats. Cancer Res 2:551-559.
l 00. Huldschinsky,K. (1933). Eye sarcoma in rats caused by abnormally long UV
(Ger). Dtsch med Wochenschr 59:530-531.
l 01. IARC. ( 1982). !ARC Monographs on the Evaluation of Carcinogenic Risks to
Humans: Chemicals, Industrial Processes and Industries Associated with Cancer
in Humans (Volumes 1 to 29). International Agency for Research on Cancer,
World Health Organization, Lyon, France. Suppl4.
102. IARC. (1987). !ARC Monographs on the Evaluation of Carcinogenic Risks to
Humans: Overall Evaluations of Carcinogenicity: An Updating of !ARC
Monographs Volumes 1 to 42. International Agency for Research on Cancer,
World Health Organization, Lyon, France. Suppl 7.
I 03. IARC. (1992). !ARC Monographs on the Evaluation of Carcinogenic Risks to
Humans. Solar and Ultraviolet Radiation. International Agency for Research on
Cancer, World Health Organization, Lyon, France 55:1-316.
104. Ito,K. and S.Kawanishi. (1997). Site-specific DNA damage induced by UVA
radiation in the presence of endogenous photosensitizer. Bioi Chern 378:1307-
1312.
105. Jagger,J. (1985). Solar-UV Actions on Living Cells. Praeger, New York.
106. Johns,H.E., S.A.Rapaport, and M.Delbruck. (1962). Photochemistry of thymine
dimers. J Mol Biol4:104-114.
107. Keijzer,W., M.P.Mulder, J.C.Langeveld, E.M.Smit, J.L.Bos, D.Bootsma,
J .H.Hoeijmakers. ( 1989). Establishment and characterization of a melanoma cell
line from a xeroderma pigmentosum patient: activation of N-ras at a potential
pyrimidine dimer site. Cancer Res 49: 1229-35.
108. Kelfkens,G., F.R.de Gruijl, and J.C.van der Leun. (1991). Tumorigenesis by
short-wave ultraviolet A: papillomas versus squamous cell carcinomas.
Carcinogenesis 12:1377-1382.
109. Kim,Y.O., H.J.Chung, S.T.Chung, J.H.Kim, J.H.Park, K.S.Kil , and D.H.Cho.
(1999). Phototoxicity of melatonin. Arch ?harm Res 22:143-50.
110. Klepp,O. and K.Magnus. (1979). Some environmental and bodily characteristics
of melanoma patients. A case control study. lnt J Cancer 23:482-486.
Ill. Kligman,L.H. , M.J.Crosby, S.A.Miller, V.M.Hitchins, and J.Z.Beer. (1990). Skin
cancer induction in hairless mice by long-wavelength UV A radiation: a process
report (Abstract). Photochem Photobiol 51: 18S-19S.
112. Klingman,L.H., M.J.Crosby, S.A.Miller, V.M.Hitchins, and J.Z.Beer. 1992.
Carcinogenesis by long wavelength UV A and failure of reciprocity. In Biological
Responses to Ultraviolet A Radiation F.Urbach ed. Valdenmar, Overland Park,
KS, pp301-308.
90
Dec. 2000
113. Kraemer,K.H. and G.D.Weinstein. (1977). Decreased thymidine incorporation in
circulating leukocytes after treatment of psoriasis with psoralen and long-wave
ultraviolet light. J Invest Dermatol69:2ll-214.
114. Kripke,M.L. and E.R.Sass, eds. (1978). International conference on ultraviolet
carcinogenesis. (50). National Cancer Institute Monograph.
115. Kunz,B.A. and J.D.Armstrong. (1998). Differences in the mutational specificities
of sunlight and UVB radiation suggest a role for transversion-inducing DNA
damage in solar photocarcinogenesis. Mutat Res 422:77-83.
116. Kusewitt,L.A., L.A.Applegate, and R.D.Ley. (1991). Ultraviolet Radiation-
induced Skin Tumors in a South American Opossum (Monodelphis domestica).
Vet Patho/28:55-65.
117. Larko,O. and G.Swanbeck. (1982). Is UVB treatment ofpsoriasis safe? A study
of extensively UVB-treated psoriasis patients compared with a matched control
group. Acta Derm Venereol62:507-12.
118. LeClerc,J.E., A.Borden, and C.W.Lawrence. (1991). The thymine-thymine
pyrimidine-pyrimidone (6-4) ultraviolet light photoproduct is highly mutagenic
and specifically induces 3'-thymine-to-cytosine transitions in Escherichia coli.
Proc. Natl. Acad. Sci. USA 88:9685-9689.
119. Lerman,S. (1988). Ocularphototoxicity [editorial]. N Engll Med319:1475-
1477.
120. Ley,R.D. (1985). Photoreactivation ofUV-Induced pyrimidine dimers and
erythema in the marsupial Monodelphis domestica. Proc Natl Acad Sci USA
82:2409-2411.
121 . Ley,R.D. (1997). Ultraviolet radiation A-induced precursors of cutaneous
melanoma in Monodelphis domestica. Cancer Res 57:3682-4.
122. Ley,R.D., L.A.Applegate,T.D.Stuart, and R.J.M.Fry. (1987). UV radiation-
induced skin tumors in Monodelphis domestica. Photodermatology 4:144-147.
123. Ley,R.D., L.A.Applegate, R.S.Padilla, and T.D.Stuart. (1989). Ultraviolet
radiation-induced melanoma in Monodelphis domestica. Photochem Photobiol
50:1-5.
124. Ley,R.D., L.A.Applegate, R.J.M.Fry, and A.B.Sanchez. (1991). Photoreactivation
of ultraviolet radiation-induced skin and eye tumors of Monodelphis domestica.
Cancer Res. 51:6539-6542.
125. Lill,P.H. (1983). Latent period and antigenicity of murine tumors induced in C3H
mice by short-wavelength ultraviolet radiation. J Invest Dermatol 81:342-346.
126. LindelofB., B.Sigurgeirsson, E.Tegner, O.Larko, A.Johannesson, B.Berne,
O.B.Christensen, T.Andersson, M.Torngren, L.Molin, et al. (1991). PUV A and
cancer: a large-scale epidemiological study. Lancet 338(8759):91-3.
91
Dec. 2000
127. Lindelof,B., B.Sigurgeirsson, E.Tegner, O.Larko, A.Johannesson, B.Beme,
B.Ljunggren, T.Andersson, L.Molin, E.Nylander-Lundqvist, and L.Emtestam.
(1999). PUVA and cancer risk: the Swedish follow-up study. Br J Dermatol
141:108-12.
128. Lippincott,S.W. and H.F.Blum. (1943). Neoplasms and other lesions of the eye
induced by ultraviolet radiation in strain A mice. J Natl Cancer lnst. 3:545-554.
129. MacKie,R.M., T.Freudenberger, and T.C.Aitchison. (1989). Personal risk-factor
chart for cutaneous melanoma. Lancet 2:487-490.
130. Madewell,B.R., J.D.Conroy, E.M.Hodgkins. (1981). Sunlight-skin cancer
association in the dog: a report of three cases. J Cutaneous Patho/8:434-443.
131. Matsui,M.S. and V.A.DeLeo. (1991). Longwave ultraviolet radiation and
promotion of skin cancer. Cancer Cells 3:8-12.
132. McMichael,A.J. and G.G.Giles. (1996). Have increases in solar ultraviolet
exposure contributed to the rise in incidence of non-Hodgkin's lymphoma? Br J
Cancer 73:945-950.
133. Maier,H., M.Schemper, B.Ortel, M.Binder, A.Tanew, and H.Honigsmann.
(1996). Skin tumors in photochemotherapy for psoriasis: a single-center follow-
up of 496 patients. Dermatology 193: 185-91.
134. Menzies,S.W., G.E.Greenoak, V.E.Reeve, and C.H.Gallagher. (1991). Ultraviolet
radiation-induced murine tumors produced in the absence of ultraviolet radiation-
induced systemic tumor immunosuppression. Cancer Res. 51:2773-2779.
135. Mizuno,T., T.Matsunaga, M.Ihara, and O.Nikaido. (1991). Establishment of a
monoclonal antibody recognizing cyclobutane-type thymine dimers in DNA: a
comparative study with 64M-1 antibody specific for (6-4)photoproducts. Mutat
Res 254:175-184.
136. Moan,J. , A.Dahlback, and R.B.Setlow. (1999). Epidemiological support for an
hypothesis for melanoma induction indicating a role for UV A radiation.
Photochem Photobiol70:243-7.
137. Morrison,H. (1985). Photochemistry and photobiology ofurocanic acid.
Photodermatology 2:158-165.
138. Morrison,W.L. (1992). Phototherapy and photochemotherapy. Adv Dermatol
7:255-270.
139. Muller,L., P.Kasper, B.Kersten, J.Zhang. (1998). Photochemical genotoxicity and
photochemical carcinogenesis--two sides of a coin? Toxicol Lett; 102-103:383-7.
140. Murata-Kamiya,N., H.Kamiya, N.lwamoto, and H.Kasai. (1995). Formation of a
mutagen, glyoxal, from DNA treated with oxygen free radicals. Carcinogenesis
16:2251-2253.
141. NASA. (2000). http:/ /imagine.gsfc.nasa.gov/docs/science/
know _11/emspectrum.html
92
Dec. 2000
142. Newton,R., J.Ferlay, G.Reeves, V.Beral, D.M.Parkin. (1996). Effect of ambient
solar ultraviolet radiation on incidence of squamous-cell carcinoma of the eye.
Lancet 347(9013):1450-1.
143. Nikula,K.J., S.A.Benjamin, G.M.Angleton, W.J.Saunders, A.C.Lee. (1992).
Ultraviolet radiation, solar dermatosis, and cutaneous neoplasia in beagle dogs.
Radiat Res 129:11-8.
144. Nomura,T., H.Nakajima, T.Hongyo, E.Taniguchi, K.Fukuda, L.Y.Li, M.Kurooka,
K.Sutoh, P.M.Hande, T.Kawaguchi, M.Ueda, H.Takatera. (1997).lnduction of
cancer, actinic keratosis, and specific p53 mutations by UVB light in human skin
maintained in severe combined immunodeficient mice. Cancer Res 57:2081-4.
145. NTP. (1997). Report on Carcinogens. Solar Radiation and Exposure to Sunlamps
or Sunbeds. U.S. DHHS, National Toxicology Program.
146. 0sterland,A., M.A.Tucker, B.J.Stone, and O.M.Jensen. (1988). The Danish case-
control study of cutaneous malignant melanoma. II. Importance ofUV-light
exposure. Int J Cancer 42:319-324.
147. Ouhtit,A., M.Ueda, H.Nakazawa, M.Ichihashi, N.Dumaz, A.Sarasin, and
H.Yamasaki. (1997). Quantitative Detection ofUltraviolet-specific p53 Mutations
in Normal Skin from Japanese Patients. Cancer Epidemiology, Biomarkers &
Prevention 6:433-438.
148. Pasker-de Jong,P. C., G.Wielink,., P.G.van der Valk, and G.J.van der Wilt.
(1999). Treatment with UV -B for psoriasis and nonmelanoma skin cancer: a
systematic review of the literature. Arch Dermatol 135:834-40.
149. Pathak,M.A. (1996). Sunscreens: Progress and Perspectives on Photoprotection of
Human Skin against UVB and UV A Radiation. The Journal of Dermatology
23:783-800.
150. Peak,M.J., C.A.Jones, B.A.Sedita, E.J.Dudek, D.R.Spitz, and J.G.Peak. (1990).
Evidence that hydrogen peroxide generated by 365-nm UV A radiation is not
important in mammalian cell killing. Radiat Res 123:220-223.
151. Peak,M.J. , J.G.Peak, and B.A. Carnes. (1987). Induction of direct and indirect
single-strand breaks in human cell DNA by far- and near-ultraviolet radiations:
Action spectrum and mechanisms. Photochem. Photobiol. 45:381-387.
152. Phillips,R. (1983). Sources and Applications of Ultraviolet Radiation. Academic
Press, London.
153. Pierceall,W.E., L.H.Goldberg, M.A.Tainsky, T.Mukhopadhyay,
H.N.Ananthaswamy. (1991 ). Ras gene mutation and amplification in human
nonmelanoma skin cancers. Mol Carcinog 4:196-202.
154. Pittelkow,M.R., H.O.Perry, S.A.Muller, W.Z.Maughan, P.C.O'Brien. (1981). Skin
cancer in patients with psoriasis treated with coal tar. A 25-year follow-up study.
Arch Dermatol117:465-8.
155. Putschar,W. and F.Holtz (1930). Production of skin tumors in rats by long
duration UV irradiation (Ger.). Z Krebsforsch. 33:219-260.
93
Dec. 2000
156. Quinn,A.G. (1997). Ultraviolet radiation and skin carcinogenesis. Br J Hosp Med
58:261-264.
157. Robert,C., B.Muel, A.Benoit, L.Dubertret, A.Sarasin, and A.Stary. (1996). Cell
survival and shuttle vector mutagenesis induced by ultraviolet A and ultraviolet B
radiation in a human cell line. J Invest Dermatol. 106:721-728.
158. RoC. (2000). 9
111
Report on Carcinogens. U.S. DHHS, National Toxicology
Program.
159. Roffo,A.H. (1934). Cancer and the sun: carcinomas and sarcomas caused by the
action of the sun in toto (Fr.). Bull Assoc fr Etude Cancer 23:590-616.
160. Roffo,A.H. ( 193 9). Physico-chemical etiology of cancer (with special emphasis
on the associaation with solar radiation) (Ger.). Strahlentherapie 66:238-350.
161. Rosenthal,F.S., S.K.West, B.Munoz, E.A.Emmett, P.T.Strickland, and
H.R.Taylor. (1991). Ocular and facial skin exposure to ultraviolet radiation in
sunlight: a personal exposure model with application to a worker population.
Health Phys 61 :77-86.
162. Runger,T.M. (1999). Role ofUVA in the pathogenesis of melanoma and non-
melanoma skin cancer. A short review. Photodermatol Photoimmunol Photomed
15:212-6.
163. Sarasin,A. (1999). The molecular pathways of ultraviolet-induced carcinogenesis.
Mutat Res 428:5-10.
164. Satyamoorthy,K, F.Meier, M.Y.Hsu, C.Berking, and M.Herlyn. (1999). Human
xenografts, human skin and skin reconstructs for studies in melanoma
development and progression. Cancer Metastasis Rev 18:401-5.
165. Sauter,E.R., S.A.K1ein, E.Atillasoy, K.T.Montone, T.Goodrow, R.L.Binder,
J.T.Seykora, and M.Herlyn. (1998). Ultraviolet B-induced squamous epithelial
and melanocytic cell changes in a xenograft model of cancer development in
human skin. Mol Carcinog 23:168-174.
166. Seddon,J.M., E.S.Gragoudas, R.J.Glynn, K.M.Egan, D.M.Albert, and P.H.Blitzer.
(1990). Host factors. UV radiation and risk ofuveal melanoma. A case-control
study. Arch Ophthalmol. 108:1274-1280.
167. Sekiya,T., M.Fushimi, H.Hori, S.Hrohashi, S.Nishimura, and T.Sugimura. (1984).
Molecular cloning and the total nucleotide sequence of the human c-Ha-ras-1
gene activated in a melanoma from a Japanese patient. Proc Natl Acad Sci USA.
81:4771-4775.
168. Setlow,R.B., A.D. Woodhead, and E.Grist. (1989). Animal model for ultraviolet
radiation-induced melanoma: platyfish-swordtail hybrid. Proc Natl Acad Sci USA.
86:8922-8926.
169. Setlow,R.B., E.Grist, K.Thompson, and A.D.Woodhead. (1993). Wavelengths
effective in induction of malignant melanoma. Proc Natl Acad Sci USA 90:6666-
6670.
94
Dec. 2000
170. Shukla,V.K., D.C.Hughes, L.E.Hughes, F.McCormick, and R.A.Padua. (1989).
ras Mutations in human melanotic lesions: K-ras activation is a frequent and early
event in melanoma development. Oncogene Res 5:121-127.
171. Siemiatycki,J. (1991). Risk Factors for Cancer in the Workplace. CRC Press,
Boco Raton, FL.
172. Smith,K.C. (1989). The Science of Photobiology, K.C.Smith ed. Plenum, New
York.
173. Soballe,P.W., K.T.Montone, K.Satyamoorthy, M.Nesbit, M.Herlyn. (1996)
Carcinogenesis in human skin grafted to SCID mice. Cancer Res 56:757-764.
174. Staberg,B., H.C.Wulf, P.Klemp, T.Poulsen, and H.Brodthagen. (1983). The
carcinogenic effect ofUV A irradiation. 1 Invest Dermatol 81:517-519.
175. Stary,A., C.Robert, A.Sarasin. ( 1997). Deleterious effects of ultraviolet A
radiation in human cells. Mutat Res 383:1-8.
176. Stenback,F. ( 1975). Studies on the modifying effect of ultraviolet radiation on
chemical skin carcinogenesis. 1 Invest Dermato/64:253-257.
177. Sterenborg,H.J. and J.C.van der Leun. (1988). Change in epidermal transmission
due to UV -induced hyperplasia in hairless mice: a first approximation of the
action spectrum. Photodermatol5:71-82.
178. Sterenborg,H.J. and J.C.van der Leun. (1990). Tumorigenesis by a long
wavelength UV-A source. Photochem Photobio/51:325-330.
179. Stem,R.S. (1990). Genital tumors among men with psoriasis exposed to psoralens
and ultraviolet A radiation (PUV A) and ultraviolet B radiation. The
Photochemotherapy Follow-up Study. N Engl 1 Med 322:1 093- l 097.
180. Stem,R.S. and N.Laird. (1994). The carcinogenic risk of treatments for severe
psoriasis. Cancer 73:2759-2764.
181. Stem,R.S., N.Laird, J.Melski, J.A.Parrish, T.B.Fitzpatrick, and H.L.Bleich.
(1984). Cutaneous squamous-cell carcinoma in patients treated with PUV A. N
EnglJ Med310:1156-61.
182. Stern,R.S., and R.Lange. (1988). Non-melanoma skin cancer occurring in patients
treated with PUV A five to ten years after first treatment J Invest Dermatol
91:120-4.
183. Stem,R.S., E.J.Liebman, and L.Vakeva. (1998). Oral psoralen and ultraviolet-A
light (PUV A) treatment of psoriasis and persistent risk of nonmelanoma skin
cancer. PUV A Follow-up Study. 1 Natl Cancer Inst 90:1278-1284.
184. Stem,R.S., K.T.Nichols, and L.H.Vakeva. (1997). Malignant melanoma in
patients treated for psoriasis with methoxsalen (psoralen) and ultraviolet A
radiation (PUV A). The PUVA Follow-Up Study [see comments] . N Eng! 1 Med
336:1041-1045.
185. Stem,R.S., L.A.Thibodeau, J.A.Parrish, and T.B.Fitzpatrick. (1979). Skin cancer
after PUV A treatment for psoriasis [letter]. N Engl J Med 301:555.
95
Dec. 2000
186. Stern, R.S. , S.Zierler, and J.A.Parrish. (1980). Skin carcinoma in patients with
psoriasis treated with topical tar and artificial ultraviolet radiation. Lancet
1(8171):732-5.
187. Stewart,M.S., G.S.Cameron, and B.C.Pence. (1996). Antioxidant nutrients protect
against UVB-induced oxidative damage to DNA of mouse keratinocytes in
culture. J Invest Dermato/106:1086-1089.
188. Strauss,G.H. (1982). PUVA: gambling with a system. Int J Dermato/21:136-137.
189. Strauss,G.H., B.A.Bridges, M.Greaves, P.Hall-Smith, M.Price, and D.Vella-
Briffa. (1980). Inhibition of delayed hypersensitivity reaction in skin (DNCB test)
by 8-methoxypsoralen photochemotherapy. Possible basis for pseudo- promoting
action in skin carcinogenesis? Lancet 2:556-559.
190. Strauss,G.H., R.J.Albertini, P.A.Krusinski, and R.D.Baughman. (1979). 6-
Thioguanine resistant peripheral blood lymphocytes in humans following
psoralen, long-wave ultraviolet light (PUVA) therapy. J Invest Dermatol73:211-
216.
191. Streilein,J.W. (1993). Sunlight and skin-associated lymphoid tissues (SALT): if
UVB is the trigger and TNF alpha is its mediator, what is the message? J Invest
Dermatoll00:47S-52S.
192. Streilein,J.W. and P.R.Bergstresser. (1981). Langerhans cell function dictates
induction of contact hypersensitivity or unresponsiveness to DNFB in Syrian
hamsters. J Invest Dermatol77:272-277.
193. Streilein,J.W., J.R.Taylor, V.Vincek, l.Kurimoto, J.Richardson, C.Tie,
J.P.Medema, and C.Golomb. (1994). Relationship between ultraviolet radiation-
induced immunosuppression and carcinogenesis. J Invest Dermatol 103(Suppl
5):107S-111S.
194. Strickland,P.T., F.J.Burns, and R.E.Albert. (1979). Induction of skin tumors in the
rat by single exposure to ultraviolet radiation. Photochem Photobiol 30:683-688.
195. Studniberg,H.M. and P.Weller. (1993). PUVA, UVB, psoriasis, and
nonrnelanoma skin cancer. JAm Acad Dermatol29: 1013-1022.
196. Swerdlow,A.J., J.S.English, R.M.MacKie, C.J.O'Doherty, J.A.Hunter, J.Clark,
and D.J.Hole. (1988). Fluorescent lights, ultraviolet lamps, and risk of cutaneous
melanoma [published erratum appears in BMJ 1988 Nov 5;297(6657): 1172]. BMJ
297:647-650.
197. Swerdlow,A.J. and M.A.Weinstock. (1998). Do tanning lamps cause melanoma?
An epidemiologic assessment. JAm Acad Dermatol38:89-98.
198. Toews,G.B., P.R.Bergstresser, and J.W.Streilein. (1980). Epidermal Langerhans
cell density determines whether contact hypersensitivity or unresponsiveness
follows skin painting with DNFB. J lmmunoll24:445-453.
96
Dec. 2000
199. Urbach,F., J.H.Epstein, and P.D.Forbes. (1974). Ultraviolet carcinogenesis:
experimental, global,and genetic aspects. In Sunlight and Man. Normal and
Abnormal Photobiological Responses. M.A.Pathak, L.C.Harber, M.Seiji, and
A.Kukita, eds. University of Tokyo Press, Tokyo. pp. 259-283.
200. van der Leun,J.C. (1984), Yearly review: UV-carcinogenesis. Photochem
Photobio/39:861-868.
201. van der Schroeff,J.G., L.M.Evers, A.J.M.Boot, and J.L.Bos. (1990). Ras
Oncogene mutations in basal cell carcinomas and squamous cell carcinomas of
human skin. J Invest Dermatol. 94:423-425.
202. van't Veer,L.J., B.M.T.Burgering, R.Versteeg, A.J.M.Boot, D.J.Ruiter, S.Osanto,
P.I.Schrier, and J.L.Bos. (1989). Effects of ultraviolet B light on cutaneous
immune responses of humans with deeply pigmented skin. J Invest Dermato/97:
729-734.
203. van Weelden,H., F.R.de Gruij l, and J.C.van der Leun. (1986). Carcinogenesis by
UV A, with an attempt to assess the carcinogenic risks of tanning with UV A and
UVB. In The Biolgical Effects of UVA Radiation, F.Urbach and R.W.Gange, eds.
Praeger , New York pp. 137-146.
204. van Weelden,H., F.R.de Gruijl, S.C.van der Putte, J.Toonstra, and J.C.van der
Leun. (1988). The carcinogenic risks of modern tanning equipment: is UV-A
safer than UV-B? Arch Dermatol Res 280:300-307.
205. van Weelden,H., S.C.van der Putte, J.Toonstra, and J.C.van der Leun. (1990).
UVA-induced tumours in pigmented hairless mice and the carcinogenic risks of
tanning with UV A. Arch Dermatol Res 282:289-294.
206. Vitasa,B.C., H.R.Tylor, P.T.Strickland, F.S.Rosenthal, S.West, H.Abbey, S.K.Ng,
B.Munoz, and E.A.Emmett. (1990). Association ofnonmelanoma skin cancer and
actinic keratosis with cumulative solar ultraviolet exposure in Mryland watermen.
Cancer 65:2811-2817.
207. Walter,S.D. , W.D.King, and L.D.Marrett. (1999). Association of cutaneous
malignant melanoma with intermittent exposure to ultraviolet radiation: results of
a case-control study in Ontario, Canada. lnt J Epidemiol28:418-27.
208. Walter,S.D., L.D.Marrett, L.From, C.Hertzman, H.S.Shannon, and P.Roy. (1990).
The association of cutaneous malignant melanoma with the use of sunbeds and
sunlamps. Am J Epidemiol 131 :232-43.
209. Walter,S.D., L.D.Marrett, H.S.Shannon, L.From, and C.Hertzman. (1992). The
association of cutaneous malignant melanoma and fluorescent light exposure. Am
J Epidemio/135:749-62.
210. Wei,Q., G.M.Matanoski, E.R.Fam1er, M.A.Hedayati, and L.Grossman. (1994).
DNA repair and susceptivility to basal cell carcinoma: a case-control study. Am J
Epidemiol 140:598-607.
97
Dec. 2000
211 . Wei,Q., G.M.Matanoski, E.R.Farmer, M.A.Hedayati, and L.Grossman. (1995).
DNA repair capacity for ultraviolet light-induced damage is reduced in peripheral
lymphocytes from patients with basal cell carcinoma. J Invest Dermatol104:933-
936.
212. Westerdahl,J., C.Ingvar, A.Masback, N.Jonsson,., and H.Olsson, H. (2000). Risk
of cutaneous malignant melanoma in relation to use of sunbeds: further evidence
for UV-A carcinogenicity. Br J Cancer 82:1593-9.
213. Westerdahl,J., H.Olsson, A,Masback, C.Ingvar, N.Jonsson, L.Brandt,
P.E.Jonsson, and T.Moller. (1994). Use ofsunbeds or sunlamps and malignant
melanoma in southern Sweden. Am J Epidemiol 140:691-699.
214. Whitmore,S.E. and W.L.Morison. (1997) Melanoma after PUVA therapy for
psoriasis [letter; comment]. N Engl J Med 337:502-3.
215. WHO. (1979). Ultraviolet Radiation. In: Environmental Health Criteria 14 World
Health Organization, Geneva.
216. Wikonkal,N.M., RJ.Berg, C.W.van Haselen, I.Horkay, E.Remenyik, A.Begany,
J.Hunyadi, W.A.van Vloten, and F.R.de Gruijl. (1997). bcl-2 vs p53 protein
expression and apoptotic rate in human nonmelanoma skin cancers. Arch
Dermatoll33:599-602.
217. Wikonkal,N.M. and D.E.Brash. (1999). Ultraviolet radiation induced signature
mutations in photocarcinogenesis. J lnvestig.Dermatol.Symp.Proc. 4:6-10.
218. Whittaker,S. (1996). Sun and skin cancer. British Journal of Hospital Medicine
56:515-518.
219. Wulff,D.L. and G.Fraenkel. (1961). On the nature of the thymine photoproduct.
Biochim Biophys Acta 51:332-339.
220. Yarosh,D.B. and M.L.Kripke. (1996). DNA repair and cytokines in
antimutagenesis and anticarcinogenesis. Mutat. Res. 350:255-260.
221. Young,A.R., S.L.Walker, J.S.Kinley, S.R.Plastow, D.Averbeckm, P.Morliere, and
L.Dubertret. ( 1990). Phototumorigenesis studies of 5-methosypsoralen in
bergamot oil: evaluation and modification of risk of human use in an albino
mouse skin model. J Photochem Photobiol B Biol7:231-250.
222. Young,A.R., C.S.Potten, O.Nikaido, P.G.Parsons, J.Boenders, J.M.Ramsden,
C.A.Chadwick. (1998). Human melanocytes and keratinocytes exposed to UVB
or UV A in vivo show comparable levels of thymine dimers. J lnvest.Dermatol.
111:936-40.
223. Zanetti,R., S.Rosso, F.Faggiano, R.Roffino, S.Colonna, and G.Martina. (1988). A
case-control study on cutaneous malignant melanoma in the province ofT orino,
Italy. Rev Epidem et Sante Publ36:309-317.
224. Ziegler,A., D.J.Leffell, S.Kunala, H.W.Sharma, M.Gailani, J.A.Sirnon,
A.J.Ha]perin, H.P.Baden, P.E.Shapiro, A.E.Bale, and D.Brash. (1993). Mutation
hotspots due to sunlight in the p53 gene of nonmelanoma skin cancers. Proc Natl
Acad Sci USA 90:4216-20.
98
Dec. 2000
Appendix A: IARC Monograph of Evaluation of
Carcinogenic Risks to Humans. Solar and Ultraviolet
Radiation. Vol. 55. Lyon, France. World Health
Organization. 1992. pp. 43-279.
99
Dec. 2000
100
1.1 Nomenclature
1.1.1 Optical radiation
1. Exposure Data
Optical radiation is radiant energy within a broad region of the electromagnetic spectrum
that includes ultraviolet (UV), visible (light) and infrared radiation. Ultraviolet radiation
(UVR) is characterized by wavelengths between 10 and 400 nm-bordered on the one side
by x rays and on the other by visible light (Fig. 1 ). Solar radiation is largely optical
radiation, although ionizing radiation (i.e., cosmic rays, gamma rays and x rays, which have
wavelengths less than approximately 10 nm) and radio-frequency radiation (i.e.,
wavelengths greater than 1 mm: microwaves and longer radio waves) are also present in the
spectrum.
The optical radiation spectrum is generally considered to fall between 10 nm and 1 mm,
and several different conventions have been developed to describe different bands within
this spectrum. It is important to recognize that no single convention is uniquely 'correct' but
that each may be useful for a particular branch of science and technology. For example, in
optics, it is convenient to separate the spectrum into different bands on the basis of the
transmission and absorption properties of optical materials (e.g .. glass and quartz). In one
optical convention, shown in Figure 1, UVR is divided into vacuum UV, extending from 10
to 180 nm; middle UV, from 180 nm to 300 nm: and near UV, from 300 nm to 380 or 400
run. Meteorological scientists typically define optical spectral regions on the basis of
atmospheric windows. Some spectral designations are based on uses, e.g .. 'germicidal' and
'black-light' regions.
For the purposes of this monograph, the photobiological designations of the Commission
Intemationale de I'Eclairage (CIE, International Commission on Illumination) are the most
relevant and are used throughout to defme the approximate spectral regions in which certain
biological absorption properties and biological interaction mechanisms may dominate
(Commission lntemationale de l'Eclairage, 1987). The CIE bands are: UVC (100-280 nm),
UVB (280-315 nm) and UVA (315-400 nm). Visible light is the region between 400 nm
and 780 nm.
It is important to recognize that these spectral band designations are merely short-hand
notations and cannot be considered to designate fine dividing lines below which an effect is
present and above which it does not occur. The reader should also be alerted to the fact that
the CIE nomenclature is not always followed rigorously and that some authors introduce
slight variations; for example, distinguishing between UVB and UV A at 320 rather than 315
nm (frequently used in the USA) and defining UVC as 200-280 nm (Moseley, 1988). The
German Industrial Standard (DIN 5031) defines UV A as radiation between 315 and 380 nm
(Mutzhas, 1986).
-43-
44 IARC MONOGRAPHS VOLUME 55
Figure 1. Electromagnetic spectrum with enlargement of ultraviolet (UV) region
Cosmic rays, gamma rays, x rays
X rays_,
100
I ~
Vacuum UV
I
100
Extreme uv Far UV
~ I ~
10 100
Visible
Violet Red
Ultra-
violet
180
I
uvc
Infrared
300
I ~ UVB +
280 315
Black
light
UVA
Middle UV Near uv
Radio waves
Visible
400
I
400
~ ~ ~ ~
180 300 380
Wavelength (nanometres)
Adapted from WHO (1979), Morison (1983a) Sylvania (undated)
From the viewpoint of photochemistry and photobiology, interactions of optical
radiation with matter are considered to occur when one photon interacts with one molecule
to produce a photochemically altered molecule or two dissociated molecules (Phillips, 1983;
Smith, 1989). In any photochemical interaction, the energy of the individual photon is
important, since this must be sufficient to alter a molecular bond. 'fee photon energy is
generally expressed in terms of electron volts (eV). A wavelength of 10 nm corresponds to
a photon energy of 194 eV, and 400 nm to an energy of 3.1 eV (WHO, 1979). The number
of altered molecules produced relative to the number of absorbed photons is referred to as
the ' quantum yield' (Phillips. 1983). The efficacy of photochemical interaction per incident
quantum and the photobiological effects per unit radiant exposure typically vary widely with
wavelength. A quantitative plot of such spectral variation, usually normalized to unity at the
most effective wavelength, is referred to as an 'action spectrum' (Jagger, 1985).
EXPOSURE DATA 45
1.1.2 Quantities and units
Two systems of quantities and units are used to describe the characteristics of light and
light sources: the radiometric and the photometric systems. Radiometry can be applied to
all optical sources and to all exposures to optical radiation (including solar radiation and
UVR). Photometry can be used only to describe visible light sources. and photometric
quantities are used in illumination engineering. The basic photometric unit is the lumen,
which is defined in terms of the spectral response of the human eye (specifically, the
spectral response of the CIE 'standard observer'), i.e., the action spectrum of vision, which is
initially a photochemical process. It is important to recognize that radiometric quantities
and units are absolute, while photometric quantities and units are related to standardized
human perception; the relationship between the two sets of units varies significantly with
the spectrum of radiation. The effects of optical radiation (including light), other than
vision, must therefore be measured and quantified in terms of radiometric units and spectral
characteristics rather than photometric units. This is particularly important in relation to the
photobiological effects of UVR. Most lamps used for illumination are rated by
manufacturers only in photometric terms (e.g., lumen output) and not in terms of UVR
emission (Phillips, 1983 ).
The most important radiometric quantities and units commonly used to describe optical
radiation are given in Table 1. Certain terms are used primarily to describe source charac-
teristics, e.g., radiance, radiant intensity; whereas other terms are generally used to describe
exposure (irradiance, radiant exposure). The term 'spectral' placed before any of the quan-
tities implies restriction to a unit wavelength band, e.g., spectral irradiance (watts per square
metro per nanometre) (Moseley, 1988). For a more detailed discussion of these parameters,
see various standard textbooks on radiometry, such as Boyd (1983).
The quantities of radiometry are expressed in terms of absolute energy (Jagger, 1985).
Radiant intensity is the power emitted per unit solid angle of a source. Radiance is the
radiant intensity per unit area of source. Thus, a fluorescent lamp dries not have very high
radiance in comparison to the filament of a flashlight bulb, even though it has a high radiant
power output. The radiometric term expressed in units of watts per square metre (dose rate)
is irradiance, which is also the power striking a unit area of surface.
The energy of UVR falling on a unit surface area of an object was defined in 1954 by the
First International Congress of Photobiology as the 'dose'; it has also been referred to as
'exposure dose'. The equivalent radiometric quantity is radiant exposure, expressed in joules
per square centimetre or per square metro. Radiant exposure has been referred to as 'energy
fluence' in some texts; however, fluence is a radiometric quantity, with the same units as
radiant exposure, but referring to energy arriving at a plane of unit area from all directions,
including backscatter. Thus, fluence is quite correctly of value in describing an exposure
dose at a depth inside tissue: it bas, however, seldom been calculated in photobiological
studies of the effects of UVR, in which the radiant exposure incident upon the skin is
normally measured. Radiant exposure is the amount of energy crossing a unit area of space
normal to the direction of propagation of a beam of UVR. If the radiant energy arrives
from many directions, as from the sky, then the fluence at one point is the sum of all the
component fluences entering a unit sphere of space. The energy fluence rate is the power
that crosses a unit area normal to the direction of propagation, or the energy per unit area per
unit time
Table 1. Some basic terminology used to quantify optical radiation
Term International Definition
symbol
Wavelength 'A
Radiant energy Qc :E(P c x dt)
Radiant flux P e dQe/dt
Irradiance Ee dP eldA
Radiant intensity Ie dPcfdQ
Radiance Lc dPcldA X dQ
SI unit Synonyms and comments
nm Nanometre = 10-
9
m (also called
millimicron, mfl)
J Joule; 1 joule = 1 watt x second; total
energy contained in a radiation field or
total energy delivered to a given receiver
by such a radiation field
W Watt; rate of delivery of radiant energy
('radiant power'); also expressed as 0
W/m
2
Radiant flux arriving over a given area
('fluence rate', 'dose rate', 'intensity',
'radiant incidence'). ln photobiology, has
also been expressed in W/cm
2
, mW/cm
2
and 11W/cm
2
W/sr Watt/steradian; radiant flux emitted by
source into a given solid angle (solid
angle expressed in steradians)
W/m
2
x sr Watt/m2 x steradian; radiant flux per unit
solid angle per unit area emitted by an
extended source
Radiant exposure He Ec x t J/nl Radiant energy delivered to a given area
('fluence', 'exposure dose', ' dose'); t =
time in seconds. Has also been expressed
as J/cm
2
, mJ/cm
2
and fl}/cm
2
Adapted from WHO ( 1979), Boyd ( 1983 ), Jagger ( 1985), Hoffman ( 1987) and Weast ( 1989)
(J/m
2
/s or W/m
2
). The terms dose (J/m
2
) and dose rate (W/m
2
) pertain to the energy and
power, respectively, striking a unit surface area of an irradiated object (Jagger, 1985).
In terms of visible light perceived by humans, the photometric analogue of the radiance
of a source is luminance (brightness), and irradiance is illuminance (measured in 'lux' or
lumen per square metre). In photometry, the lumen is the unit of luminous power (Jagger,
1985).
1.1.3 Units of biologically effective ultraviolet radiation
In addition to general radiometric quantities, specialized quantltles of effective irra-
diance relative to a specified photochemical action spectrum are used in photochemistry and
photobiology. Effective radiant exposures to produce erythema (Jagger, 1985) or photo-
keratitis are examples. Effective irradiance or radiant exposure is not limited to photo-
biology, and a similar approach has been used to quantify the photocuring of inks, in photo-
polymerization (Phillips, 1983) and in assessing the hazards ofUVR. In order to weight a
EXPOSURE DATA 47
source spectrally, the general formula involves an action spectrum and a spectral
radiometric quantity. The effective irradiance of a given photobiological process is defined
as:
expressed in W/m
2
, where E
1
, is the spectral irradiance (W/m
2
x nm) at wavelength A (nm)
and 1:!:.
1
, is the wavelength interval (A.
1
-4A
2
) used in the summation (in nm). So, is a measure
of the effectiveness of radiation of wavelength A. (nm), relative to some reference
wavelength, in producing a particular biological end-point. As it is a ratio, s ~ has no units
(American Conference of Governmental Industrial Hygienists, 1991 ).
Effective irradiance is equivalent to a hypothetical irradiance of monochromatic radia-
tion with a wavelength at which s ~ is equal to unity. The time integral of effective
irradiance is the effective radiant exposure (also called the 'effective dose').
A unit of effective dose commonly used in cutaneous photobiology is the 'minimal
erythema dose' (MED). One MED has been defined as the lowest radiant exposure to UVR
that is sufficient to produce erythema with sharp margins 24 h after exposure (Morison,
1983a). Another end-point often used in cutaneous photobiology is a just-perceptible
reddening of exposed skin; the dose of UVR necessary to produce this 'minimal perceptible
erythema' is sometimes also referred to as an MED. In unacclimatized, white-skinned
populations, there is an approximately four-fold range in the MED of exposure to UVB
radiation (Diffey & Farr, 1989). When the term MED is used as a unit of exposure dose,
however, a representative value is chosen for sun-sensitive individuals. If, in the above
expression for effective irradiance. So is chosen as the reference action spectrum for
erythema (McKinley & Diffey, 1987) and a value of 200 J/m
2
at wavelengths for which s ~ is
equal to unity is assumed for the MED, the dose (expressed in MED) received after an
exposure period oft seconds is t x L E).. x s,, x ilJ 200.
Notwithstanding the difficulties of interpreting accurately the magnitude of such an
imprecise unit as the MED, it has the advantage over radiometric units of being related to
the biological consequences of the exposure.
1.2 Methods for measuring ultraviolet radiation
UVR can be measured by chemical or physical detectors, often in conjunction with a
monochromator or band-pass filter for wavelength selection. Physical detectors include
radiometric devices, which depend for their response on the heating effect of the radiation,
and photoelectric devices, in which incident photons are detected by a quantum effect such
as the production of electrons. Chemical detectors include photographic emulsions,
actinometric solutions and UV-sensitive plastic films.
1.2.1 Spectroradiometry
The fundamental way of characterizing a source of UVR is on the basis of its spectral
power distribution in a graph (or table) which indicates the radiated power as a function of
wavelength. The data are obtained by a technique known as spectroradiometry. Spectral
measurements are often not required as ends in themselves but are used to calculate
biologically weighted radiometric quantities. A spectroradiometer comprises three essential
components (Gibson & Diffey, 1989):
(i) input optics, such as an integrating sphere or Teflon diffuser, which collects the
incident radiation and conducts it to
(ii) the entrance slit of a monochromator, which disperses the radiation by means of
one or two wavelength dispersive devices (either diffraction grating or prism).
The monochromator also incorporates mirrors to guide the radiation from the
entrance slit to the dispersion device and on to the exit slit, where it is incident on
(iii) a radiation detector, normally a photodiode or, for higher sensitivity, a photo-
multiplier tube.
Spectroradiometry is generally considered to be the best way of specifying UV sources,
although the accuracy of spectroradiometry, particularly with respect to the UVB waveband
of terrestrial radiation, is affected by a number of parameters including wavelength cali-
bration, band width, stray radiation, polarization, angular dependence, linearity and
calibration sources. It is therefore essential to employ a double monochromator for accurate
characterization of terrestrial UVR and particularly UVB (Garrison et al., 1978; Kostkowski
et al., 1982; Gardiner & Kirsch, 1991 ).
1.2.2 Wavelength-independent (thermal) detectors
General-purpose radiometers incorporate detectors that have a flat response over a wide
range of wavelengths. Such thermal detectors operate on the principle that incident
radiation is absorbed by a receiving element, and the temperature rise of the element is
measured, usually by a thermopile or a pyroelectric detector. A thermopile, which
comprises several thermocouples connected in series for improved sensitivity, must have a
window made of fused silica for measuring UVR at wavelengths down to at least 250 nm.
Pyroelectric detectors rely on a voltage generated by temperature changes in a lithium
tantalate crystal. Thermal detectors are normally used to measure the total radiant power of
a source rather than just the UV component (Moseley, 1988).
Instruments for measuring broad-band solar radiation fall into three categories: pyro-
heliometers, pyranometers and pyranometers with a shading device (Iqbal. 1983). These
types of instrument find their applications in meteorology rather than in UV photobiology.
1.2.3 Wavelength-dependent detectors
Detectors of this type have a spectral response that varies widely depending on the types
of detector and filters that may be incorporated. Detectors can be designed to have a
spectral response that matches a particular action spectrum for a photobiological end-point.
The success with which this is achieved is variable. The most widely used device,
particularly for measuring solar UVR, has been the Robertson-Berger meter (Robertson,
1972; Berger, 1976), which incorporates optical filters, a phosphor and a vacuum phototube
or photovoltaic cell. This device measures wavelengths of less than 330 nm in the global
spectrum with a spectral response that rises sharply with decreasing wavelength. lt has been
used to monitor natural UVR continuously at several sites throughout the world (Berger &
Urbach, 1982; Diffey, 1987a).
EXPOSURE DATA 49
Detectors incorporating a photodiode or vacuum photocell in conjunction with optical
filter(s) and suitable input optics (e.g. , a quartz hemispherical detector) have been produced
to match a number of different action spectra. One such detector is the International Light
Model 730 UV Radiometer, which has a spectral response close to the action spectrum
designated by the American Conference of Governmental Industrial Hygienists for eval-
uating the hazard to health of exposure to UVR, and has been used to measure irradiance
over different terrains (Sliney, 1986).
Wavelength-dependent detectors with spectral responses largely in the UVA waveband
are used, for example, in measuring the output of irradiation units for the treatment of
psoriasis by psoralen photochemotherapy (Morison, 1983a).
A different yet complementary approach is the use of various photosensitive films as UV
dosimeters. The principle is to relate the degree of deterioration of the films, usually in
terms of changes in their optical properties, to the dose of incident UVR. The principal
advantages of the film dosimeter are that it provides a simple means of integrating exposure
continuously and allows simultaneous comparison of numerous sites that are inaccessible to
bulky, expensive instruments (Diffey, 1987a). The most widely used photosensitive film is
polymer polysulfone (Diffey, 1989a). Personal dosimeters of polysulfone film have been
developed and used in a number of dosimetric studies (Challoner et at .. 1976, 1978; Leach et
al.. 1978: Holman et al., 1983a: Larko & Diffey, 1983; Diffey, 1987a; Schothorst et al..
1987a; Slaper, 1987; Rosenthal et al., 1990).
It is difficult to achieve a prescribed UVR spectral response with wavelength-dependent
detectors. Accurate results can be achieved only if the detectors are calibrated against the
appropriate source spectrum using a spectroradiometer (Gibson & Diffey,l989). Unless this
is done, severe dosimetric errors can arise, particularly with measurements of solar UVR
(Diffey, 1987a; Sayre & Kligman, 1992).
Accurate measurement of UVB radiation is far more difficult than would appear
initially. The primary problem is that the UVB produced by most optical sources-the sun
as well as incandescent and fluorescent lamps used for illumination-is only a very small
fraction (i.e., less than 0.3%) of the total radiant energy emitted. Additionally, biological
action spectra (e.g., for erythema and photokeratitis) typically decrease dramatically within
the same waveband in which the source spectrum increases (Diffey & Parr, 1991a). This
means that either a spectroradiometer or a direct-reading filtered 'erythemal' or 'hazard'
meter must reject out-of-band radiant energy to better than one part in 1 0
4
or even 10
5
. The
spectral band-width of a monochromator can also greatly affect measurement error: too
large a band-width can reduce the steepness of reported action spectra.
1.3 Sources and exposures
In the broadest sense, UVR may be produced when a body is heated (incandescence) or
when electrons that have been raised to an excited state return to a lower energy level , as
occurs in fluorescence, in an electric discharge in a gas and in electric arcs (optical plasma)
(Sliney & Wolbarsht, 1980; Phillips, 1983; Moseley, 1988). The characteristics of
exposures to both terrestrial solar radiation (an incandescent source) and artificial light
sources are discussed in the following sections.
1.3 .1 Solar ultraviolet radiation
Optical radiation from the sun is modified significantly as it passes through the Earth's
atmosphere (Fig. 2), although about two-thirds of the energy from the sun that impinges on
the atmosphere penetrates to ground level. The annual variation in extra-terrestrial radiation
is less than 1 0%, but the variation in the modifying effect of the atmosphere is far greater
(Moseley, 1988). Measurements corrected for atmospheric absorption show that the visible
portion comprises approximately 40% of the total radiation received at the surface of the
Earth. While UVR comprises only a small proportion of the total radiation (approximately
5%), this component is extremely important in various biological processes. The principal
effect of infrared radiation is to warm the earth; approximately 55% of the solar radiation
received at the surface of the earth is infrared (Foukal, 1990).
Fig. 2. Spectral irradiance from the sun outside the Earth's atmosphere (upper curve)
and at sea level (lower curve)
Wavelength ( n m)
From Moseley (1988)
On its path through the atmosphere, solar radiation is absorbed and scattered by various
constituents of the atmosphere. It is scattered by air molecules, particularly oxygen and
nitrogen (Rayleigh scattering), which produce the blue colour of the sky. It is also scattered
by aerosol and dust particles (Mie scattering) and is scattered and absorbed by atmospheric
pollution. Total solar irradiance and the relative contributions of different wavelengths vary
with altitude. Clouds attenuate solar radiation, although their effect on infrared radiation is
greater than on UVR. Reflection of sunlight from certain ground surfaces may contribute
significantly to the total amount of scattered UVR. An effective absorber of solar UVR is
ozone in the stratosphere (Moseley, 1988). An equally important absorber in the longer
wavelengths (infrared) is water vapour (Diffey, 1991 ); a secondary absorber in this range is
carbon dioxide. These two filter out much of the solar energy with wavelengths longer than
1000 nm (Sliney & Wolbarsht, 1980).
EXPOSURE DATA 51
The quality (spectral distribution) and quantity (total UV irradiance) of UVR reaching
the Earth's surface depend on the radiated power from the sun and the transmitting
properties of the atmosphere. Although UVC exists in the extra-terrestrial solar spectrum, it
is filtered out completely by the ozone layer in the atmosphere. UVB radiation, which
represents about 5% of the total solar UVR that reaches the Earth (Sliney & Wolbarsht,
1980), has been considered to be the most biologically significant part of the terrestrial UV
spectrum. The levels of UVB radiation reaching the surface of the Earth, although heavily
attenuated, are also largely controlled by the ozone layer.
Ozone (0
3
) is a gas which comprises approximately one molecule out of every two
million in the atmosphere. It is created by the reaction of molecular oxygen (0
2
) with
atomic oxygen (0), fonned by the dissociation of 02 by short-wavelength UVR (< 242 nm)
in the stratosphere at altitudes between about 25 and 100 km. Absorption of UVR at
wavelengths up to about 320 nm converts the ozone back to 0
2
and 0, and it is this
dissociation of ozone that is responsible for preventing radiation at wavelengths less than
about 290 nm from reaching the Earth's surface (Moseley, 1988; Diffey, 1991). Molina and
Rowland (1974) first proposed that chlorofluorocarbons and other gases released by human
activity could alter the natural balance of creative and destructive processes and lead to
depletion of the stratospheric ozone layer. Substantial reductions, ofup to 50%, in the ozone
column observed in the austral spring over Antarctica were first reported in 1985 and may
continue. There are, however, serious limitations in our current understanding of and ability
to quantify ozone depletion at the present levels of contaminant release and in our ability to
predict the effects on stratospheric ozone of any further increases (United Nations
Environment Programme, 1989; United Kingdom Stratospheric Ozone Review Group,
1991).
A number of factors influence terrestrial UVR levels:
Variations in stratospheric ozone with latitude and season (United Nations Environment
Programme. 1989)
Time of day: In summer, about 20-30% of the total daily amount ofUVR is received between 11:00
and 13:00 hand 75% between 9:00 and 15:00 h (Diffey, 1991: Table 2 and Fig. 3). Although the
amount of visible light falling on the ground in the summer may vary by only 30% between 12:00
and 15:00 h (local solar time), the short-wavelength component of the UVB spectrum undergoes a
dramatic change during
Table 2. Percentage of daily UVB and W A radiation received during
different periods of a clear summer's day. Solar noon is assumed to
be at 12:00 h, i.e., no allowance is made for daylight saving time
Latitude
CN)
UVB
11:00-13:00 h
20
40
60
From Diffey (1991)
30
28
26
9:00-15:00 h
78
75
69
UVA
11:00-13:00 h
27
25
21
9:00-15:00 h
73
68
60
Fig. 3. Daily variation in ultraviolet radiation: erythemal effective irradiance falling on
a horizontal earth surface at Denver, CO, USA, on one summer's day
100
c:
ns
-:.e-
ns ns'" 10
E ... E
0
.r:. .......
>. 1.0
... . _
W()-
Q)
::: 0.1
Q)
From Machta et al. (1975)
Clear day
........... /
10.00 12.00 14.00 16.00 18.00
Standard time (h)
this period. At a wavelength of 300 nm, the spectral irradiance decreases by 10 fold,
from approximately 1.0 to 0.1 11W/(cm
2
x nm) (Sliney, 1986).
Season: Seasonal variation in terrestrial UV irradiance, especially UVB, at the Earth's
surface is significant in temperate regions but much less nearer the equator (Table 3).
Table 3. Typical values for ambient daily and annual UVB
radiation expressed in minimal erythema dose (MED)
Latitude CN)
Diurnal UVB (MED)
Winter Autumn Summer Annual
20 (Hawaii, USA)
14 20 25 6000
30 (Florida, USA)
5 12 15 4000
40 (Spain) 2 7 12 2500
50 (Belgium2 0.4 3 10 1500
From Diffey (1991)
Geographical latitude: Annual UVR exposure dose decreases with increasing
distance from the equator (Table 3).
Clouds: Clouds reduce UV ground irradiance; changes in UVR are smaller than those
of total irradiance because water in clouds attenuates solar infrared radiation much
more than UVR. Even with heavy cloud cover, the scattered UVB component of
sunlight (often called skylight) is seldom less than 10% of that under clear sky;
however, very heavy cloud cover can virtually eliminate UVB even in summer.
Light clouds scattered over a blue sky make little difference in sunburning
effectiveness unless they directly cover the sun. Complete light cloud cover prevents
about 50% of UVB energy, relative to that from a clear sky, from reaching the
surface of the Earth (Diffey, 1991 ).
EXPOSURE DATA 53
Surface reflection: The contribution of reflected UVR to a person's total UVR expo-
sure varies in importance with a number of factors (Table 4). A grass lawn scatters
about 37 of incident UVB radiation. Sand reflects about 10-15%, so that sitting
under an umbrella on the beach can lead to sunburn both from scattered UVB from
the sky and reflected UVB from the sand. Fresh snow has been reported to reflect up
to 85-90% of incident UVB radiation, although reflectance of about 30-50% is
probably more typical. Ground reflectance is important, because parts of the body
that are normally shaded are exposed to reflected radiation (Diffey, 1990a).
Table 4. Representative terrain reflectance factors for horizontal surfaces
measured with a UVB radiometer at 12:00 h (290-315 nm) in the USA
Material Reflectance (%)
Lawn grass, summer, Maryland, California and Utah 2.0-3.7
Lawn grass, winter, Maryland 3.0-5.0
Wild grasslands, Vail Mountain, Colorado 0.8-1.6
Lawn grass, Vail, Colorado 1.0-1.6
Flower garden, pansies 1.6
Soil, clay/humus 4.0-6.0
Sidewalk, light concrete 10-12
Sidewalk, aged concrete 7.0-8.2
Asphalt roadway, freshly laid (black) 4.1-5.0
Asphalt roadway, two years old (grey) 5.0-8.9
House paint, white, metal oxide 22
Boat dock, weathered wood 6.4
Aluminium, dull, weathered 13
Boat deck, wood, urethane coating 6.6
Boat deck, white fiberglass 9.1
Boat canvas, weathered, plasticized 6.1
Chesapeake Bay, Maryland, open water 3.3
Chesapeake Bay, Maryland, specular component of 13
reflection at Z = 45 N
Atlantic Ocean, New Jersey coastline
Sea surf, white foam
Atlantic beach sand, wet, barely submerged
Atlantic beach sand, dry, light
Snow, fresh
Snow, two days old
From Sliney (1986)
8.0
25-30
7.1
15-1 8
88
50
Altitude: In general, each 300-m increase in altitude increases the sunburning
effectiveness of sunlight by about 4%. Conversely, places on the Earth's surface
below sea level have lower UVB exposures than nearby sites at sea level (Diffey,
1990a).
- Air pollution: Tropospheric ozone and other pollutants can decrease UVR,
particularly in urban areas (Frederick, 1990).
(a) Measurements of terrestrial solar radiation
Since UVR wavelengths between about 295 and 320 nm (UVB radiation) in the terres-
trial solar spectrum are thought to be those mainly responsible for adverse health effects, a
number of studies have concentrated on measuring this spectral region (Sliney, 1986).
Accurate measurements of UVR in this spectral band are difficult to obtain, however,
because the spectral curve of terrestrial solar irradiance increases by a factor of more than
five between 290 and 320 run (Fig 4). Nevertheless, extensive measurements of ambient
Fig. 4. Action spectrum designated by the American Conference of Governmental
Industrial Hygienists (ACGIH) for assessing the hazard of ultraviolet radiation (very
similar to erythemal action spectrum from 300-230 nm) and the solar spectrum. The
ACGIH action spectrum, which is unitless, is closely fit by some radiometers; however,
because of the small overlap of the terrestrial solar spectrum with the action spectrum,
problems of stray light must be dealt with by constant checks with a filter that blocks
wavelengths of less than 320 nm
1cY
-E
~
><
10
1
~ E
0
-
Q)
--
5;
:l
:=1.
~
1cf
>
Q)
1 Q)
0
c::
~ ro
:a
ro
ro ~
,_
10
1
,_
,_
:I:
~ a
.!:: (..)
(.)
<:
Q)
c..
"'
-2
,_ 10
~
0
r/)
1&
1 6
4
~ ~ ~ ~ ~ ~ ~ 10
4
280 300 320 340 360 380 400
Wavelength (nm)
Adapted from Sidney et al. (1990)
EXPOSURE DATA 55
UVR in this spectral band have been performed worldwide (Schulze, 1962; Schulze &
Grafe, 1969; Henderson, 1970; Sundararaman eta/., 1975; Garrison et al., 1978; Doda &
Green, 1980; Mecherikunnel & Richmond, 1980; Kostkowski et al., 1982; Ambach &
Rehwald, 1983; Blumthaler et al., 1983; Livingston, 1983; Blumthaler et al., 1985a,b;
Kolari et al., 1986; Hietanen, 1990; Sliney et al., 1990). Longer-wavelength UVR (UV A)
was measured at the same time in many of these studies. Measurements of terrestrial solar
UV A radiation are less subject to error than measurements of UVB, since the spectrum does
not vary widely with zenith angle and the spectral irradiance curve is relatively flat.
Maps of annual UVR exposure, such as that shown in Figure 5, have been compiled for
epidemiological studies of skin cancer and other diseases (Schulze, 1962, 1970; Scotto et
al., 1976). Despite the large numbers of measurements, their interpretation in relation to
human exposure has been complicated by three factors: (i) the considerable variation in
UVB spectral irradiance with solar position throughout the day and with season; (ii) the
effect of the geometry of exposure of individuals; and (iii) va1iation between humans in
outdoor exposure and the parts of their bodies that are exposed.
Fig. 5. Global distribution of ultraviolet radiation
From Schulz (1970); WHO (1979)
The total solar radiation that arrives at the Earth's surface is termed 'global radiation',
and measurements of terrestrial UVR most frequently pertain to this quantity, i.e. , the
radiant energy falling upon a horizontal surface from all directions (both direct and scattered
radiation). Global radiation comprises two components, referred to as 'direct' and 'diffuse'.
Approximately 70% of the UVR at 300 nm is in the diffuse component rather than in the
direct rays of the sun (Fig. 6). The ratio of diffuse to direct radiation increases steadily from
less than 1.0 at 340 nm to at least 2.0 at 300 nm (Garrison eta., 1978).
Fig. 6. Diffuse and direct solar spectral irradiance (solar zenith angle, 45)
Spectral irradiance !-1W/ (cm
2
x nm)
~ ~ ~ ~ o ~ ~ o ~ o ~ ~ . 5 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ o ~ ~ 5 ~
w
C)
C)
0
0
0
. :--.. ......... .
l'::. ::; .::
... -:;
,..
From Garrison et al. (1978)
0
8
-
....
0
0
Ratio diffuse: direct
<D
0
w
0
UVR reflected from the terrain (the albedo) may also be important; however, essentially
all measurement programmer have been limited to the direct and total diffuse components of
sunlight. While such measurements are of interest in calculating the exposure dose of UVR
of a prone individual, they are of very limited value in estimating exposure of the eye and
shaded skin surfaces (e.g., under the chin), where the UVB radiation incident upon the body
from terrain reflectance and horizon sky is of far greater impotiance. Sliney (1986) and
Rosenthal et al. (1988) reported measurements of outdoor ambient UVR that included the
reflected component to the eye. Exposure data for different anatomical sites is of value in
developing biological dose-response relationships (Diffey et al., 1979). The fact that ocular
exposure differs significantly from cutaneous exposure is emphasized by the finding that
photokeratitis is seldom experienced during sunbathing yet the threshold for UV photo-
keratitis is less than that for erythema of the skin (Sliney, 1986).
EXPOSURE DATA 57
Measurements of the angular distribution of UVR relative to solar position and cloud
distribution have been reported (Sliney, 1986; Fig. 7). A cloud obscuring the sun had no
effect upon the UV radiance of open blue sky or the horizon sky; however, when the sun
was 'out' (i.e., in an open sky), clouds near the horizon opposite the sun apparently reflected
more UVR than would otherwise be present from the blue sky. This confirms the findings
of studies of photographs of the sky taken through a narrow-band filter at 320 nm
(Livingston, 1983), which revealed that the sky looks almost uniformly bright even when
clouds are present and the clouds disappear into a uniformly hazy sky. Only the sun stands
out, as would be expected from the plots on Figure 7. When the sun is near the horizon and
can be looked at without great discomfort (i.e., at Z = 75-90 ), the effective UV irradiance
is again of the order of 0.3 ll W/cm
2
, e.g., about 0.08-1.1 ll W/cm
2
at an elevation angle of
12- 15 (Sliney, 1986).
Fig. 7. Semilogarithmic plots of the angular dependence of skylight for 290-315 nm
ultraviolet radiation (UVR) with the sun at zenith angle of about 45 o. A narrow
field-of-view detector was scanned from zenith to the horizon. Uppermost curves show
that direct UVR from the sun is more than 10 times greater than scattered UVR
normally incident upon the eye at near-horizon angles where the zenith angle Z =
70-90 . Most surprising is the similarity of blue sky and cloudy sky UV irradiances at
zenith or near the horizon
-
N
E
(.)
........
3:
~
-
C1>
~
-(.)
C1>
-
-UJ
- s
10
10 - e
10 -l
Sun's position
Hazy- ,... ... -- -.. ..,. Sunny-
toward /. ._ _toward
sun \ sun
'
, .
'
--....___'
~ .
------ - ~ - ,,;.--- -t
-;_.:.._ , __ ~ - +
Clear sky-away ~ ,
from sun '- ' -- *
1
"-----. I
'i
Cloudy bright-toward sun
10 -
6
L ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
0 30 60 90
Zenith angle (degrees)
Adapted from Sliney (1986)
(b) Personal exposures
The exposure of different anatomical sites to solar UVR depends not only on ambient
UVR and orientation of sites with respect to the sun but also on cultural and social
behaviour, type of clothing and whether spectacles are worn.
Measurements of ambient UVR are useful in that they provide upper limits on human
exposure (Scotto et al., 1976). They are of lesser value for assessing exposure doses
received by groups of individuals. Polysulfone film has been used to monitor personal
exposure to solar UVR (seep. 49). The wide variations in recorded exposure doses reflect
diversity of behaviour and, in most cases, the small numbers ( < 30) of subjects monitored.
Nevertheless, it can be estimated that recreational (excluding vacations) exposure to the sun
of people in northern Europe (where most of these studies were carried out) results in an
annual solar exposure dose to the face of 20-100 MED, depending on the propensity for
outdoor pursuits. The annual weekday UV exposure dose of indoor workers is around 30
MED; as a two-week outdoor vacation can result in a further 30-60 MED, the total annual
exposure dose to the face of most indoor workers is probably in the range 40-160 MED.
Outdoor workers at the same latitudes receive about two to three times these exposure
doses, typically around 250 MED (Diffey, 1987b; Slaper, 1987).
An alternative approach to estimating personal exposure is to combine measured data on
ambient UVR with a behavioural model of exposure. This approach was applied to a group
of more than 800 outdoor workers in the USA ( 40 ON) by Rosenthal et al. (1991 ). These
investigators estimated annual facial exposure doses of 30-200 MED, which are
considerably lower than those estimated for outdoor workers in northern Europe, perhaps
because Rosenthal et al. assumed facial exposure to be about 5-l 0% of ambient. A number
of researchers have used polysulfone film badges on both human subjects (Holman et al.,
1983a; Rosenthal et al., 1990) and mannequins (Diffey et al., 1977. 1979; Gies et al., 1988)
to measure solar UVR exposure on the face relative to ambient exposure. The results vary
considerably, reflecting factors such as positioning of film badges, behaviour of individuals,
solar altitude and the influence of shade. Examination of the data suggests, however, that the
exposure of an unprotected face is probably close to 20% of the ambient. Using this
estimate, the annual facial exposure doses in the outdoor worker group studied by Rosenthal
et al. (1991) would be about 80-500 MED. These data demonstrate clearly the current
uncertainties associated with estimates of population exposure doses.
1.3.2 Exposure to artificial sources of ultraviolet radiation
(a) Sources
Six artificial sources that often produce UVR incidental to the production of visible light
(Sliney & Wolbarsht, 1980; Phillips, 1983; Moseley, 1988) are described below.
(i) Incandescent sources
Optical radiation from an incandescent source appears as a continuous spectrum. Incan-
descent sources are usually ascribed a certain 'colour temperature', defined as the tempe-
rature of a black body that emits the same relative spectral distribution as the source. UVR
is emitted in significant quantity when the colour temperature exceeds 2500 K (2227 C).
Tungsten-halogen lamps in a quartz envelope (colour temperature, 3000 K [2727 OC]) may
emit significant UVR, whereas the US emission of an ordinary tungsten light bulb is
negligible.
EXPOSURE DATA 59
(ii) Gas discharge lamps
Another method of producing optical radiation is to pass an electric current through a
gas. The emission wavelengths are determined by the type of gas present in the lamp and
appear as spectral lines. The width of the lines and the amount of radiation in the interval
between them (the continuum) depend on the pressure in the lamp. At low pressures, fine
lines with little or no continuum are produced; as pressure is increased, the lines broaden
and their relative amounts alter. Low-pressure discharge lamps, commonly containing
mercury, argon, xenon, krypton or neon, are useful for spectral calibration.
Medium-pressure mercury lamps operate at an envelope temperature in the region of
600-800 C.
(iii) Arc lamps
Arc lamps operate at high pressures (20-100 atm [2020-10133 kPa]) and are very intense
sources ofUVR. Commonly available lamps contain xenon, mercury or a mixture of the two
elements, which are effective sources of UVR. Xenon arc lamps operate at a colour
temperature of 6000 K (5727 OC); they are often used as the light source in solar
simulation or are combined with a monochromator in spectral illumination systems.
Deuterium arc lamps provide a useful source of UVC radiation and find their main use in
spectrophotometers and as a calibration source for spectroradiometers.
(iv) Fluorescent lamps
The primary source of radiation in a fluorescent lamp arises from a low-pressure
mercury discharge, which produces a strong emission at 254 nm, which in tum excites a
phosphor-coated lamp to produce fluorescence. By altering the composition and thickness
of the phosphor and the glass envelope, a wide variety of emission spectral characteristics
can be obtained. The output is thus chiefly the fluorescent emission spectrum from the
coating, with a certain amount of breakthrough of UVB mercury lines at 297,303 and 313
nm, as well as those in the UVA and visible regions (WHO, 1979).
( v) Metal halide lamps
The addition of other metals (as halide salts) to a mercury discharge lamp allows for the
addition of extra lines to the mercury emission spectrum. Most such tubes are basically
medium-pressure discharge lamps with one or more metal halide additives, usually iodide.
Advantage has been taken of the strong lead emission lines at 364, 368 and 406 nm in the
lead iodide lamp, in which there is a 50% increase in output in the region between 355 and
380 nm compared to a conventional mercury lamp. Antimony and magnesium halide lamps
provide spectral lines in the UVB and UVC regions.
(vi) Electrodeless lamps
A type of lamp recently introduced on a large scale is the electrodeless lamp. In this
design, the discharge tube absorbs microwave energy fed, via waveguides, into a microwave
chamber containing the tube. Two 1500-W magnetrons generate microwave energy at 2450
MHz. The life of such lamps is longer than that of electrode lamps, and a greater range of
metal halides is available. Electrodeless lamps are used extensively for UV curing of inks
and coatings, particularly when a short lamp length is adequate for the area to be irradiated.
They have often been the first choice for curing prints on containers such as two-piece cans,
plastic pots and bottles, and tubes.
(b) Human exposure
Although the sun remains the main source of UVR exposure for humans, the advent of
artificial UVR sources has increased the opportunity for both intentional and unintentional
exposure.
Intentional exposure is most often to acquire a tanned skin, frequent ly using sunbeds and
solaria emitting principally UV A (315-400 nm) radiation (Diffey, 1987 c). Another reason
for intentional exposure to artificial UVR is the treatment of skin diseases, notably psoriasis.
Unintentional exposure is most often the result of occupation, and workers in many
industries (seep. 66) may be exposed to UVR from artificial sources. The general public is
exposed to low levels of UVR from sources such as fluorescent lamps used for indoor
lighting and may be exposed in shops and restaurants where UV A lamps are employed in
traps to attract flying insects.
(i) Cosmetic use
To some individuals, a tanned skin is socially desirable. A 'suntanning industry' has
grown up, particularly in northern Europe and North America, in which artificial sources of
UVR supplement exposure to sunlight.
Description of UVR sources used for tanning: Prior to the mid- 1970s, the source of
UVR was usually an unfiltered, medium- or high-pressure mercury arc lamp which emitted
a broad spectrum of radiation, from UVC through to visible and infrared radiation (Diffey &
Farr, 1991 b). The units often incorporated one or more infrared heaters and were commonly
called 'sunlamps' or ' health lamps' (Anon., 1979). One disadvantage of this type of unit
was that the area of irradiation was limited to a region such as the face and so whole-body
tanning was tedious. By incorporating several mercury arc lamps into a 'solarium', whole
body exposure was achieved. Tanning devices based on mercury arc lamps emit relatively
large quantities of UVB and UVC radiation, resulting in a significant risk of burning and
acute eye damage. Solaria that incorporate unfiltered mercury arc lamps are therefore now
less popular (Diffey, 1990a).
So-called UVB fluorescent lamps (e.g., Westinghouse FS Sunlamp, Philips TL12) emit
approximately 55% of their UV energy in the UVB and approximately 45% in the UVA
regions (Diffey & Langley, 1986). They were often used in tanning booths, more
commonly in the USA than in Europe.
Sunbeds, incorporating high-intensity UVA fluorescent lamps, were developed in the
1970s. These devices consist of a bed and/or canopy incorporating 6-30 fluorescent lamps
150-180 em in length. The earliest type of UV A lamp used in sunbeds is typified by the
Philips TL09, Wotan L100/79 and Wolff Solarium lamps (Diffey, 1987c). The spectral
power distribution from this type of lamp is shown in Figure 8a. The emission spectrum
comprises the fluorescence continuum, extending from about 315 to 400 nm and peaking at
350-355 nm, together with the characteristic lines from the mercury spectrum down to 297
nm (UVB) (Diffey & McKinlay, 1983). The UVA irradiance at the skin surface from a
typical sunbed or suncanopy containing these lamps is between 50 and 150 W 1m
2
(Bowker
& Longford, 1987; Bruyneel-Rapp et al., 1988).
EXPOSURE DATA 61
Fig. 8. Spectral emissions of different lamps used for cosmetic tanning: (a) Philips
TL09 (Diffey, 1987c); (b) Philips TLlOR (Diffey, 1987c); (c) Wolff Bellarium S (B.L.
Diffey, unpublished data); (d) optically filtered high-pressure metal halide lamp
(Diffey, 1987c)
0.1
0. 1
0.01 -
0.01
... 0.001
0.001
Q>

...
Q>
g 0.0001

0 0.0001 -
(ij
a.
!; (ij
()
280 310 340
...
Q> 370 400
- 280 340 370
(.)
400 a.
Wavelength (nm)
Ql
V>
a. Wavelength (nm)
Q>
<ll
>
Q>
-
> (d)

0.1
0.1
a:

0.01
a:
0.01
0.001
0.001
0.0001
0 .0001
310 340 370
Wavelength (nm)
400
280 310 340 370
Wavelength (nm)
400
In the mid-1980s, another type of UV A fluorescent lamp (Philips TLIOR) was
introduced especially for cosmetic tanning. The principal features of this type of lamp were
a reflector intrinsic to the lamp envelope and a fluorescence spectrum extending from about
340 to 400 nm, peaking at 370 nm (Fig. 8b ): note also the presence of characteristic mercury
lines in the UVB region. The skin surface itTadiance from a sunbed or suncanopy
incorporating Philips TLIOR lamps is typically around 250 W/m
2
(Diffey, 1987c).
Another type of UV fluorescent lamp that has been used in sun beds is the so-called 'fast
tan' tube. This type of lamp is typified by the Wolff Bellarium S, the spectral power
distribution of which is shown in Figure 8( c). The spectrum extends from about 290 to 400
nm and peaks at around 350 nm (Diffey & Farr, 1987).
Optically filtered, high-pressure mercury lamps doped with metal halide additives are
also used in cosmetic tanning. The spectral emission lies entirely within the UV A
waveband (Fig. 8d), and irradiances at the skin surface of more than 1000 W/m
2
can be
achieved. The best known of this type of unit is probably the UVASUN (Mutzhas, 1986).
A summary of the physical and photobiological emissions from these different types of
lamps is given in Table 5 (Diffey & Farr, 1991 a).
Table 5. Characteristics of different ultraviolet (UV) lamps used for tanning
Lamp Radiation emission(%)
UVA UVB uvc
Mercury arc sunlamp 40 40 20
Simulated sunlight lamp 95 5 0
Type 1 UV A lamp 99 1 0
Type II UV A lamp > 99.9 < 0.1 0
Optically filtered high-pressure lampa 100 0 0
Summer UV sunlightb 95 5 0
From Diffey & Farr (1991b) unless otherwise specified
aFrom Mutzhas (1986)
bFrom Sliney & Wolbarsht (1980)
Contribution to tanning(%)
UVA UVB uvc
0 35 65
20 80 0
60 40 0
> 90 < 10 0
100 0 0
20 80 0
Exposure to UVR sources used for tanning: Telephone surveys carried out in the
Netherlands (Bruggers et al., 1987) and in the United Kingdom (Anon., 1987) in the
mid-1980s showed that 7-9% of the adult population in each country had used sunbeds in
the previous one to two years. A more recent market survey in the United Kingdom (R.
McLauchlan, personal communication), with a sample size of 5800, gave a slightly higher
figure, with 10% of the population having used a sunbed during the previous year (1988)
and 19% of the sample admitting to having used a sunbed at some time in the past. In these
and other surveys in the United Kingdom (Diffey, 1986) and the USA (Dougherty et al.,
1988), women accounted for 60-85% of users, about half of the subjects being young
women aged between 16 and 30. The commonest reason given for using tanning equipment
was to acquire a pre-holiday tan (Anon., 1987; R. McLauchlan, personal communication);
other reasons included perceived health benefits, reduction of stress and improved
relaxation, protection of the skin before going on holiday, sustaining a holiday tan and
treatment of skin diseases such as psoriasis and acne (Diffey, 1986; Dougherty et al., 1988).
In the Dutch survey (Bruggers et al., 1987), about half of the users interviewed used
tanning equipment at home and the other half used facilities at commercial premises, such
as tanning salons, hairdressers, sports clubs and swimming pools. Most people had used
UV A equipment; 24% had used either UVB mercury arc sunlamps or solaria incorporating
these lamps. A more recent survey in the United Kingdom (McLauchlan, 1989) confirmed
the Dutch finding that the amount of use at home and at commercial premises was
approximately the same. A survey carried out at commercial establishments in the United
Kingdom indicated that all the equipment used emitted primarily UV A radiation, mostly
from fluorescent UV A lamps and 10% from optically filtered high-pressure metal halide
lamps (Diffey, 1986). Sales of tanning appliances in the United Kingdom increased rapidly
during the 1980s, but by the end of the decade there appeared to be a steady, or possibly
reduced, level of sales (Diffey, 1990a).
The mean number of tanning sessions per year in the Dutch study was 23 (Bruggers et
al., 1987). In the United Kingdom, half-hour sessions were the most popular (Diffey, 1986).
Each tanning session with UV A equipment normally results in an erythemally-weighted
EXPOSURE DATA 63
Exposure of about 0.8 MED (150 J/m
2
), whereas exposure to mercury arc lamps results in
about 2 MED per session ( 400 J/m
2
). In the Dutch survey, it was estimated that the median
annual exposure was 24 MED (4.8 kJ/m
2
) (Bruggers et al., 1987).
(ii) Medical and dental applications
UVR has both diagnostic and therapeutic applications in medicine and dentistry. The
diagnostic uses are confined largely to fluorescing of skin and teeth, and the UVR source is
normally an optically filtered medium-pressure mercury arc lamp producing radiation
mainly at 365 nrn (so-called 'Wood's lamps') (Caplan, 1967). Radiation exposure is limited
to small areas (< 15 em in diameter), and the UVA radiation dose per examination is
probably no more than 5 J/cm
2
The therapeutic uses of UVR, which result in considerably
higher doses, are mainly in the treatment of skin diseases and occasionally the symptomatic
relief of pruritus.
Phototherapy: The skin diseases that are most frequently treated with UVR are psoriasis
and eczema. Phototherapy of psoriasis at hospital may include the use of tar and related
derivatives and other substances, such as anthralin, on the skin (Morison, 1983a; see also
IARC, 1987a).
The first treatment of psoriasis with an artificial source of UVR is credited to
Sardemann, who used a carbon arc lamp of the type developed by Finsen at around the tum
of the century. These lamps were unpopular in clinical practice because they emitted noise,
odour and sparks, and they were superseded by the development of the medium-pressure
mercury arc lamp. In the 1960s, a variety of metal halides were added to mercury lamps to
improve emissions in certain regions of the UV and visible spectra. Fluorescent lamps were
developed in the late 1940s; since then, a variety of phosphor and envelope materials have
been used to produce lamps with emissions in different regions of the UV spectrum, such
that, today, there exists a wide range of lamps for the phototherapy of skin diseases (Diffey
& Fan, 1987).
Lamp systems can be classified into one of five categories in terms of suitability for
phototherapy (Diffey, 1990b ):
Type A: a single, medium-pressure mercury arc or metal halide lamp;
Type B: one or more vertical columns containing five or six optically filtered
high-pressure metal halide lamps;
Type C: a canopy or cubicle containing fluorescent sunlamps which emit
predominantly UVB but also significant amounts of radiation at
wavelengths below 290 nrn (e.g., Westinghouse FS sunlamp, Philips TL12
and Sylvania UV21 lamps);
Type D: a canopy, sunbed or cubicle incorporating fluorescent lamps which emit
predominantly UVB radiation and negligible amounts of radiation at wave-
lengths below 290 nm (e.g., the WolffHelarium);
Type E: a newly developed fluorescent lamp that emits a nanow band of radiation
around 311-312 nm (Philips TLOl).
The spectral power distributions characteristic of each of these five types of lamp are
shown in Figure 9. The therapeutic radiation for psoriasis lies principally within the UVB
waveband (Panish & Jaenicke, 1981), and the cumulative UVB dose required for clearing
Fig. 9. Spectral power distributions of different types of phototherapy lamp (Diffey,
1990b). Type A: unfiltered medium-pressure mercury arc lamp; type B: optically
filtered iron iodide lamp; type C: fluorescent sunlamp (Philips TL12); typeD: Wolff
Helarium lamp; type E: narrow-band UVB fluorescent lamp (Philip.s TLOI)
IC" 10 n
....
TYPE
TYPE B

IV -1
10 -I
10
;:)
\\
0.
0
0.
iG 10 -
2
- 10 -l
!:;
Cll
>. /
... !;j
(.,)
1)
Cl) -)
/\J
)
0. !0
/
:o -3
C/) .I
I
C/)
!'
ell
rJ' :o _,
'
r: .. ,
f/
Cll

Cll
J Qi
o: l:l -s
0: 10 -s
!
i
10 -s
10
200 250 300 350 40:) 2C:l 250 300 350
Wavelength (nm) Wavelength (nm)
10 IC
0
...
TYPE C ....
TYPED I

Cl) - 1
10 I !0 I
0 0
I
I
0. 0.
I
10 -z
""
iG :o -<
{
.'=
(.,)
d
(,)
10 -l 10 -;
!
i C/) C/)
I
Cl) Cl)
i
10 _,
> -
IV
';:: 10
Cll Cll
Gi
o: ID-s 0: :o -5
10-
6
I 0 -s
200 250 300 350 400 200 250 300 350 400
Wavelength (nm)
Wavelength (nm)
10 c
TYPE E
10 -
200 250 300 350 400
Wavelength (nm)
EXPOSURE DATA 65
psoriasis is typically 100-200 MED (Diffey, 1990a), usually delivered over a course
consisting of 10-30 exposures over 3-10 weeks (van der Leun & van Weelden, 1986).
Annual doses received by 90% of patients given UMB phototherapy for psoriasis range
from about 60 to 670 MED, with a typical dose in a single course being between 200 and
300 MED (Slaper, 1987).
Psoralen photochemotherapy (see also IARC, 1980, 1986a, 1987b): This form of treat-
ment, known colloquially as PUV A, involves the combination of photoactive drugs, pso-
ralens (P), with long-wave UVR (UVA) to produce a beneficial effect. Psoralen photo-
chemotherapy has been used to treat many skin disease in the past decade, although its
principal success has been in the management of psoriasis (Parrish et al. , 197 4 ), a disorder
characterized by an accelerated cell cycle and rate of DNA synthesis. Psoralens may be
applied to the skin either topically or systemically; the latter route is generally preferred, and
the psoralen most commonly administered is 8-methoxypsoralen. The patient is usually
exposed to OVA radiation from banks of fluorescent lamps with the spectral power
distribution shown in Figure 8a. Values for UV A irradiance in clinical treatment cubicles
have been found to range from 16 to 140 W/m
2
(Diffey et al., 1980; Diffey, 1990b ),
although an irradiance of 80 W/m
2
is probably typical. The UV A dose per treatment session
is usually in the range 1-10 J/cm
2
(Diffey et al., 1980).
Generally, approximately 25 treatments over a period of 6-12 weeks, with a cumulative
UV A dose of 100-250 J/cm
2
, are required to clear psoriatic lesions (Melski et a!. , 1977;
Henseler et al., 1981). PUVA therapy is not a cure for psoriasis, and maintenance therapy is
often needed at intervals of between once a week to once a month to prevent relapse (Gupta
& Anderson, 1987).
Neonatal phototherapy for hyperbilirubinaemia: Phototherapy is sometimes used in the
treatment of neonatal jaundice or hyperbilirubinaemia. The preferred method of treatment is
to irradiate the baby for several hours a day for up to one week with visible light,
particularly blue light (Sisson & Vogl, 1982). The lamps used for phototherapy, although
intended to emit only visible light, may also have a UV component: One commercial
neonatal phototherapy unit was found to emit not only visible light and UV A but also
radiation at wavelengths down to 265 nm (Diffey & Langley, 1986).
Fluorescence in cutaneous and oral diagnosis: Wood's light-a source of UV A obtained
by filtering optically a mercury arc lamp with 'blackglass'-is used by dermatologists as a
diagnostic aid in skin conditions that produce fluorescence (Caplan, 1967; Diffey, 1990a).
As irradiation of the oral cavity with a Wood's lamp can produce fluorescence under certain
conditions, this has been used in the diagnosis of various dental disorders, such as early
dental caries, the incorporation of tetracycline into bone and teeth, dental plaque and
calculus (Hefferren et al., 1971 ).
Polymerization of dental resins: Pits and fissures in teeth have been treated using an
adhesive resin polymerized with UV A. The resin is applied with a fine brush to the surfaces
to be treated and is hardened by exposure to UVA radiation at a minimal irradiance of 1 00
W/m
2
for 30 s or so (Eriksen et al., 1987; Diffey, 1990a).
(iii) Occupational exposures
Artificial sources of UVR are used in many different ways in the working environment.
In some cases, the UV source is well contained within an enclosure and, under normal
circumstances, presents no risk of exposure to personnel. In other applications of UVR, it is
inevitable that workers are exposed to some radiation, normally by reflection or scattering
from adjacent surfaces. Occupational exposure to UVR is also a consequence of exposure
to general lighting in the workplace.
Industrial photoprocesses: Many industrial processes involve a photochemical com-
ponent. The large-scale nature of these processes often necessitates the use of high-power
(several kilowatts) lamps such as high-pressure metal halide lamps (Diffey, 1990a).
The principal industrial applications of photopolymerization include the curing of
protective coatings and inks and photoresists for printed circuit boards. The curing of
printing inks by exposure to UVR is now widespread; as the cure takes only a fraction of a
second, UV drying units can be installed between printing stations on a multicolour line, so
that each colour is dried before the next is applied. Another major use of UV curing has
been for metal decorating in the packaging industry (Phillips, 1983). UV A is also used to
inspect printed circuit boards and integrated circuits in the electronics industry (Pauw &
Meulemans, 1987).
Artificial sources of UVR are used to test the weathering capability of materials such as
polymers. Xenon-arc lamps are often the light source because their emission spectra is
similar to the spectrum of terrestrial sunlight, although some commercial weathering
chambers incorporate carbon-arc lamps, high-pressure metal halide lamps or fluorescent
sunlamps (Davis & Sims, 1983).
Sterilization and disinfection: Radiation with wavelengths in the range 260-265 nm is the
most effective for this use, since it corresponds to a maximum in the DNA absorption
spectrum. Low-pressure mercury discharge tubes are thus often used as the radiation source,
as more than 90% of the radiated energy lies in the 254 nm line. These lamps are often
referred to as 'germicidal lamps', 'bactericidal lamps' or simply 'UVC lamps' (Diffey, 1990a).
UVC radiation has been used to disinfect sewage effluents, drinking-water, water for the
cosmetics industry and swimming pools. Germicidal lamps are sometimes used inside
microbiological safety cabinets to inactivate airborne and surface microorganisms (Diffey,
1990a). The combination of UVR and ozone has a very powerful oxidizing action and can
reduce the organic content of water to extremely low levels (Phillips, 1983).
Welding (see also IARC, 1990): Welding equipment falls into two broad categories: gas
welding and electric arc welding. Only the latter process produces significant levels of
UVR, the quality and quantity of which depend primarily on the arc current, shielding gas
and metals being welded (Sliney & Wolbarsht, 1980).
Welders are almost certainly the largest occupational group with exposure to artificial
sources of UVR. It has been estimated (Emmett & Horstman, 1976) that there may be as
many as half a million welders in the USA alone. The levels of UV irradiance around
electric arc welding equipment are high; effective irradiance (relative to the action spectrum
of the American Conference of Governmental Industrial Hygienists) at 1 mat an arc current
of 400 A ranged from 1 to 50 W/m
2
(Table 6), and the unweighted UV A irradiance ranged
from 3 to 70 W/m
2
, depending on the type of welding and the metal being welded (Cox,
EXPOSURE DATA 67
1987; Mariutti & Matzeu, 1987). It is not surprising therefore that most welders at some
time or another experience 'arc eye' or 'welder's flash' (photokeratitis) and skin erythema.
The effective irradiance at 0.3 m from many types of electric welding arcs operating at 150
A is such that the maximum permissible exposure time for an 8-h working period on
unprotected eyes and skin varies from a few tenths of a second to about 10 s, depending on
the type of welding process and the material used (Cox, 1987).
Table 6. Limits of exposure to ultraviolet radiation
and radiation effectiveness
Wavelength
(run)
180
190
200
205
210
215
220
225
230
235
240
245
250
254b
255
260
265
270
275
280b
285
290
295
297b
300
303b
305
308
310
313b
315
316
317
318
319
Exposure limit
(J/M2)
2500
1600
1000
590
400
320
250
200
160
130
100
83
70
60
58
46
37
30
31
34
39
47
56
65
100
250
500
1200
2000
5000
l.Ox 10
4
1.3 X 10
4
1.5x 10
4
1.9x 10
4
2.5x 10
4
Relative spectral
effectiveness S ~ c t
0.012
0.019
0.030
0.051
0.075
0.095
0.120
0.150
0.190
0.240
0.300
0.360
0.430
0.500
0.520
0.650
0.810
1.000
0.960
0.880
0.770
0.640
0.540
0.460
0.300
0.120
0.060
0.026
0.015
0.006
0.003
0.0024
0.0020
0.0016
0.0012
Table 6 ( contd)
Wavelength
(nm)
320
322
323
325
328
330
333
335
340
345
350
355
360
365b
370
375
380
385
390
395
400
Exposure limit
(J/m2)
2.9x 10
4
4.5 X 10
4
5.6 X 10
4
6.0x 10
4
6.8x 10
4
7.3 X 10
4
8.1 X 10
4
8.8x 10
4
1.1 X 10
5
1.3xl0
5
1.5 X 10
5
1.9 X 10
5
2.3 X 10
5
2.7 X 10
5
3.2x 10
5
3.9x 10
5
4.7 X 10
5
5.7 X 10
5
6.8x 10
5
8.3 X 10
5
1.5x10
6
Relative spectral
effectiveness (SJ..t
0.0010
0.00067
0.00054
0.00050
0.00044
0.00041
0.00037
0.00034
0.00028
0.00024
0.00020
0.00016
0.00013
0.00011
0.000093
0.000077
0.000064
0.000053
0.000044
0.000036
0.000030
From American Conference of Governmental industrial
Hygienists (1991); wavelengths chosen are representative,
and other values should be interpolated at intennediate
wavelengths.
aFor explanation, see pp. 46-47
bEmission lines of a mercury discharge spectrum
In a survey of electric arc welders in Denmark, 65% of those questioned had experienced
erythema; however, as no indication of the frequency of skin reactions was reported, it is not
possible to estimate annual exposure (Eriksen, 1987). Monitoring of the exposure to UVR
of non-welders working in the vicinity of electric arc welding apparatuses showed that their
daily exposure dose exceeded the maximum permissible exposure limits by almost an order
of magnitude (Barth et al., 1990).
Phototherapy: Although there is a trend to the use of enclosed treatment cubicles, some
of the lamps used to treat skin disease (see the section on medical and dental applications)
are unenclosed, emit high levels of UVR and can present a marked hazard to staff; at 1 m
from these lamps, the recommended 8-h occupational exposure limits can be exceeded in
less than 2 min (Diffey & Langley, 1986).
In a study of the exposure of staff in hospital phototherapy departments (Larko & Diffey,
1986), annual exposure to UVR could be estimated from the number of occasions per year
on whi ch staf:fuad experienced at least minimal erythema (Diffey, 1989b). Estimated annual
EXPOSURE DATA 69
occupational exposures to UVR were 15,92 and 200 MED, corresponding to a frequency of
erythema of once per year, once per month and once per week, respectively.
Operating theatres: UVC lamps have been used since the 1930s to decrease the levels of
airborne bacteria in operating theatres (Berg, 1987). The technique requires complete
protection of the eyes and skin of staff and patients; for this and other reasons, filtered air
units are often preferred.
Research laboratories: Sources of UVR are used by most experimental scientists
engaged in aspects of photobiology and photochemistry and in molecular biology. These
applications, in which the effect of UV irradiation on biological and chemical species is of
primary interest to the researcher, can be differentiated from UV fluorescence by absorption
techniques where the effect is of secondary importance (Diffoy, 1990a).
UV photography: There are two distinct forms of UV photography: reflected or
transmitted UV photography and UV fluorescence photography. In both applications, the
effective radiation lies within the UV A waveband (Lunnon, 1984 ).
UV lasers: High-power lasers which emit in the UV region, used in nuclear and other
research laboratories, are far less common than those that emit in the visible or infrared
regions of the electromagnetic spectrum.
Nitrogen lasers emit at a wavelength of 337 nm (Phillips, 1983), and instruments with a
peak power output of up to 2.3 MW per pulse are available. Nitrogen lasers can be used in
conjunction with fluorescent dyes to produce spectral emissions of 360-900 nm, with a
power pulse of 200-480 kW. If frequency doubling crystals are used in conjunction with a
nitrogen laser, UV emissions down to 260 nm are possible.
An alternative laser source of UVR is the excimer laser. (The term 'excimer' denotes a
homonuclear molecule which is bound in an electronically excited state but is dissociative in
the ground state [Phillips, 1983].) The wavelength of the pulsed UVR from this type of
laser depends on the excimer molecules, such as ArF, F
2
, XeCI and KrF, which emit at 193,
157, 308 and 248 nm, respectively (Phillips, 1983; Bos & de Haas, 1987). On the basis of
worst-case assumptions, the estimated annual risk for skin cancer for workers exposed to
UV lasers in medical applications is equivalent to about one additional day of sunbathing,
and that for workers exposed to UV lasers in laboratories is comparable to the risk for
outdoor workers (Sterenborg et al., 1991).
Quality assurance in the food industry: Many contaminants of food products can be
detected by UV fluorescence techniques. For example, the bacterium Pseudomonas
aeruginosa, which causes rot in eggs, meat and fish, can be detected by its yellow-green
fluorescence under UV A irradiation. One of the longest established uses of UV A
fluorescence in public health is to demonstrate contamination with rodent urine, which is
highly fluorescent (Ultra-Violet Products, Inc., 1977).
Insect traps: Many flying insects are attracted by UV A radiation, particularly in the
region around 350 nm. This phenomenon is the principle of electronic insect traps, in which
a UV A fluorescent lamp is mounted in a unit containing a high-voltage grid. The insect,
attracted by the UV A lamp, flies into the unit and is electrocuted in the air gap between the
high-voltage grid and a grounded metal screen. Such units are commonly found in areas
where food is prepared and sold to the public (Diffey, 1990a).
Sunbed salons and shops: The continuing popularity of UVA sunbeds and suncanopies
for cosmetic tanning has resulted in the establishment of a large number of salons and shops
selling sunbeds for use at home. Some shops may have 20 or more UV A tanning
appliances, all switched on, thus exposing members of the public and staff to high levels (>
20W/m
2
) ofUVA radiation (Diffey, 1990a).
Discotheques: UV A 'blacklight' lamps are sometimes used in discotheques to induce
fluorescence in the skin and clothing of dancers. The levels of UV A emitted are usually low
(< 10 W/m
2
) (Diffey, 1990a).
Offices: Signatures can be verified by exposing a signature obtained with colourless ink
to UV A radiation, under which it fluoresces. UV A exposure of office staff is normally to
hands, and irradiance is low(< 10 W/m
2
) (Diffey, 1990a).
(iv) General lighting
Fluorescent lamps used for general lighting in offices and factories emit small quantities
of both UV A and UVB. A UV A irradiance of 30 mW/m
2
(Diffey, 1990a) and a UVB irra-
diance of 3 m W 1m
2
(McKinley & Whillock, 1987) were found for bare fluorescent lamps
with a typical illuminance of 500 lux. These UV levels give rise to an annual exposure of
indoor workers to no more than 5 MED, and this dose can be reduced appreciably by the use
of plastic diffusers (McKinley & Whillock, 1987). A study of the personal doses of UVR
received by workers in the car manufacturing industry who were engaged in inspecting
paintwork of new cars under bright fluorescent lamps indicated a similar annual exposure
(Diffey et al., 1986). Most plastic diffusers reduce erythemally effective irradiance to 0.2%
or less of that of the bare lamp. An exception is clear acrylic diffusers, which absorb only
about 20% of the erythemally effective radiation. The absorption of UV A radiation by
diffusers is less effective, transmission ranging from 1% for opal polycarbonate to 74% for
clear acrylic (McKinley & Whillock, 1987). Spectroradiometric measurements of the UV
levels from indoor fluorescent lamps carried out in the USA, however, indicated much
higher annual doses for people exposed occupationally for 2000 h per year: The annual
estimated exposure dose ranged from 8 to 30 MED for an illuminance level of 500 lux from
bare lamps (Cole et al., 1985).
Desk-top lights which incorporate tungsten-halogen (quartz) lamps may result in
exposure to UVR of the hands and arms, if the lamps are used in excess of recommended
occupational exposure levels (McKinley et al., 1989). Experimental studies have shown
that erythema can be induced in susceptible individuals after a 15-min exposure at I 0 em
from a 1 00-W tungsten-halogen source, principally by the UVB component of the emission
(Cesarini & Muel, 1989). Tungsten-halogen lamps are also used for general lighting (e.g. ,
spotlights, indirect lighting, floor lamps) in some countries.
(c) Regulations and guidelines
(i) Cosmetic use
The most comprehensive guidelines for the use of sunlamps and sun beds in cosmetic
tanning are those published by the international Electrotechnical Commission ( 1987, 1989).
The guidelines classify tanning appliances into one of four types according to the effective
irradiance at short (A. s_320 nm) and long (320 <A. s_ 400 nm) UV wavelengths (Table 7).
EXPOSURE DATA
Table 7. Classification of tanning appliances
Type Effective irradiance (W /m )
/.. < 320 nm 320 < A< 400 nm
1 < 0.0005 2:0.15
2 0.0005-0.15 2: 0.15
3 < 0.15 < 0.15
4 > 0.1 < 0.15
From International Electrotechnical Commission
(1989)
Effective radiance is defmed as:
400
lSO
71
where E;. is the spectral irradiance (W/m
2
x nm) at wavelength A. (nm) at the shortest
recommended exposure distance; b.t.. is the wavelength interval used in the summation; and
S;. is the relative erythemal effectiveness recently adopted by the Commission Internationale
de l'Eclairage (McKinley & Diffey, 1987), specified as shown in Table 8. The guidelines
recommend that the exposure time for the first session on untanned skin should correspond
to an effective dose not exceeding 100 J/m
2
; this is approximately equivalent to 1 MED for
subjects with sun-reactive skin type I. The annual exposure should not exceed an effective
dose of 25 kJ/m
2
(International Electrotechnical Commission, 1989).
Table 8. Specifications of relative erythemal effectiveness
Wavelength(/..; nm) Relative erythemal effectiveness (Sf..)
(weighting factor)
A< 298 1
298 <A< 328 1 0
0094
(
298
'-l
328 < A::; 400 1 00.015(139-A)
From McKinlay & Diffey (1987); International Electrotechnical
Commission ( 1989)
Although these guidelines form the basis of several national standards on sunlamp and
sunbed use, it should be noted that variations exist; for example, in the Netherlands, Norway
and Sweden, certain UV appliances are not permitted. Regulations concerning the use of
tanning appliances are in force in only a few countries, but many others have published
advice on sunbed use, including information on adverse effects, as well as guidelines on
manufacturing standards.
(ii) Occupational exposure
Guidance on the maximal limits of exposure to UVR as a consequence of occupation is
given by the International Non-ionizing Radiation Committee of the International Radiation
Protection Association. These exposure limits, which apply only to incoherent (i.e., non-
laser) sources, represent conditions under which it is expected that nearly all individuals
may be repeatedly exposed without adverse effects and are below levels which would be
used for medical or cosmetic exposure to UVR. The limits for occupational exposure to
UVR incident upon the skin or eye were considered separately for the UVA spectral region
(315-400 nm) and the actinic UV spectral region (UVC and UVB, 180-315 nm). In 1984,
the limit provided an equal spectral weighting between 315 and 400 nm, a maximal I 000-s
radiant exposure of 10 KJim
2
and a maximal irradiance of 10 W 1m
2
for longer periods
(International Non-ionizing Radiation Committee of the International Radiation Protection
Nssociation, 1985). Studies of skin and ocular injury resulting from exposure to UV A led
the Committee to issue revised exposure limits in 1988: For the UVA spectral region
(315-400 nm), the total radiant exposure incident upon the unprotected eye should not
exceed 1.0 Jlcm
2
(I 0 kJim
2
) within an 8-h period, and the total 8-h radiant exposure incident
upon the unprotected skin should not exceed the values given in Table 6. Values for the
relative spectral effectiveness S
1
, are given up to 400 nm to expand the action spectrum into
the UV A region for determining the exposure limit for skin exposure. For the actinic UV
spectral region (UVC and UVB, 180-315 nm), the radiant exposure incident upon the
unprotected skin or eye within an 8-h period should not exceed the values given in fable 6
(International Non-ionizing Radiation Committee of the International Radiation Protection
Association, 1989).
'l'he effective irradiance (Eerr) in Wlm
2
of a broad-band source weighted against the peak
of the spectral effectiveness curve (270 nm) is determined according to the formula:
E eff = L EA. X s ~ . . X ~ A .
where Et... is the spectral irradiance (W 1m
2
x nm) from measurements, s ~ . . is the relative
spectral effectiveness (Table 6) and ~ ~ is the band-width (nm) of the calculation or
measurement interval (International Non-ionizing Radiation Committee of the International
Radiation Protection Association, 1985).
The maximal permissible exposure time in seconds for exposure to UVR incident on the
unprotected skin or eye within an 8-h period is computed by dividing 30 Jl m
2
by the value
of Eerr in Wl m
2
(American Conference of Governmental Industrial Hygienists, ]991). A
worker receiving the maximal permissible exposure of 30 11m
2
per 8-h day will, in the
course of a working year, have a cumulative dose of 60-70 MED (Diffey, 1988), a value
comparable with the natural exposure of non-occupationally exposed indoor workers
(Diffey, 1990a).
Occupational exposure limits to lasers were also defmed by the International Non-
Ionizing Radiation Committee of the International Radiation Protection Association in 1989,
at 3 mJi cm
2
and 40 mJi cm
2
over 8 h for argon-fluoride and xenon-chloride lasers,
respectively (Sliney, 1990).
CONTINUE
2. Studies of Cancer in Humans
2.1 Solar radiation
2.1.1 Nonmelanocytic skin cancer
Nonmelanocytic skin cancer is classified into two major histological types: basal-cell
carcinoma and squamous-cell carcinoma. Basal-cell carcinoma is the commoner type in
white populations. No information was available to the Working Group on other types of
nonmelanocytic skin cancer.
(a) Case reports
In general, case reports were not considered, owing to the availability of more infor-
mative data.
(i) Studies of xeroderma pigmentosum patients
Xeroderma pigmentosum is a rare autosomal-recessive genetic disease in which there is
an excision repair defect, as observed in cultured skin fibroblasts damaged by UVR
(Cleaver, 1968). Patients display cellular and clinical hypersensitivity to UVR (Kraemer,
1980). The disease is present in about one in 250 000 people in the USA and Europe
(Cleaver & Kraemer, 1989), and as many as 1 in 100 000 (Takebe et al., 1987) or even 1 in
40 000 (Cleaver & Kraemer, 1989) people may be affected in Japan.
In a survey of 830 cases located through published case reports (Kraemer et al., 1987),
45% had malignant skin neoplasms. Most of the patients were young, and the median age
of development of the first skin cancer in the 186 patients for whom information was
available was eight years: this observation presumably represents a substantial excess over
the expected number. Only 259 neoplasms were specifically categorized as basal- or
squamous-cell carcinoma in the published reports. Of these 97% were on constantly
exposed sites (face, head and neck) by comparison with 80% of similar tumours in the US
general population. [The Working Group recognized that data collected from previously
published case rep011s is not uniform and may not be typical of a true incidence or
prevalence series.]
(ii) Studies of transplant recipients
Australian renal transplant recipients were reported to have an increased risk for non-
melanocytic skin cancer (Hardie et al., 1980). Among 875 male and 669 female
Australasian recipients, aged 35-64, 47 squamous-cell carcinomas and 27 basal-cell
carcinomas were observed among males and 27 squamous-cell and 15 basal-cell carcinomas
were observed among females (Kinlen et al., I 979). The rates/1 0
5
person-years for
squamous-cell carcinoma were 2680 in males and 171 0 in females, or 3. 0 and 5. 9 times the
rates observed among residents of the same age distribution surveyed in Gerald ton, Western
Australia (Kricker et al., 1990). For basal-cell carcinoma, the rates for 1540 (males) and
940 (females) were 1.154 and 1.150 times the Geraldton rates, respectively.
-73-
By February 1980, a registry in Denver, Colorado (USA), had received data on 906
organ transplant recipients who had developed 959 types of cancer: 42% arose in the skin,
of which 47% were squamous-cell carcinomas (Penn, 1980). While several studies from
areas with lower solar radiation are available (Boyle et al., 1984), neither singly nor
collectively do they contain enough observations to permit a comparable calculation.
(b) Descriptive studies
Nonmelanocytic skin cancer is often not recorded in cancer registries (e.g., in the USA
and in most parts of Australia), and when it is registered case ascertainment is likely to be
incomplete since many patients are treated in consulting rooms, frequently without histo-
logical verification (Doll et al., 1970). Thus, descriptive studies of the incidence of non-
melanocytic skin cancer can be difficult to perform because of the absence of routinely
collected data or difficult to interpret because of incomplete registration. Studies in
Australia and the USA have relied upon special surveys, while in the United Kingdom and
the Nordic countries data from cancer registries have been used. Studies of mortality rates
are also difficult to interpret because nonmelanocytic skin cancer is rarely fatal, and many
deaths are incorrectly attributed to skin cancer (Muir et al., 1987).
A number of features of the occurrence of nonmelanocytic skin cancer as revealed by
descriptive studies have been taken as evidence that exposure to the sun is a major cause of
the disease. These include features presumed to be related to sun exposure such as sex,
anatomical site, latitude of residence (or annual dose of UVB radiation), migration from
places of low insolation to places of high insolation, occupation and features related to
sensitivity to the sun such as race (i.e., degree of skin pigmentation).
(i) Host factors
The occurrence of nonmelanocytic skin cancer according to host factors such as race
provides indirect evidence that sunlight is a cause. In most white populations, non-
melanocytic skin cancer occurs more commonly in men than in women (Muir et al., 1987).
The highest incidence rates have been recorded among Australians, who are largely of
British (Celtic) descent (Giles et al., 1988). Populations with greater skin pigmentation
have low rates of nonmelanocytic skin cancer, for instance, in South Africa (Oettle, 1963)
and Singapore (Shanmugaratnam et al., 1983).
Albinism is an inherited disorder of melanin metabolism, with a decrease or complete
absence of melanin. Large numbers of skin cancers (mostly squamous-cell carcinomas)
have been reported in albinos (Luande et al., 1985; Kromberg et al., 1989).
(ii) Anatomical distribution
The majority of cases of skin cancer recorded in cancer registries (Haenszel, 1963
[USA]: Whitaker et al., 1979 [United Kingdom]; Swerdlow, 1985 [United Kingdom]: Levi
et al., 1988 [Switzerland]; 0sterlind et al., 1988a [Denmark]; Moanetal., 1989 [Norway])
and in special surveys in the USA (Haenszel, 1963; Scotto et al., 1983) occurred on the head
and neck. In contrast, in two studies in Australia-one of incidence (Giles et al., 1988) and
the other of prevalence (Kricker et al., 1990)-the proportions of cancers on the head and
neck were lower. [The Working Group noted that the contrasting results may be due to time
differences.] In the incidence survey, 43% of squamous-cell carcinomas and 66% of
STUDIES OF CANCER IN HUMANS 75
basal-cell carcinomas were on the head and neck. In the prevalence survey, about one-third
of all basal-cell carcinomas were on the head and neck, whereas the trunk accounted for
about half of these lesions. The density of tumours was five times greater in men and eight
times greater in women on usually exposed sites than on sites which were sometimes
exposed. Squamous-cell carcinomas occurred almost exclusively on exposed sites. The site
distributions of both types of nonmelanocytic skin type are generally similar in the two
sexes (0sterlind et al., 1988a; Moan et al .. 1989; Kricker et al.. 1990).
A distinctive feature of the site distribution of basal-cell carcinoma is a virtual absence
on the dorsa of the hands and infrequent occurrence on the forearms, compared with the
distribution of squamous-cell carcinoma (Haenszel, 1963; Silverstone & Gordon, 1966; Levi
et al., 1988; Magnus, 1991). Basal-cell carcinoma also occurs frequently on parts of the
face that receive comparatively little sun exposure (Urbach et al., 1966).
[The Working Group noted that cancers on the head and neck may be more likely to be
diagnosed than cancers at other sites.]
(iii) Geagraphical variation
Nonmelanocytic skin cancer incidence and mortality have long been known to increase
with increasing proximity to the equator. Gordon and Silverstone (1976) demonstrated a
negative correlation between incidence of nonmelanocytic skin cancer in various countries
and latitudes by tabulating the incidence according to latitudinal zones. Much of the early
evidence came from surveys conducted in the USA. Tn the first of these, Dom (1944a,b,c)
reported the results of the US First National Cancer Survey conducted in 10 urban areas in
1937-38. [Nonmelanocytic] skin cancer incidence was greater among whites living in the
south than in the north of the country. Blum (1948) subsequently reanalysed these data,
substituting latitude for place of residence, and showed a strong inverse relationship
between incidence of mostly nonmelanocytic skin cancer and latitude. No other cancer,
with the exception of the buccal cavity (including the lip), showed a similar latitude
gradient.
Auerbach (1961), using data from the US Second National Cancer Survey conducted in
1947-48 in the same areas as the previous survey, calculated that the age-adjusted rates for
skin cancer doubled for each 3 48 ' (approximately 265 miles) of latitude towards the
equator; similar gradients were seen for men and women and in all age groups. Haenszel
(1963) reanalysed data from this survey for four southern and four northern cities. The
inverse gradient with latitude was present for both basal-cell and squamous-cell carcinoma.
In addition, there was some evidence that the gradient was strongest for head, neck and
upper limbs (sites which are usually exposed).
A similar latitude gradient was seen in the US Third National Cancer Survey (Scotto et
al., 1974). Inverse latitude gradients have also been reported in Australia (Silverstone &
Gordon, 1966; Giles et al., 1988) and in the Nordic countries (Teppo et al. , 1980; Moan et
at., 1989; Magnus, 1991 ).
Several authors have correlated nonmelanocytic skin cancer incidence (or mortality)
with estimates of UVR. Green et al. (1976) reported a positive correlation between
estimates of annual UV dose and of incidence rates in the USA, the United Kingdom,
Canada and Australia. Estimates of UV dose were derived from models relating latitudinal
and seasonal ozone distributions, adjusted for cloud cover. [The Working Group noted that
no allowance was made in the analysis for different methods of case ascertainment. It is not
clear how well the predicted values were correlated with actual levels ofUVR.]
A positive correlation, stated to be stronger than that for latitude, was seen between
UVR, as measured by Robertson-Berger meters, and the incidence of nonmelanocytic skin
cancer in four cities in the US Third National Cancer Survey (Scotto et al. , 1982). Scotto et
a!. (1983) examined incidence data collected in eight cities in 1977-78 and again showed an
inverse relationship with latitude and a positive correlation with measurements of UVR.
The gradient was steeper for squamous-cell than for basal-cell carcinoma.
Moan et al. (1989) examined nonmelanocytic skin cancer incidence in six regions of
Norway from 1976 to 1985, excluding the area around Oslo to reduce bias due to possible
differences in reporting and diagnosis. Two measures of UVR, one weighted according to
the action spectrum for erythema and the other according to the action spectrum for
mutagenesis in cells in the basal layer of the skin, were derived from atmospheric models.
Similar, positive relationships between UVR and nonmelanocytic skin cancer incidence
were obtained with each method.
Elwood et al. (1974) conducted a study of mortality from nonmelanocytic skin cancer in
the contiguous states of the USA and in all of the provinces of Canada fit 1950-67. The
correlation between latitude and mot1ality was as strong as that between mortality and an
index of UVR derived from a model relating erythemal dose according to latitude with
adjustments for cloud cover.
(iv) Migration
Studies of migrants to Australia (and other countries with high exposure to the sun) offer
the opportunity to examine, indirectly, the effect of exposure to the sun. Most migrants to
Australia come from higher latitudes which have lower levels of exposure to the sun than
Australia. The effect of exposure to the sun is most readily examined in migrants from the
British Isles to Australia, from whom most Australians are descended.
Armstrong et al. ( 1983) found that the age-adjusted mortality rate among men born in
England or Wales was 0.55 (95% confidence interval (CI), 0.43-0.71) times that in
Australian-born men. There was little evidence that rates in migrants increased with
duration of residence in Australia, although the numbers of deaths were small and the rates
unstable.
Giles et al. (1988) found age-adjusted incidence rates of 402 per 100 000 person-years
among immigrants from the British Isles and 936 in the Australian-born population.
(v) Occupation
Death certificates for 1911-44 in England and Wales were used in an analysis of cancer
of the skin, excluding melanomas, in male agricultural workers, miners and quarriers and
professionals (Atkin et al. . 1949). During part of the period (1911-16), cancers of the penis,
scrotum and skin were classified together, and the numbers of cancers of the skin alone were
estimated from the proportions occurring in the later period. The standardized mortality
ratios (SMRs) were greater for those engaged in agriculture (142.4 [137.4-147.6]) than for
those in mining (94.4 [88.8-l 00.3]), and lowest of all for professionals ( 47.5 [ 42.6-52.9]).
Whitaker et al. (1979) examined occupations among cases of squamous-cell carcinoma
reported to the Manchester Regional Cancer Registry, United Kingdom, in 1967-69. The
occupations of 23% of cases were not ascertained. In men, standardized registration ratios
(SRRs) were elevated for textile workers (238;p < 0.001) and farmers (243;p < 0.001). The
SRR was also high for female farmers (690;p < 0.001). Male fishermen, chemical workers
and paper/printing workers had high SRRs for squamous-cell carcinoma of the arm, and
building workers for squamous-cell carcinoma of the ear.
The association between occupation and nonmelanocytic skin cancer was examined in
England and Wales in 1970-75 in a 10% sample of all male incident cases for which occu-
pation was recorded (Beral & Robinson, 1981 ). Individuals were assigned, on the basis of
stated occupation, to one of three groups: outdoor workers, indoor office workers and other
indoor workers, according to the classification of occupations of the Office of Population
Censuses and Surveys. The SRRs for men aged 15-64 were 110 [95% CI, 1 09-116] for
outdoor work, 97 [92-103] for office work and 92 [86-89] for other indoor work. Since
place of work may be confounded with social class, the analyses were repeated for men
aged 15-64 years in social class III; the SRRs were ll2 [102-122] for outdoor work,
111[100-123] for office work and 85 [78-92] for other incioor work.
Vagero et al. (1986) linked cancer incidence data in Sweden from 1961 to 1979 with
census data from 1960 to determine the occupations of cases of nonmelanocytic skin cancer.
Occupations were classified into three main groups: office workers, other indoor workers
and outdoor workers. SRRs standardized for age, county of residence and social class, were
slightly higher for outdoor workers (106; 95% CI, 101-112) than for office workers (103;
96-110) and other indoor workers (95; 91-100). The authors noted that registration may
have been more complete among high socioeconomic groups.
(c) Cross-sectional studies
Design features of cross-sectional studies of exposure to the sun are summarized in
Table 9, and the results are shown in Table 10.
A population-based survey of the prevalence of nonmelanocytic skin cancer [types not
separated] was conducted in County Galway, Ireland (O'Beim et al., 1970). Exposed areas
of skin were examined for the presence of cancers. In the 26 cases found, there was no
significant association with frequent severe sunburn for basal-cell or squamous-cell skin
cancer; among males, there was a positive relationship between cumulative hours of
exposure to sunlight and the prevalence of nonmelanocytic skin cancer.
Silverstone and Gordon (1966) and Silverstone and Searle (1970) reported the results of
three surveys in Queensland, Australia. Exposed areas of the skin were examined, and
subjects were asked to report previously treated nonmelanocytic skin cancer [types not
separated]. Women performing home duties were classified as indoor workers. Outdoor
occupation showed a weakly positive association with past and present incidence in men
and a negative association in women.
Holman et al. (1984a) conducted a population-based survey of 1216 subjects in western
Australla. After controlling for age, cutaneous sun damage (as assessed by
microtopography) was strongl y related to a past history of nonmelanocytic skin cancer.
Engel et al. (1988) analysed data on basal-cell epithelioma (carcinoma) from the First
National Health and Nutrition Examination Survey in the USA (1971-74). Dermatologists
diagnosed skin cancers and assessed actinic skin (solar) damage, but histological con-
fitmation of the diagnosis was not obtained routinely. Strong associations between the
prevalence of basal-cell epithelioma and solar skin damage were seen in both men and
women.
Green et al. (1988a) conducted a survey of the prevalence ofnonmelanocytic skin cancer
[types not separated for calculation of RR] in Queensland, Australia. Information about
exposure to the sun was obtained from questionnaires; dermatologists diagnosed skin
cancers and assessed signs of actinic damage (solar lentigines, telangiectasia of the face,
solar elastosis of the neck and solar keratoses). After adjustment for age, sex, skin colour
and ability to tan, outdoor occupation and number of sunburns were both weakly associated
with increased prevalence. Stronger associations were seen for cutaneous indicators of sun
exposure, particularly for solar lentigines on the hands and telangiectasia on the face.
Recreational exposure was not associated independently with nonmelanocytic skin cancer.
In a later report (Green, 1991), the occurrence of nonmelanocytic skin cancer was posi-
tively correlated with grade of cutaneous microtopography.
In a subsequent study (Green & Battistutta, 1990), subjects were asked to report
nonmelanocytic skin cancer treated between 1 December 1985 and 30 November 1987,
around the survey in 1986. Meclical records were searched to confirm the diagnoses.
Subjects who had had a skin cancer diagnosed at the prevalence survey were excluded.
Outdoor occupation, outdoor leisure activities and number of sunburns showed little
association with basal-cell carcinoma in an analysis including past history of skin cancer.
All three variables were related to incidence of squamous-cell carcinoma. [The Working
Group noted that the exclusion of subjects found to have skin cancer during the prevalence
survey makes interpretation of these results difficult. The inclusion of past history of skin
cancer in the analysis would have weakened any association with exposure to the sun.]
Vitasa et al. (1990) conducted a survey of the occurrence of nonmelanocytic skin cancer
among men engaged in traditional fishing practices ('waterman') in Maryland, USA.
Subjects were examined by dermatologists and interviewed about their history of exposure
to the sun. Estimates of individual annual and lifetime doses of UVB radiation were made
by weighting the ambient UVR by a history of occupation and outdoor activities and by
taking into account relative doses recorded by film dosimeters on the fierce. Patients with
squamous-cell carcinoma aged 15-60 had had an 11% higher annual dose of UVB radiation
and those with basal-cell carcinoma had had an 8% lower annual dose than that of
age-matched watermen without cancers. The effect of cumulative UVB radiation was
examined after adjustment for age, eye colour, childhood freckling and skin reaction to
sunlight, all of which were positively associated with occurrence of both types of
nonmelanocytic skin cancer. Cumulative UVB radiation dose was not associated with
basal-cell carcinoma but was positively associated with squamous-cell carcinoma. The
latter association was significant in a comparison of the top quarter of cumulative UVB
versus the bottom three-quarters but not in a comparison of exposures above and below the
median. [The Working Group noted that the results for the two types of cancer are not
necessarily incompatible, both because of the small number of cases and the fact that the
diagnosis was confirmed histopathologically in only 62%.]
Table 9. Design features of cross-sectional studies of sun exposure and nonmelanocytic skin cancer
Reference Place Period of Population Sample size Response Cases Histological
diagnosis rate confirmation
O'Beim et al. (1970) County 1960s Population-based 1338 Approx. 13 BCC; 13 SCC on Incomplete; 57%
Galway, 81% exposed sites only had biopsies
Ireland
Silverstone & Queensland, 1961-63 Population-based About 2200 87% 221 BCC or SCC on Incomplete
Gordon ( 1966); Australia exposed surfaces
Silverstone & Searle
(1970)
Holman eta!. Busselton, 1981 Population-based 1216 1 02, type not stated No
(1984a) Western
Australia
Engel et al. (1988) USA 1971-74 Population-based 20 637 74% BCC, number not stated Incomplete [small
proportion]
Green et al. (1988a) Nambour, 1986 Population-based 2095 70-78% 42 BCC or SCC [90% of Yes
Australia subjects examined on
head/neck/hands/foreann
s only]
Green & Battistutta Nambour, 1985-87 Population-based 1770 84% 66BCC; 21 SCC self- Incomplete
(1990) Australia reported ( confim1ed from
medical records)
Vitasa et al. ( 1990) Maryland, 1985-86 Male fishermen > 838 70% 33 Bee; 35 sec Incomplete
USA 30 years old
BCC, basal-cell carcinomas; SCC, squamous-cell carcinoma
Table 10. Summary of results of cross-sectional studies of nonmelanocytic skin cancer
Reference Index of exposure Categories Odds ratio (95% Cl)
O'Beirn et at. ( 1970) Sunlight hours (lifetime) < 30 000 h 1.00
Silverstone & Searle
(1970)
Holman eta/. (1984a)
Engel et al. (1988)
Green et al. ( 1988a)
Occupation
Occupation
Cutaneous microtopography
Solar skin damage
Occupational exposure
Painful sunburns
Solar lentigines on hands
Telangiectasia on face
> 50 000 h [8.10 (1.2-348.2)]
Indoors 1.0
Outdoors [ 1.29]
Indoors 1.0
Outdoors [0.6]
Grades 1-3 1.0
Grade 4 3.9
Grade 5 3.6
Grade 6 9.2
None 1.0
Any [8.0]
None 1.0
Any [6.0]
Indoors 1.00
Indoors and outdoors 1.01 (0.44-2.31)
Outdoors 1.76 (0.77-4.05)
None 1.00
1 0.77 (0.22-2.61)
2-5 1.09 (0.41-2.95)
26 1.66 (0.59-4.64)
None 1.00
1- 10 1.61 (0. 78-3.35)
11-20 1.43 (0.43-4.77)
26 3.78 (1.06-13.41)
None 1.00
Mild 1.63 (0.58-4.57)
Moderate 2.74 (0.89-8.40)
Severe 3.67 (0.79-17.11)
Comments
Mean aged > 60 years;
calculated from raw data [p =
0.02]
Men, chi-square = 1.4 [p > 0.1];
calculated from raw data, no
adjustment
Women, chi -square = 0.3 [p >
0.1 ]; calculated from raw data,
no adjustment
p = 0.004, trend adjusted for age
BCC, men, age-adjusted
prevalence ratio, p < 0. 01
Adjusted for age, sex, skin
colour and propensity to sunburn
Adjusted for age, sex and other
signs of actinic damage
Table 10 (contd)
Reference Index of exposure
Green et al. (1988a) Actinic elastosis on neck
(contd)
Solar keratoses on face
Green & Battistutta BCC
(1990) Occupational exposure
Leisure exposure
No. of painful sunburns
sec
Occupational exposure
Leisure exposure
No. ofpainful sunburns
Categoties
None
Mild to moderate
Severe
None
1-5
6-20
21-50
2: 51
Mainly indoors
Indoors and outdoors
Mainly indoors
Mainly outdoors
None
1
2-5
2:6
Mainly indoors
Mainly outdoors
Mainly indoors
Mainly outdoors
0- 1
2-5
2: 6
Odds ratio (95% Cl)
1.00
1.42 (0.53-3.80)
1.75 (0.56-5.45)
1.00
1.55 (0.67-3.59)
1.86 (0.69-5.04)
3.00 (0.54-16.69)
2.72 (0.73-10.15)
1.0
1.5 (0.8-2.9)
1.3 (0.6-2.8)
1.0
1.0 (0.4-2.2)
0.6 (0.3-1.3)
1.0
0.5 (0.2-1.4)
0.6 (0.3-1.5)
1.0 (0.4-2.5)
1.0
4.4 (0.9-20.9)
5.5 (1. 1-28.2)
1.0
2.0 (0.2-19.9)
3.9 (0.5-30.9)
1.0
3.3 (0.9-12.3)
3.0 (0.7-12.2)
Comments
Adjusted for age, sex and other signs of
actinic damage
Adjusted for age, sex and other signs of
actinic damage
Adjusted for age, sex, skin colour and
past history of skin cancer
past hi story of skin cancer
past hi story of skin cancer
Table 10 (contd)
Reference Index of exposure
Vita sa eta/. ( 1990) sec
Cumulative UVB dose to face
BCC
Cumulative UVB dose to face
Categories Odds ratio (95% CI)
Below median 1.0
Above median 2.05 (0.84-5.01)
Below 75 percentile 1.0
Above 75 percentile 2.53 (1.18-5.40)
Below median 1.0
Above median 0.69 (0.3 1-1.53)
Below 75 percentile 1.0
Above 75 percentile 1.11 (0.50-2.44)
Comments
Proportionate odds ratios; adjusted for age, eye
colour, freckling and sunburn reaction
Proportionate odds ratios; adjusted for age, eye
colour, freckling and sunburn reaction
BCC, basal-cell carcinoma; SCC, squamous-cell carcinoma; unless otherwise specified, all analyses are for the two types together
( d ) Case-control studies
Design features of the case-control studies of exposure to the sun and the occurrence of
nonmelanocytic skin cancer are summarized in Table 11. Most of the studies employed
hospital- or clinic-based controls, which introduces potential for selection bias. The results
are summarized in Table 12. The methods of analysis and of measurements of exposure to
the sun, particularly in the earlier studies, were crude. Neither sensitivity to the sun, usually
measured as the ability to tan or propensity to bum, nor pigmentary characteristics (such as
skin colour and hair colour), which are likely to be confounding variables, were taken into
account in most of the analyses.
The hospital-based study of Lancaster and Nelson (1957) in Sydney. Australia, was
primarily a case-control study of melanoma (described in detail on p. 100). It can also be
considered to be a case-control study of nonmelanocytic skin cancer, however, because it
included two control groups-one of patients with basal-cell carcinoma, squamous-cell
carcinoma or solar keratosis and the second of patients with leukaemia or cancer at a site
other than the skin. All groups were matched by age and sex. Among males, long duration
of occupational exposure to the sun was associated with an increased risk for
nonmelanocytic skin cancer or solar keratosis. A summary of total exposure to the sun was
devised by assigning scores to a number of factors considered to be related to exposure to
the sun. Risk was highest among subjects judged to have excessive exposure to the sun.
[The Working Group noted that the proportion of cases who had a solar keratosis is not
stated, that no account was taken of matching in the analyses, and that the effect of exposure
to the sun was not adjusted for sensitivity to the sun.]
Gellin et al. (1965) conducted a study in a single hospital in New York, USA, on 861
patients with basal-cell carcinoma and 1938 non-cancer dermatolcygical patients attending
the same clinic. Since 95% of cases and 43% of controls were 40 years old and over, the
study was limited to these patients, resulting in 771 cases and 783 controls. The skin cancer
patients spent more time outdoors per day than did control patients and were significantly
more likely than controls to have light hair, fair complexion, blue eyes and an inability to
tan. [The Working Group noted that the analyses were not adjusted for age, sex or
sensitivity to the sun, and that confounding by age is likely because controls were younger
than cases.]
Urbach et al. (1974) conducted a hospital-based study in Philadelphia, USA, and
compared exposure to the sun of 392 patients with histologically confirmed basal-cell carci-
noma, 59 patients with histologically confirmed squamous-cell carcinoma and 281 out-
patients receiving treatment for a skin disease other than cancer. Controls were matched to
cases by age and sex. Among male patients, those with basal-cell or squamous-cell
carcinoma had more cumulative hours of exposure than did controls. Skin cancer patients
also reported more sunburns. [The Working Group noted that the analyses were not
adjusted for ability to tan, age or sex (apart from the sex-specific analysis).]
Vitaliano (1978) subsequently reanalysed the data of Urbach et al. (1974) and showed
that, after adjustment for complexion (dark versus pale), ability to tan and age ( < 60, 2:60),
the cumulative time spent outdoors was related to both types of nonmelanocytic skin cancer.
For basal-cell carcinoma, the odds ratio for 2: 30 000 h of exposure relative to < 10 000 h
was 3.19; for squamous-cell carcinoma it was 22.8. [The Working Group noted that confi-
dence intervals were not given. Part of the apparently stronger effect for squamous-cell
carcinoma could be due to confounding by age: the controls were matched by age to the
basal-cell carcinoma cases, who were younger than the squamous-cell carcinoma cases.]
A hospital-based case-control study was conducted in Montreal, Canada (Aubry &
MacGibbon, 1985), in which patients with histologically confirmed squamous-cell carci-
noma were identified in hospitals in 1977-78. Two patients with other conditions were
matched as controls to each case by age, sex and hospital. Information on exposure to the
sun was obtained from a postal questionnaire. Among 306 eligible cases, 94 (31 %) replied,
as did 186 (30%) of the eligible controls; 92 cases and 17 4 controls completed the
questionnaire. Most of the controls who replied had been seen for seborrheic keratoses
(61 %) or intradermal naevi (16%). Scores for nonoccupational and occupational exposures
were estimated, and the two scores were divided into thirds for analysis, which was based
on logistic regression. The odds ratios, adjusted for each other and for host factors, were
1.08 and 1.64 for the middle and upper thirds of occupational exposure and 1.93 and 1.58
for the same levels of nonoccupational exposure, respectively. [The Working Group noted
the low response rate and that the complexity of the recreational exposure to sun indices and
the nature of the control group make the results difficult to interpret.]
O'Loughlin et al. (1985) conducted a case-control study in a hospital in Dublin, Ireland.
Patients with histologically confirmed nonmelanocytic skin cancer [types not separated]
were compared w ith age- and sex-matched patients who had cancers of other organs. There
was no statistically significant difference between cases and controls in eight measures of
exposure to the sun summarized in a single index of exposure and either type of
nonmelanocytic skin cancer. [The Working Group noted that the measures of exposure to
the sun were crude and likely to be subject to considerable misclassification. No adjustment
was made for sensitivity to the sun.]
Herity et al. (1989) conducted a case-control study in the same hospital in Dublin of 396
histologically confirmed nonmelanocytic skin cancers in 1984-85. An equal number of age-
and sex-matched patients with other cancers, attending the same hospital, were used as
controls. More cases than controls lived in rural areas (p = 0.007), and cases reported more
frequently spending more than 30 h outdoors per week, but the difference was
nonsignificant. For other indices of exposure to the sun, there was little difference between
cases and controls. [The Working Group noted that results were not adjusted for reaction to
sunlight.]
In a case-control study (reported as an abstract) conducted in I 983-84 in Alberta, Canada
(Fincham & Hi ll, 1989), 225 men with basal-cell carcinoma and 181 men with
squamous-cell carcinoma were compared with 406 age-matched male controls. Sunburn in
adult life gave an odds ratio of 2.33 (p < 0.05) for all nonmelanocytic skin cancer; for basal-
cell carcinoma, childhoood sunburn gave an odds ratio of 2.48 (p < 0.05) and peeling an
odds ratio of 1.85 (p < 0.05).
A population-based case-control study was conducted in Saskatchewan, Canada (Hogan
et al., I 989), which included all patients diagnosed with basal-cell carcinoma in the
Province in 1983. Two controls, matched by year of birth, sex and municipality of
residence, were selected for each case from a universal Provincial health insurance plan.
Replies to mailed questionnaires were received from 55.5% of the cases and 43.7% of the
controls. A number of measures of exposure to the sun were associated with incidence of
basal-cell carcinoma. In a stepwise logistic regression analysis, occupation as a farmer,
history of severe sunburn and working outdoors for more than 3 h per day in winter were
independently associated with basal-cell carcinoma, after adjustment for freckles in
childhood, family history of skin cancer, 'Celtic' mother, skin colour and hair colour. [The
Working Group noted that the measures of exposure were crude and that the estimates do
not appear to have been adjusted for the matching variables. The low response rate makes
interpretation of the results difficult.]
On the basis of a population-based survey in Western Australia in 1987 of skin cancer
among residents aged 40-64 years of age (Kricker et al., 1990), Kricker et al. (1991a)
conducted a case-control study of 226 confirmed cases of basal-cell carcinoma and 45 of
squamous-cell carcinoma; two sets of 1015 controls with no lesions, who had completed an
interview, were available for each type of cancer. The response rate among those eligible to
participate was identical for cases and controls: 89%. Separate analyses were undertaken
for basal-cell carcinoma and squamous-cell carcinoma using unconditional logistic
regression analysis. Risks for both cancers were higher in native-born Australians than in
migrants, and the risk for basal-cell carcinoma decreased with increasing age at arrival in
Australia. Only four of the subjects with squamous-cell carcinoma had been born outside
Australia-an insufficient number to examine the effects of age at arrival. Indicators of sun
damage to the skin (facial telangiectasia, solar elastosis of the neck, facial solar lentigines
and number of solar keratoses), assessed by dermatologists during the prevalence survey,
were examined in models adjusted for age, sex, ethnicity and migrant status and including
all other sun damage indicators except solar keratoses, which were considered to be
preneoplastic lesions and thus inappropriate for inclusion in models concerned with
etiology. Cutaneous microtopography, an objective measure of actinic skin damage, graded
without knowledge of the person's skin cancer status, and solar elastosis of the neck had
significant residual effects for basal-cell carcinoma, while solar elastosis and facial
telangiectasia had significant residual effects for squamous-cell carcinoma. The
independently significant indicators of sun damage were analysed in models which included
adjustment for age, sex, ethnicity and migrant status as well as measures of sun sensitivity.
Solar elastosis of the neck remained an independent predictor of risk of basal-cell carcinoma
(odds ratios,> 1.50;p = 0.003) and squamous-cell carcinoma (odds ratios, > 2.00;p = 0.04).
A subsequent analysis of individual su1r exposure was published as an abstract (Kricker et
al.. 1991 b). A positive association was found between nonmelanocytic skin cancer and
life-time potential for exposure to the sun, but no evidence of increasing risk for either
basal-cell carcinoma or squamous-cell carcinoma with increasing total hours of actual
exposure to the sun as recalled by subjects. Risk for basal-cell carcinoma on the trunk was
increased substantially in association with maximal exposure of the trunk to the sun, but
there was no consistent pattern of association of site-specific basal-cell or squamous-cell
carcinoma with exposure of the head and neck or limbs. Neither basal-cell nor
squamous-cell carcinoma showed evidence of an association with sun exposure on working
days; however, there was persuasive evidence of increased risk for both types of skin cancer
with intermediate and high levels of accumulated exposure to the sun on non-working days.
Moreover, there was evidence of an association, stronger for basal-cell carcinoma than for
squamous-cell carcinoma, with a measure of intermittent exposure to the sun.
Gafa et al. ( 1991) conducted a case-control study of nonmelanocytic skin cancer in
Sicily, Italy, in which 133 cases identified from a population-based registry (response rate,
94%) were compared with 266 sex- and age-matched controls. For each case, one control
was selected randomly from among patients with non-neoplastic diseases at the same
hospital as the case, and a second control was selected randomly from among friends or
relatives of the case. After adjustment for family history of skin cancer, 'cancer-related
cutaneous disease', skin colour and skin reaction to sunlight, sun exposure for at least 6 h
per day and residence for at least 10 years at more than 400 m above sea level were
significantly related to risk for nonmelanocytic skin cancer. In cmde analyses in which the
two types of cancer were separated, sun exposure for at least 6 h per day without a hat was
strongly associated with risk for squamous-cell carcinoma [site unspecified] (odds ratio, 6.4;
95% CI, 1.9-21.1) but not for basal-cell carcinoma (1.4, 0.7-2.6). [The Working Group
noted that the nature of the control group, the assessment of exposure and the failure to
account for age in the analysis make the results difficult to interpret. The crude analysis of
the type-specific results, the lack of data on the site of the tumours and the small numbers
may explain the different results for the two types.]
(e) Cohort studies (Tables 13 and 14)
In a study in Chicago, IL (USA), Robinson (1987) investigated the incidence of second
nonmelanocytic skin cancer among a group of 1000 patients who had had basal-cell carci-
noma. Among 978 who were followed for five years after the initial diagnosis, 22% deve-
loped a second basal-cell carcinoma at the end of the first year and 36% within five years.
There was no significant correlation between developing a second cancer and frequent expo-
sure through sunbathing or outdoor leisure activities, work or currently living in an area
with heavy exposure to the sun, or according to estimated number of hours of daily exposure
to the sun. Among those with skin types I and II (always bum easily and never or
minimally tan) who reported frequent sun exposure, there was an increased risk of second
cancer (p < 0.03). [The Working Group noted that the methods of assessing exposure and
the methods of analysis were not described, and that no numbers were reported. Risk
factors for second cancers might not be the same as for the first.]
Marks et al. ( 1989) conducted a longitudinal series of examinations of the head, neck,
forearms and hands of a population in Maryborough, north-central Victoria, Australia, for
one week annually between 1982 and 1986. The incidence rates of squamous-cell and
basal-cell carcinoma were higher in outdoor workers than in indoor workers. In an analysis
of the two types combined, occupation was not significantly associated after adjustment for
age, sex and reaction to sunlight (p = 0.09). [The Working Group noted that no account was
taken of lesions that might have been removed between surveys.]
Hunter et al. (1990) conducted a study of basal-cell carcinoma in a cohort of female
nurses in the USA. A total of 771 cases were identified from responses to follow-up ques-
tionnaires sent to the women two and four years after the initial exposure questionnaire was
given. In a sample of 29 women, the diagnosis was confirmed for 28; confirmation of the
diagnosis was not obtained routinely. Residents of California and Florida had the highest
Table 11. Design features of case-control studies of sun exposure and nonmelanocytic skin cancer
Reference Place Periods of Cases Controls
diagnosis
No. Source No. Source
Lancaster & Sydney, Australia Unknown 173 BCC, SCC or Major hospitals 173 Other cancers, same
Nelson (1957) solar keratosis hospital
Gellin et al. New York, USA 1955-59 771 BCC One skin hospital 783 Other diagnoses, same
(1985) ~ 40 years old ~ 4 skin clinic
Urbach et al. Philadelphia, USA 1967-69 392BCC One skin and 281 Other diagnoses, same
(1974) 59 sec cancer cl inic clinic
Aubry & Montreal, Canada 1977-78 92 sec 12 hospitals 174 Skin conditions, same
MacGibbon hospitals
(1985)
O'Loughlin et al. Dublin, Ireland Unknown 63 sec One hospital 121 Other cancers, same
(1985) 58 BCC hospital
Herity et al. Dublin, Ireland 1984-85 396 BCC and One hospital 396 Other cancers, same
(1989) sec hospital
Hogan eta!. Saskatchewan, 1983 538 BCC Population 738 Population
( 1989) Canada
Kricker et at. Geraldton, 1987 226BCC Population 1015 Population
(1991a) Australia 45 sec 1015
Gafa et al. (1991) Ragusa, Sicily, 1987-88 133 BCC and Cancer registry 133 Non-neoplastic
Italy sec 133 disesases, same
hospital; friends or
relatives
BCC, basal-cell carcinoma; SCC, squamous-cell carcinoma
Table 12. Summary of results of case-control studies of nonmelanocytic skin cancer
Reference Exposure Categories Odds ratio Comments
(95% CI)
Lancaster & Nelson Years of occupational exposure < 5 1.0 [p < 0.001 , trend; p and odds ratio calculated
5-10 [1.9] from raw data]
> 10 (4.2]
Total sun exposure Minimal 1.0 [p = 0.13; p and odds ratio calculated from raw
Moderate [ 1.8] data]
Excessive (2.4]
Gellin et al. ( 1965) Hours per day outdoors 0-2 1.0 BCC [p < 0.001]
3-5 (4.9 (3.8-6.3)]
~ [7.7 (5.6-1 0.6)]
Urbach eta/. (1974) Cwnulative hours < 30 1.0 BCC
(x 1000) 30-50 [3.5 (20.-6.6)]
> 50 [9.3 (3.2-37.4)]
< 30 1.0 sec
30-50 [4.0 (1.7-9.6)]
> 50 [11.1 (2.8-53.6)]
Aubry & MacGibbon Non-occupational exposure score Low 1.0 sec [p = 0.07] for continuous variable, adjusted
( 1985) Medium 1.23 for occupation and host factors
High 1.58
Occupational score Low 1.0 sec [p = 0.008], for continuous variable,
Medium 1.08 adjusted for non-occupational score and host
High 1.64 factors
Use of sunlamps Never 1.0 sec [p = 0.008], adjusted for sun exposure and
Ever 13.4 (1.38-130.48) host factors
O'Loughlin et al. Outdoor occupation No 1.0 Not significant (McNemar's test) [odds ratio
(1985) Yes (1.5] calculated from raw data ignoring matching]
Hours per week outdoors < 10 1.0 Not significant
~ 10 [1.4]
Sunbathing > 4 h per day on No l.O Not significant
vacations Yes [ 1.0]
Herity et al. (1989) Living in rural area > 30 h [1.4] p = 0.007
outdoors/week [1.1] p =0.7
Table 12 ( contd)
Reference Exposure
Hogan eta!. ( 1989) Farmer
Severe sunburn
Working outdoors > 3 h per day
winter
Krickeretal. ( l99la) BCC
Age at migration (years)
Solar elastosis of the neck
Cutaneous microtopography
sec
Migrant to Australia
Permanent colour
difference between neck and
adjacent ski n
Telangiectasia of face
Solar elastosis of the neck
Categories
No
Yes
No
Yes
No
Yes
Australian
born
< 10
< 10
None
Mild
Moderate
Severe
Grades 1-3
Grade 4
Grade 5
Grade 6
No
Yes
No
Yes
None/mild
Moderate
Severe
None/mild
Moderate
Severe
Odds ratio
(95% CI
1.0
1.29 [ 1.12-1.46)
1.0
1.19 [1.04-1.35)
1.0
1.13 [1.01 -1.27]
1.0
1.37 (0.55-3.42)
0.32 (0.18-0.59)
1.00
1.85 (0.80-4.26)
2.75 (1.16-6.50)
3.96 (1.58-9.93)
1.0
2.01 (1.00-4.07)
2.42 (1.17-5.01)
2.15 (0.99-4.70)
1.0
0.46 (0.15-1.38)
1.0
2.58 (1.03-6.47)
1.0
2.22 (1.06-4.67)
1.88 (0.72-4.90)
1.00
2.31 ( 1.00-5.34)
3.33 (1.23-9.04)
Comments
BCC, adjusted for each other, plus freckles,
family history of skin cancer, Celtic mother, skin
colour, hair colour
BCC
BCC
p < 0.001, adjusted tor other variables below and
for ethnicity, ability to tan, freckling as a child
and number of moles on back
p = 0.003, comments as above
p = 0.13, adjusted for variables below plus ability
to tan, skin colour, freckling as a child
p = 0.1 0, comments as above
p = 0.04, conm1ents as above
Table 12 ( contd)
Reference
Gafa et al. (1981)
Exposure Categories
Residence > 400 m above sea level No
Yes
Sun exposure 2: 6 h/day No
Yes
Odds ratio
(95% CJ
1.0
2.0 (1.2-3.2)
1.0
1.9 (1.2-3. 1)
Commentsa
Adjusted for family history of skin cancer,
cutaneous-related conditions, skin colour, skin
reaction to sunlight and sun exposure
Adjusted for family hi story of skin cancer,
cutaneous-related conditions, skin colour, skin
reaction to sunlight and residence > 400 m above
sea level
BCC, basal-cell carcinoma; SCC, squamous-cell carcinoma; unless otherwise specified, analyses are for the two types together
incidence rates. There was a trend of increasing incidence with increasing number of sun-
burns. With respect to time spent outdoors during the summer, nurses who spent more than
8 h per week outside and who used sunscreens had the highest incidence rates. The rates in
women who spent the least time outdoors were similar to those who spent more time
outdoors and did not use sunscreens. [The Working Group noted that the high incidence
rate in nurses using sunscreens, despite control for reaction to sunlight, might be due partly
to confounding.]
Table 13. Design feature of cohort studies of sun exposure and nonmelanocytic skin
cancer
Reference
Robinson
(1987)
Mark
et al.
(1989)
Hunter
et al.
(1990)
Place
Chicago, IL,
USA
Maryborough,
Australia
USA
Period of
diagnosis
Not
stated
1982-86
1980-84
Population
Patients
with
previous
BCC
Sample
size
1 000
Population- 1 981
based
Female 73 366
nurses
Response
rate
98%
74%
74%
Cases
BCC,
approx. 350
35 SCC; 113
BCC on light-
exposed
surfaces only
771 BCC
(self-reported)
BCC, no. ofpeople with basal-cell carcinoma; SCC, no. of people with squamous-cell carcinoma
(f) Collation of results
Histological
confirmation
Not stated
Yes
Not routinely
[records of28
out of sample
of29
con finned]
The results discussed in this section come from cross-sectional studies by Holman et al.
(1984a), Angel et al. (1988), Green et al. (1988a) and Vitasa et al. (1990), a case-control
study by Kricker et al. (1991a) and cohort studies by Marks et al. (1989) and Hunter et al.
(1990), all of which included information pertinent to the association between
nonmelanocytic skin cancer and different aspects of sun exposure. Other studies described
individually were not considered to provide useful information because of various
methodological deficiencies. No data were avai lable on short periods of residence and
intermittent exposure, issues which are addressed for melanoma of the skin.
(i) Total sun exposure: potential exposure by place of residence
Consistent with descriptive data in a case-control study, migrants to Australia had a
lower risk for squamous-cell carcinoma than did native-born Australians, after adjustment
for host factors related to risk for nonmelanocytic skin tumours. Late age at arrival in
Australla was associated with a lower risk for basal-cell carcinoma (Kricker et al., 199la).
(ii) Biological responses to total sun exposure
Cross-sectional studies and a case-control study are consistent in showing a strong
relationship between cutaneous indicators of sun damage and both types of nonmelanocytic
skin cancer. In most studies, the indicators of damage and diagnoses of skin cancer were
made by the same examiner, but cutaneous microtopography, graded without knowledge of
outcome, also showed strong associations.
Table 14. Summary of results of cohort studies of nonmelanocytic skin cancer
Reference Exposure Categories RR (95% CT) Comments
Marks et al. Occupation BCC
(1989) Indoors 1.0 Adjusted for age, p = 0.03
Outdoors 1.6
sec
Indoors 1.0 Adjusted for age, p = 0.109
Outdoors 1.7
Hunter et al. Severe sunburns on face or am1s BCC
(1990) None 1.0 Adjusted for age; p (trend)= 0.001
1-2 1.40 (1.13-1.75)
3-5 1. 78 (1.42-2.25)
~ 2.91 (2.37-3.58)
Severe sunburns on face or arms None 1.0 Adj usted for age, time period, region, time
1-2 1.18 (0.94-1.48) spent outdoors, sunscreen habit, hair co-
3-5 1.34 (1.05-1.71) lour, childhood tendency to sunbwn; p
~ 1.90 (1.50-2.40) (trend) < 0.001
Time spent outdoors during summer ~ 8 (sunscreen) 1.0 Adjusted for age
(h/week) ~ 8 (no sunscreen) 0.59 (0.50-0.69)
<8 0.71 (0.58-0.88)
Time spent outdoors during summer ~ 8 (sunscreen) 1.0 Adjusted for age, time period, region,
(hlweek) ~ 8 (no sunscreen) 0.70 (0.60-0.82) number of sunburns, hair colour, childhood
<8 0.73 {0.59-0.902 tendency to sunburn
aBCC, basal-cell carcinoma; SCC, squamous-cell carcinoma
(iii) Total sun exposure assessed by questionnaire
No effect of time spent outdoors during summer was seen in a cohort study of basal-cell
carcinoma (Hunter et al., 1990). In a cross-sectional study of fishermen, cumulative
exposure to UVB radiation was positively associated with the occurrence of squamous-cell
carcinoma but not of basal-cell carcinoma (Vitasa et al., 1990). The different results may be
attributable in part to small numbers and incomplete histopathological confirmation of
diagnoses.
(iv) Occupational exposure
In two studies from Australia, outdoor occupation was not significantly associated with
the prevalence of the two types of carcinoma combined (Green et al., 1988a) or with the
incidence of squamous-cell carcinomas (Marks et al., 1989).
(v) Sunburn
A cohort study of basal-cell carcinoma in the USA showed a trend of increasing risk
with increasing number of sunburns after adjustment for various factors, including tendency
to sunburn (Hunter et a/., 1990). Number of sunburns showed a nonsignificant positive
association with risks for basal-cell and squamous-cell carcinoma of the skin after
adjustment for various constitutional variables, including propensity to bum (Green et a/.,
1988a).
2.1.2 Cancer of the lip
Assessment of the carcinogenicity of solar radiation for the lip is complicated by the fact
that carcinoma at this site is actually diagnosed as a mixture of cancers of the external lip
and cancers of the buccal membranes (oral cavity). Use of alcohol and tobacco are known
causes of the latter tumours (IARC, 1985, 1986b, 1988).
While there are wide variations in the apparent incidence of cancer of the lip with
latitude, evaluation of the association is difficult because of inconsistency in the definitions
of the boundaries of the lip. 'Cancer of the lip' is defined as cancer of the vermilion border
and adjacent mucous membranes and thus excludes cancers of the skin of the lip (WHO,
1977). Most are squamous-cell carcinomas and are located on the lower lip (Keller, 1970;
Lindqvist, 1979), which is more heavily exposed to sunlight than is the upper lip (Urbach et
a/., 1966).
In general, case reports were not considered, because of the availability of more infor-
mative data. One case report from Nigeria described the occunence of two lip tumours in
albinos (Onuigbo, 1978).
(a) Descriptive studies
The incidence of lip cancer is 4-10 times higher in men than in women in most white
populations, and higher in whites than in populations of darker skin complexions living in
the same geographical areas (Muir eta/., 1987).
(i) Geographical variation
The incidence of lip cancer is higher in rural than in urban areas, in particular among
men (Doll, 1991 ).
Mortality from and incidence of lip cancer are substantially lower in migrants to
Australia than in native-born Australians (Armstrong et al., 1983; McCredie & Coates,
1989). Groups of migrants to Israel all show lower risks for lip cancer than the locally born
population (Steinitz et al., 1989).
(ii) Occupation
As reviewed by Clemmesen (1965), several observations during the nineteenth century
pointed to an increased risk of lip cancer among people in outdoor occupations, in particular
farmers and farm labourers. In England and Wales, increased risks for lip cancer were
reported among agricultural labourers, fishermen, other dock workers and railwaymen
employed outdoors (Young & Russell, 1926). Atkin et al. (1949) studied the occupations of
1537 men in England and Wales who died from lip cancer between 1911 and 1944. They
reported that mortality from cancer of the lip was 13 times higher among men employed in
agriculture than in men with professional jobs. Excess risks for lip cancer have also been
observed in farmers in western Canada (Gallagher et al., 1984) and in Denmark (Olsen &
Jensen, 1987; Lynge & Thygesen, 1990).
(b) Case-control studies
Keller ( 1970) compared 301 men with lip cancer admitted to veterans' hospitals in the
USA between 1958 and 1962 with two groups of white age-matched controls admitted to
the same hospitals, comprising 301 oral cancer controls and 265 general controls.
Altogether, 59.9% of the lip cancer cases, 3 7.1 % of the cancer controls and 40.6% of the
general controls had been born in the south of the USA. Farming was recorded as the
occupation of 27% of the lip cancer cases but of only to of cancer controls and 47 of the
general controls [crude odds ratios, 4.0 and 8.4, respectively]. Any type of outdoor work
was recorded for 39% of cases of lip cancer, for 20% of cancer controls and for 12% of the
general controls [crude odds ratios, 2.6 and 4.8, respectively]. Risk estimates were not
adjusted for smoking, another risk factor identified in the study.
Spitzer et al. (1975) obtained information by personal interview on 339 men with
squamous-cell carcinoma of the lip registered with the Newfoundland (Canada) Cancer
Registry between 1961 and 1971 and 199 male controls chosen from the electoral register,
matched for age and geographical location in nine census divisions; the overall response rate
was 93%. An association was found between lip cancer and outdoor work (odds ratio,
1.52;p < 0.05); an odds ratio of 1.50 (p < 0.05) was found for occupation as a fisherman for
at least eight full seasons, after adjustment for outdoor work, pipe smoking and age. No
positive association was found for specific fishing activities, such as use of mouth as a third
hand or of cast nets.
Lindqvist (1979) obtained information by mailed questionnaires from 171 cases (149
men, 22 women; 74% response rate) of epidermoid carcinoma of the lip registered with the
Finnish Cancer Registry in 1972-73 and from a control group of 124 patients (56 men, 68
women; 77% response rate) registered with squamous-cell carcinoma of the skin of the head
and neck. Risk estimates were adjusted for age. Odds ratios for men working outdoors
ranged from 2.2 to 3.2 according to the calendar period during which the subjects had
worked outdoors. The odds ratio was significantly increased only for those who both
worked outdoors and smoked. [The Working Group noted that the choice of head and neck
skin cancer patients as controls would lead to an underestimate of the odds ratio for outdoor
work.]
Dardanoni et al. (1984) obtained information by personal interviews from 53 men with
lip cancer registered in the Ragusa Cancer Registry in Italy and from 106 male controls
matched for age and municipality of residence and admitted to the same hospitals for
non-neoplastic diseases. An association was found between lip cancer and working or
spending at least 6 h each day outdoors (odds ratio, 4.9; p < 0.001). After control for socio-
economic level, the odds ratio was 1.7 (p < 0.001). [The Working Group noted that the
latter p value is inconsistent with the number of subjects.]
2.1.3 Malignant melanoma of the skin
Melanoma of the skjn is divided into three major histological types. The majority of
melanomas in white-skinned populations (of European origin) are superficial spreading and
nodular melanomas. Lentigo maligna melanoma- also known as Hutchinson's melanotic
freckle- occurs later in life than the other types, and more specifically on exposed sites;
however, the body site and evidence of sun damage in surrounding skin may influence its
pathological classification (McGovern et al., 1980). Acral lentiginous melanoma has not
been studied epidemiologically; it is rare in white-skinned populations, although it
comprises a substantial proportion of melanomas in Japan (Elwood, 1989a).
(a) Case reports
In general, case reports were not considered, owing to the availability of more
informative data.
In a survey of 830 cases of xeroderma pigmentosum located through published case
reports (Kraemer et at., 1987), melanomas were reported in 37 patients (5%). As the median
age at last follow-up of these cases was only 19 years, this observation is likely to represent
a substantial excess over the number expected, although the exact nature of the study popu-
lation precludes an accurate comparison. Site was specified for 29 of the 37 cases; 65% of
these were on the face, head and neck (normally constantly UVR-exposed sites) as
compared with 19.4% on this site among affected members of the US general population.
[The Working Group recogmzed that data collected from previously published case reports
are not uniform and may be atypical of a true incidence or prevalence series. Furthermore,
no information is available on the relationship between solar exposure and the occurrence of
malignant cutaneous melanoma in these patients.]
(i) Sex distribution
The sex distribution of melanoma, adjusted for age, varies widely between populations.
In many, it occurs as often as or more commonly in women than in men (Lee & Storer,
1980; Lee, 1982), in contrast to other types of skin cancer which are unifonnly commoner in
men (Muir et al., 1987).
(ii) Age distribution
Age distributions of melanoma in human populations vary with sex (Lee, 1982). They
cannot easily be interpreted because they represent a variable combination of the different
patterns of melanomas at different sites as well as a combination of time trends and trends in
the experience of birth cohorts.
(iii) Anatomical distribution
Melanoma is proportionately commonest on the back and face in men and on the legs in
women (Crombie, 1981); however, the incidence of melanoma per unit of body area is
similar on fully exposed sites, such as the face, and on patiially exposed sites, such as the
lower limbs in women and the back in men. The frequency on body sites that are usually
covered, such as the buttocks, is much lower (Elwood & Gallagher, 1983).
(iv) Ethnic origin
Melanoma is predominantly a disease of white-skinned populations. Rates in dark-
skinned populations are much lower, the age-standardized incidence rate in India being 0.2
per 100 000 compared to around 30 in Queensland, Australia. In Los Angeles, USA, rates
were less than I per l 00 000 in Japanese and Chinese subjects and 11-12 in white subjects
(Muir et al., 1987: Whelan et al., 1990). The site and histological distribution of melanoma
are different in non-white populations and have been little studied epidemiologically. The
remainder of this section deals only with melanoma in white populations.
The incidence of melanoma is substantially lower among Hispanics than among other
whites in the USA. For example, the incidence among Hispanics in New Mexico is less
than 2 per 100 000 person years. but in other whites it is about 11 per 100 000 (Muir et al. ,
I 987). In several case-control studies (described in detail below), subjects with a southern
or eastern European background had lower risks than those with northern European or
British origins (Elwood et al., 1984; Holman & Armstrong, 1984a).
In a Canadian study (Elwood et al., 1984), people with an eastern or southern European
background had a crude odds ratio of 0.5 relative to those with an English background. This
effect was not changed appreciably after adjustment for constitutional factors of hair, eye
and skin colour and the skin's reaction to sun exposure. In contrast, the effect of ethnic
origin observed in Western Australia was substantially reduced after adjustment for
pigmentation characteristics (Holman & Armstrong, 1984a).
(v) Geographical variation
Armstrong ( 1984) showed that the relationship between melanoma incidence in
Caucasians and latitude of residence decreases from around 35 o to a minimum around 55
and then rises with latitude due to high rates in Scandinavian and Scottish populations. This
pattern is likely to be due to both latitudinal and pigmentation factors. Within countries,
inverse relationships of incidence or mortality with latitude have been seen in England and
Wales (Swerdlow, 1979), Norway (Magnus, 1973), Sweden (Eklund & Malec, 1978) and
Finland (Teppo et al., 1978).
In the first comprehensive analysis of the geography of melanoma in whites, Lancaster
(1956) noted that mortality from the disease was higher in Australia and South Africa than
in the parts of Europe from which their populations originated: that mortality in Australia,
New Zealand and the USA increased with proximity to the equator; but that within Europe it
was higher in Norway and Sweden in the north than in France and Italy in the south. These
patterns are also evident in more recent data (Armstrong, 1984).
Geographical variation in relationship to ambient UV irradiation levels: Several studies
have compared melanoma incidence and mortality rates in different areas of North America
to estimated or measured levels of ambient UVR, and Elwood ( 1989b) estimated the change
in rate for a 10% change in UVR level (Table 15). [The Working Group noted that these
studies did not assess any other component of the solar spectrum.]
Elwood et al. (1974) showed, using mortality data for US states and Canadian provinces,
that the correlation coefficients with latitude were 0.79 for men and 0.72 for women. A
variation in latitude of 2 , which is equivalent to 138 miles, was associated with a change in
death rates from melanoma of about 10%. Annual UV flux at erythema-producing wave-
lengths was calculated from information on latitude and meteorological data on cloud cover.
This calculated index of exposure was very strongly correlated with latitude (correlation
coefficient, 0.89), so melanoma mortality rates were strongly related to this index; a 10%
increase in received UVR dosage would be expected to give an increase of 3.7-4.5% in the
death rate from melanoma at latitude 50 , and 6.8-10.3% at latitude 30 (Table 15). These
values were somewhat higher for men than for women; for example, 4.4% in men compared
with 3.0% in women at latitude 50 o using the exponential model.
Fears et al. (1976) related melanoma incidence to latitude and to a calculated measure of
UVR. Their data cover a slightly narrower range of latitude, and they calculated that a 10%
increase in UVR would cause an increase in melanoma mortality of 7-12%, the higher
figure applying to more southerly latitudes, which already have higher rates. Incidence rates
vary more rapidly with latitude than do mortality rates, and therefore they predicted that a
10% increase in UVR would be likely to give a 14-24% increase in the incidence of
melanoma (see Table 15).
Estimates using calculated UVR levels: Fears et al. (1977) used measurements from
Robertson-Berger meters for four areas and a power model, in which the calculated
percentage changes are not dependent upon the initial latitude. These calculations showed
considerably stronger effects, with an estimated 25% increase in incidence for a 10%
increase in solar UVR (see Table 15).
Scotto and Fears (1987) used annual UVR counts from Robertson-Berger meters in
seven areas of the USA (Detroit, Seattle, Iowa, Utah, San Francisco, Atlanta and New
Mexico) and data on melanoma from incidence registries (the Surveillance Epidemiology
and End Results system). They fitted a power model and presented analyses by sex and by
body site of the melanoma divided into trunk and lower limb versus head, neck and upper
limb. They obtained data on covariates, including ethnic origin, pigmentation
characteristics, hours spent outdoors during weekdays and during weekends and use of
suncreens, suntan lotion and protective clothing, from telephone interviews with at least 500
households in each area. Data on the melanoma patients were not available, however. The
results predict greater increases for females than for males, unlike the earlier work. The
overall effects of a 10% increase in UVR are a 5.5% increase for trunk and lower limb
turnouts and a 9% increase for head, neck and upper limb tumours, averaged over the two
sexes. Adjustment for the various covariates reduces the predicted increases to a 3.5 %
increase for trunk and lower limb tumours, and 5.5% for head, neck and upper limb turnouts
(see Table 15).
Pitcher and Longstreth (1991) used data on melanoma mortality over a 30-year period
and calculated UV flux on the basis of satellite data from the US National Aeronautics and
Space Administration, including measurements of ozone concentrations at high atmospheric
Table 15. Estimates by Elwood (1989b) of percentage increase in frequency of melanoma among whites with a 10% increase in solar
ultraviolet radiation, based on differences with latitude in Canada and the USA
Ultraviolet radiation level
derived froma
Model 50
latitude
30 latitude
Incidence Mortality Incidence
Calculation of erythema-
weighted index
weighted index
RB meter (1974)
RB meter ( 1978-81)
weighted estimate from
NASA including satellite
ozone column measurements
Linear
Exponential
Exponential
Power
Power
Trunk and lower limb
Crude
Adjusted
Head, neck and upper
limb
Crude
Adjusted
Total
Crude
Adjusted
Power
Annual
Peak
Exponential
Annual
Peak
Both sexes (simple average of sex-specific results)
14.0
25.0
5.5
3.5
9.0
5.5
6.7
4.2
3
RB, Robertson-Berger; NASA, National Aeronautics and Space Administration
bMortality data, USA and Canada 1950-67 by state/province; 58 areas
4.5
3.7
7.0 23.5
25.0
5.5
3.5
9.0
5.5
6.7
4.2
3.2
7.0
2.1
5.8
Mortality
Reference on which
estimates based
6.8 Elwood et al. (1974/
10.3
12.0
3.2
7.0
4.5
8.2
Fears eta/. (1976t
Fears et al. ( 1977t
Scotto & Fears ( 1987)e
Pitcher & Longstreth
(199l)f
<Incidence data. Third National Cancer Survey (1969-71) for nine areas; US mortality by state. Calculation based on latitude equivalent to change in ultraviolet radiation
dlncidence data, Third National Cancer Survey ( 1969-71) for four areas
erncidence data, Surveillance Epidemiology and End Results Program for seven areas. Crude results take account only of age; adjusted results are controlled for ethnic origin, hair or
skin colour, suntan lotion use and hours spent outdoors; total, for comparison, is based on 67% trunk and lower limb and 33% bead, neck and upper limb tumours
1
Mortality data by US county 1950-79; estimates of changes in mean annual dose and in peak doses (clear day in June); estimates using DNA action spectrum were also made and were
1-8% higher than those shown.
conditions. The models fitted are complex, as they are fitted for the two sexes, for three
different places covering a range of latitudes, and separately for changes in the annual UV
flux and changes in the peak levels in clear summer conditions. Larger effects were again
found for males than for females, and a larger effect when using the peak measurements
than when using the annual measurements. The overall estimates of the percentage increase
in melanoma mortality associated with a 5% decrease in ozone level, on the assumption that
this is roughly equivalent to a 10% increase in solar UVR, ranged from 2.1 to 7.0 at 50 oN
and from 3.2 to 8.2 at 30 N (see Table 15).
[The Working Group noted that, despite the sophistication of some of the mathematical
models, these results are derived from population-based descriptive data and not from
individual measurements and are restricted to N otih America.]
(vi) Migration
The most informative data on risk in migrants come from Australia, New Zealand, Israel
and the USA. Native residents of Australia (McCredie & Coates, 1989; Khlat et al., 1992)
and New Zealand (Cooke & Fraser, 1985), mostly of British origin, experienced incidence
and mortality rates of melanoma roughly twice those of British immigrants. Native Israelis
had a risk at least twice that of immigrants to Israel from Europe for at least 30 years after
immigration (Steinitz et al., 1989).
The higher incidence in white immigrants to Hawaii from the US mainland compared
with white natives has been attributed to a difference in skin colour (Hinds & Kolonel ,
1980). Non-Hispanic migrants to Los Angeles County (California, USA) from higher
latitudes in the USA are still substantially protected against melanoma of all histological
types decades after migration. Similar relative protection is enjoyed by native residents of
more northerly US communities in comparison with co-resident migrants from the
south-western USA (Mack & Floderus, 1991).
(vii) Socioeconomic status and occupation
Melanomas are much commoner in higher socioeconomic groups, as shown in data from
the United Kingdom since 1949-51. In the United Kingdom, the distribution of melanoma
in married women by social class (categorized by their husbands' social class) is similar to
that of men, indicating that this is a social rather than a specific occupational factor (Lee,
1982). In the USA, the risk increases with income for men aged 30-69; at age 70 and above,
the trend is reversed, suggesting a role for long-term exposure to the sun (Kirkpatrick et al.,
1990). In case-control studies, the effect of socioeconomic status is weakened after adjust-
ment for measures of exposure to the sun (Gallagher et al., 1987; 0sterlind et al., 1988b).
Assessment of outdoor exposure on the basis of routine data on job descriptions showed
that melanoma is commoner in indoor than in outdoor workers, even within the same socio-
economic group (Lee & Strickland, 1980; Lee, 1982). Cutaneous melanoma incidence rates
during 1972-76 in New Zealand showed no pattern according to outdoor workplace (Cooke
et al., 1984). An analysis of 3991 cases of cutaneous melanoma registered during 1971-78
in England and Wales and of 5003 cases registered during 1961-79 in Sweden suggested an
elevated incidence in professional occupations. The incidence among farmers was close to
that expected (Vagero et al., 1990).
Garland et al. (1990) reported 176 incident cases of melanoma among US Navy
personnel. The rate for indoor occupation was higher than that for outdoor workers.
(c) Case-control studies
Elements of each case-control study described below are given in Table 16.
(i) Australia
Lancaster and Nelson (1957) carried out a case-control study on 173 patients aged over
14 years treated for malignant melanoma in hospitals in Adelaide, Melbourne and Brisbane,
and 173 hospital controls with cancers other than of the skin, matched for sex and age.
Information was obtained by interviews [response rate not given], and analysis was done by
single factor cross-tabulations only. Unmatched crude odds ratios were calculated by the
Working Group. Skin [odds ratio, 1.95 for fair versus olive and medium], hair colour [odds
ratio, 1.7 for fair and red versus black and brown], eye colour [odds ratio, 1.75 for blue and
green-gray versus brown and hazel] and skin reaction to sunlight [2.9; 95% CI, I .9-4.5 for
red versus brown reaction] were significantly associated with risk for malignant melanoma.
Among the other factors studied were birth outside Australia [0.8; 0.4- 1.6], 10 years' or
more occupational exposure to sunlight in males [1.4; 0.7-2.7], sunbathing [1.5; 0.9-2.4] and
moderate [1.2; 0.5-3.1] and excessive [2.3; 0.8-6.3] total exposure to the sun compared to
minimal exposure. There were only eight cases and 11 controls in the latter category of sun
exposure.
Beardmore ( 1972) studied 468 cases of histologically confirmed malignant melanoma
and 468 sex- and age-matched hospital controls (including patients with skin cancer) at one
hospital in Brisbane. Information was obtained by interview [response rate and method of
evaluation of hair, skin and eye colour not given]. Hair, skin and eye colour and skin
reaction to sunlight were not associated with risk for malignant melanoma. Comparison of
exposure to sunlight from mainly outdoor occupations to that from mainly indoor
occupations resulted in a crude odds ratio of [1.42; 95% CI, 1.03-1.97]; a similar
comparison for recreational activities gave a crude odds ratio of [ 1.03: 0.75-1.42]. Fewer
cases than controls had a history of treatment for keratosis and/or skin cancer or currently
had keratosis and/or skin cancer [crude odds ratios, 0.51, 0.38-0.69; and 0.16, 0.12-0.22,
respectively].
In the Western Australia Melanoma study (Holman & Armstrong, 1984a,b ), 511 cases
aged 10-79 years and 511 population controls matched for sex, age and area of residence
were interviewed at home using a questionnaire based on that of the Western Canada study,
which included objective measurements and naevi counts. The study also included a review
of pathology slides. Analyses were presented for superficial spreading, nodular and lentigo
maligna melanomas and for a fourth, unclassifiable group. Response rates were 76% for
cases and 62% for controls, and adjustment was made for chronic and acute skin reaction to
sunlight, hair colour, ethnic origin and age at arrival in Australia using a multiple logistic
regression model. Hair colour, acute and chronic reaction to sunlight, number of naevi and
family history of melanoma were significantly associated with risk; skin and eye colour
were significantly associated in a crude analysis only. Duration of residence in Australia
was strongly, positively associated with risk for all melanomas and for all sub-types except
for unclassifiable melanoma. After control for ethnic origin, the odds ratios for superficial
spreading melanoma were 1.2 (95% CI, 0.25-5.5) for people arriving in Australia at age 0-4,
1.7 (0.34-8.0) for those arriving at age 5-9, 0.74 (0.17-3.3) for those arriving at age 10-14,
0.25 (0.05-1.4) for those arriving at age 15-19 years or older (< 30 years) and 0.38
(0.19-0.78) for those arriving at age ~ 3 years (p for trend, < 0.0001) compared to those
born in Australia. A lifetime residential history was used to calculate the mean annual hours
of bright sunlight based on place of residence as a measure of potential exposure to the sun.
An analysis restricted to native-born Australians showed positive associations for all
melanomas and for each subtype except nodular melanoma. An analysis dichotomizing
exposure at an annual mean of > 2800 h sunlight at different ages showed that the highest
risk ratio for all melanomas and for the superficial spreading subtype were for high exposure
at ages 10-24. Cutaneous microtopography was used to measure skin damage; a positive
association was found with all melanomas, being strongest for lentigo maligna melanoma.
In a further analysis by individual habits of exposure to the sun (Holman et al., 1986a),
no significant association was seen for total outdoor exposure. Analysis by recreational
outdoor exposure, expressed as a proportion of total exposure, at ages 10-24 years showed
no significant association. For superficial spreading melanoma, analysis by specific activity
showed positive associations with boating (p = 0.04) and fishing (p = 0.07) and weaker,
nonsignificant associations with swimming and sunbathing at ages 15-24 or 0-9 years before
diagnosis. For other types of melanoma, no clear positive association was found; regular
swimmers had a lower risk of lentigo maligna melanoma (trend test significant). Occupa-
tional exposure was analysed on the basis of whether the site of the melanoma was usually
covered by clothing and compared to that of a referent group for whom the site was usually
covered: subjects for whom the site was exposed showed a significant positive association.
In comparison with the same referent group, patients who had never worked outdoors had
significantly increased risks for all melanomas. The type of bathing suit usually worn by
females in summer was assessed, and a positive association was found for wearing bikinis
or for nude bathing, which was significant for all trunk melanomas and for superficial
spreading melanoma on the trunk. When previous sunburns were classified by severity, no
significant trend was observed for all melanomas; but there was a positive trend for lentigo
maligna melanoma (p = 0.06) and a significant negative association for nodular melanoma.
In the smaller Queensland Melanoma study (Green, 1984; Green et al. , 1985a), 183
patients with histologically confirmed melanoma, other than lentigo maligna melanoma or
acral lentiginous melanoma, and 183 population controls matched for sex, age and area of
residence were interviewed at home using a standardized questionnaire, which included
objective measurements and naevi counts. The response rates were 97% and 92%, respec-
tively. Adjustment was made using a multiple logistic regression model. Hair colour, acute
sun reactions and naevi were significantly associated with risk. Skin colour, eye colour,
chronic sun reaction, freckling and family history of melanoma were significant in a crude
analysis only. Hours of occupational and recreational exposure to the sun from 10 years of
age across three categories gave risks of 1, 3.2 (95% CI, 0.9-12.4) and 5.3 (0.9-30.8) after
adjustment for naevi, hair colour and propensity to sunburn. Average levels of exposure to
UVB radiation were also allocated by residential history but showed no association with risk
for melanoma. People born in Queensland had moderately higher risks than those who
arrived there later in life or who had lived somewhere else at any time. Melanoma patients
had more kerotoses or skin cancers on their faces (odds ratio, 2.8; 1.1-7.2). Sunburn (Green
et al., 1985a) was defined as pain persisting longer than 48 h, with or without blistering, and
was recorded as the number of episodes in each decade. Risk increased with the number of
severe sunburns and was 1.9 and 5.0 in the two higher categories on matched analysis,
decreasing to 1.5 (0.7-3.2) and 2.4 (1.0-6.1), respectively, when adjusted for naevi and exact
age. An additional analysis of 49 cases of lentigo maligna melanoma and 49 controls
showed no association with sunburn (Green & O'Rourke, 1985; Green et al., 1986).
In a more detailed review of these data (Green et al., 1986), no association was observed
with occupational exposure to the sun. Analyses of recreational hours spent on the beach in
the sun were made for lifetime exposures, exposures at 10-19 years of age and exposures in
the five years prior to diagnosis; no strong or consistent association was seen in either crude
or adjusted analyses. Associations with total accumulated hours of exposure to the sun
(calculated by adding occupational and total recreational exposures) showed a positive trend
for lifetime exposure and exposure at ages 10-19 (odds ratio, 4.4; 95% CI, 1.8-184.5), but
no association was seen for exposure during the previous five years. Analysis of levels of
UVR by lifetime residential history showed no major association and no site-specific
association.
(ii) Europe
In a case-control study of residents of Oslo, Norway (Klepp & Magnus, 1979), 78 mali-
gnant melanoma patients over 20 years of age were compared with 131 unmatched hospital
controls with other cancers. Both cases and controls with advanced disease were excluded.
Information was obtained by questionnaire [response rate not given]. Hair and eye colour
were recorded independently by the interviewer and subject but were not associated with
risk for the disease, whereas skin reaction to sunlight and freckling were. A nonsignificant
odds ratio of [ 1.5] was found for men working outdoors for more than 3-4 in/day; the odds
ratio for taking sunbathing holidays in southern Europe was 2.4 (p = 0.05). No significant
association was seen with degree of exposure of different body sites, classified from 'as
often as possible' to 'hardly ever'.
Adam et al. ( 1981) conducted a population-based case-control study in the United
Kingdom of 111 female cases of malignant melanoma aged 15-49 traced from registries and
342 female controls randomly selected from general practitioners' lists and matched for age
and marital status. Information was obtained by postal questionnaire; response rates were
66% for cases and 68% for controls. Hair colour and skin reaction to sunlight, but not skin
colour, were significantly associated with risk for malignant melanoma. Slightly more cases
than controls reported deliberately tanning their legs or trunk, either at home or abroad. No
difference was reported in the amount of work, leisure or total time spent outdoors. [The
Working Group noted that the study concentrated on oral contraceptive use and that
information on exposure to the sun was very limited.]
MacKie and Aitchison (1982) conducted a case-control studyin western Scotland of 113
malignant melanoma patients aged 18-76 years and 113 sex- and age-matched hospital
controls with conditions not related to the skin. Cases of lentigo maligna melanoma were
excluded. Information about exposure to the sun within the previous five years was
obtained by questionnaire [response rate not given] and included occupational and
recreational exposure ~ 16 h versus < 16 h outdoor exposure per week) and history of
severe sunburn, defmed as either 'blistering sunburn' or 'erythema persisting for a week or
longer'. Other factors included in the multivariate analysis were social class and skin type.
A significant negative association was observed for recreational exposure and for
occupational exposure to the sun in males. A significant positive association was observed
for severe sunburn. No significant difference was observed for the number of continental
holidays taken or total number of days spent in sunnier climates.
Sorahan and Grimley (1985) studied 58 patients aged 20-70 years with cutaneous mali-
gnant melanoma (other than lentigo maligna melanoma) in two hospitals in the United
Kingdom and 182 hospital controls with diseases other than of the skin and 151 unmatched
controls from electoral rolls. The response rates were 64% for cases and 60% for each
control group. Information was obtained by postal questionnaire, and analyses were
adjusted using a multiple logistic regression model. A significant positive association was
observed for number of bouts of painful sunburn ever experienced, with an odds ratio
reaching 7.0 for five or more bouts compared to none. A significant positive association
was also seen with the number of holidays ever spent abroad in a hot climate, reaching 6.5
for 21 holidays or more, compared to none. Both associations were weakened, and the latter
became nonsignificant, after adjustment for propensity to sunburn, number of moles and
history of sunburn.
In another study in the United Kingdom (Elwood et al., 1986), 83 histologically con-
firmed cases over 18 years of age and 83 hospital controls (in- and out-patients), matched
for sex, age and area of residence, were interviewed at home using a questionnaire which
included objective measurements and naevi counts. The responses were validated by replies
to a postal questionnaire. The response rates were 74% for cases and 92% for controls.
Adjustment was made using a multiple logistic regression model. Skin reaction to sunlight,
freckling and naevi were significantly associated with risk. A history of sunburn causing
pain for two days or more gave a significant odds ratio of 3.2 (95% CI, 1.7-5.9). Past
outdoor occupational exposure showed a significantly reduced odds ratio of 0.2 (0.1-0.9) for
the second highest category but a nonsignificant odds ratio of 1.7 (0.3-8.6) for the highest
category and no overall trend.
In northern Italy, Cristofolini et al. ( 1987) compared 103 patients aged 21-79 under
treatment for cutaneous malignant melanoma at one hospital with 205 hospital controls with
diseases other than skin tumours. Subjects were interviewed [response rate not given] and
assessed by a dermatologist. Adjustment was made using a multiple logistic regression
model. Hair and skin colour and family history were significantly associated with risk, but
eye colour, freckling and number of naevi were not. A history of frequent sunburn as an
adult gave an odds ratio of 1.2 (95% CI, 0.7-2.1) and that of severe sunburn in early life an
odds ratio of 0.7 (0.4-1.2). Heavy or frequent exposure to sunlight during the previous 20
years, categorized as yes or no, gave a significantly reduced odds ratio of 0.6 (0.4-0.95).
Outdoor compared to indoor occupation gave a nonsignificant odds ratio of 0.9 (0.5-1.7),
and a history of carcinoma of the skin gave a risk ratio of 0.4 (0.02-2.9), based on small
numbers. Melanoma at exposed sites showed positive associations with heavy sun exposure
(1.44; 0.8-2.8) and outdoor occupation (1.8; 0.9-3.7), while melanoma at normally
unexposed sites showed a significant negative association with heavy exposure to the sun
(odds ratio, 0.25; 95% CI, 0.13-0.47).
In a study of melanoma in eastern Denmark (0sterlind et al., 1988b,c; 0sterlind, 1990),
474 cases of melanoma, excluding lentigo maligna melanoma patients, aged 20-79 were
compared with 926 population controls and matched for sex and age. Subjects were inter-
viewed at home using a questionnaire which included objective measurements and naevi
counts, and adjustment was made using a multiple logistic regression model. Response
rates were 92% for cases and 82% for controls. The number of sunburns (defined as those
causing pain for two days or longer) before age 15, from age 15 to 24 and over the previous
10 years were all significantly associated with risk: crude odds ratios for the maximal
categories, 3.7 (95% CI, 2.3-6.1), 2.4 (1.6-3.6) and 3.0 (1.6-5.4), respectively. Adjustment
for sex and host factors, including naevi, freckles and hair colour, reduced the risk ratios,
but they remained significant. Adjustment for sunburns before age 15 rendered the
associations with later sunburn weak and nonsignificant. Joint analysis of sunburns and
naevi suggested independent, additive risks. Significantly increased risks were seen with
residence near the coast before age 15 or for more than 30 years. Specific recreational
activities were investigated and categorized by the number of years of regular participation,
adjusted for sex and host factors, including number of naevi, and for other activities.
Significant positive associations were observed with sunbathing, boating, winter skiing and
swimming, the latter becoming nonsignificant after adjustment. Regular participation in
gardening, ball games, golf, horseback riding or hiking was not associated with risk for
melanoma. A positive trend was seen with vacations spent in beach resorts in southern
Europe (odds ratio, 1.7; 95% Cl, 1.2-2.4), which was weakened after adjustment for
sunbathing and sunburn (1.4; 1.0-2.1). Socioeconomic status showed a strongly positive
association in men, which became nonsignificant when adjusted for sunburn and
recreational exposure to the sun. Occupational exposure outdoors for at least six months
was associated with a significantly reduced odds ratio of 0.7 (0.5-0.9) in men; the protective
effect was most pronounced in men who started working outside at an early age and
continued for at least 10 years. No association was seen with skin grading categories
defined by microtopography.
In a study in nmthern Italy (Zanetti et al., 1988), 208 cases of histologically confirmed
malignant melanoma were identified from the regional tumour registry and were compared
with 416 controls chosen from the National Social Service Registry. Response rates were
87% for cases and 68% for controls. An increased risk was observed with light hair colour,
tendency to burn and a history of sunburn in childhood. No significant effect of region of
origin was observed. Exposure to the sun was assessed by activity: for outdoor work, a
nonsignificant increased risk was seen with the maximal duration of exposure (2 33 years)
in men, but the overall trend was nonsignificant. Outdoor sports, assessed by years of
participation, showed an increased risk at the maximal level in men and women (significant
for men). A significantly increased risk was found for men participating in sports
categorized as involving the greatest exposure to the sun. A nonsignificantly increasing
trend in men was observed for total number of weeks' holiday, but little effect was seen in
women; a significant positive trend was observed in men, but not for women, for the
number of weeks spent at the seaside in childhood. Similar exposure in adult years resulted
in a nonsignificant positive trend.
Garbe et al. (1989) studied 200 malignant melanoma patients at a dermatological
follow-up clinic in Berlin, Germany, in 1987 and 200 controls from the same clinic who had
any other skin disease (response rate, 90%). Subjects of non-German origin were excluded,
as were those seeking consultation for pigmented naevi or who had been treated previously
by UVR (10%). Occupational exposure to the sun, assessed as none, sometimes or nearly all
the time, showed a strongly increased risk up to an odds ratio of 5.5 (1.2-25.3). No
significant relationship was found with duration of leisure-time exposure to the sun or
number of sunburns [The Working Group noted that little detail was given about exposure
and that the control group consisted of patients with other skin disease.]
Weiss et al. (1990) studied 1079 cases of malignant melanoma reported to the Gennan
Dermatological Society Registries in 1984-87 and 778 hospital controls from the same
clinics. Positive associations were seen with occupational exposure to the sun, which
increased with the number of years of exposure. No association was seen with exposure to
the sun during leisure time or with sunbathing. [The Working Group noted that this study
appears to overlap with that of Garbe et al. (1989) and that the data were presented with
relative risks but with no test of significance.]
Beitner et al. (1990) studied 523 incident cases of malignant melanoma seen at a hospital
in Stockholm, Sweden (representing 64% of all cases registered in Stockholm County), and
505 controls selected from the population register for Stockholm County. Cases completed
a questionnaire while waiting at the clinic, and controls received the questionnaire by mail
(response rates, 99.6% and 96.2%, respectively). A significant positive effect was seen for
the number of sunbathing sessions each summer, with a history of erythema after sunbathing
and with sunbathing vacations abroad. Residence in countries around the Mediterranean or
in a sub-tropical or tropical climates for more than one year during the previous 10 years
gave a significant odds ratio of 1.9 [95% CI, 1.0-3.6]. There was no increase in risk with
sunbathing during winter vacations at high altitudes. Outdoor workers had a significantly
reduced risk of 0.6 (0.4-1.0) after adjustment for age, sex and hair colour.
Elwood et al. (1990) studied 195 cases of superficial spreading or nodular melanoma in
people aged 20-79 from five pathology laboratories in the United Kingdom and 195 controls
chosen from among all in- and out-patients in the region. Cases and controls underwent an
interview and a limited examination by an interviewer in their homes (participation rate-
cases and controls, 73%; voluntary response rate-cases, 91%: controls, 78%). Risk was
significantly increased with sunburn at age 8-12 (odds ratio, 3.6; 1.4-11.2), but no
significant increase was observed with sunburn at age 18-22 or with sunburn received 18-20
or five years prior to diagnosis. No other sun exposure variable was reported.
Grob et al. (1990) compared 207 consecutive white patients, 18-81 years old, with histo-
logically confirmed invasive melanoma (at least level 2: lentigo melanoma and acral
lentiginous melanoma excluded) seen in one dermatology clinic in Marseilles, France, with
295 controls. Controls under 65 years of age were chosen from among subjects interviewed
after reportedly random selection and examined at a public health centre; those over 65 were
chosen from among out-patients with non-cancer and non-dermatological conditions.
Patients and controls were examined and interviewed by the same dermatologist. Multiple
logistic model analysis was used. The risk for melanoma was increased significantly in
association with annual outdoor leisure exposure during the previous two years (odds ratio,
8.4; 95% Cl, 3.6-19.7), outdoor occupation (6.0; 2.1-17.4) and total lifetime sun exposure
(odds ratio for maximum category, 3 .4; 1.6-7.1 ). There was a nonsignificant association
with sunburns in recent years (1.7; 0.63-4.6) after adjustment for number of naevi, maximal
depth of suntan, hair colour, social level, complexion and age. [The Working Group found
the study difficult to interpret because of the nature of the control group and the relative
recency of measurements of exposure to the sun.]
In a report designed to produce a risk prediction model, MacKie et al. (1989) studied
280 cases of invasive cutaneous malignant melanoma (level 2 or deeper) from Scottish
melanoma registries. Controls were 280 hospital patients with non-dermatological diseases.
Response rates were 76% for cases and unknown for controls. An increased risk was
observed for history of severe sunburn (adjusted odds ratio, 7.6 (95% CI, 1.8-32.0) for men
and 2.3 (0.95.6) for women). A significant positive association for tropical residence was
noted for men, which became nonsignificant after adjustment. [The Working Group noted
that, apart from tropical residence, no data were presented on exposure to the sun.]
(iii) North America
Gellin et al. (1969) studied 79 patients, aged 30-79, with histologically confirmed mali-
gnant melanoma at one hospital in New York, USA, and compared them with 1037 hospital
controls with skin conditions other than cancer. Information was obtained by interview and
examination [response rate not given]. The odds ratios for duration of daily outdoor activity
were [2.8 (95% CI, 1.3-5.8)] for 6 h or more and [ 4.1 (2.5-6.8)] for 3-5 h, compared to 0-2
h. [The Working Group noted that the controls had skin diseases.]
Paffenbarger et al. (1978) reported on cases found by follow-up of subjects first exa-
mined when entering Harvard University in 1916-50 and the University of Pennsylvania in
1931-40. Out of a total of 50 000 male subjects and 1.71 million person-years of
observation, 45 deaths from melanoma were observed and each compared to four controls
born in the same year, who were classmates and who had survived as long as the case
subjects. Of the many factors investigated, only outside remunerative work was associated
with a significant risk for melanoma (odds ratio, 3.9;p = 0.01). Within the cohort, students
from New England had a 50% lower risk for melanoma than other students, presumably
owing to more northerly residence.
Lew et al. (1983) carried out a study in Massachusetts on Ill cases of cutaneous mali-
gnant melanoma, aged 23-81 , followed at one hospital and I 07 controls who were friends of
cases, matched by age and sex. Information was obtained by interview at the clinic;
response rates were 99% for cases and 90% for controls, and analysis was made using a
logistic regression model. Cases showed poorer tanning ability, and a significant
association was observed with blistering sunburn during adolescence (odds ratio, 2.1; 95%
CI, 1.2-3.6) and with 30 days or more vacation in sunny, warm places during childhood
(2.5; 1.1-5.8). The association with history of sunburn persisted after controlling for tanning
ability. [The Working Group noted that the nature of the controls and the simplicity of the
analyses presented make interpretation of the results difficult.]
Rigel et al. (1983) analysed data on 114 melanoma patients (out of a total of 328) seen in
a referral centre in New York between 1978 and 1981, and on 228 controls who were staff
and patients at the centre. Significantly increased risks were seen with > 2 h per day sun
exposure 11-20 years previously (odds ratio, 2.5;p = 0.005) and outdoor versus indoor
recreation (2.4; p = 0.01). [The Working Group noted that the selection of subjects and the
nature of the control group make these results difficult to interpret.]
In the Western Canada Melanoma case-control study (Elwood et al .. 1984, 1985a,b ),
carried out in four Canadian provinces, 595 cases of malignant melanoma, aged 20-79, and
595 population controls, matched for sex, age and province of residence, were questioned by
trained interviewers at their homes (response rates: cases, 83%; controls, 48-59%). Cases of
lentigo maligna melanoma and acrallentiginous melanoma were excluded. Analyses were
made using a multiple logistic regression model. Significant positive associations were
found after adjustment for host factors and ethnic origin for frequent recreational (odds
ratio, 1.7; 95% CI, 1.1-2.7) and holiday exposure (1.5; 1.0-2.3) and with the number of
sunny vacations per decade (1.7; 1.2-2.3). No overall trend was observed for occupational
exposure, but a significantly increased risk was associated with moderate occupational
exposure, defined as seasonal or short-term occupational exposure. Maximal occupational
exposure was associated with a significantly reduced odds ratio in men (0.5 [CI not given])
but not in women (1.5 [CI not given]). Analysis of total annual exposure to the sun from all
sources showed no overall trend (odds ratio, 1.0-1.6 in various categories above the minimal
exposure referent group). Severe or frequent sunburn in childhood resulted in a
nonsignificant odds ratio of 1.3, after adjustment for host factors and sun sensitivity. From
variables relating to sunburn on vacation and the usual degree of suntan in winter and
summer, positive associations were observed for increasing sunburn and with decreasing
usual tan. Cross-tabulation of sunburn with tendency to sunburn (skin type) did not change
the significant positive effect of tendency to bum, but the odds ratio for sunburn fell from
1.8 in the maximal category to 1.4 (p > 0.2) after adjustment for sun reaction. Simi larly,
cross-tabulation of usual degree of suntan against skin type gave little difference in the
positive association with reaction to the sun, but a weakening of the association with usual
degree of suntan was seen which became nonsignificant. A multivariate analysis including
history of sunburn, usual degree of suntan, skin type and host factors showed significance
for the two latter factors, nonsignificant positive effects of holiday sunburn and a significant
negative effect of usual degree of suntan. These results are interpreted as showing a primary
association with tendency to bum easi ly or to tan poorly rather than with history of either
sunburn or suntan. For men, a significant negative association was seen with outdoor
occupation, but this weakened and became nonsignificant when adjusted for recorded
exposure to the sun. Similarly, the crude odds ratio for upper compared to lower
socioeconomic groups was 3.8 (2.0-7.4) but was reduced to 2.3 (1.0-5.1) after adjustment
for host factors and for occupational, recreational and holiday sun exposure (Gallagher et
al. , 1987).
Elwood et al. ( 1987) made an analysis separating superficial spreading melanoma,
nodular melanoma and lentigo maligna melanoma in the western Canada study, based on
415,128 and 56 cases, respectively. Recreational exposure, holiday exposure and the
number of sunny vacations per decade were positively and significantly (trends) associated
with superficial spreading melanoma (odds ratios, 1.4, 2.0 and 2.2; 95% CI, 1.0-2.0, 1.4-2.9
and 1.5-3.3, respectively); recreational exposure was also positively associated with nodular
melanoma (2.4; 1.3-4.5), but neither holiday exposure nor the number of sunny vacations
showed an association. None of these measures of intermittent exposure was significantly
associated with lentigo maligna melanoma. Occupational exposure showed no significant
association with any of the three types. History of sunburn showed positive but
nonsignificant associations with superficial spreading and lentigo maligna melanomas but
not with nodular melanoma.
Brown et al. (1984) identified 120 men who had been aged 18-31 during the Second
World War from among 1067 patients seen at a melanoma clinic in New York City in
1972-80 and sent them questionnaires (response rate, 74%). Controls were 65 age-matched
subjects attending the same dermatology department with skin diseases other than
melanoma [response rate unknown]. Within the total of 74 cases and 49 controls who had
been in the armed services, the odds ratio for service in the tropics as compared to service in
the USA or Europe was [7.7; 95% CI, 2.5-23.6].
In a hospital-based study in Buffalo, NY, USA (Graham et al., 1985), 404 cases of cuta-
neous malignant melanoma referred to the Roswell Park Memorial Institute, aged from
under 30 to over 65, were compared with 521 controls with other neoplasms at the same
institute, using questionnaires completed on admission. There was a weak negative trend
with total number of hours of exposure to the sun, which was significant in men; a simi lar
trend was observed for average annual exposure to the sun. Occupational exposure to the
sun gave a nonsignificant reduction in risk in men in the highest exposure group after
adjustment for tendency to burn. Multivariate analysis showed a negative association with
cumulative exposure to the sun, which was significant in men when adjusted for tendency to
burn, freckling and light complexion. Results specific to recreational or holiday exposure to
the sun were not presented.
Dubin et al. (1986) compared 1103 cases of melanoma seen at the New York University
Medical Center from 1972 to 1982 (mostly in 1977 -79) to 5 85 controls interviewed in 1979-
82 at the skin clinic for conditions excluding cancer. Both cases and controls were inter-
viewed by physicians; response rates were 98% for cases and 78% for controls. In order to
complete the data on risk factors, a postal questionnaire was sent requesting information on
exposures to fluorescent lights and to the sun and on skin colour (response rates, 45% of
cases and 30% of controls). Mostly outdoor compared to mostly indoor work gave an odds
ratio of 2.5 (95% CI, 1.4-4.4) and mostly outdoor compared with mostly indoor recreation
gave an odds ratio of 1.7 (1.2-2.3), although mixed indoor and outdoor recreation gave a
significantly reduced risk of 0.6 (0.5-0.8). Overall exposure to the sun (three categories)
showed no trend. A history of the presence of solar keratosis gave a significant risk ratio of
5.0 (2.3-10.5). Quantitative total sun exposure was assessed for 623 cases and all 585 con-
trols: there was no significant trend with total hours of exposure to the sun per day 0-5, 6-10
or 11-20 years before diagnosis. [The Working Group noted that the cases and controls
were not interviewed over the same period.]
In a study based on a subset of the above (Dubin et al. , 1989), 289 cases and 527 controls
were interviewed using the same method (response rates, 100% of eligible cases; 70% of
controls [ 19% of potential controls were excluded because of diagnosis of a lesion known to
be caused by exposure to the sun]). Mostly outdoor occupation gave a nonsignificant ele-
vated risk. Mostly outdoor recreation was associated with a significantly elevated risk in
light tanners but a nonsignificant elevated risk in dark tanners (interaction nonsignificant).
Overall exposure to the sun was associated with significantly increased risks in all groups.
A history of sunburn was associated with a significantly increased risk in light tanners and
in all subjects but had a nonsignificant protective effect m dark tanners (interaction
significant).
When analysed by age group, a history of sunburn gave a positive association at age
20-39, a weak association at 40-59 and a negative association at 60 or over (interaction
significant). Prior skin cancer or solar keratosis had a significant effect, which was stronger
in men than in women (interaction nonsignificant).
In a study in San Francisco, Holly et al. ( 1987) compared 121 patients with nodular or
superficial spreading melanoma at a university melanoma clinic with 139 controls from a
medical screening clinic or from an orthopaedic clinic at the same centre. Response rates
were 'over 95%'. Sunburn score, based on the number of blistering sunburns during school
and young adult years, showed a significant odds ratio of 3.8 (95% CI, 1.4-10.4) after
controlling for naevi, hair colour and previous skin cancers. A positive association was seen
with previous skin cancer (3 .8; 1.2-12.4).
Weinstock et al. (1989) reported a case-control study within a cohort of US nurses (see
Hunter et al., 1990, p. 86). Data on 130 cases and 300 controls (response rates to post-
diagnosis questlonnaire, 85% and 81%, respectively) were analysed using multivariate
models. Following adjustment for skin sensitivity, significant positive effects were seen for
sunburn at ages 15-20 (odds ratio, 2.2; 95% CI, 1.2-3.8), but not at age 2: 30 (1.3; 0.7-2.3),
and for residence at a southern latitude at age 15-20 (2.2; 1.1-4.2), but not at age 2 30 (1.6,
0.9-2.8). No direct recording of exposure to the sun was reported.
A further analysis (Weinstock et al., 1991a) assessed the use of swimsuits in these
subjects. There was a significant positive association of melanoma risk with the frequency
of use of swimsuits of any type in sun-sensitive women (odds ratio, 6.4; 95% CI, 1.7-23.8)
but not in sun-resistant women (0.3; 0.1-1.0). After controlling for type of swimsuit and
sensitivity factors, melanoma risk was increased with increasing hours per day of outdoor
swimsuit use (any type) after age 30, but no association was seen with intensity of exposure
or with the number of winter vacations in warm and sunny locations. The use at age 15-20
of a bikini compared to high backline, one-piece swimsuits, gave an odds ratio for all
melanomas of 1.9 (1.0-3.7) and for trunk melanoma specifically of 0.8 (0.3-2.6); the risks
were 3.5 [CI not given] among sun-sensitive women and 1.3 [CI not given] among less
sun-sensitive women, but the interaction was not significant.
In a case-control study of patients attending a pigmented lesion clinic in Boston, USA
(Weinstock et al., 1991b), 186 had cutaneous melanoma; the 239 controls had other
dermatological diagnoses, the most frequent of which were common naevus and solar
keratosis. Data were obtained from medical records and from a self-administered
questionnaire completed before clinical examination and were analysed by a multivariate
method. Significantly increased risks for melanoma were associated with lack of tan after
repeated exposures as a teenager (odds ratio, 2.3; 95% CI, 1.0-4.9). A nonsignificant trend
towards increased risk was observed for residence in southerly areas. [The Working Group
noted that the paper dealt primarily with dysplastic naevi and the results on melanoma are
not given in detail, and that the controls also had dermatological conditions.]
Table 16. Case-control studies of melanoma in which exposure to the sun and/or artificial ultraviolet radiation was
assessed
Place Period of No. of Source of cases Melanoma No. of Type of control Reference
diagnosis cases type controls
Australia
East Australia NS 173 3 hospitals All types 173 Other cancers Lancaster & Nelson
(1957)
Queensland, 1963-69 468 1 hospital All types 468 Hospital patients, Beardmore ( 1972)
Australi a including skin
cancers
Western Austraia 1980-81 51 1 Population All types 511 Population Holman & Armstrong
(1984a,b)
Queensland, 1979-80 183 Popul ation NoLMM 183 Population Green (1984); Green et
Australia al. (1985a)
Europe
Oslo, Norway 1974-75 78 1 hospital All types 131 Other cancers, same Klepp & Magnus (1979)
hospital
United Kingdom 1971-76 11 1 Popul ation All types 342 General practice Adam eta/. (1981)
lists
Western Scotland 1978-80 113 Hospital NoLMM 113 Hospital, non-skin MacKie & Aitchison
(1982)
Birmingham, UK 1980-82 58 2 hospitals No LMM 333 Hospital and Sorahan & Grimley
population (1985)
Nottingham, UK 1981-84 83 Population All types 83 Matched hospital Elwood et a/. (1986)
(2 hospitals)
Trento, Italy 1983-85 103 1 hospital All types 205 Hospital Cristofolini et al. (1987)
East Denmark 1982-85 474 Population NoLMM 926 Matched population 0sterlind et al.
(I988a,b); 0sterlind
(1990)
Turin, Italy 1984-86 208 Population All types 416 Population Zanetti et al. (1988)
Berli n, Germany 1987 200 1 hospital All types 200 Skin clinic patients Garbe et al. (1989)
Table 16 ( contd)
Place Period of No. of Source of cases Melanoma No. of Type of control Reference
diagnosis cases controls
Scotland 1987 280 Population Invasive 280 Hospital, excluding MacKie eta/. (1989)
MMat skin
least type 2
Germany 1984-87 1079 6 dermatology All types 778 Skin clinic patients Weiss et al. ( 1990)
clinics
Stockholm, Sweden 1978-83 523 1 hospital All types 505 Matched population Beitner et al. (1990)
Midlands, Uk 1984-86 195 Population SSM and 195 Hospital in-/out- Elwood eta!. ( 1990)
NM patients
Southeast France 1986-88 207 Hospital Invasive, 295 Health centre Grob et al. ( 1990)
all types
North America
New York, USA 1955-67 79 l hospital All types 1037 Other skin diseases, Gellin et a!. (1969)
non-cancer
Boston, MA, USA NS 45 Cohort of All types 180 Classmates Paffenbarger et al.
Philadelphia, P A, university (1978)
USA alumni
Boston, MA, USA 1978-79 Ill 1 hospital All types 107 Friends of cases Lew et al. (1983)
New York, USA 1978-81 114 1 hospital All types 228 Patients and staff Rigel eta/. (1983)
New York, USA 1972-80 74 1 melanoma All types 49 Skin clinic patients Brown et al. (1984)
clinic
Western Cauda 1979-81 595 Population SSM, NM 595 Population Elwood eta!. ( 1984,
orUCM 1985a,b)
Buffalo, NY, USA 1974-80 404 Hospital patients All types 521 Cancer patients Graham et al. (I 986)
New York, USA 1972-82 1103 3 hospitals All types 585 Skin clinic patients Dubin et al. (1986)
Western Canada 1979-81 415 Population SSM 415 Population Elwood et al. (1987)
128 NM 128
56 LMM 56
San Francisco, CA, 1984-85 121 I melanoma NMand 139 Clinic patients Holly eta!. (1987)
USA clinic SSM
Table 16 ( contd)
Place Period of No. of Source of cases Melanoma No. of Type of controls Reference
diagnosis cases type controls
New York, USA 1979-82 289 3 hospitals All types 527 Non-cancer skin Dubin eta/. ( 1989)
patients
USA 1976-84 130 Nurses cohort AM 300 Nurses cohort Weinstock eta/.
excluded (1989)
Boston, MA, USA 1982-85 186 1 hospital All types 239 Skin clinic patients Weinstock eta/.
(1991 b)
NS, not specified; SMM, superficial spreading melanoma; NM, nodular melanoma; UCM, unclassifiable melanoma; LMM, lentigo maligna melanoma (or
Hutchinson's melanotic freckle); AM, acrallentiginous melanoma
(d) Collation of results
The studies summarized above show that a range of host characteristics are related to
melanoma risk, including ethnic origin, skin, hair and eye pigmentation, and, importantly, a
tendency to sunburn or suntan, often expressed clinically as skin type. These factors can be
assumed to reflect genetic sensitivity to cutaneous effects of sun exposure and, in addition to
the indirect evidence of a role of exposure to the sun in melanoma that they provide, should
be considered as confounders in a relationship between sun exposure and melanoma. The
numbers of acquired benign naevi and of dysplastic naevi have been shown to be very
strong risk factors for melanoma in several studies; the density of freckling on the skin has
also been shown to be a risk factor. Because there is evidence that these outcomes are
themselves related to sun exposure, and in the case of naevi may be intermediate steps in the
genesis of melanoma, they should not be considered confounding factors (Armstrong,
1988). Most of the studies relied on a wide range of questions to assess different aspects of
sun exposure. Armstrong (1988) developed a useful classification of such questions,
dividing them into those that assess potential exposure, such as place of residence and time
of migration, those that record actual exposure and those that record response to exposure,
such as questions on sunburn and suntanning.
(i) Total sun exposure: potential exposure by place of residence (Table 17)
Consistent with the descriptive studies, Holman and Armstrong (1984b) showed that the
risk in migrants arriving in Australia before age 10 (odds ratio, 0.89; 95% CI, 0.44-1.80) is
as high as that of the Australian born (1.00), and the risk in those arriving at age 10 or above
is much less (0.34. 0.16-0.72 for age 10-29; 0.30; 0.08-1.13 for age 2 30). These data are an
improvement on descriptive data as they allow control for ethnic background and pigmen-
tation. In the same study, an association was seen with annual hours of bright sunlight
averaged over all places of residence.
In the USA, two case-control studies (Graham et al., 1985; Weinstock et al., 1989)
showed increased risks for people who had lived at southerly latitudes.
Increased risks in people who have lived near the coast were seen in Denmark (0sterlind
et al., 1988b) and in Queensland, Australia (Green & Siskind, 1983). It was assumed in the
Danish study that coastal residence would involve more exposure to the sun. In
Queensland, living near the coast is not related to annual ambient UVR, which varies with
latitude, so that peak summer UV irradiance is higher in the interior than on the coast
(Green & Siskind, 1983). The observations are thus due either to different behavioural
patterns with geographical location or to differences in exposure to UVR.
(ii) Biological response to total sun exposure
It has been assumed that a history of nonmelanocytic skin cancer, solar keratoses, actinic
turnouts or changes on cutaneous microtopography are all indicators of cumulative sun
damage. Positive associations are seen with these measures in studies in Australia and in
the USA, although 0sterlind et al. (1988b) in Denmark saw no relationship with micro-
topographical change (Table 17).
Table 17. Results of case-control studies on melanoma: place of residence, biological markers
Place Direction of
ORa
95% CI p value Measurement of exposure Reference
association
Potential exposure by place of residence
Australia Up 5 Residence near coast; mortality Green & Siskind (1983)
rate/ 1 00 000 (incidence
rate/ 100 000, 37)
Australia Down 0.3 (0.1-1.1) < 0.001 Age at arrival in Australia; OR Holman & Armstrong
given for g e ~ 30 years; p (1984b)
value for trend
Australia Up 2.8 (1 .8-4.8) < 0.001 Mean annual hours of bright Holman & Armstrong
sunlight at places of residence; (1984b)
p for trend
Graham et al. (1985)b USA Up 1.4 (0.9-2.0) > 0.05 Ever resided below 40 N
latitude
Australia Down 0.3 (0.1-1.4) > 0.05 Length of residence in Green et al. ( 1986)
Australia; risk associated with
migration to Australia
Denmark Up 1.7 (1.1-2.7) 0.006 Residence near coast; crude OR 0sterlind et al. ( 1988b)
USA Up 2.2 (1.1-4.2) 0.02 Residence in southerly latitude Weinstock et al. (1989)
at age 15-20, OR for 12.6
Biological markers of cumulative sun exposure
Australia Up 2.7 (1.4-5.0) 0.003 Cutaneous microtopography; p Holman & Armstrong
for trend (1984b)
Australia Up 3.7 (2.1-6.6) <0.001 History of nonmelanocytic skin Holman & Armstrong
cancer (1984b)
Australia Up 3.6 (1.8-7.3) < 0.001 Actinic tumours on face Dubin et al. ( 1986)
USA Up 5.0 (2.3-10.5) < 0.01 History of solar keratosis Green & O'Rourke
(1985)
USA Up 3.8 (1.2-12.4) 0.03 History or nonmelanocytic skin Holly et al. ( 1987)
cancer, adjusted
Denmark Flat 1.1 (0.7-1.8) > 0.05 Cutaneous microtopography; 0sterlind et al. (I 988b)
crude OR
"Odds ratio for maximal catego1y
bResults calculated by Armstrong (1988)
(iii) Total sun exposure assessed by questionnaire
The results of studies in which total sun exposure was assessed using questionnaires,
either over lifetime or at different periods of life, have been mixed (Table 18). Positive
associations were seen by Green (1984) in Queensland, Australia; no consistent overall
association was seen in western Canada, and in Western Australia the association was
negative. The results of the other studies are similarly mixed. This inconsistency, in
contrast to the results noted above by place of residence and by biological response, could
be due either to the difficulty of assessing total sun exposure by questionnaires (Armstrong,
1988) or to different effects of differing patterns of exposure to the sun.
(iv) Short periods of residence implying high potential exposure
Several case-control studies have reported, usually as incidental findings, that subjects
who have had a short period of residence in tropical or sub-tropical environments have an
increased risk for melanoma (Table 19).
(v) Occupational exposure
Regular outdoor occupational exposure is probably the most convenient measure of
relatively constant sun exposure and has been assessed with differing degrees of detail, from
simple questions on ever/never or a basic amount of outdoor exposure, to detailed
assessments involving assessments of clothing habits, geographical location of work and so
on. The results appear to be inconsistent (Table 20). The more detailed studies, however,
show more consistency, with a significant negative association, particularly in men, who
constitute most of the highly exposed subjects (Table 21).
An overall irregular pattern was seen in western Canada, probably because individuals
with relatively little occupational exposure are those who perform outdoor work seasonally
or for short periods, often in early life, so that this exposure may be an indication of inter-
mittent rather than constant exposure (Elwood et al., 1985b). Such results are consistent
with the effects of a short period of residence in a sunny place, as reviewed earlier.
Paffenbarger et al. (1978) also showed that students who recorded outdoor work before
college [presumably summer employment] had a significantly increased risk of melanoma
in later life.
(vi) Intermittent exposure
To assess the effects of intermittent exposure, investigators have asked questions about
specific activities that would be likely to represent relatively severe intermittent exposure,
such as sunbathing, or asked particularly about holidays in sunny places, or used more
complex questionnaires to attempt to assess total intermittent exposure through recreational
or holiday activities. Most of these studies show positive associations, but few show large
effects (Table 22).
In general, the more detailed studies show reasonably consistent positive results. For
example, in western Canada, significant positive associations were seen with recreational
and holiday sun exposures in activities involving reasonably intense sun exposure, such as
beach activities (Elwood et al., l985b ). In Denmark, rather simi lar relative risks of 1.5-1.9
were seen with regular participation in activities such as sunbathing, boating, skiing,
swimming and vacations in sunny places (0sterlind et al., 1988b). Significant positive
associations with sunbathing were seen in the Swedish study of Beitner et al. (1990). In the
study of Zanetti et al.
Table 18. Results of case-control studies on melanoma: total sun exposure assessed by questionnaire
Place Direction of OR" 95%CI p value Measurement of exposure Reference
association
USA Up 2.5 NA < 0.001 Sun exposure 2 h/day, 11-20 years Rigel et al. ( 1983)
previously
Australia Up 5.3 0.96-30.8 NA Total sun exposure throughout life > Green (1984)
50 000 h, adjusted
Canada Weakly up 1.2 0.7-2.0 > 0.1 Hours of sun exposure per year, p for Elwood et al. ( 1985b)
trend
USA Down 0.6 0.4-0.9 < 0.05 Total sun exposure throughout life Graham et al. (1985)b
USA Weakly up 1.1 0.6-2.1 >0.05 Hours of sun exposure 0-5 years Dubin et al. ( 1986)
previously,> 5 h/day
USA Down 0.85 0.5-1.4 > 0.05 Hours of sun exposure 11-20 years Dubin et al. (1986)
previously, > 5 h/day
USA Weakly up 1.1 0.8-1 .6 > 0.05 Lifetime sun exposure Dubin et al. (1986)
Australia Down 0.7 0.4-1.1 0.13 Mean total outdoor hours/week in Holman et al. (1986a)
summer,> 23 h/week; p for trend
Italy Down 0.7 0.4-1.1 > 0.05 Heavy or frequent exposure in Cristofolini et al. ( 1987)
previous 20 years
France Up 3.4 1.6-7.1 < 0.05 Total lifetime outdoor sun exposure, Grob et al. (1990)
adjusted
aOdds ratio for maximal category
bResults calculated by Armstrong ( 1988)
Table 19. Evidence of melanoma risk with short periods of residence implying high potential exposure
Place Direction of Odds ratio 95% CI p value Measurement of exposure Reference
association
USA Up [7.7 2.5- 0.0002 US service: tropics versus USA/Europe Brown et al.
23.6] (1984)
UK Up 1.8 0.6-5.1 >0.05 2: 1 year living in tropics, subtropics Elwood ( 1986)
Scotland Up 2.6 (males) 1.3-5.4 < 0.05 2:5 years living in tropics, subtropics; crude MacKie et al.
1.8 (females) 0.8-4.0 > 0.05 OR (1989)
Sweden Up 1.9 1.0-3.6 < 0.05 Living in Mediterranean, tropics, Beitner eta!.
subtropics > 1 year in last 1 0 years ( 1990)
Table 20. Results of case-control studies on melanoma: occupational exposure
Place Direction of
ORa
95%CI p value Measurement of exposure Reference
association
USA Up 3.9 NR 0.001 Outdoor work recorded at college medical Paffenbarger et al. (1978)
examination; prospective
Norway Up 1.4 0.6-3.5 0.37 At least 3-4 h of outdoor work a day Klepp & Magnus (1979)b
Scotland Down 0.5 0.2-1.2 > 0.05 Hours of outdoor occupation a week MacKie & Aitchison (1982)b
USA Up 1.2 NR > 0.05 Outdoor occupation versus indoor Rigel et al. (1983)
Canada Irregular 0.9 0.6-1 .5 < 0.01 Hours of outdoor occupation a week in Elwood et al. (1985b)
summer
USA Down 0.7 0.3-1.3 > 0.05 Lifetime hours of outdoor occupation Graham et al. (1985)
USA Up 2.5 1.4-4.4 < 0.05 Mostly outdoors; multiple logistic OR= Dubin et al. (1986)
2.4, p < 0.05
UK Irregular 1.7 0.3-8.6 0.5 Lifetime hours of outdoor occupation Elwood et al. ( 1986)
Australia Down 0.5 NR 0.04 Mean hours of outdoor occupation a week Holman et al. (1986a)
111 summer
Denmark Down 0.7 0.5-0.9 < 0.05 Outdoor occupation versus indoor 0sterlind et al. ( 1988b)
Italy Irregular 2.1 0.6-6.8 0.32 Outdoor occupation Zanetti et al. ( 1988)
Germany Up 5.5 1.2-25.3 < 0.05 Outdoor occupation; adj usted OR = 11.6 Garbe et al. (1989)
(2.1-63.3)
Sweden Down 0.6 0.4-1.0 NR Outdoor occupation, yes/ no Beitner et al. (I 990)
France Up 6.0 2.1-17.4 < 0.05 Outdoors occupation versus indoor Grob et at. (1990)
NR, not reported
bCalculated by Armstrong (1988)
Table 21. Results of case-control studies on different types of melanoma and occupational exposure
Place Type of melanoma Odds ratio 95%CI p value Measurement of exposure Reference
Canada Excluding LMM 0.5 [0.3-1.0] NR > 32 h outdoor occupation a Elwood et al. (1985b)
and ALM week in summer (men)
Queensland, Excluding LMM No association Outdoor occupation Green et al. ( 1986)
Australia and ALM
Western SSM 0.5 NR 0.04 for Top quattile, hours of outdoor Holman et al. (1986a)
Australia trend occupation a week in summer
Denmark Excluding LMM 0.7 0.5-0.9 < 0.05 Outdoor occupation (men) 0sterlind et al. (1988b)
andALM
LMM, lentigo maligna melanoma; ALM, acrallentiginous melanoma; SSM, superficial spreading melanoma; NR, not reported
Table 22. Results of case-control studies on melanoma: intermittent exposure
Place Direction of
ORa
95% cr p value Measurement of exposure Reference
association
Norway Up 2.4 1.0-5.8 0.06 Sunbathing holidays in southern Europe in Klepp & Magnus
previous 5 years (1979)b
UK Up 1.5 0.9-2.5 0.16 Spent some time deliberately tanning their legs Adam et al. (1981 )b
Up 1.6 1.0-2.5 0.05 Spent some time deliberately tanning their trunk Mackie & Aitchison
(1982)
11
Scotland Down 0.4 0.2-0.9 < 0.05 Hours a week in outdoor recreation Lew et al. (1983)
USA Up 2.5 1.1-5.8 < 0.05 Days of vacation in a sunny warm place in Rigel et al. (1983)
childhood
USA Up 2.4 NR 0.01 Outdoor versus indoor recreation Elwood et a/. (1985b)
Canada Up 1.7 1.1-2.7 < 0.01 Hours of high exposure in recreational activities
per week in sununer
Up 1.5 1.0-2.3 < 0.01 Hours of high and moderate exposure in
recreational activities per day in summer
vacations
Up 1.7 1.2-2.3 < 0.001 Number of sunny vacations per decade
UK Up 5 NR > 0.05 Number of holidays abroad in hot climate; Sorahan & Grimley
adj usted (1985)
USA Irregular 1.7 1.2-2.2 < 0.01 Recreation type; multiple logistic OR, 1.0 Dubin et al. ( 1986)
Australia Irregular 1.9 0.5-7.4 0.62 Recreational hours spent in sun on beach over Green et al. (1986)
whole life; crude RR
Australia Up 1.3 0.9-1.9 0.25 Proportion of recreational outdoor exposure in Holman et al. (1986a)
summer at I 0-24 years of age; p for trend
Up 2.4 1.1-5.4 0.04 Boating in summer; p for trend
Up 2.7 1.2-6.4 0.07 Fishing in summer; p for trend
Irregular 1.1 0.7-1.8 0.66 Swimming in summer; p for trend
Up 1.3 0.8-2.2 0.26 Sunbathing in summer at 15-24 years of age; p
for trend
Denmark Up 1.9 1.3-2.9 0.004 Sunbathing; crude RR; p for trend 0sterlind et at. (1988b)
Up 1.7 1.1-2.8 0.012 Boating; crude RR; p for trend
Up 1.5 0.9-2.4 0.006 Skiing; crude RR; p for trend
Up 1.5 1.2-2.0 0.004 Swimming (outdoors); crude RR; p for trend
Up 1.7 1.2-2.4 < 0.01 Vacations in sunny resorts; crude RR; p for trend
Table 22 ( contd)
Place Direction of OR 95%CI p value Measurement of exposure Reference
association
Italy Irregular 2.6 1.0-6.9 0.003 Years of outdoor sport (men); p for trend Zanetti et al. (1988)
Up 3.8 1.1-13.0 NR High-exposure sports (men)
Inegular 1.9 0.6-5.8 0.27 Total weeks' vacation (men); p for trend
Up 3.7 1.4-9.7 0.001 Weeks' vacation near sea; early life (men); p for
trend
Up 1.6 0.7-3.6 0.77 Weeks' vacation near sea; adult life (men); p for
trend
Irregular 2.1 0.6-7.9 0.37 Years of outdoor sport (women); p for trend
Up 2.3 0.6-9.1 NR High-exposure sports (women)
Irregular 1.1 0.5-2.4 0.56 Total weeks' vacation (women); p for trend
Up 1.2 0.6-2.5 0.56 Weeks' vacation near sea; early life (women); p
for trend
UP 1.5 0.9-2.7 0.16 Weeks' vacation near sea; adult life (women); p
for trend
Germany No NR NR NR Free-time sun exposure Garbe et al. (1989)
association
Sweden Up 1.8 1.2-2.6 <0.05 Number of sunbaths per summer Beitner et al. ( 1990)
UP 2.4 1.5-3.8 < 0.05 Sunbathing vacations abroad
France UP 8.4 3.6-19.7 < 0.05 Outdoor leisure exposure Grob eta/. (1990)
NR, not repotted
bCalculated by Armstron (1988)
(1988) in Turin, Italy, positive associations were seen with doing an outdoor sport for many
years and with number of weeks of holidays spent near the sea. These consistently positive
associations contrast with the less consistent pattern seen in Australia. In Western Australia,
stronger associations are seen with boating and fishing than with swimming and sunbathing,
which would be expected to involve more intense exposure to the sun, and only a weak
association was seen with the proportion of outdoor time spent on recreational activities in
teenage and early adult years (Holman et al. , 1986a). In Queensland, Green et al. ( 1986)
found only irregular associations with recreational hours spent at the beach or in other
activities with intense exposure to the sun. This finding might be consistent with the
concept that, in a sunny environment, recreational activities may involve sufficient
frequency or intensity of sun exposure to result in a constant rather than an intermittent dose
pattern.
(vii) Sunburn
Most of the studies show positive associations between risk for melanoma and a history
of sunburn (Table 23). The questionnaires usually defined very severe sunburn as a bum
that causes pain lasting for at least two days or blistering. The greater consistency of this
relationship compared to that with intermittent exposure may indicate a specific association
with sunburn per se or that sunburn is simply a more easily remembered measure of
intermittent and/or intense exposure to the sun.
A history of sunburn indicates both unusually intense exposure and skin sensitivity, and
therefore studies which assess sunburn while controlling for sensitivity through a separate
question on tendency to bum are important. Both the western Canada and Western
Australia studies when analysed in this way show that the association is primarily with
tendency to bum rather than with a history of sunburn (Elwood et al., 1985a; Holman et al.,
1986a). The studies in Queensland, Demnark and Scotland, however, show strong
associations with sunburn history even after controlling for tendency to bum and other
measures of skin sensitivity.
Because sensitivity to the sun and sunburn are likely to be highly correlated and both are
likely to be measured with a degree of error, it is difficult to distinguish their effects.
Similarly, sunburn is likely to be confounded with intermittent exposure of a less intense
nature, from which it cannot readily be distinguished because of measurement error
(Armstrong, 1988).
The study in England by Elwood et al. (1990) assessed sunburn at different ages and
showed the strongest association with sunburn at ages 8-12; a stronger association with
sunburns at young age was also seen by Weinstock et al. (1989) and by 0sterlind et al.
(1988b).
2.1.4 Malignant melanoma of the eye
(a) Case reports
In general. case reports were not considered, owing to the availability of more infor-
mative data.
Kraemer et al. (1987) reported on 830 cases of xerodetma pigmentosum, with a median
age of 12 years at last observation, located through a survey of published case reports.
Ocular abnormalities were found in 328 of 337 patients on whom information was available.
Table 23. Results of case-control studies on melanoma: history of sunburn
Place Direction of
ORa
95% CI p value Measurement of exposure Reference
association
Scotland Up 2.8 1.1-7.4 < 0.05 Blistering sunburn or erythema persisting > I MacKie & Aitchison
week (1982)
USA Up 2.1 1.2-3.6 < 0.05 Blistering sunburn during adolescence (yes/no) Lew et al. (1983)
Canada Up 1.8 1.1-3.0 < 0.01 Vacation sunburn score Elwood et al. ( 1985a)b
Australia Up 2.4 1.0-6.1 < 0.05 Number of severe sunburns throughout life Green et al. (1985a)
UK Up 4.2 NR < 0.01 Bouts of painful sunburn; adjusted Sorahan & Grimley
(1985)
Canada Up 3.2 1.7-5.9 < 0.001 Sunburn causing pain for 2: 2 days Elwood et al. ( 1986t
Australia lnegular 0.9 0.5-1.5 0.43 Sunburn causing pain for 2: 2 days, during last Holman et al. (1986a)b
10 years
Up 1.2 0.6-2.3 0. 1 Sunburn causing pain for 2: 2 days, < 10 years of
age
Up 1.7 1.0-2.9 0.003 Blistering sunburn
Italy Down 0.7 0.4-1.2 > 0.05 Severe sunburn in adolescence or early adult life Cristofolini et al. (1987)
(yes/no)
Up 1.2 0.7-2.1 > 0.05 Sunburn as an adult (yes/no)
USA Up 3.8 1.4-10.4 NA Number of blistering sunburns up to adult age, Holly et al. (1988)
adjusted
Denmark Up 3.7 2.3-6.1 < 0.001 Sunburn causing pain for 2 days, < 15 years of 0sterlind et al. (1988)
age
Up 3.0 1.6-5.4 < 0.001 Sunburn causing pain o r ~ 2 days, during
previous 10 years
Italy Up( men) 4.1 1.8-9.2 < 0.05 Sunburn in childhood (yes/no) Zanetti eta/. (1988)
Up( women) 2.7 1.3-5.6 < 0.05
Germany No association NR NR NR Number of sunburns Garbe eta/. ( 1989)
Scotland Up( men) 7.6 1.8-3.2 NR Number of episodes of severe sunburn, and age, MacKie et at. (1989)
adjusted
Up( women) 2.3 0.9-5.6 NR Number of episodes of severe sunburn, any age,
adjusted
Table 23 ( contd)
Place Direction of
ORa
95% CI
association
USA Up 2.2 1.2-3.8
Sweden Up 1.7 1.0-2.9
UK Up 3.6 1.4-11.2
No association 1.0 0.6-2.0
Up 1.8 0.9-3.7
Up 1.2 0.6-2.3
NR, not reported
bData calculated by Armstrong ( 1988)
p value
0.01
NR
< 0.05
> 0.05
> 0.05
> 0.05
cExposure to fluorescent and other Lighting sources
Measurement of exposure Reference
Number ofb1istering sunburns at ages 15-20 Weinstock et at. ( 1989)
Erythema after sunbathing Beitner eta/. ( 1990)
Moderate sunburn at ages 8-12 (yes/ no) Elwood et al. (1990)
Moderate/ maximum sunburn 18-20 yrs (yes/no)
Moderate/ maximum sunburn 18-20 yrs before
diagnosis (yes/ no)
Moderate/ maximum sunburn 5 years before
diagnosis (yes/ no)
Of these, 88 were reported to have some form of ocular neoplasm, mostly in the limbus,
cornea and conjunctive. Five of these patients were reported as having ocular melanoma;
only one was specified as being of uveal origin. [The Working Group recognized that data
collected from previously published case reports is not uniform and may not be typical of a
true incidence or prevalence series. Furthermore, no information is available on the
relationship between solar exposure and the occurrence of ocular melanoma in these
patients.]
As there is no separate lCD code for intra-ocular melanoma, descriptive data for cancer
of the eye (ICD-9 190) as a whole have been used as a surrogate. Intra-ocular melanoma
comprises some 80% of tumours of the orbit of the eye (0sterlind, 1987), and cancer of the
eye has been used as a surrogate for adult ocular melanoma in previous studies (Swerdlow,
1983a,b).
(i) Ethnic origin
Examination of incidence figures from many parts of the world reveals higher rates of
ocular tumours in whites than in blacks or Asians residing at the same latitude and under
similar conditions (Waterhouse et al., 1976; Muir et al., 1987).
(i i) Place of birth and residence
When rates for whites are evaluated separately, no variation in incidence rates for ocular
tumours is seen with decreasing latitude in the northern hemisphere (Table 24). Similarly,
no incidence grading was seen among whites in the USA (Table 25). The more northerly
states of Australia do not show higher incidence rates for ocular tumours than the southern
states (Table 25).
Table 24. Trends in cancer of the eye for whites by latitude and by time period (rates
per 100 000 age standardized to UICC 'world population')
Latitude Area

Men Women Men Women Men Women
56 -61 N Denmark 1.4 1.2 0.8 0.7 1.0 0.7
Finland 0.9 1.0 0.9 0.7 1.0 0.7
Sweden 1.3 1.2 0.9 0.8 0.9 0.6
47 -55 N Canada
British Columbia 1.0 0.8 0.9 0.6 0.7 0.4
Alberta 0.8 0.6 0.8 0.9 0.7 0.7
Saskatchewan 1.3 0.8 1.1 1.0 1.0 0.7
Manitoba 1.7 0.9 1.2 1.0 0.8 0.8
46 N Geneva, 0.4 0.2 0.8 1.1 0.6 1.1
Switzerland
38N San Francisco, CA, 0.9 0.9 0.9 0.5 0.9 0.8
USA
35N New Mexico, USA 1.0 0.7 1.3 0.7 0.9 0.9
32 -38 os Australia
New South Wales NA NA 0.8 0.8 0.9 0.5
South NA NA 0.9 1.0 0.7 0.6
Table 24 (contd)
Latitude Area - 1968-72a - 1977-82c
Men
22 os Hawaii, USA 0.4
3 $ Cali, Colombia 0.6
NR, not reported
8
From Waterhouse et al. (1976)
bFrom Waterhouse et at. (1982)
cFrom Muir et al. (1987)
Women Men Women Men
0.2 1.2 0.2 1.0
0.2 0.4 0.5 0.5
Women
0.0
0.5
Table 25. Incidence of cancer of the eye (ICD-9 190) in US and
Australian whites 1978-82 in various locations by latitude
Latitude Location Male rate/ Female rate/
100 000 100 000
USA
47 N Seattle 0.9 0.8
42N Detroit 0.7 0.6
42 N Iowa 1.0 0.7
41 N Connecticut 0.6 0.3
41 N New York City 0.5 0.4
41 N Utah 1.4 1.1
38N San Francisco Bay Area 0.9 0.8
35N New Mexico 0.9 0.9
34N Los Angeles 0.7 0.6
33 N Atlanta 0. 7 0.8
22N Hawaii 1.0 0.0
Australia
43 os Tasmania 1.2 0.8
38 os Victoria a 1.1 0.4
34 os South Australia 0.7 0.6
33 os New South Wales 0.9 0.5
32 os Western Australia 1.6 0.5
28 os Queenslanda 0.6 0.7
From Muir et al. (1987); rates standardized to UICC 'world
population'
aData available only for 1982
Schwartz and Weiss (1988) compared the state of birth of 763 white (not of Spanish
origin) US patients with uveal melanoma diagnosed between 1973 and 1984 and identified
in nine cancer registries with those of the whites covered by the registries as recorded in the
1980 census. Patients with unknown or foreign birthplace or non-uveal ocular melanomas
were excluded. Risk estimates were adjusted for age, sex and residence. The odds ratio for
subjects born in the southern USA (south of 40 N) was 1.1 (95% CI, 0.8-1.5). When states
were classified according to average daily global solar radiation, a nonsignificant gradient
was observed, only among women (odds ratio for> 15 500 kJ/m
2
versus:::; 12 300 kJ/m
2
,1.6;
95% Cl, 0.7-3.6).
Mack and Floderus (1991) examined birthplace and residence of patients diagnosed with
intra-ocular melanoma among non-latino whites in 1972-82 in Los Angeles County. The
proportional incidence ratio was not higher for cases born in California and Arizona than for
those born in more northerly areas.
Doll (1991) observed a small rural excess in the incidence of cancer of the eye compared
with urban residence, in a number of countries.
(iii) Occupation
Four studies of occupational mortality and one of incidence gave inconsistent results
with regard to ocular cancer. Two investigations using proportional mortality ratios
demonstrated more deaths from ocular cancer than expected among male farmers (Saftlas et
al., 1987; Gallagher, 1988), a group likely to have substantial exposure to solar UVR.
These findings were not confirmed, however, in two other studies using similar methods
(Milham, 1983; Office of Population Censuses and Surveys, 1986).
An investigation of ocular melanoma carried out on data from the cancer registry of
England and Wales did not show an elevated incidence in farmers, but an increased risk was
seen for professionals (relative risk, 124; 95% CI, 99-153), which was significant for
teachers (177; 120-248) (Vagero et al., 1990).
(iv) History of skin cancer
Cancer registry-based studies (0sterlind et al., 1985; Tucker et al., 1985a; Holly et al.,
1991) found no or a nonsignificant (Lischko et al., 1989) association between the
occurrence of cancer of the eye and cutaneous melanoma or nonmelanocytic skin cancer. A
single investigation of 400 sequential cases of uveal melanoma (Tumer et al., 1989)
suggested that intra-ocular melanoma patients have an elevated frequency of prior cutaneous
melanoma. Thus, although one study indicated a possible association, the overall evidence
does not support an association between ocular melanoma and either melanoma or
nonmelanocytic skin cancer.
( c ) Case-control studies
Four case-control studies were evaluated. The first study (Gallagher et al., 1985)
evaluated all ocular melanomas, while the other three (Tucker et al., 1985b; Holly et al. ,
1990; Seddon et al., 1990) studied uveal melanomas (excluding conjunctival melanomas).
Gallagher et al. (1985) conducted a study of ocular melanoma in patients diagnosed in
the four westem provinces in Canada between 1 April 1979 and 31 March 1981. Of the 90
ascertained cases, 87 were eligible by age for interview (20-79 years); of these, 65 cases
(75%) were actually interviewed. For each case, a single control was randomly selected
from the general population, matched by age ( 2 years), sex and province of residence.
Response rates for controls were 59N, for Alberta, Saskatchewan and Manitoba and 48%
for British Columbia. Personal interviews were conducted in subjects' homes, and
conditional logistic regression was used to control for matching variables and eye, hair and
skin colour. No significant association was seen between ocular melanoma and either
intermittent (occupational, recreational and holiday) or cumulative exposure to solar UVR.
A strong association was detected between ocular melanoma and blue or grey iris colour
(crude odds ratio, 3.0; p = 0.04) and blond or red hair colour (crude odds ratio, 7.7; p =
0.03). (In a multivariate analysis, these odds ratios became nonsignificant.) A
nonsignificantly elevated risk (crude odds ratio, 2.8; p = 0.08) for ocular melanoma was also
seen for subjects with light skin colour by comparison with subjects with darker skin.
A case-control study conducted by Tucker et al. ( 1985b) evaluated risk factors in 444
white patients with intra-ocular (uveal) melanoma treated at the Wills Eye Hospital in
Philadelphia, USA, and 424 controls with detached retinas seen at the same centre. [The
Working Group noted that use of a single disease category for the controls could introduce
spurious associations with risk factors for that condition.] Response rates were 89% for
cases and 85 % for controls. Interviews were conducted by telephone; interviews were with
next-of-kin for 17% of the cases and 14% of the controls. Logistic regression models were
fitted which included sun-exposure variables, age, sex, eye colour and presence of cataracts,
which was included to reduce bias in view of the association between cataracts and detached
retina. Sunbathing appeared to increase the risk of intra-ocular melanoma, although no
gradient of risk was noted with frequency of exposure (frequent versus never, odds ratio,
1.5; 95% Cl, 0.9-2.3). A significantly elevated risk was detected for those who engaged in
gardening (1.6; 1.0-2.4), but similar associations were not seen for other recreational
outdoor activities, such as fishing, camping and hunting. Cases of intra-ocular melanoma
also reported increased exposure to the sun during vacations in comparison with controls'
with an odds ratio of 1.5 (95% CI, 0.97-2.3) for subjects 'frequently' experiencing increased
exposure versus subjects never exposed (test for linear trend over four strata, p = 0.01).
Cases reported less frequent use of eye protection (sunglasses, headgear, visors) when
outdoors as compared with controls, but there was no dose-response relationship with
frequency of use of these protective devices. A gradient of risk was seen with use of any
eye shading when iris melanomas were examined separately, suggesting that eye shading
may have been specifically important for lesions at the front of the eye (never versus
occasional use of eye protection, odds ratio, 4.9; 95% CI, 1.4-13.7). [Numbers of iris
melanomas were not given.] Subjects who were born in the southern USA (lower than 40
0
N latitude) were found to have a significantly elevated risk of intra-ocular melanoma (2.7;
1.3-5.9) after adjustment for number of years spent in the south and for the presence of
cataracts; with adjustment for all other sun-related variables, the odds ratio was 3.2 (95%
CI, 1.8-5.7). The association persisted after excluding subjects not living close to
Philadelphia. There was no relation between the number of years spent in the south and the
risk of intraocular malignant melanoma, after adjustment for having been born in the south.
Blue-eyed subjects had the highest risk of intra-ocular melanoma, with gray-green and
hazel-eyed subjects at intermediate risk, and brown-eyed subjects at lowest risk (unadjusted
odds ratio for brown- versus blue-eyed subjects, 0.6; 95 % CI, 0.4-0.8). Cases were more
likely than controls to have fair skin and blond or brown hair, although no odds ratios are
given and the differences disappeared when eye colour was taken into account. Cases were
also more likely to have 25 or more freckles (used as an indirect measure of sun exposure
and sensitivity) than controls (odds ratio, 1.4; 95% CI, 1.0-2.0).
A case-control study by Holly et al. (1990) involved 407 white cases of uveal melanoma
and 870 controls. The cases were diagnosed between January 1978 and February 1987 at
the for Ocular Oncology Unit of the University of California, San Francisco, USA, were
aged 20-74 at diagnosis and lived in 11 western states. Controls were selected by random
digit dialling and were matched to cases on age and area of residence. Telephone interviews
were conducted by interviewers unaware of the study hypotheses, most cases being
interviewed within four years of their diagnosis. The response rate was 93% of cases and
77% of eligible controls. No clear association was seen between uveal melanoma and
vacation time spent in sunny climates or high proportion of leisure time spent outdoors.
Individuals who spent 50% of their leisure time indoors and 50% outdoors had a reduced
risk for uveal melanoma (odds ratio, 0.6; 95% Cl, 0.4-0.9) when compared to subjects who
stayed mainly indoors. Significantly elevated risks were seen in subjects with grey, green,
hazel or blue eyes, compared to those with brown eyes, with increasing frequency of large
naevi (27 mm) (p = 0.04 for trend) and with a propensity to burn rather than tan in the sun.
Seddon et al. (1990) compared 197 white patients with uveal melanoma diagnosed in
1984-87, who were resident in the six New England states close to the Massachusetts Eye
and Ear Infirmary, with 385 controls obtained through random digit dialling and matched to
cases by age ( 8 years), sex and area of residence. All subjects were interviewed by
telephone using a standard questionnaire. The response rate was 92% among cases, and
85% of the eligible controls contacted agreed to participate in the study. Matched logistic
regression techniques were employed to evaluate potential associations between exposme to
UVR and risk of uveal melanoma, adjusting for age, sex, constitutional factors and socio-
economic variables. An inverse association with southern birthplace (south of 40 N
latitude) was detected (odds ratio, 0.2; 95% CI, 0.0-0.7) after adjustment for constitutional
and other factors. When cumulative lifetime residence in the south was examined, subjects
who had lived for more than five years south of 40 N had an odds ratio of 2.8 (95% CI,
1.1-6.9) after adjustment for birthplace. Several indices of sun exposure were computed for
each subject. The first combined duration of residence in the north or south with self-
reported severity of sun exposure (low, medium, high). Subjects in the highest exposure
group appeared to have a higher risk of uveal melanoma by comparison with those in the
lowest exposure category (1.7; 0.9-3.0) although no dose-response relationship was seen
over the three categories of exposure. A further index was obtained by taking average
values of solar radiation for each state in which the subject has resided and multiplying this
value by the duration of residence within the state and the reported amount of time spent in
the sun. No association was seen between this index and risk of uveal melanoma.
Individuals who reported having spent a great deal of time working outdoors 15 years prior
to diagnosis showed a somewhat lower risk of uveal melanoma than those who worked
minimally outdoors or were retired (odds ratio, 0.6; 95% CI, 0.3-1.4) after control for age,
skin, eye colour and southern residence. No association was seen with sunbathing, use of
sunglasses or visors, or outdoor hobbies all conducted 15 years prior to diagnosis. Use of
eye glasses was not related to uveal melanoma risk. Cases reported more cutaneous naevi
and lighter skin colour than controls and were more likely to be of northern European or
British ancestry than controls. An expanded analysis comparing 387 cases of uveal
melanoma with 800 sibling controls was also conducted. There was a gradient of risk with
cumulative years of intense sun exposure; the odds ratio for the highest exposure was 2.1
(1.4-3.2).
2.1.5 Other cancers
No adequate data were available to the Working Group.
2.2 Artificial sources of ultraviolet radiation
Epidemiological investigations that have attempted to assess exposure to artificial
sources of UVR have neither measured actual UVR nor considered the emission spectra. It
is presumed that in the studies described below, subjects were exposed to sources that varied
in intensity and emission spectra.
2.2.1 Nonmelanocytic skin cancer
Three case-control studies, described in detail on p. 84, addressed this issue. In the study
in Montreal, Canada, of Aubry and MacGibbon (1985), any use of a sunlamp gave an odds
ratio of 13.4 [95% CI, 1.4-130.5] after adjustment for sun exposure and constitutional
factors. O'Loughlin et al. (1985) in Ireland found that fewer cases than controls reported
frequent exposure to 'artificial sunlight' (nonsignificant). In the study of Herity et al.
(1989) in Ireland, a smaller proportion of cases than of controls reported ever having used
sunlamps or sunbeds (p = 0.2).
2.2.2 Malignant melanoma of the skin
1
The results of case-control studies of exposure to fluorescent light and melanoma are
summarized in Table 26.
Beral et al. (1982) conducted a case-control study in Sydney, Australia, of 274 female
cases aged 18-54 identified at a melanoma clinic between 1978 and 1980 and 549 hospital
and population controls matched by age and, for population controls, residence. The res-
ponse rate for cases was 71% [response rates for controls not given]. Each job lasting 12
months or longer was recorded, together with information about whether the work had been
carried out predominantly indoors or outdoors, whether fluorescent lighting was present,
and whether the fluorescent lights were switched on most of the time or less frequently.
Among women who always worked indoors, the odds ratio increased with duration of
working with fluorescent lights most of the time to a maximum of 2.6 (95% CI, 1.2-5.9) for
20 or more years' exposure. The effect was greater for office workers (odds ratio, 4.3) than
for other indoor workers (2.0). Stratification by amount of time spent outdoors, main
outdoor activity and amount of clothing worn, history of sunburn, place of birth, hair colour
and skin colour did not diminish the association. Among cases exposed to fluorescent
lights, there was a relative excess of melanomas on the trunk (a site likely to be covered at
work); 24% in exposed cases versus 4% in unexposed cases. [The Working Group noted
that crude estimates of sun exposure were used.]
Rigel et al. (1983) conducted a case-control study in New York, USA, described on p.
106. Cases had had shorter average daily exposure to fluorescent lights (4.9 h) than had
1
After the meeting, the Secretariat became aware of a study by Walker et al. (1992) on the risk of cutaneous
malignant melanoma associated with exposure to fluorescent light.
controls (5.4 h). Among office workers, average daily exposures were similar for cases and
controls. The crude odds ratio for any exposure was 0.7 among all subjects and 0.6 among
office workers.
English et al. (1985) conducted a studyin 1980-81 of the exposure to fluorescent light of
337 cases and 349 age-matched controls who had already participated in a population-based
case-control study in Western Australia (see Holman and Armstrong (1984a), p. 100). The
response rate was 68% for cases and 91% for controls. Detailed information was obtained
from telephone interviews about lifetime hours of residential and occupational exposure, the
distance to the nearest light fixture and the presence of diffusers. Neither the duration of
occupational exposure, the rate of total exposure (hours/year) nor cumulative total exposure
was associated with risk for melanoma. Analyses by body site showed no consistent
association with exposure to lights without diffusers. Adjustment for measures of total and
intermittent exposure to the sun did not alter the results. Subjects were also asked about
exposure to plan printers, laboratory equipment emitting UVR, insect tubes, black lights and
photocopiers. No association was seen with any of these sources, although the number of
exposed subjects was small. The odds ratio for any use of sunlamps was 1.1 (95% CI,
0.6-1.8), although few subjects had used sunlamps (Holman et al., 1986b ).
Sorahan and Grimley (1985) examined fluorescent light exposure in 1980-82 in a case-
control study in the United Kingdom, described in detail on p.1 03. Information on exposure
was confined to whether lights were 'mainly on' or 'sometimes on' at work. After
adjustment for age and sex, no consistent association was seen for duration of exposure
when cases were compared with electoral register controls.
Dubin et al. (1986) examined fluorescent light exposure in a subset of subjects in a case-
control study in New York, USA, described on p. 108. Subjects were interviewed and/or
sent postal questionnaires. In data obtained from interview, but not in data obtained from
postal questionnaires, the odds ratios increased with average daily exposure in the five years
before interview, after adjustment for age and sex (p value for linear trend, < 0.05). A
similar pattern was seen for exposure 6-11 years and 11-20 years previously.
Elwood et al. (1986) examined fluorescent light exposure in their case-control study in
the United Kingdom in 1981-84, described in detail on p. 103. Subjects were interviewed
and later sent postal questionnaires to validate the responses. From the interview data,
exposure to undiffused lights at work was associated with an odds ratio of 4.0 (95% CI,
0.8-19 .2) for those maximally exposed (p value for trend = 0.2). Control for constitutional
factors did not change the results. From the questionnaire data, the odds ratio for maximal
exposure (undiffused lights) was 1.9 (95% CI, 0.4-8.4). No association was seen with
exposure at home, and no association was seen for use of sunlamps. Subjects were also
asked about exposure to particular or unusual light sources, such as vacuum or discharge
lamps, insecticidal or germicidal lamps or welding equipment. The odds ratio for exposure
to any such source was 2.2 (95% CI, 1.0-4.9). [The Working Group noted that the use of
open-ended questions about lighting sources may have introduced recall bias.]
In the Western Canada case-control study in 1979-81 (see Elwood et al .. 1984, 1985a,b,
p. 1 07), no association was seen with use of sunlamps Cx2 = 6.1 ,5 df) (Gallagher et al., 1986).
0sterlind et al. (1988b) examined exposure to fluorescent lighting at work and use of
sunlamps and sunbeds in their case-control study in Denmark in 1982-85, described on pp.
103-104. The same proportions of cases and controls reported having been exposed to
fluorescent lights at work, and no association was seen with age at first exposure, duration
of exposure or type of work place. Past use of sunlamps was also not associated with
melanoma, and a smaller proportion of cases than controls had ever used sun beds (odds
ratio, 0.7: 95% CT. 0.5-1.0).
In a case-control study in Scotland (Swerdlow et al., 1988), 180 cases aged 15-84 from
three clinics during 1979-84 were compared with 197 age- and hospital-matched patients
with various non-malignant diseases. Subjects were interviewed about exposure to fluo-
rescent lights and UV lamps, use of sunbeds, sun exposure and constitutional factors.
Controls with skin conditions were excluded from the analysis of UV lamps and sunbeds.
No consistent association was seen with exposure to fluorescent lights at home or at work,
with or without adjustment for constitutional factors and sun exposure. Significant, positive
associations were seen for duration of use of UV lamps and sunbeds (p value for trend, <
0.05). The odds ratio for use for more than one year was 3.4 (95% CI, 0.6-20.3) after
adjustment for constitutional factors and sun exposure. Amount of use within five years
(1.9; 0.6-5.6) of the interview and more than five years (9.1; 2.0-40.6) before the interview
were both positively associated with the risk for melanoma.
MacKie et al. (1989) examined use of sunbeds and sunlamps in their case-control study
in Scotland described on p. 106. Use was associated with melanoma in men (odds ratio, 2.6;
95% CI, 0.9-7.3) but showed little association in women (1.5; 0.8-2.9). The effect on men
largely disappeared after adjustment for sun exposure and constitutional factors.
In the study of Zanetti et al. (1988) from Turin, Italy, described in detail on p. 104, an
odds ratio of 0.9 (0.4-2.0) was found for use of UV A lamps, although few subjects reported
exposure.
A large population-based case-control study on occupational exposures was conducted
during 1979-85 in Montreal, Canada (Siemiatycki, 1991). Overall, there were 3730 male
cases of cancer aged 35-70, including 124 cutaneous melanoma cases; the participation rate
was 82%. Each cancer site was compared with the other cancer sites. Exposure to 293
agents, including arc welding fumes and UVR, was assessed by a team of chemists and
industrial hygienists on the basis of each individual's occupational history. Neither arc
welding fumes nor exposures to UVR was associated with the risk for cutaneous melanoma
(odds ratios, 0.5; 90% CI, 0.3-l.l and 0.3; 0.1-1.5, respectively).
In a population-based study in southern Ontario, Canada (Walter et al., 1990), 583 cases
identified from pathology laboratories and from the cancer registry between 1984 and 1986
were compared with 608 controls randomly sampled from property tax rolls. Participation
rates were 90% for cases and 80% for controls. Odds ratios for any use of sunbeds or
sunlamps were 1.9 (95% CI, 1.2-3.0) in men and 1.5 (0.99-2. 1) in women. Adjustment for
constitutional factors did not affect the results. The odds ratios increased with duration of
use; for more than 12 months' use, the odds ratios were 2.1 (0.9-5.3) in men and 3.0
(1.1-9.6) in women.
Table 26. Case-control studies of melanoma of the skin and exposure to fluorescent lights
Countrl Cases/ controls Odds ratio 95% CI Definition of exEosure Reference
Australia 274/549
2.6a,6
1.2-5.9 Indoor workers, 2: 20 years' occupational exposure Beral et al. (1982)
4.3a.b
NR Office workers, 2: 20 years' occupational exposure
USA 114/228 0.7 NS Any exposure Rigel et al. (1983)
0.6 NS Any exposure, office workers
Australia 337/349
1.
2
a,b
0.8-1.9 2: 35 000 h exposure English et al.
1.2a,b
0.7-1.9 2: 1600 h per year (1985)
1.3a,b
0.8-1.9 2: 22 500 h undiffused lights
1.2a,b
0.8-1.9 2: 1300 h per year undiffused lights
1.2a.b
0.6-2.6 2: 22 500 h head, neck, upper limbs, undiffused lights
United 58/333 0.6" NR 2: 20 years, occupational exposure (mainly on) Sorahan &
Kingdom 0.5
3
NR 2: 20 years, indoor workers only (maily on) Grimley ( 1985)
USA 1103/585 2.3
3
1.0-5.8 2: 9 h per day, 0-5 years previously (interview) Dubin et al.
508/222 0.6
11
0.3-1.3 2: 9 h per day, 0-5 years previously (postal (1986)
questionnaire)
United 83/83 1
.4a,b
0.4-5.1 2: 50 000 h occupational exposure (total fluorescent Elwood et al.
Kingdom
4.0a.b
0.8-19.2 light, postal questionnaire) (1986)
2: 50 000 h occupational exposure (undiffused lights,
interview)
67/66
1.2a.b
0.3-5.7 2: 50 000 h occupational exposure (total fluorescent
1.9a,b
0.4-8.4 light, postal questionnaire)
2: 50 000 h occupational exposure (undiffused lights,
postal questionnaire)
Denmark 474/926 No association Duration of exposure, age at first exposure, type of 0sterlind et al.
workplace (1988b)
Scotland 180/ 197
1.2b
0.7-1.9 Any occupational exposure < 5 years previously Swerdlow et al.
United
0.8b 0.4-1.4 Any exposure at home < 5 years previously (1988)
Kingdom
1.6b
0.9-2.6 2: 5 h per day < 5 years previously at work and at
1.4b
0.9-2.3 home
0.8b 0.4-1.4 Any occupational exposure > 5 years previously
Any residential exposure > 5 years previously
NR, not reported; NS, not significant
"Odds ratio for category with highest level of exposure
b Adjusted for sun exposure
2.2.3 Malignant melanoma of the eye
In the case-control study carried out in Philadelphia, USA, which is described in detail
on p. 128, cases of uveal melanoma were more likely to report use of sunlamps than
controls. After adjustment for age, eye colour and a history of cataracts, there was a trend to
increasing risk with frequency of use (odds ratio for frequent veesus never, 2.1; 95% CI,
0.3-17.9; test for linear trend over four levels: p = 0.10). The odds ratios for those who had
ever worked as welders was 10.9 (2.1-56.5) (Tucker et al., 1985b ).
In the case-control study from San Francisco, USA, described on pp. 128-129, exposure
to artificial UV light or 'black light' [details not given] conferred over three-fold risks for
intra-ocular melanoma after adjustment for other significant factors (odds ratio, 3.7; 95%
CI, 1.6-8.7). The odds ratios were 2.9 for 1-5 years of exposure and 3.8 for 6 or more years
(Holly et al., 1990).
In the case-control study from Boston, USA (Seddon et al., 1990), described on p. 129,
exposure to fluorescent lighting was associated with an elevated risk of uveal melanoma
(odds ratio, 1.7; 95% CI, 1.1-2.5 for 40 h or more per week as compared to no exposure) in
the larger data set, based on case-sibling comparison. In the population-based comparison,
the corresponding odds ratio was 1.2 (95% CI, 0.6-2.1 ). A history of working with welding
arcs was reported with similar frequency among cases and controls in both comparisons.
Cases reported more frequent use of sunlamps in comparison with both sets of controls.
After adjustment for constitutional factors and exposure to the sun, the odds ratios for
frequent/occasional use versus never were 3.4 (1.1-10.3) in the population comparison and
2.3 (1.2-4.3) in the sibling comparison.
In the large Canadian study on occupational exposure, described on p. 132, 23 cases of
ocular melanoma were included. Analysis only of French Canadians revealed four cases of
eye melanoma with exposure to arc welding fumes (odds ratio, 8.3; 90% Cl, 2.5-27.10)
(Siemiatycki, 1991). No increase was found for substantial exposure; no increase in risk
was reported for exposure to UYR.
2.3 Premalignant conditions
2.3.1 Basal-cell naevus syndrome
Basal-cell naevus syndrome is a hereditary condition (Godin, 1987) in which affected
family members may show, among other major manifestations, an apparent excess of basal-
cell carcinomas. These seem to occur more commonly in sun-exposed parts of the body or
in unusual patterns. There is no other evidence that solar radiation plays a role in their
development.
2.3 .2 Dysplastic navvus syndrome
Dysplastic naevus syndrome is a hereditary condition in which affected family members
have multiple dysplastic naevi and a greatly increased risk of malignant melanoma (Green et
al., 1985b ). The distribution of tumours conforms to the usual distribution, and there is
anecdotal evidence that solar radiation plays a role in their development (Kraemer Greene,
1985).
2.4 Molecular genetics of human skin cancers
Analysis of mutations in DNA isolated from tumours and believed to be relevant to
carcinogenesis can potentially help in making a causal link with exposures to carcinogens.
Two important qualifications must, however, be borne in mind. Firstly, the changes
detected may have arisen late in tumour development (whether or not the tumour is the
result of exposure to UVR) and may not be involved in initiation or other early steps.
Secondly, the spectrum of mutations that is seen may be constrained to those changes that
can lead to a functional gene product. This qualification applies, for example, to mutations
that activate ras genes but to only a lesser extent to tumour suppressor gene mutations in
which inactivation of gene function is involved.
Experimental studies indicate that UV -induced mutations have a distinctive pattern of
base-substitution mutations (see section 4.5):
Virtually all mutations occur at dipyrimidine sites, especially 5'TC and 5'CC
sequences.
The majority of the base substitution mutations involve cytosille with the C IT
transition predominating.
Tandem 5'CC----+5'TT mutations occur.
2.4.1 ras Gene mutations
Primary melanomas, metastases and cell lines derived from melanomas which developed
at body sites characterized as exposed 'rarely', 'intermittently' or 'continuously' to the sun
were analysed for the presence of N-ras mutations. Of 37 cutaneous melanomas, seven had
N-ras mutations; all were from 'continuously' exposed sites. All mutations in the N-ras
gene were at TT or CC sites, which are potential locations for mutagenic UV photoproducts,
suggesting a role of sun exposure in N-ras mutation (van't Veer et al., 1989).
In several investigations, base-substitution mutations were found in Ha-, Ki- and N-ras
genes in human skin melanomas (Table 27) and in squamous-cell and basal-cell carcinomas
(Table 28) from xeroderma pigmentosum and normal patients. In single studies, Ha- and
N-ras gene amplification was found in squamous-cell carcinomas of the skin
(Ananthaswamy & Pierceall, 1990), and loss of the Ha-ras allele was seen in basal-cell and
squamous-cell carcinomas (Ananthaswamy et al., 1988). Whether exposure to the sun was
involved in tumour induction in these studies is, however, less clear.
2.4.2 p53 Gene mutations
Brash et al. (1991) found p53 mutations at various codons in 14 out of 24 (58%) invasive
squamous-cell carcinomas from sun-exposed skin (Table 29). The mutations found were
predominantly C----+T (5 of 14 total mutants, 36%) and CC----+TT (3 of 14, 21 %) transitions
exclusively at tandem pyrimidine stretches. This finding is consistent with the hypothesis
that these mutations are induced by UV irradiation. CC----+ TT double-base changes in the
p53 gene have not yet been found in turnouts in any internal organ. These results strongly
suggest that solar radiation plays a role in the induction of p53 gene mutations.
Pierceall et al. (1991) found p53 mutations in exon 7 in 2 out of 10 squamous-cell
carcinomas from sun-exposed body sites; one was a C----+ T transition and the other a C----+A
transversion.
Table 27. ras Gene mutations detected in human naevi and primary and secondary melanomas that developed at sites
subject to sun exposure
Oncogene Base change Base-substitution Site of original tumour Reference
codon mutation
N-ras-61 GGACAAGAA
AAA C to A Neck van 't Veer et at. (1989)
AAA CtoA Lower leg van't Veer et al. (1989)
AAA C to A Nose van 't Veer et al. ( 1989)
AAA C to A Cheek van't Veer et al. (1989)
CGA [T to C] Lower leg van't Veer et al. ( 1989)
CAT [T to A/G) Xeroderma pigmentosum patiene Keijzer et al. ( 1989)
CAT [T to A] Site unspecified, probably metastasis Sekiya et at. (1984)
N-ras-13 GGT GGT GIT
GAT [C to T] Finger van't Veer et al. (1989)
GTT [C to A] Finger van 't Veer et al. (1989)
GTT [C to A] Lower leg van't Veer et al. (1989)
N-ras-12 GAT [C toT) Leg van't Veer et al. (1989)
N-ras-61 CAT/C [T to A/G] Back Shukla eta!. ( 1989)
Ki-ras-61 GGACAAGAA
AAA CtoA Lower leg Shukla et at. ( 1989)
Ki-ras-12 GCT GGT GGC
TGT [C to A] Abdomen Shukla et at. ( 1989)
TGT [C to A) Knee Shukla et al. (1989)
TGT [C to A] Site unspecified, probably metastasis Shukla et al. ( 1989)
TGT [C to A] Site unspecified, probably metastasis Shukla et al. (1989)
[C toT] Buttock Shukla et al. ( 1989)
[C to T] Site unspecified, probably metastasis Shukla eta!. (1989)
[C toT] Forearm (naevus) Shukla et al. ( 1989)
[C to T) Abdomen (naevus) Shukla et al. ( 1989)
Ha-ras-12 GCCGGCGGT
TGC [C to A] Abdomen Shukla et al. (1989)
Italics indicate potential pyrimidine dimer site including neighbouring condon; [ ], base changes occurring in anti-sense strand
aMalignant melanoma probably resulting from metastasis of a primary skin tumour
Table 28. ras Gene mutations detected in human keratoacanthomas (KA), basal-cell carcinomas (BCC) and
squamous-cell carcinomas (SCC) that developed at sites to sun exposure
Oncogene Base change Base-substitution Tumour Site Reference
condon mutation
Ki-ras 12 GCT GGT GGC
TGT [C to A] sec Lip van der Schroeff et al. ( 1990)
BCC Shoulder van der Schroeff eta/. (1990)
BCC Neck van der Schroeff et al. ( 1990)
GAT [C toT] BCC Face van der Schroeff et al. ( 1990)
Ha-ras 61 GGCCAGGAG
CTG [T to A] sec Not specified Corominas et al. (1989)
CTG [T to A] KA Not specified Corominas et al. ( 1989)
CAT [C to A] BCC Face van der Schroeff et al. (1990)
AAG Cto A KA Not specified Corominas et al. (1989)
Ha-ras 12 GCCGGCGGT
AGC [C to T] sec Not specified Corominas et al. ( 1989)
AGC [C toT] KA Not specified Corominas et al. ( 1989)
AGC [C to T] KA Not specified Corominas et al. (1989)
TGC [C to A] sec Not specified Corominas et al. (1989)
TGC [C to A] sec Not specified Corominas et al. ( 1989)
Italics indicate potential pyrimidine dimer site including neighbouring codon; [ ], base changes occurring in anit-sense strand
Table 29. p53 Tumour suppressor gene mutations in human squamous-cell carcinomas
that developed at sites subject to sun exposure
Codon Nucleotide Base-substitution Incidence a Site of tumour origin Reference
sequence mutation
7 TCT TGT;C-G 1/14/24 Preauricular Brash et al. (1991)
56 T TCA TAA;C-A 1/ 14/24 Chest Brash et al. (1991)
104/105 CG CCT deletion of a C 2/ 14/24 Preauricular/temple Brash et al. (1991)
151 ccccc CAC; c - A 1/14/24 Scalp Brash eta!. (1991)
152 CC CCC CAC;C- T 1/ 14/24 Hand Brash et al. (1991)
179 A CCA CAA;C-A 1/14/24 Scalp Brash et al. (1991)
244 CCGG TCG;C- T 1/2/ 10 Face Pierceall et al. ( 1991)
245 G CCG CAG;C- A 1114/24 Cheek Brash et al. (1991)
245 G CCG r T;cc-rT 1114/24 Chest Brash et al. (1991)
247/248 AC CG r T;cc-Tr 1114/24 Nose Brash et al. (1991)
248 GCC GAC;C- A 1/2/ 10 Face Pierceall et al. (1991)
258 T TCC TTC;C-T 1114/24 Face Brash et al. (1991)
278 T CCT TCT;C-T 1/14/24 Cheek Brash et al. (1991)
285/286 TC CT T T;CC- TT 1114/24 Face Brash et al. (1991)
286 TCCT crr; c - r 1/14/24 Forehead Brash et al. (1991)
317 cc CCA TCA; C-T 1/14/24 Postauricular Brash et al. (1991}
Italics indicate potential pyrimidine dimer site
aNo. of specific mutations/no. of total mutations found/Total number of samples tested only from sites
continuously exposed to the sun
CONTINUED
3. Studies of Cancer in Animals
3.1 Experimental conventions
3.1 .1 Species studied
The experimental induction of skin cancers in mice following exposure to a mercury-arc
lamp was flrst reported by Findlay (1928). Initially, haired albino mice were used, but
hairless and nude mice are now preferred.
An important development was the use of the hairless mouse as a model (Winkelmann et
al., 1960, 1963). In haired animals, the fur provides effective protection of the skin against
UVR. This limits investigations to sparsely haired skin regions, mainly the ears, as, in long-
term experiments with frequent exposures, the mechanical trauma caused by shaving might
influence the process of tumorigenesis. The skin of hairless mice differs, however, from
human skin in many respects. It is, for instance, much thinner and has abnormal hair
follicles. The hairless mouse does, however, have a thymus and a functioning immune
system, in contrast to the nude mouse (Eaton et al. , 1978; Hoover et al., 1987). Many recent
studies on carcinogenesis induced by UVR used the hairless mouse model (Forbes et al.,
1981; de Gruijl et al., 1983; Gallagher et al., l 984b). The changing designations of 'Skin'
mice are listed in Table 30. Skin tumorigenicity has been evaluated experimentally in only
a relatively small number of species other than the mouse.
Table 30. Alternative designations used for 'Skin' outbred stocks of hairless mice
Phenotype 1970-86 After 1986 Synonyms used Inbred strains derived
Pigmentedc
(any colour)
Skh: hairless-!
Skh: hairless-2
aFrom Forbes et al. (1990)
Skh:hr I
Skh:hr II
bForbes et al. (1981); de Gruijl et al. (1983)
cDavies & Forbes (1988)
3.1.2 Wavelength ranges
in the literature from Sk:hr stock
3
Sk-1; Skh-1; Skh/Hr-1; HRA/Skh (Temple Uni-
Skh:HR; HRA/Skh-1; versity, Philadelphia,
Skh-hr 1 P A, USA)
Sk-2; Skh-2; Skh/Hr-2 HRA/Skh-1 (University
of Sydney, Sydney,
Australia)
As noted in section 1.1 , for the purposes of this monograph, the UV wavelength range is
subdivided according to the convention of the Commission Intemationale de l'Eclairage
(1987) into: UVA (315-400 nm), UVB (280-315 nm) and UVC ( l 00-280 nm). The UVB
-139-
range is generally found to be most effective in inducing skin cancer, i.e., tumorigenesis
may be achieved with smaller doses of radiant exposure than with UVA and UVC. A
complete discussion of wavelength ranges is given in section 1.1.
3.1.3 Measured doses
Many investigators of the carcinogenicity of UVR have reported the type of lamps they
used, which are frequently broad-spectrum lamps, sometimes in combination with filters.
When estimates of the doses of UVR adminjstered are given, the measuring instrument is
usually mentioned and the result is given in terms of irradiance or dose, with no further
detail. Such information is of some value, especially for comparing the results of
experiments in which the same type of ramps were used.
The action spectrum (see section 1) given in Figure 10 shows that the carcinogenic
effectiveness of UVR in hairless mice changes steeply, even by orders of magnitude, over a
wavelength range of 10 or 20 nm. This pattern indicates that irradiance must be spectrally
specified in order to be meaningful , and not integrated into one value over a broad spectrum.
One approach is to give irradiance weighted according to the action spectrum for UV
carcinogenesis, but this is available only in provisional form (see Fig. 10 and discussion on
pp. 46-47). Another approach is to provide data on erythemally weighted irradiance, since
the action spectrum for erythema corresponds approximately to that for carcinogenesis
(Forbes et al. , 1978). A simple, direct way of calculating this is to relate the doses admi-
nistered to the minimal elythema dose or to the minimal oedemic dose for the animal being
investigated. When investigators supplied such measures of effect, they are mentioned in
the summaries below.
In experimental situations, there is never a perfectly sharp cut-off of wavelengths. The
expression 'mainly UVA' is of questionable value, because even if UVB represents only
0.1% of the emission spectrum, it may still dominate the effect (see pp. 144-147, 151 and
Fig. 10). Terms such as 'mainly UVB' are used below only when there are good reasons to
assume that the effects considered are due mainly to UVB radiation.
3.1.4 Protocols
Experimental investigations on the carcinogenicity of UVR, conducted mostly on mice,
have been reviewed (Blum, 1959; Urbach et al., 1974; Kripke & Sass, 1978: WHO, 1979;
van der Leun, 1984; Epstein, 1985).
Hundreds of studies have been reported. Most were not designed to test whether or not
the radiation used was carcinogenic per se but to investigate the process of UV carcino-
genesis. The methods used in these studies differ in many respects from those in standard
lifetime studies to evaluate the carcinogenicity of chemicals. For example, many studies do
not give complete details of the UVR emission spectrum used or exposure dose, do not
enumerate all tumours, do not provide data on survival or do not provide histological details
of tumours. Control groups are not always included; however, spontaneous skin tumours
are rare in mice and rats. In many of the studies presented in detail below, appropriate
statistical analyses have been done demonstrating clear dose-related trends in numbers of
tumourbearing animals, number of tumours per animal and/or median time to first tumour.
STUDIES OF CANCER IN ANIMALS 141
Fig. 10. Sterenborg-Slaper action spectrum for ultraviolet-induced skin carcinogenesis
(1.0-mm tumours) in albino hairless mice. Effectiveness is defined as the reciprocal of
the daily dose at each wavelength that leads to tumours of 1-mm diameter in 50% of
animals in 265 days, relative to the corresponding value at the wavelength of maximal
effectiveness. The effectiveness between 340 and 400 nm represents an average value
for that wavelength range.

(/)
(/)
Q)
c
Q)
10-
1
>
u
Q)
10-
2
'-
'+-
Q)
Q)
>

-
10-
3
0

Q)
~

__ .......... -
10-
1 0 ~ ~ ~ ~ ~
250 300
350
400
Wavelength ( nm)
From van der Leun (1987a)
3.2 Broad-spectrum radiation
3.2. 1 Sunlight
In one study by Roffo (1934), 600 rats [sex and strain unspecified] were exposed to solar
radiation (sunlight) at a latitude of 35 S in Buenos Aires, Argentina. The average exposure
was for 5 h per day, with avoidance of the hours around solar noon in the summer. In the
first days, 365 rats died from sunstroke. Of the 235 remaining animals, 165 (70%)
developed tumours. There were 140 tumours of the ear (58% squamous-cell carcinomas;
36% spindlecell sarcomas; 6% carcinosarcomas): 58 eye tumours (tumours of the
conjunctive, 100% spindle-cell sarcomas; tumours of the eyelid, 50% squamous-cell
carcinomas and 50% spindle-cell sarcomas); and 15 other tumours, mainly squamous-cell
carcinomas, at sites including the nose, tail, paw and neck. In complementary experiments
reported in the same paper, groups of animals were exposed either to sunlight filtered
through various colours of glass, to radiation from various types of lamp (quartz mercury,
glass mercury, neon gas and filament lamps) or to short Hertzian wavelengths. Tumours
[types and sites unspecified] were observed in all 150 animals exposed to quartz mercury
lamps; no tumour was induced in any other experimental group. On the basis of this
evidence, the author concluded that the carcinogenicity of sunlight could be attributed to
UVR.
In another report by Roffo ( 1939), 2000 white rats and mice [exact numbers unspecified]
were exposed to sunlight for an average of 5 h per day. After three to six months, benign
neoplasms and, after seven to nine months, malignant neoplasms of the skin of the ear (88%
of all malignant tumours), the forepaw (7.25%), the tail (my) and nose (one tumour) deve-
loped in 600 animals; 25% of the tumours were seen on the eyes. The ear tumours were
diagnosed as squamous-cell carcinomas (58%), spindle-cell sarcomas (36%) and carcinosar-
comas (6%) by detailed histological examination. Similarly, the paw tumours were
diagnosed as squamous-cell carcinomas (42%) and spindle-cell sarcomas (58%); the
tumours of the tail were all squamous-cell carcinomas. The distribution of tumours of the
eye was similar to that in the study of Roffo (1934). [The Working Group considered that
these are exceptional studies which fully document the carcinogenicity of solar radiation in
rats and mice, even though quantitative detail is lacking. The resulting neoplasms are
described and photographically illustrated in exact detail. The Working Group accepted the
weight of evidence contained in these studies as to the carcinogenicity of solar radiation to
rats and mice.]
Domestic and other animals of many species (cows, goats, sheep (reviewed by Emmett,
1973), cats (Dom et al., 1971) and dogs (Madewell et at., 1981; Nikula et al., 1992))
develop skin tumours, and there are good indications that sunlight is involved. The tumours
described generally developed in sparsely haired, light-coloured skin. Cancers of the eye
occur in many species, including dogs, horses, cats, sheep and swine, but are particularly
frequent in cattle (Russell et at., 1956).
3.2.2 Solar-simulated radiation
In several investigations on carcinogenesis by UVR, 'solar-simulated radiation' was
used (Forbes et al., 1982; Staberg et at., 1983a; Young et al., 1990; Menzies et al., 1991).
In one large, particularly informative experiment (Forbes et al., 1982), more than 1000
hairless albino Skh-hrl mice were exposed to solar-simulated radiation from a xenon arc
lamp, with various Alters to make the spectral distribution in the UV region similar to that
of sunlight under various thicknesses of the ozone layer. The exposures lasted for up to 80
weeks. More than 90% of the mice developed skin tumours, predominantly squamous-cell
carcinomas. The time to development of 50% of first tumours was shorter after exposure to
the spectra that included higher irradiance in the wavelength range 290-300 nm. The other
experiments mentioned were more limited and dealt with more specialized aspects of UV
. .
carcmogenests.
3.2.3 Sources emitting UVC, UVB and UVA radiation
Sources emitting radiation in the entire UV wavelength range were used in experiments
on UV carcinogenesis mainly between 1930 and 1960.
(a) Mouse
Grady et at. (1943) exposed 605 strain A mice to broad-spectrum UVR at a wide range
of doses and irradiances (weekly doses, 3.6-43 x 107 ergs/cm
2
[ 40-430 kJ/m
2
]; Blum &
Lippincott, 1942). The investigation dealt primarily with skin tumours (mainly spindle-cell
sarcomas). About 5% of the mice developed tumours of the eye. Histological examination
by Lippincott and Blum (1943) showed that the eye tumours arose mostly in the cornea end
were spindle-cell sarcomas or fibrosarcomas; haemangioendotheliomas were also found.
A particularly large, informative series of investigations was carried out with unfiltered
medium-pressure mercury arc lamps which emitted UVC, UVB and UV A (Blum, 1959).
More than 600 strain A mice were irradiated (daily dose, 0.32-8.6 x I 07 ergs/cm
2
[3-86
kJ/m
2
]) in a series of investigations deal ing with various aspects of UV carcinogenesis; the
dose-effect relationship was addressed particularly. In most of the experiments, more than
90% of mice developed skin tumours, mainly of the ears, the only site for which quantitative
data were given.
(b) Rat
Findlay (1930) exposed six epilated albino rats to broad-spectrum UVR from a mercury-
vapour lamp at a distance of 18 in [46 em] for 1 min three times a week. Rapidly growing
papillomas were reported in one rat. The time required was, however, much longer than in
mice exposed similarly, namely, 21 months as compared to eight months for mice.
Putschar and Holtz (1930) exposed 35 rats [strain unspecified] with very low sponta-
neous turnout incidence to almost continuous irradiation with broad-spectrum UVR from a
quartz mercury lamp for 11 months. They reported regular occurrence of skin tumours,
including papillomas, squamous-cell carcinomas and, occasionally, basal-cell carcinomas.
The tumours were first seen after 27 weeks of exposure.
Huldschinsky (1933) exposed seven white rats to UVR from a solar lamp for 2 h per
day, six days per week for one year or more. Another group of five rats was exposed to a
quartz lamp emitting a predominantly UVC waveband ( < 270 nm). The doses given per
session were about 10 times higher than those used in phototherapy. Spindle-cell sarcomas
of the eye were found in 217 and 5/ 5 rats in each group, respec6vely.
Hueper (1942) reported squamous-cell carcinomas and, rarely, spindle-cell carcinomas
and sarcomas. round-cell carcinomas and basal-cell carcinomas of the skin in 20 rats [strain
unspecified] exposed for up to 10 months to broad-spectrum UVR from a mercury vapour
burner (a Hanovia SuperS Alpine lamp) at a distance of75 em.
In a study by Freeman and Knox (I 964), a group of 78 rats (66 pigmented and 12 un-
pigmented) was exposed to broad-spectrum UVR from mercury lamps at 50 em from the
skin on five days a week for one year; the doses per session corresponded to approximately
1 MED for rat skin. A total of 98 eye tumours developed, with more tumours in pigmented
rats. The tumours arose in the corneal stroma; two-thirds were diagnosed as fibrosarcomas
and one-third as haemangioendotheliomas.
(c) Hamster
Hamsters exposed to an irradiation regimen similar to that described above also
developed eye tumours (Freeman & Knox, 1964). In 19 animals (9 pigmented, 10 un-
pigmented) exposed for one year, haemangioendotheliomas and fibrosarcomas developed in
14 eyes.
(d) Guinea-pig
Guinea-pigs were exposed to the same regimen as described above. None of 17 animals
developed a tumour of the eye (Freeman & Knox, 1964).
3.3 Sources emitting mainly UVB radiation
Many experiments have been carried out with sources emitting mainly UVB radiation, in
which increases in the number of tumour-bearing animals and/or in the number of tumours
per animal were seen (Blum, 1959; Winkelmann et al., 1963; Freeman, 1975; Stenback,
1975a; Daynes et al., 1977; Kripke, 1977; Spikes et al., 1977; Forbes et al., 1981; de Gruijl
et al., 1983; Gallagher et al., 1984b). The most informative studies are described below.
3.3. 1 Mouse
Freeman (1975) studied carcinogenesis induced by chronic exposure to narrow-band
UVB produced by a high-intensity diffraction grating monochromator with a half-power
band-width of 5 nm. Exposure was three times per week to one ear of each haired albino
mouse. Four wavelengths were used, and the doses were determined as the MED. Of a
group of 30 mice exposed to 300 nm (weekly dose, 60 mJ/cm2), 16 developed
squamous-cell carcinomas of the ear. Of a group of 30 mice exposed to 310 nm (weekly
dose, 750 mJ/cm
2
) , 16 survived to 450 days and eight developed five squamous-cell
carcinomas, two fibrosarcomas and one angiosarcoma of the ear. No skin tumour was
observed among 30 mice irradiated with UVR at 290 nm (weekly close, 42 mJ/cm
2
); of five
mice irradiated with 320 nm (weekly dose, 4950 mJ/cm
2
), two developed squamous-cell
carcinomas of the ear.
Two fibrosarcomas and one unspecified tumour of the eye were reported in 24 C3H/
HeN mice bearing 25 skin tumours (mostly fibrosarcomas) after exposure to UVR (168 J/m
2
three times a week) from Westinghouse FS40T12 sunlamps (280-340 nm) (Kripke, 1977).
In the experiment of Forbes et al. ( 1981 ), groups of 24 male and female hairless albino
Skh:HR mice (the changing designations of sources of 'Skin' mice are listed in Table 30),
six to eight weeks old, were irradiated on five days per week with Westinghouse FS40Tl2
sunlamps (see Fig. 9c, p. 64), emitting mainly UVB (with < 1% below 280 nm; two-thirds at
280-320 nm and one-third at > 320 nm). All animals had developed tumours by the end of
the experiment (up to 45 weeks), and a dose-response effect was demonstrated, as assessed
by time to tumours in 50% of animals (Table 31 ). Histological examination showed
tumours of 4 mm or more in diameter to be squamous-cell carcinomas; those of about 1-4
mm formed a continuum from carcinoma in situ to squamous-cell carcinoma, and those less
than 1 mm comprised epidermal hyperplasia and squamous metaplasia tending toward
carcinoma in situ. Less than 1% of tumours were fibrosarcomas.
Six groups of 22-44 male and female Skh-hr 1 hairless albino mice (total, 199), six to
eight weeks of age, were exposed to daily doses ranging from 57 to 1900 J/m
2
of mainly
UVB radiation from Westinghouse FS40TL12 sunlamps; this dose range encompassed a
factor of 33. Most of the animals developed skin tumours, although even the highest daily
dose was sub-erythemic. A clear-cut relationship was shown between daily dose and time
required for 50% of animals to develop skin tumours, which were predominantly
squamous-cell carcinomas (Fig. 11 ). Squamous-cell carcinomas developed in 71 % of the
STUDIES OF CANCER IN ANIMALS
Table 31. Dose-response to ultraviolet radiation
of hairless Skh:HR mice
Daily dose
(J/m2)
420
587
822
1152
1613
2259
Time to 50% tumour
incidence (weeks)
38.6
33.3
29.2
20.0
17.6
12.9
From Forbes et a! ( 1981)
Terminated
at week
45
45
45
36
36
25
145
Fig. 11. Dose-effect relationship for the induction of< 1-mm skin tumours in hairless
mice by exposure to UVB radiation over a wide range of daily doses; tm, median
induction time
1000 i - --- - ---
800 r
soo r
r
t::;,.._
,......_
(II
300 I
>-
0
"0
,_
100 :
E
.....,
-
80 ~
I
50 I_
' I
-
L . ... _ ___, . ~ 1 ~ . ~ L _ _ _ ... l-.. . --L..-..1.. . . l .1...-. L.l-
3 5 10 30 50 80100
Dose (as % of maximal dose)
From de Gruijl eta/. (1983)
mice in the lowest dose group, and two skin tumours were reported in a total of 24
nonirradiated control mice (de Gruijl et al., 1983).
In albino hairless Skh:Hr-1 mice irradiated with UVB or UVB plus UV A radiation three
times a week for 16 weeks, with a 17-week recovery period, the spectrum for UV tumori-
genesis was sharp and had a maximum near 300 nm (Bissett et al., 1989).
3.3.2 Rat
Skin tumour induction was studied in a group of 40 shaven female NMR rats,8-10weeks
old at the start of the experiment. The animals were irradiated chronically at a distance of
37.5 em for 60 weeks with Westinghouse FS40Tl2 sunlamps (Fig. 9c), emitting mainly
UVB (weekly dose, 5.4-10.8 x 104 J/m
2
). A total of25 skin tumours, most of which were
papillomas of the ears, developed in 16/40 animals (Stenback, 1975a).
3.3.3 Hamster
Stenback (1975a) inadiated 40 shaven female Syrian golden hamsters, 8-10 weeks of
age, using the same protocol as described above. A total of 30 skin tumours developed in
14/40 animals; 22 were papillomas (14 animals), four were keratoacanthomas (three
animals), one was a squamous-cell carcinoma of the skin and three were papillomas of the
ear (one animal).
3.3.4 Guinea-pig
Stenback (1975a) exposed guinea-pigs using the same protocol as above and found skin
tumours in 2/25 animals (a fibroma and a trichofolliculoma).
3.3.5 Fish
Two hybrid fish strains susceptible to melanocytic neoplasms by UVR were developed
by Setlow et al. (1989) by crossing platyfish and swordtails. A group of 460 fish were
exposed to mainly UVB radiation from Westinghouse FS40 sunlamps, filtered with acetate
sheets transmitting > 290 nm or > 304 nm at various doses ( 150 and 300 J/m
2
per day for >
290 nm; 850 and 1700 J/m
2
per day for > 304 nm) for 1-20 consecutive days. There were
103 controls. Depending on the wavelength, the level, the number of days of exposure and
the strain, 19-40% of the irradiated fish developed melanocytic tumours; 13 and 2% of the
controls in the two strains, respectively, developed such tumours.
3.3.6 Opossum
Monodelphis domestics, a South American opossum, is unusual in showing the pheno-
menon of photoreactivation (see Glossary) of pyrimidine dimers and erythema (Ley, 1985);
it also developed actinic keratoses and skin tumours (mainly fibrosarcomas and
squamous-cell carcinomas) on exposure to UVR from an FS-40 sunlamp (280-400 nm) (toy
et al., 1987). Animals were shaved regularly and exposed to mainly UVB radiation from
Westinghouse FS40 sunlamps, with relative emissions of 0.04, 0.27, 0.69, 1.0 and 0.09 at a
dose of 250 J/m
2
(which is approximately half of an average MED; see Fig. 9c) at 280, 290,
300, 313 and 360 run, respectively. Eight of 13 animals developed localized melanocytic
hyperplasia; 100 weeks after the start of the experiment, melanomas were found in 5/ 13
surviving animals. M. domestica do not develop spontaneous melanomas, as was apparent in
a much larger colony not exposed to UVR. Exposure of another group to photoreactivating
light after UV inadiation reduced the incidence of melanocytic hyperplasia (3/ 1 7); this was
considered to be a precursor lesion of the melanomas, although photoreactivation could not
be demonstrated in the melanoma (Ley et al., 1989).
[The Working Group noted that the melanocytic lesions induced in fish and the South
American opossum differ histologically from human melanoma: they grow to a larger size
and do not metastasize readily.]
Ley et al. (1991) exposed groups of M. domestic a to UVR from fluorescent sunlamps
(Westinghouse FS40; 280-400 nm with a peak at 313 nm) three times a week for 70 weeks
at a dose of 250 J/m
2
. Besides skin tumours, tumours of the anterior eye were observed
beginning 30 weeks after the start of exposure. At 69 weeks, 50% of the animals had eye
tumours, which were classified as fibrosarcomas of the corneal stroma. In animals exposed
to UVR followed immediately by photoreactivating light, tumours appeared later and in
reduced numbers.
'Cancer eye' in cattle, which includes squamous-cell carcinoma of the eye and the
circumocular skin, is thought to be caused by solar UVR. In an attempt to confirm this
relationship experimentally (Kopecky et al., 1979), four Hereford cattle (which lack pigment
around the eyes) were exposed to UVB radiation from Westinghouse FS40 lamps. Three
cows developed grossly observable tumours of the eye, one of which was
histopathologically confirmed as a preneoplastic growth.
3.4 Sources emitting mainly UVC radiation
3.4.1 Mouse
Carcinogenicity studies have been performed mainly in mice, but no study is available in
which animals were exposed solely to UVC radiation. Several studies have been reported in
which the source of UVC radiation was low-pressure mercury discharge germicidal lamps,
which emit 90-95% of their radiation at wavelength 254 nm and weaker spectral lines in the
UVB, UVA and visible light regions (Rusch et al., 1941; Blum & Lippincott, 1942; Forbes
& Urbach, 1975; Lill, 1983; Joshi et at., 1984; Sterenborg et al., 1988). In all of these
investigations, the exposures induced tumours. Two of the most informative studies are
described in more detail below.
A group of 40 female C3H/HeNCrlBr mice were irradiated with these lamps at a weekly
dose of 3 x 10
4
J/m
2
. Three animals died without tumours after 9, 43 and 63 weeks of
irradiation; all of the other animals had tumours. By 52 weeks, 97% of the animals had
developed skin tumours, with a median time to appearance of 43 weeks. The mean number
of tumours per tumour-bearing mouse was 2.9. Tumour histology was carried out in 29/37
mice. Of a total of 83 suspected tumours, 66 were squamous-cell carcinomas, 10 were
proliferative squamous lesions and 6 were invasive fibrosarcomas; one had the appearance
of a cystic dilatation (Lill, 1983). [The Working Group that resulted in !ARC Monographs
volume 40 (IARC, 1986a) noted that the 4% UVB content of the source, representing a
weekly dose of 1170 J/m
2
, could not be excluded as contributing to the induction of skin
tumours.]
Sterenborg et al. (1988) presented evidence that the tumours they induced in albino
hairless mice were indeed due to UVC radiation. Groups of 24 male and female hairless
albino mice (Skh-hrl), 6-10 weeks of age, were exposed to UVC radiation from Philips
germicidal TUV 40W low-pressure mercury discharge lamps (mainly 254 nm) on seven
days a week for 75 min per day at 230, 1460 or 7000 J/m
2
(30 times the MED); this dose
was 60% less during the first seven days of the experiment. A total of 65 squamous-cell
carcinomas of the skin were found [number of animals with tumours not specified]. Both the
percentage of tumour-bearing animals and the number of tumours per mouse were strongly
dose-related. By comparing their results with those of experiments with UVB, the
investigators concluded that (i) the UVB emitted by the low-pressure mercury discharge
lamps was insufficient to account for the induction of tumours at the rate found, as at least
850 days of exposure to the UVB radiation present would be required to induce skin
tumours at the rate observed, as compared to 161 days with the low-pressure mercury
discharge lamp used; (ii) there is a qualitative difference between the effects of low-pressure
mercury discharge and UVB lamps, in that the tumours induced by the mercury discharge
lamps were scattered more widely over the skin of the mice than in the experiments with
UVB; and (iii) the dose-effect relationship for tumorigenesis was less steep with the
mercury discharge lamps than with UVB sources. [The Working Group noted that the
evidence given to exclude UVB as contributing to the induction of skin tumours does not
obviate the possibility that some interaction between UVC and UVB radiation led to turnout
induction.]
3.4.2 Rat
Nine groups of 6 or 12 male CD-1 rats, 28 days of age, were shaved and exposed to
varying doses of UVC from Westinghouse G36T6L sterilamps emitting predominantly 254
rim (dose range, 0.08-26.0 x 104 J/m
2
) . Survival ranged from 75 to 92% for the nine
experimental groups. Keratoacanthoma-like skin tumours developed at a yield that was
approximately proportional to dose throughout the dose range 0.65-26.0 x 104 J/m
2
,
although no tumor was observed at 0.32 x 104 J/m
2
or below (Strickland et al., 1979).
3.5 Sources emitting mainly UV A radiation
The carcinogenic properties of UV A radiation received little attention before the
introduction of UV A equipment for tanning, which led to the development of powerful
sources ofUV A. Many experiments have now been performed, using mainly hairless mice,
to examine the possible carcinogenicity of UV A radiation (Zigman et al. , 1976; Forbes et
al., 1982; Berger & Kaase, 1983; Staberg et al. , 1983a,b; Kaase et al., 1984; Santamaria et
al., 1985; Strickland, 1986; van Weelden et al., 1986; Slaper, 1987; Kligman, 1988
[abstract]; van Weelden et al., 1988; Kligman et al., 1990 [abstract]; Sterenborg & van der
Leun, 1990; van Weelden et al., 1990a; Kelikens et al., 1991a; Kligman et al., 1992). Some
have shown no induction of tumours (Staberg et al., 1983a,b; Kaase et al.. 1984; Kligman,
1988 [abstract]). The Working Group noted that the doses may have been too small (daily
doses in the range of 160 kJ/m
2
) (Staberg et al., 1983b) or the exposure period too short
(Berger & Kaase, 1983; Kaase et al., 1984; Kligman, 1988 [abstract]), as noted by the
authors in a subsequent report (Kligman et al., 1992).] In the other experiments, tumours
were induced. [The Working Group noted that in some of the latter experiments either it is
unclear whether UVB radiation was sufficiently excluded from the spectrum (Zigman et al. ,
1976; Berger & Kause, 1983; Staberg et al., 1983a; Santamaria et al., 1985) or the exclusion
of UVB radiation was not fully convincing (Strickland, 1986).]
Studies in which the exclusion of UVB radiation was documented to be sufficient and
which led to the induction of tumours by UV A in hairless mice were reported by van
Weelden et at. (1986, 1988, 1990a), Slaper (1987), Kligman et al. (1990 [abstract], 1992),
Sterenborg and van der Leun (1990) and Kelttens et al. (1991a). A few of the most
informative studies are described below.
Groups of 24 male and female albino hairless Skh-hr 1 mice were exposed to UVA
radiation from a bank of Philips TL40W/09 fluorescent tubes, filtered through a 10-mm
glass plate selected for strong absorption of UVB radiation, for 12 h a day on seven days a
week for about one year, at which time the experiment was terminated. The daily dose was
220 kJ/m
2
. Most animals developed scratching lesions before they contracted skin tumours,
which occurred in all animals; the median time to tumour appearance was 265 days. At the
end of the experiment, the larger lesions were examined histologically: 60% were classified
as squamous-cell carcinomas, 20% as benign tumours, including papillomas and keratoacan-
thoma-like lesions, and 20% as mild cellular and nuclear atypia. The histological findings
were similar to those observed in a parallel experiment with UVB, but the tumours in the
UV A-exposed group appeared over a longer time span. Residual UVB radiation was
excluded as the cause of tumours in UV A-exposed mice on quantitative considerations: the
authors concluded that more than 100 000 times the UVB present would have been required
in order to induce tumorigenesis at the rate observed (van Weelden et al., 1986, 1988).
Groups of 48 male and female hairless albino Skh-hr 1 mice were exposed to 220 kJ/m
2
UVA radiation ( > 340 nm) from four high-pressure mercury metal-iodine lamps (Philips
HP A 400 W), passed through liquid filters, for 2 h per day on seven days per week for up to
400 days. The spectrum matched that of a damp used for tanning (the UV ASUN 5000);
UVB was effectively excluded by the filters. Skin tumours developed in most of the
animals, and 31 developed tumours before any scratching was observed. The largest
tumours were examined histologically at the end of the experiment: 15/20 tumours
examined were squamous-cell carcinomas (Sterenborg & van der Leun, 1990).
The desire to tan safely has raised interest in the possible carcinogenicity of long-wave-
length UV A (340-400 nm). In some experiments, UVB was excluded so rigorously that
there was also very little UV A in the range 3 15-340 nm; exposure was therefore mainly to
wavelengths in the region of 340-400 nm (van Weelden et al., 1988; Sterenborg & van der
Leun, 1990; van Weelden et al., 1990a). These experiments yielded squamous-cell carci-
nomas in most animals. [The Working Group noted that if these were to be ascribed to the
small proportion of shorter-wavelength UVA present in the spectra, a sharp peak in the
action spectrum for UV carcinogenesis would have to occur between 330 and 340 nm,
which does not appear likely.] In experiments by Kligman et al. (1990 Abstract], 1992),
wavelengths shorter than 340 nm were filtered out rigorously. Female hairless albino
Skh-hr l mice were exposed several times per week for 60 weeks to UV A at wavelengths of
340-400 nm at daily doses of 360 and 600 kJ/m
2
, as used in artificial suntanning. Eighteen
weeks later, 44 surviving mice had 19 skin tumours, mostly papillomas. At week 100, 22
surviving mice had 40 tumours. many of which were considered clinically to be
squamous-cell carcinomas.
The carcinogenicity of short-wavelength UV A (315-340 nm) was investigated in one
experiment. Groups of 24 male and female albino hairless Skh: hr 1 mice were exposed to
average daily doses of 20 or 56 kJ/m
2
radiation from specially developed fluorescent tubes
with peak emission near 330 nm (UVB radiation was filtered out efficiently using a glass
filter) on seven days a week for 650 days. All mice in the high-dose group developed
multiple tumours, first mainly papillomas and later predominantly squamous-cell
carcinomas. In the lower-dose group, three mice developed skin tumours, all of which were
papillomas. The lamps also emitted long-wavelength UV A (340-400 nm), but in a
proportion considered by the authors to be too small to account for the rate of tumorigenesis
observed (Kelfkens et al., 1991a). The investigators estimated the carcinogenic
effectiveness of short-wavelength UVA (315-340 nm) to be approximately five times
greater than that of long-wavelength UV A (340-400 nm)
3.6 Interaction of wavelengths
In daily life, the skin is exposed frequently to several wavelength ranges (UV A, UVB,
UVC) simultaneously, or to different combinations at different times. The simplest explana-
tion of an effect of such combined exposures is 'photoaddition', i.e., each exposure contri-
butes to the effective dose in an additive way. The validity of this hypothesis is one of the
assumptions underlying widely used concepts such as ' erythemal effective energy' and the
derivation of the action spectrum shown in Figure 10 (p. 141 ). It implies that any additional
exposure to an effective dose, in any wavelength region, increases the carcinogenic effect.
Several studies provide indications, however, that the situation is more complicated.
Interactions are seen between the effects of different wavebands that result in deviations
from photoaddition (for reviews, see van der Leun, 1987b, 1992). The literature on this
topic is controversial and cannot be summarized in detail here. The following two sections
fmm an attempt to give an overview and interpretation.
3 .6.1 Interaction of exposures given on the same day
Several types of interactions have been reported between different wavelength ranges
administered simultaneously or in close temporal proximity. These have led to concepts of
processes such as:
photorecovery: the effect of UVB or UVC is reduced by simultaneous or
immediately subsequent exposure to UVA or visible light [The Working Group noted
that photoreactivation is a special case of photorecovery but applies only to species
that have the 'photoreactivating enzyme', photo lyase (see Glossary).];
- photoprotection: the effect of UVB or UVC is reduced by prior administration of
UV A or visible light;
- photoaugmentation: the effect of UVB or UVC is enhanced by prior, simultaneous or
subsequent administration ofUV A or visible light.
Photoaugmentation of UVB carcinogenesis by UV A was suggested by several investi-
gators (Urbach etal., 1974; Willis eta!., 1981, 1986; Kligman, 1988 [abstract]; Talve etal.,
1990) but could not be confirmed by others (Forbes et al., 1978; van Weelden & van der
Leun, 1986). The latter investigators found evidence of photorecovery: the effect of UVB
plus UV A was smaller than that of the same UVB exposure given alone. The reduction was
small; however, UV A reduced the carcinogenic effective dose of UVB by 16%.
Interactions of different wavelength ranges when given simultaneously, prior to or
immediately after each other appear to be either nonexistent or unproven, as in the case of
photoaugmentation, or small, as in the case of photorecovery. Such interactions currently
play a small role in the evaluation of risks (see, for example, Health Council of the
Netherlands, 1986). Other uncertainties in the estimates, such as the dose received, are
likely to have a greater influence than interactions. Photoreactivation, is, however, a
well-defined process in those species which possess photolyase and may result in reduction
of effects.
3.6.2 Long-term interactions
A different type of interaction occurs when exposures to one wavelength band are
separated temporally from exposures to another. For example, a prolonged course of UVB
exposures, by itself sufficient to induce tumours, is compared with an identical UVB course
that is preceded or followed by a course ofUV A exposures, usually over several weeks.
Forbes et al. (1978) exposed hairless mice to tumorigenic UVB or to UVB followed by
UV A and visible light for 30 weeks. The longer-wavelength exposures reduced the tumori-
genic effect of the UVB. Staberg et al. (1983b) gave mice a tumorigenic combination of
UVB and UV A and found that subsequent exposures to UV A increased the tumorigenic
effect. The UV A was derived from Philips TL40W/09 lamps filtered through 2-mm plain
glass to remove the UVB. [The Working Group noted that since the glass transmitted some
UVB the increased carcinogenic effect may have been due to added UVB radiation.]
Bech-Thomsen et al. (1988a) pretreated lightly pigmented hairless female hr/hr C3H/Tif
mice with UV A for four weeks before exposure to broad-spectrum UVR. The UV A
reduced the carcinogenic effect of the broad-spectrum UVR. This result was not
corroborated in a subsequent, similar experiment by the same investigators (Bech-Thomsen
et al., 1988b ), in which mice were pretreated with radiation from various UV A sources.
The purest UV A radiation neither increased nor decreased the carcinogenic effect of UVB.
Slaper (1987) exposed one group of mice daily to UVB and a second group daily to
UV A at doses matched for approximately equal carcinogenic effect. In a third group of
mice that received the two regimens altemately every week, the carcinogenic effect was less
than that in the UVA- or the UVB-exposed group. The effective dose in the alternating
regimen was estimated to be 80% that in the UVB regimen. The investigator concluded that
both UV A and UVB contributed to the carcinogenic effect of the alternating regimen.
[The Working Group noted that the effect of long-term interactions appears to be simi Jar
to that of interactions of exposures given on the same day. Photoaddition gives a reasonable
prediction, but the combined effects tend to be slightly less than would be predicted.]
3. 7 Additional experimental observations
3. 7.1 Tumour types
Skin tumours in UV -exposed animals are commonly epidermal, benign papillomas and
malignant squamous-cell carcinomas; adnexal neoplasms, mainly basal-cell carcinomas, are
less common. Attempts have been made to induce naevi and malignant melanomas. Many
tumours are found, since the animals are followed for long periods of time; however,
tumours coalesce and regress, and all tumours are not examined histologically.
Squamous-cell carcinoma is the commonest type of tumour found after exposure to
UVR. These tumours have been reported in mice exposed to predominantly UVB radiation
(Winkelmann et al., 1960, 1963; Epstein & Epstein, 1963; Freeman, 1975; Forbes et al.,
1981; de Gruijl et al., 1983), to predominantly UVA radiation (van Weelden et al., 1988;
Sterenborg & van der Leun, 1990) and to predominantly UYC radiation (Lill, 1983;
Sterenborg et al., 1988). They have also been found in rats (Putschar & Holtz, 1930; Roffo,
1934, 1939; Hueper, 1942), hamsters (Stenback, 1975a) and opossums (Ley et al., 1989)
following exposure to broad-spectrum UVR.
Papillomas were reported to be the commonest tumour after exposure of hairless mice to
UVR consisting of UVB and UV A (Gallagher et al., 1984b ). Papillomas were also reported
to precede or accompany squamous-cell carcinomas induced in hairless mice by UVA (van
Weelden et al., 1988), UVB (Stenback, 1978) or UYC radiation (Sterenborg et at., 1988).
Papillomas were also common in rats (Findlay, 1930; Putschar & Holtz, 1930; Stenback,
1975a) and hamsters (Stenback, 1975a) exposed to broad-spectrum UVR.
The main type of tumour diagnosed after exposure of haired mice to broad-spectrum
UVR was fibrosarcomas (Grady et al., 1941, 1943). Squamous-cell carcinomas were less
common, but the ratio of carcinomas to sarcomas increased with the number of exposures
per week (Grady et al., 1943). Spikes et al. (1977) reported many squamous-cell
carcinomas in clipped C3Hf mice irradiated with UVB, especially at low doses; the
high-dose group had a much higher proportion of fibrosarcomas. The investigators
suggested that the type of tumour induced might be dose-dependent. Norbury and Kripke
(1978) found that the type of tumour might depend on immunological factors. They
compared UVB tumorigenesis in normal C3H/HeN (MTV) mice, in T cell-depleted mice
and in T cell-depleted mice reconstituted with thymus grafts. In the normal mice,
fibrosarcomas predominated: in the T -cell depleted, reconstituted mice, squamous-cell
carcinomas predominated. Spindle-cell sarcomas were reported in rats irradiated with
sunlight (Roffo, 1934), and fibrosarcomas were seen in opossums irradiated with UVB (Ley
et al., 1989).
The diagnosis of fibrosarcoma was questioned by Morison et at. (1986). After C3H/
HeNCr (mammary tumour virus-free) haired pigmented mice were exposed to mainly UVB
radiation, the tumours induced were almost all squamous-cell carcinomas. The investigators
noted that the same type of tumour had been diagnosed in many previous reports as
fibrosarcoma: they diagnosed squamous-cell carcinomas by studying specific markers for
cell differentiation in the tumours. In a study by Phelps et al. (1989) in which hairless
albino Skh/hr-1 mice were exposed to UV A and UVB at 0.3 J/cm
2
[30 kJ/m
2
], all mice
developed epidermal neoplasia and 25% of animals developed spindle-cell tumours that
resembled human atypical fibroxanthoma. [The Working Group noted that earlier studies
did not use presently available cellular markers.]
Keratoacanthomas and similar benign epidermal neoplasms have been reported in mice
exposed to UVB (Stenback, 1978), rats exposed to UVB and UYC (Strickland et at., 1979)
and hamsters exposed to UVB (Stenback, 1975a).
Actinic keratosis, or solar keratosis, a precursor lesion of squamous-cell carcinomas, has
been reported in hairless mice exposed to UV A and UVB (Kligman & Kligman, 1981) and
in haired mice exposed to UVB (Stenback, 1978).
Basal-cell carcinomas have not been reported in studies in mice. A few studies on UV
carcinogenesis in nude mice, which have a deficient immune system, have been reported
(Eaton et at., 1978; Anderson & Rice, 1987; Hoover et at., 1987). The skin tumours
induced by mainly UVB radiation in these studies were mostly squamous-cell carcinomas,
but in the experiments reported by Anderson and Rice (1987) in nude mice of BALB/c
background there were several basal-cell carcinomas. Basal-cell carcinomas were found
occasionally in rats exposed to broad-spectrum UVR (Putschar & Holtz, 1930; Hueper,
1942). [The Working Group noted that the classification of these neoplasms and their
relation to the corresponding neoplasms in humans is not clear.]
There is no report in which cutaneous malignant melanoma was induced in mice by
UVR alone (Epstein, 1990; van Weelden et al., 1990b; Husain et al., 1991), in spite of
concerted attempts to achieve this.
No study was found in which the primary objective was to examine the susceptibility of
the eye to UVR; rather, eye tumours were found incidentally in studies designed to
investigate skin carcinogenesis. All of the tumours of the eye identified in these reports
involved superficial parts of the eye (cornea and conjunctive); no turnout of the interior eye
was reported.
Studies of the effect of UVR on turnout induction in other organs (lymphoma in mice)
are few and were not designed to determine this effect (Ebbesen, 1981; Joshi et al., 1986).
[The Working Group considered that the data were inadequate for evaluation and that data
on survival among treated and control groups, sample selection and analysis of data were
limited.]
3.7.2 Dose and effect
Quantitative information is available mainly on the induction of squamous-cell carci-
noma in mice. In most of the experiments, exposure was regular, several times per week or
every day, until tumours developed. The daily doses of UVR required for skin
tumorigenesis are usually well below those present outdoors in the environment, and most
experiments have been conducted with UVB doses lower than those required to elicit acute
reactions in mouse skin (erythema or oedema). In one experiment in hairless mice, with a
UVB dose 33 times lower than that required for acute reactions, 71 by of the skin tumours
were squamous cell carcinomas (de Gruijl et al., 1983). The effectiveness of UVB radiation
is increased at lower dose rates (Kelfkens et at. , 1991 b).
The higher the dose given, the less time it takes for tumours to appear. In most
experiments. the time required for 50% of mice to develop tumours ranged between a few
months and one year. By maximizing the exposure regimen in hairless mice (escalating
doses ofUVB radiation), the time could be reduced to 18 days (Willis et al. , 1981). In a few
experiments, in both mice and rats, skin tumours resulted from a single exposure to UVB
radiation (Hsu et al., 1975; Strickland et al., 1979); in mice, this required a dose that first
caused skin ulceration: hairless mice, 60 kJ/m
2
(Hsu et al., 1975); Sencar mice, 29 kJ/m
2
(Strickland, 1982).
Quantitative dose-effect relationships have been derived for mice exposed regularly
(usually daily) to UVR. The median time to first tumour, tnl> has been used as a measure of
the effect and is related to dose level. Dose-effect relationships of the form
tm = CD-r,
where c is a constant incorporating the susceptibility of the strain of mice as well as the
effectiveness of the radiation spectrum, Dis the daily dose of radiation and r is a numerical
exponent giving the steepness of the relationship, have been proposed by several authors.
Estimates of r vary from 0.2 (Sterenborg et al. , 1988) for small tumours of the skin induced
by UVC radiation in hairless mice, to 0.5 (Blum et al. , 1959) for large tumours on the ears of
haired mice induced by broad-spectrum UVR and to 0.6 (de Gruijl et al., 1983) for small
tumours induced by broad-band UVB in hairless mice. Figure 11 (p. 145) illustrates the
shape of this dose-response relationship for r = 0.6; other forms of the relationship have
been proposed (Forbes et at., 1982). All of them provide adequate descriptions of the
dose-response within the range of the available data, although extrapolations outside this
range differ substantially.
3.7.3 Dose delivery
The tumorigenic effect of UVR depends not only on the dose but also on the temporal
pattern of exposure. In general, the effectiveness of treatment increases with the number of
fractions of the dose per week (Forbes et al. , 1981), for both daily and accumulated doses.
A daily dose administered over 12 h is more effective than the same daily dose administered
in 1 h (Kelfkens et al, 199lb ). The same weekly dose is more effective when given over
three to five days than if given in one day (Forbes et al., 1981).
3. 7.4 Action spectra
Ideally, the carcinogenic effectiveness of UVR can be expressed as a continuous func-
tion of wavelength. That function, called the action spectrum for UV carcinogenesis, is not
yet completely delineated. Freeman (1978) made an early attempt to determine this
spectrum and found that it was limited to a few narrow bands around the wavelengths 290,
300, 310 and 320 nm. Narrow-band monochromatic sources are difficult to achieve.
Since that time, various action spectra have been proposed to weight the spectral irra-
diance of a source. Forbes et al. (1982) and Cole et al. (1986) determined dose-effect
relationships similar to that shown in Figure l 1 for many different UV spectra. By
weighting these lamp spectra with various existing action spectra for photobiological
effects, effective doses were computed for each experiment. In this way, the investigators
tried to align the results from the experiments with different UV spectra into one dose-effect
relationship. One of the action spectra (MEE48), originally determined for the induction of
oedema in mice 48 h after exposure to UVR and which is similar to the human erythema
action spectrum, fitted well. The authors concluded that the mouse oedema spectrum was
also appropriate for describing skin cancer induction (Cole et al. , 1986).
Sterenborg and van der Leun ( 1987) attempted to determine an action spectrum directly
from observations on UV carcinogenesis. They exposed hairless albino mice to seven
different lamp spectra under otherwise identical circumstances. The lamp spectra over-
lapped to some extent, and the action spectrum was derived by mathematical fitting. The
analysis yielded an action spectrum for the wavelength range 250-360 nm. Slaper (1987)
added observations in the UV A region and extended the action spectrum throughout the
UVA range (see Fig. 10, p. 141).
The action spectrum shown in Figure 10 is for albino hairless Skh-hr 1 mice with an
end-point of 1.0-mm tumours. Although different end-points may yield different action
spectra, this curve shows good agreement in the UVB range with the MEE48 spectrum and
also with the observations of Freeman (1978) for wavelengths 300, 310 and 320 nm. [The
Working Group noted that the action spectrum for UV carcinogenesis in the wavelength
range 300-320 nm may be considered a good approximation.] The different shapes of
Figure 10 and MEE48 in the UVC reflect a scarcity of data in this wavelength range. [The
Working Group noted that the action spectrum for carcinogenesis by UVC is still highly
uncertain.] The MEE48 left widely different options open for the action spectrum of
long-wavelength UVA: the effectiveness in the wavelength range 330-400 nm could be
either zero or as high as 0.0002 (Cole et al., 1986). More recent data on the carcinogenesis
ofUVA, used to construct the curve in Figure 10, indicate a mean effectiveness of0.00015
in this range (Slaper, 1987). [The Working Group noted that this value for the carcinogenic
effectiveness for UV A may be regarded as an estimate of the order of magnitude.]
3.7.5 Pigmentation
Pigment was reported to be protective against tumours arising from the conjunctive in
cattle (Anderson, 1963).
Freeman and Knox (1964) also examined the influence of pigmentation in a group of78
rats composed of 66 pigmented rats of various strains (black, black and white, grey-brown,
grey and white) and 12 albinos. Under the same irradiation regimen, the pigmented rats
developed tumours in 73% of eyes and the albinos in only 8%. The tumour yield was
consistently higher in the pigmented strains than in the albinos. In nine pigmented and 10
albino hamsters exposed for one year, 50% of pigmented animals and 25% of nonpigmented
animals developed eye tumours.
Davies and Forbes (1988) exposed closely related albino hairless Skh-hr 1 mice and
pigmented hairless Skh-hr 2 mice to broadband UVR from a filtered xenon arc lamp.
Especially at high doses, the latent period until 50% of animals had first tumours was longer
in Skh-hr 2 mice.
van Weelden el al. ( 1990a) derived mice of different degrees of pigmentation-'browns'
end 'blacks'-by selective breeding from Skh-hr 2 stock and exposed 24 albinos (Skh-hr 1)
(van Weelden et al., 1988), 16 'browns' and eight 'blacks' to UVA radiation. The brown
mice were less susceptible to skin tumours than the albinos, but the more heavily pigmented
blacks were as susceptible as the albinos: the median times for tumour induction were 265
days for albinos, 267 days for blacks and 375 days for browns (van Weelden et al. , 1990a).
3.8 Administration with known chemical carcinogens
Since UVR alone produces tumours, it is a 'complete' carcinogen and may thus be
involved in cocarcinogenicity. Several investigators have attempted to determine whether
UVR has tumour 'initiating' and/or tumour 'promoting' activity when tested in a traditional
two-stage protocol. For the purposes of this monograph, a 'tumour initiators' is defined as
an agent that, at a stated amount and upon administration once, is incapable of causing
tumours in the population of animals unless the skin is subsequently treated with a 'tumour
promoter'. A 'tumour promoter' is defined as an agent that, under stated conditions is
incapable of causing tumours unless the skin was previously treated with a 'tumour initiator'.
The test systems used embody a number of variables, not all of which were necessarily
considered by the authors. For example, UVR has also been shown to influence the immune
system, and polycyclic aromatic hydrocarbons are photochemically active.
3.8.1 Administration with polycyclic aromatic hydrocarbons
Most of the studies summarized below demonstrate that UVR bas a cocarcinogenic
action with other carcinogens. Other reports provide additional information on cocarcino-
genesis, on photolysis ot polycyclic aromatic hydrocarbons and on other interference with
chemical carcinogenesis (Clark, 1964; Ito, 1966; Santamaria et al. , 1966; Davies et al., 1972
[abstract]; Shabad & Litvinova, 1972; Stenback & Shubik, 1973; Stenback, 1975b; Roberts
& Daynes, 1980; Gensler & Welch, 1992).
(a) 3,4-Benzo[a]pyrene
Groups of 18 female SPF (specific pathogen-free) BALB/c mice, six weeks of age,
received 30-min exposures on the shaved dorsal skin to UVB from a Westinghouse FS40
sunlamp (280-320 nm) five times a week for 13 weeks (total close, 7.0 x 105 J/m
2
) or no
UVB exposure followed one week later by twice weekly applications of 0, 0.1 or 1.0 mg
3,4-benzo[a]pyrene in acetone on the shaved ventral skin for 20 (acetone only), 20 or 10
weeks, respectively. Pre-exposure to UVB enhanced tumour growth in the high-dose group:
29 tumours (of 20 examined histologically, 90% were squamous-cell carcinomas and 10%
undifferentiated sarcomas) in the UVB-pretreated group compared to two (squamous-cell
carcinomas) in the non-irradiated 3,4-benzo[a]pyrene-treated animals 18 weeks after the
first treatment with 3,4-benzo[a]pyrene. No such effect was seen in the low-dose group
(Gensler & Bowden, 1987; Gensler, 1988a).
(b) 7, 12-Dimethylbenz[a]anthracene
In an attempt to assess the promoting effects of UVR, groups of 15-31 male and 16-22
female Swiss albino mice, 11- 18 weeks of age, received a single application of two drops
(0.1 ml) of 0 or 0.5% 7,12-dimethylbenz[a]anthracene (DMBA) in acetone on the posterior
half of the dorsal skin, followed 14 days later by exposures to UVB (280-320 nm;
high-pressure Hanovia hot quartz contact lamp) twice a week for 67 weeks (total dose,
13.33 x 107 ergs/cm
2
[133 kJ/m
2
]) or no exposure. At the end of the UVB treatment, 16/31
mice treated with DMBA and UVB had developed 19 skin tumours, compared to 4/41 and
0/47, respectively, among mice treated with DMBA alone and UVB alone. Exposure to
UVB also enhanced the multiplicity and degree of malignancy of DMBA-induced tumours
(Epstein & Epstein, 1962).
Groups of 26-42 male and female outbred hairless mice, 7-12 weeks old, received a
single application of two drops (0.1 ml) of 0 or 0.5% DMBA in acetone, followed six weeks
later by exposures to UVB (280-320 nm; high-pressure Hanovia hot quartz contact lamp)
three times a week for 29 weeks (total dose, 15.34 x 107 ergs/cm
2
[153 kJ/m
2
]) or no
exposure. All animals were observed for 63 weeks. UVB exposure produced skin tumours
in 22/26 animals, and DMBA treatment alone in 3/41; acetone alone produce no skin
tumour. Exposure to UVB following DMBA treatment enhanced carcinogenicity with
regard to appearance time (first tumour observed at 14 weeks compared to 30 in the group
treated with DMBA alone and 20 in that given UVB alone), multiplicity at 58 weeks after
DMBA treatment (40 in 24 animals compared to 22 in 26 animals treated with UVB alone
and 3 in 41 animals treated with DMBA alone) and degree of malignancy. Two
'melanomas' appeared in the group receiving the combined treatment (Epstein, 1965).
Groups of 18-46 outbred hairless pigmented mice [sex unspecified], 8-11 weeks old,
received a single application of 0.05 ml of 0.4% DMBA (0.2 rna) in acetone or no DMBA.
After 13 months, mice treated with DMBA had developed pigmented lesions ('blue naevi')
in the treated areas. For the following seven months, mice received UVB (280-320 nm;
high-pressure Hanovia hot quartz contact lamp) three times a week or no UVB treatment.
Exposure to UVB following DMBA treatment enhanced the growth of naevi into malignant-
appearing pigmented tumours ('melanomas'): 5/ 18 versus 0/41 in the group treated with
DMBA alone and 0/39 in the group treated with UVB alone (Epstein et al., 1967). [The
Working Group noted the limited reporting on metastases.]
A group of 56 B6D2F
1
/J mice [sex unspecified], six weeks of age, was irradiated with
UVB (280-340 nm; Westinghouse FS40 sunlamp) dorsally for 30 min per day on five
daysper week (Roberts & Daynes, 1980) for 11.5 weeks (total dose, 6.2 x 105 J/m
2
) . A
control group of 41 mice received no irradiation. Both groups subsequently received a
single application of 100 ~ DMBA in 0.1 ml acetone on the shaved ventral skin, followed
four days later by applications of 5 ~ 12-0 -tetradecanoylphorbol 13-acetate (TP A) three
times a week for 32 weeks. Tumour yield was significantly decreased at 32 weeks (2.2
versus 4.8 tumours/mouse) in the pre-irradiated mice (Gensler, 1988b).
Groups of 20-24 female hairless Skh-hr 2 mice, six to eight weeks old, received a single
application of 0 or 0.5 % DMBA in acetone on the dorsal skin. Two weeks later, the animals
were irradiated with UVB (290-320 nm; Westinghouse FS40-T12 sunlamp), UVA (320-400
nm; GTE-Sylvania fluorescent black light tubes) or a combination of UVA plus UVB three
times a week for 30 weeks or were not irradiated, and were observed for 12 months. All
mice receiving DMBA treatment developed multiple 'blue naevi'; vittually none of the
untreated mice or mice that received UVR treatment only showed this effect. Irradiation of
DMBA-treated animals induced a higher incidence of papillomas (70-100%), squamous-cell
carcinomas (30-80%), melanomas (25-33%) and lymphomas (21-50%), than exposure to
UVA alone (0-32% papillomas, 0-47% squamous-cell carcinomas, no melanoma and no
lymphoma) or to DMBA alone (90, 25, 0 and 5% of these tumours, respectively). The
authors also examined selected lesions induced by DMBA alone or by DMBA with UVR
for the presence of H- or N-ras mutations. Mutations at codon 61 in N-ras were present in
three (two induced by DMBA plus UVR, one by DMBA alone) out of eight of the early
pigmented lesions examined and in one out of three of the malignant melanomas examined
(induced by DMBA plus UVR); no H-ras mutation was observed (Husain et al., 1991).
[The Working Group noted that lesions were not induced by UVR alone.]
3.8.2 Administration with other agents with promoting activity
These studies were designed to evaluate the action of UVR as a tumour initiator.
(a) Croton oil
Groups of 15-53 male and 9-30 female random-bred hairless mice, 9-12 weeks old,
received a single exposure to UVB (280-320 nm; high-pressure Hanovia hot quartz contact
lamp) for 30 s (1.3 x 107 ergs/cm3 [13 kJ/m
2
]) or no exposure, followed two weeks later by
applications to the dorsal skin of 0 or 0.1 ml croton oil in acetone twice a week for 18
months. Neither UVB exposure nor croton oil alone produced any skin tumour over the
course of the study. The group of 79 mice that received both UVB exposure and croton oil
had eight persistent skin tumours (one per mouse) (Epstein & Roth, 1968).
Groups of 30 female Swiss mice, eight weeks old, received UVB once (5.5 x 107
ergs/cm
2
[55 kJ/m
2
]) from Westinghouse FS40Tl2 lamps or croton oil (0.02 ml of a 2.5%
solution, twice a week for 30 weeks); a group of 60 mice received UVB followed after 10
days by croton oil for life. UVB alone produced no tumour; croton oil alone produced
regressing tumours, and the combination produced 11 tumours (four papillomas, four
fibromas and three regressing tumours) in seven mice (Stenback, 1975c).
Groups of 40 male haired mice (random-bred 'Hall' strain), 18 weeks of age, were
clipped and exposed once to UVC (medium-pressure mercury discharge lamp). One group
received no further treatment; the other received one appllcation of croton oil one day before
irradiation and, beginning two weeks later, received applications of 0.25 ml croton oil (0.5%
solution) once a week for 30 weeks. By 35 weeks, the groups had 20 and 23 survivors, with
0 and 12 skin tumours, respectively (Pound, 1970).
(b) 12-0-Tetradecanoylphor 1 bol13-acetate
Six groups of 25 eight-week-old female C3H/HeNCr(MTV-) mice were irradiated with
UVB (Westinghouse FS40 sunlamps) on the shaved dorsum for 30 min, five times a week
for two weeks (total dose, 1.44 x 105 J/m
2
), followed two weeks later by 'promotion' with
applications of 0 or 5 ~ TP A in acetone twice a week. Ventral irradiation for 30 min, three
times a week for 12 weeks (total dose, 4.54 x 105 J/m
2
) (to produce a 'systemic' effect) was
begun two weeks after completion of dorsal initiation. At 70 weeks, UVB exposure of the
dorsum alone had produced no tumour, and dorsal applications of TP A alone had produced
a 5% incidence of tumours. The combination of these treatments produced a 41% tumour
incidence. Ventral irradiation of animals that had received TPA only produced a 33%
incidence, and ventral irradiation of mice that had received both UVB and TP A produced a
100% incidence. The authors suggested that these findings reflect a systemic effect-
possibly suppression of immune surveillance or a biochemical influence on the epidermal
growth regulatory system (Strickland et al., 1985).
(c) Benzoyl peroxide
Benzoyl peroxide is considered to be a prototype promoter of two-stage chemical
carcinogenesis in the skin (Slaga et al., 1981 ). The studies summarized below were
motivated, however, by concerns about the safety of using this compound for treating acne
vulgaris.
Groups of Used (Hr) stock hairless albino mice (total, 148) [sex unspecified], three to
four months old, were exposed on the posterior half of the back to UVR (Hanovia hot quartz
contact lamp emitting primarily UVB; 270 mJ/cm
2
[2.7 kJ/m
2
]) three times a week for eight
weeks. Four weeks later, the mice were divided into four groups. The final skin tumour
incidences at the irradiated sites were: 38% in the group that received applications of 0.1 ml
of a 0.1% solution of croton oil in acetone on the back skin five times a week for the
duration of the experiment ( 62 weeks); 5% in the group that received applications of acetone
alone; 8% in mice that received applications of the benzoyl peroxide base; and 8% in those
that received applications of a 5% lotion of benzoyl peroxide in water five times a week for
the duration of the study (Epstein, 1988).
Five groups of Oslo hairless mice (16 males and 16 females) were inadiated under
Philips HP3114 sunlamps (mostly UVB) twice a week for 52 weeks (total dose, 26.5 J/cm
2
[265 kJ/m
2
]). The mice were treated before or after each exposure with 5% benzoyl
peroxide in gel, with the gel alone or with no chemical. Throughout the study, the groups
were indistinguishable in terms of the proportion with one of more tumours (median latent
period, approximately 40 weeks) and of the total number of turnouts per survivor
(approximately 1.5 at 40 weeks and approximately 4 at 48 weeks). Thus, benzoyl peroxide
did not enhance photocarcinogenesis. The study also included several groups of SENCAR
mice treated topically with DMBA once (51.2 J.Lg) or with vehicle followed by benzoyl
peroxide twice a week. Benzoyl peroxide reduced the number of DMBA-induced tumours
(Iversen, 1988). Two unresolved concerns were raised by the author: Firstly, the fact that
benzoyl peroxide reduced the tumorigenicity of DMBA was contrary to the author's
previous experience (Iversen, 1986) and to that of several others; secondly, the UVR dose
used in this study was lower (total dose, 265 kJ/m
2
) than that used in the 1986 study (total
close, 480 kJ!m\ but the tumour response was significantly greater.
(d) Methyl ethyl ketone peroxide
A postulated mechanism for tumour promotion involves the generation of free radicals,
possibly with reactive oxygen species, leading to enhanced lipid peroxidation and DNA
damage and/or cell phenotype. A study was therefore designed to test whether methyl ethyl
ketone peroxide (ME.KP), which is known to produce lipid-peroxidizing activity in vivo,
acts as a tumour promotor in skin 'initiated' by UVR. Furthermore, since glutathione has
been shown to be a major endogenous reducing agent which protects against lipid
peroxidation, the study also tested diethyl maleate (DEM), which is known to deplete the
intracellular level of glutathione in mouse skin.
Groups of 24 male and female hairless albino mice (14-16 weeks old) were inadiated
with UVB (280-320 nm; Westinghouse FS40 fluorescent sunlamps; 2054 J/m
2
daily) for 18
weeks. Three weeks later, topical application of MEKP (20 Jll containing 0 or 10 Jlg
MEKP) was begun and continued twice a week for 25 weeks. Other groups received DEM
(0 or 1 J.Lg in dibutyl phthalate) 1 h before each MEKP application. Otherwise identical
control groups received either the chemical treatments or UVB alone. At 46 weeks, the
groups that did not receive UVB inadiation had at most two tumours on two mice (among
21 survivors in mice exposed to MEKP plus DEM). Exposure to UVB produced five
turnouts in four mice exposed to the solvent, out of 19 survivors; 11 tumours in eight mice
exposed to MEKP, out of21 survivors; and 18 tumours in nine mice exposed to MEKP plus
DEM, out of 16 survivors. Using tumour onset rate analysis (Peso et al., 1980), the overall
effect of MEKP was statistically significant. Tumour enhancement by MEKP was greater in
the presence ofDEM (Logani et al., 1984).
3.9 Interaction with immunosuppressive agents
Investigations have been reported on agents known to influence immunological
responses in humans and on agents chosen to test some aspect of immunological response in
mice. [The Working Group noted that in most cases the effect on the immune system of the
animas was not evaluated directly; these agents have effects other than immunosuppression,
which may explain their interaction with photocarcinogenesis.]
Three groups of 12 male Skh-Hrl hairless mice, eight weeks of age, were irradiated with
280-320 nm UVB (Westinghouse FS40T12 sunlamps) on five days per week for 30 weeks
at daily doses of 470 J/m
2
. Two weeks after the first UVB exposure, one group received
subcutaneous injections of 0.1 ml anti-mouse lymphocytic serum twice a week for 20 weeks;
a second received intraperitoneal injections of 12 mg/kg bw6-mercaptopurine (Purinethol)
five times a week for 20 weeks; and a third received intraperitoneal injections of 0.1 ml
isotonic saline five times a week for 20 weeks. Treatment with anti-mouse lymphocytic
serum resulted in an earlier appearance and a greater numbers of tumours than did treatment
with saline; in contrast, 6-mercaptopurine appeared to delay the appearance of tumours
(Nathanson et al. , 1976).
Groups of 24-28 female albino HRA/Skh-1 hairless mice, 21-35 weeks of age, were
irradiated with UVR (UVB from an Oliphant FL40SE tube and UVA from six Sylvania
40BL tubes) to simulate the UVR portion of terrestrial sunlight on five days per week for 10
weeks to achieve a MED. At the same time, the animals received intraperitoneal injections
of 15 mg/kg bw azathioprine in 0.1 ml glycine buffer, 10.6 mg/kg bw cyclophosphamide in
0.1 ml glycine buffer or 0.1 ml vehicle alone. At day 200, mice receiving UV irradiation
alone had a tumour incidence of 77%; those also receiving azathioprine had an incidence of
96% (marginally significant enhancement of tumour growth); and those receiving cyclo-
phosphamide had an incidence of 85% (nonsignificant increase) (Reeve et al., 1985).
Groups of 15 female albino HRS/J hairless hr/hr mice, eight weeks old, were irradiated
with UVB (280-320 nm; Westinghouse FS40 sunlamps) on five days a week for 24 weeks;
further groups also received injections of 4 or 8 mg/kg bw azathioprine or 10 or 25 mg/kg
bw cyclosporine three times a week. The mean latent period for tumour development was
16 weeks in the group receiving UV irradiation only and 12-13 weeks in the groups also
receiving azathioprine or cyclosporine, indicating enhancement of photocarcinogenesis by
both drugs (Nelson et al., 1987).
Groups of female C3H/HeN(MT\f ) mice [initial numbers unspecified], four to six
weeks of age, received grafts of fragments of an antigenic ('regressor') tumour (fibro-
sarcoma) previously induced in a host animal by UVB. Some animals received no further
treatment; other groups received UVB irradiation (Westinghouse FS40; 5 kJ/m
2
per day on
five days a week for four to six weeks), subcutaneous injections of 25 or 75 mg/kg bw
cyclosporine once a day on eight consecutive days, or injections of 20 mg/kg bw
cyclophosphamide 1, 3, 6, 9 and 13 days after tumour challenge. Tumours grew
progressively in the groups treated with UVB or cyclosporine, but not in the groups
receiving no further treatment or cyclophosphamide (Servilla et al., 1987).
Groups of six female albino HRNSkh-1 hairless mice, 10-12 weeks of age, were irra-
diated with OVA plus OVB (one Oliphant FL40SE tube and three Sylvania F4/350 BL
tubes)
on five days a week until death (about 35 weeks). During that time, they were also
injected intraperitoneally with 15 mg/kg bw azathioprine, 20 mg/kg bw prednisolone or 15
mg/kg bw cyclophosphamide in 0.1 ml saline or given 60 mg/kg bw cyclosporine in 0.1 ml
peanut oil by gavage or 0.1 ml vehicle alone. Azathioprine, cyclophosphamide and
cyclosporine all significantly enhanced photocarcinogenesis with regard to median latent
periods and tumour multiplicity. Prednisolone did not enhance this effect, nor did it
interfere with the enhancement by other drugs when given in combination with them (Kelly
et al., 1987).
Groups of 15-32 female albino Skh-hr 1 hairless mice, 10-12 weeks of age, were irra-
diated with UV A plus UVB (250-700 nrn; one Oliphant FL40SE tube, three Sylvania
F40/350 BL tubes and two True-Lite [Duro-Test Corp] tubes) on five days per week for 12
weeks. Two weeks after the first irradiation, mice received intraperitoneal injections on five
days a week of 15 mg/kg bw azathioprine or 6-mercaptoptlrine in 0.1 ml saline or 0.1 ml
vehicle alone. Both compounds significantly enhanced skin photocarcinogenesis with
regard to median latent period, proportion of malignant:benign growths and tumour
multiplicity (Kelly et al., 1989).
3.10 Molecular genetics of animal skin tumours induced by ultraviolet radiation
Three skin papillomas and three skin carcinomas produced in female SENCAR mice
after a single exposure to UVB (280-315 nm; Westinghouse FS20; 70 kJ/m
2
) were
examined for ras gene alterations. A five- to 10-fold increase in cHa-ras RNA gene
expression associated with the gene amplification was found in papillomas and carcinomas,
while DNA from carcinomas, but not from papillomas, induced foci in the NIH-3T3 cell
transfection assay (Husain et al., 1990).
CONTINUED
4. Other Relevant Data
4.1 Transmission and absorption in biological tissues
UVR may be transmitted, reflected, scattered or absorbed by chromophores in any layer
of tissue, such as the skin and eye. Absorption is strongly related to wavelength, as it
depends on the properties of the responsible chromophore(s). Accordingly, transmission is
also wavelength-dependent. Transmission and other optical properties are affected by
changes in the structure of the tissue and, especially in the case of the lens of the eye, by
agemg.
Absorption of radiation by a tissue chromophore is a prerequisite for any photochemical
or photobiological effect; however, absorption does not necessarily have a biological con-
sequence.
4 .1.1 Epidermis
Since UVR-induced skin cancer is an epidennal phenomenon, this section focuses on
epidermis and excludes the dermis.
The epidermis, a tissue with a high replication rate, can be divided functionally into two:
an inner, living part ( 60-160-j..im thick in humans) of cells at various stages of differentiation
and the outermost, non-living, terminally differentiated stratum corneum (8-15-j.lm thick in
humans). The dividing cell population is located in the innermost basal layer of the living
epidermis. Optical properties have usually been studied using isolated strateum corneum or
whole epidermis. Absorption and scattering of UVR by the stratum corneum afford some
protection to the living part of the epidermis from UVR exposure.
Human and mouse epidermis have impotiant structural differences. The living part and
the stratum corneum of human epidermis have about 10 cell layers each. In mice, the living
part has two to three cell layers and the stratum cornea one to two cell layers. The
interphase of human epidermis and dermis is highly undulated (i.e., epidermal thickness
varies), whereas in the mouse it is flat.
Skin contains sebaceous glands which secrete lipid-containing sebum, which forms a
film on the stratum corneum.
(a) Humans
The optical properties of human skin have been reviewed (Anderson & Parrish, 1981,
1982).
Everett et al. ( 1966) used a variety of methods to obtain whole epidermal and stratum
corneum preparations of human skin. Transmission characteristics (from 240 to 700 nm)
were measured using a recording spectrophotometer via an integrating sphere which permits
the measurement of forward scattered radiation. Transmission values of whole epidermis in
-163-
white skin ranged from 1% at 250 nm to 44% at 320 nm, while transmission at 400 nm was
about 50%.
Kaidbey et al. (1979) compared the optical properties (250-400 nm) of whole epidermis
and stratum corneum from black and white skins. In general, the absorption spectra from
the stratum corneum were similar in shape and magnitude; however, the absorption spectra
for whole epidermis were clearly different: At about 300 nm, the absorbance (accounting for
scattering) ofblack epidermis was twice that of white epidermis.
Anderson and Parrish (1981, 1982) presented data which show that epidermal trans-
mission between 260 and 290 nm will be overestimated if no correction is made for tissue
fluorescence (330-360 nm). This is most evident at about 280 nm and is consistent with
tryptophan or tyrosine fluorescence.
Bruls et al. (1984a) measured transmission in whole human epidermis and stratum
corneum of UVR between 248 and 546 nm, using a solar blind detector which corrects for
fluorescence, and found results different from those of Everett, in particular, that UVC
transmission was one to two magnitudes lower. The transmission spectra of whole
epidermis and stratum corneum showed a similar general shape but with differences in
minima and magnitude. The minimum for epidermis was 265 nm and that for stratum
corneum was 275 nm, presumably reflecting different chromophores in those tissues. At
254 nm, transmission in stratum corneum was about two orders of magnitude greater than
that in whole epidermis. At about 300 nm, this difference was only one order of magnitude.
The transmission in stratum corneum from previously sun-exposed skin was about one order
of magnitude less than that in unexposed epidermis at 254 nm. The difference was less at
wavelengths > 290 nm. The minimal transmission in stratum corneum from previously
sun-exposed skin was shifted from 275 to 265 nm. The authors also showed that the
relationship between tissue thickness and transmission of UVR and visible light (log scale)
is linear.
Bruls et al. (l984b) studied the relationship between the MED ofUVB (filtered mercury
arc) and UVC (germicidal lamp) and epidermal transmission. A clear linear (log-log)
relationship was demonstrated; the MED increased with decreased transmission. Repeated
exposure to UVB resulted in higher MEDs ofUVB and UVC and decreased transmission of
UVB (only epidermis measured) and UVC (epidermis and stratum corneum measured).
Beadle and Burton (1981) extracted skin lipids from human scalps and measured their
transmission spectra in hexane. They estimated that lipid concentrations normally present
on the skin surface of the forehead would reduce transmission at 300 nm by about 10%.
(b) Experimental systems
No data are available on transmission in the stratum corneum of mice. Sterenborg and
van der Leun (1988) measured transmission of 246-365 nm in Skh-hr 1 mouse epidermis in
vitro. Minimal transmission (about 2%) was observed at 254 nm and 270 nm, 10% was
transmitted at 290 nm, 50% at 313 mn and 70% at 365 nm. Agin et al. (1981a) studied
changes in optical properties of the epidermis of six to eight Skh-1 albino and Skh-2
pigmented (ears and tails) hairless mice irradiated dorsally with a single, 125-h exposure to
a UVA source (GE F8T5-BL) with and without a 3-mm glass filter. When unfiltered, 1.4%
of the radiation was < 320 nm and when filtered, 0.12% was < 320 nm. The mid-back
OTHER RELEVANT DATA 165
(whole epidermis) was examined by forward scattering absorption spectroscopy (250-400
nm) at 48 h, 96 h, nine days and 23 days. With the filtered source, there was an increase in
absorbance across the spectrum at 48 h, and the absorption spectrum was similar to that of
control skin. Transmission returned to the control baseline by 23 days. With the unfiltered
source, there was a smaller increase towards baseline absorbance at 48 h. With time, there
was a general decrease in absorbance, except at 250-280 nm at which there was an increase
at nine and 23 days. At 23 days, the spectrum had not returned to baseline level, despite a
normal histological appearance.
de Gruijl and van der Leun (1982a) studied the effect of repeated exposure to UVR on
epidermal transmission in Skh-hr 1 hairless albino mice. Groups of 11-40 mice were
exposed to daily doses of UVR ranging from 0.11 to 1.9 kJ/m
2
from Westinghouse FS-40
sunlamps. Transmission measurements corrected for fluorescence of the epidermis were
made at 313, 302 and 297 nm. After six weeks' exposure, the higher daily doses resulted in
decreased transmission at all wavelengths. The optical density (the negative logarithm of
transmission) ratios for the three wavelengths were fairly constant with each dose. There
was a simple linear relationship between duration of treatment, increased optical density at
297 nm and epidermal thickness, measured microscopically from frozen sections, which
indicates that increased optical density is a result of UVR-induced epidermal hyperplasia.
These data show that UVR-induced changes in epidermal transmission may modify the
UVR dose-response relationship for skin cancer.
(c) Epidermal chromophores
The influence of chromophores on the optical properties of the epidermis has been
reviewed by Anderson and Parrish (1981). The main chromophores are urocanic acid (A,nalV
277 nm at pH 4.5), DNA (A-max, 260 nm at pH 4.5), the aromatic amino acids tryptophan
0-max, 280 nm at pH 7) and tyrosine 0-max, 275 nm at pH 7), and melanins (Morrison, 1985).
Urocanic acid is the deamination product of histidine and is present in human and
guinea-pig epidermis (mainly stratum corneum) at about 35 11glcm
2
dry weight. It exists in
two isomers, trans (E) and cis (Z); the trans-isomer is converted to the cis-isomer upon UV
irradiation. The absorption spectra of the two isomers are virtually superimposable, but the
extinction coefficient of the cis isomer at Amax is 20% lower (Morrison, 1985). Norval et al.
(1988) quantified urocanic acid isomers in mouse (C3Hf Bu/Kam) skin during development
and after exposure to UVB radiation. Fetal dorsal mouse skin had a low total urocanic acid
content, which increased in neonatal and older animals. Exposure to UVR increased the
proportion of the cis-isomer within 16 h from 4.7% in nonirradiated mice to 31%, and this
was maintained for days (16% after seven days). The photostationary state for in-vivo
isomerization in guinea-pig skin is 45% cis-/55% trans-isomer (Baden & Pathak, 1967).
DNA is not present to any extent in the stratum corneum of guinea-pigs (Suzuki et al.,
1977). Bruls et al. (1984a) attributed the differences in transmission minima between whole
epidermis (265 nm) and stratum corneum (275 nm) in humans to the lack of DNA.
Absorption by protein occurs throughout the epidermis.
Melanins are stable protein polymers packaged in melanosomes, produced by mela-
nocytes and transferred to keratinocytes. Melanins absorb broadly over the UV and visible
spectrum although they are not neutral density filters of the skin. For example,
3,4-dihydroxyphenylalanine ( dopa)-melanin shows a steady decline in optical density
between 210 and 340 nm (Anderson & Parrish, 1981). There is no significant racial
difference in the number of melanocytes/unit area of a given body site (Szabo et al., 1972),
so that differences in the transmission properties of black and white skin are believed to be
due to differences in melanin content and in the packaging and distribution of melanosomes
in the epidermis (Kaidbey et al., 1979).
(b) Enhancement of epidermal penetration of ultraviolet radiation
Prolonged exposure of skin to water increases sensitivity to UVB. This effect is thought
to be due to the removal of UVR-absorbing compounds, especially urocanic acid, from the
stratum corneum (Anderson & Parrish, 1981 ).
Spectral remittance at 300-400 nm has been measured in normal and psoriatic white skin
after the application of mineral oil. No effect was observed in normal skin, but remittance
in psoriatic skin was reduced within seconds after application of oil, implying greater
transmission (Anderson & Parrish, 1982). A similar enhancement of transmission was
proposed to explain the observation that topicall y applied arachis oil enhances tumori-
genesis by solar-simulated radiation in hairless albino mouse skin (Gibbs et al., 1985).
4.1.2 Eye
(a) Humans
Boettner and Wolter (1962) measured transmission of direct and forward scattering UVR
(220-400 nm) in the cornea, aqueous humour, lens and vitreous humour from nine freshly
enucleated normal eyes. There was no corneal transmission of < 300 nm, beyond which the
transmission spectrum showed a very steep increase to about 80% transmission at 380 nm
(the curve was almost vertical between 300 and 320 nm). Aqueous humour transmitted >
220 nm, with a steep rise to 90% transmission at 400 nm and no evidence of scattering. In a
young ( 4.5-year-old) lens, transmission started at 300 nm with a peak at 320 nm, declining
sharply to no measurable transmission between 370 and 390 nm; thereafter, it showed a
steep increase. A similar but slower pattern was reported for two older lenses (53 and 75
years old), with greater light scattering. Transmission in the vitreous humour began at 300
nm with a steep increase to 80% transmission at 350 nm. Lerman (1988) showed that
transmission of UV at 300-400 nm in normal human lenses decreases with age between
three days and 82 years. A review by Sliney (1986) stated that 1% of incident radiant
energy in the 300-315 nm range reaches the human retina early in life.
Kinsey (1948) measured transmission of direct UVR [no mention of instrumentation to
detect scattering] in the corneal epithelium, whole cornea, aqueous humour, lens and
vitreous humour of young adult albino rabbits. The cornea, aqueous and vitreous humor
absorbed virtually all radiation at < 300 nm; the lens absorbed > 90% radiation at wave-
lengths < 370 nm.
Bachem (1956) measured absorption ofUVR at 293-435 nm by the lens and cornea from
rabbit eyes. Few technical details were given, but the author indicated that scattering was
taken into account. The cornea absorbed all radiation at 293 nm, and the lens absorbed
all radiation < 334 nm. Calculation of absorption by the lens in situ gave a maximum at 365
nm, with little or no absorption at> 400 and< 300 nm.
Ringvold (1980) studied the absorption of UVR at 200-330 nm by cornea from young
adult albino rabbits, rats, guinea-pigs and domestic cats. In contrast to the results of other
studies, the comea did not completely absorb wavelengths < 300 nm; depending on the
species, absorption at 300 nm ranged from about 30 to 80%. [The Working Group noted
that this discrepancy cannot be explained by scattering, as presumed failure to take its effect
into account would overestimate absorption.]
4.2 Adverse effects (other than cancer)
This section deals generally with adverse effects of UVR; however, beneficial effects
also occur in humans. The vitamin D3 precursor, previtamin D3, is formed in the epidermis
and dermis through the photochemical action of UVB (Holick et al., 1980). The total daily
requirement of vitamin D3 (cholecalciferol) is supplied in most people by the combination
of synthesis in the skin and contribution from dietary sources of animal origin. Older people
are at particular risk for developing vitamin D3 deficiency, patily because the capacity for
its formation decreases with age (MacLaughlin & Holick, 1985). The sunscreen
para-aminobenzoic acid efficiently blocks the photosynthesis of previtamin D3 in the skin
(Matsuoka et al., 1987). It has been estimated that exposure of the cheeks for 10-15 min in
the midday sun in Boston, USA, would be sufficient to provide the daily requirement of
vitamin D.
4.2.1 Epidermis
(a) Humans
The most prominent acute effects of UVR on human skin are erythema ('sunburn') and
pigmentation, with cellular and histological changes.
(i) Erythema and pigmentation (sunburn and suntanning)
Dose-response curves for erythema were constructed for four radiation wavelengths,
254, 280, 300 and 313 nm, by Fan and Diffey (1985); the erythemal response on the back
was assessed quantitatively by a reflectance instrument. At 254 nm, erythema was maximal
approximately 12 h after irradiation at doses up to about five times the MED. At higher
doses, erythema was more persistent, with little change in intensity from about 12 h to at
least 48 h after irradiation.
At 313 nm, with doses around the MED, the maximal response was seen 7 h after
irradiation; with doses of two to three times the MED, the maximal response occurred at
about 4 h. The MED at 254 and 280 nm was substantially lower than that at 300 and 313
nm; however, the slopes of the dose-response curves for erythema with 254 nm and 280 nm
radiation were much flatter than those at 300 nm and 313 nm (Farr & Diffey, 1985).
The time-course of UV A erythema following inadiation with a high-intensity UV A
source (predominantly 360-400 nm) was found to be biphasic. Erythema, which may be due
to heat, was present immediately. It was minimal at about 4 h then increased between 6 and
24 h. The intensity of the early phase was dose-rate dependent, whereas the intensity in the
latter phase depended on dose only. The slope of the log dose-erythema response to UVA at
24 h did not differ from that to UVB (Diffey et al., 1987).
A number of variables affect the observation of erythema. including anatomical site,
time of observation after irradiation, size of irradiated area, method of recording erythema
and season (Diffey, 1982).
The pharmacological changes that may be responsible for erythema have been studied.
Plummer et al. (1977) examined suction blisters raised on UVB-intlamed human abdominal
skin. Bioassayable prostaglandin activity was elevated 6 and 24 h after irradiation, and
levels of prostaglandin F
2
a, measured by radioimmunoassay, were elevated at 24 h: levels
had returned to normal at 48 h, but erythema persisted. Greaves et al. (1978) extended these
observations. Following UVC irradiation, arachidonic acid and prostaglandin E
2
and F
2
levels were elevated at 6 hi reached a maximum between 18 and 24 h, when erythema was
most intense, but returned to control levels by 48 h, at which time the erythema had
subsided. Indomethacin substantially reduced blood flow, with a good correlation between
the reduction in visible erythema and prostaglandin E
2
and F
2
activity in irradiated skin.
The results are compatible with the view that UVC-induced erythema is mediated by
products of arachidonic acid metabolism. Changes in UVB-induced erythema were similar
to those with UVC at 24 h, but by 48 h the levels of arachidonic acid and of metabolites had
returned to normal, although erythema persisted. Further, although indomethacin
suppressed prostaglandin f01mation, it altered blood flow only slightly, indicating that other
factors must play an important role in inflammation following UVB irradiation. Elevated
histamine levels have also been observed, but antihistamines have little effect in diminishing
erythema (Gilchrest et al. , 1981 ).
Increased pigmentation of the skin by UVR occurs in two distinct phases: immediate
pigmentation and delayed tanning (Hawk & Parrish, 1982; Gange, 1987), immediate
pigmentation, thought to result from oxidation and redistribution of melanin in the skin,
begins during irradiation and is maximal immediately aftetwards: it occurs following
exposure to UV A and visible light and may fade within minutes or, after greater doses to
people with darker skin, may last up to several days. Delayed tanning is induced maximally
by exposure to UVB and becomes visible about 72 h after irradiation. It is associated with
an increase in the number of melanocytes as well as with increased melanocytic activity,
elongated dendrites, increased tyrosinase activity and increased transfer of melanosomes to
keratinocytes. Small freckles may be formed, particularly in fair-skinned individuals.
Not all pigmentary changes induced by UVR are localized at the site of irradiation.
Experimental exposures to UVB three times a week for eight exposures at the MED
increased the number of melanocytes and produced larger, more dendritic melanocytes in
both exposed skin and, to a much lesser extent, areas of skin shielded from the radiation.
The increase in melanocyte number in both exposed and covered areas was greater in
individuals whose melanocyte density was lower prior to exposure than in individuals with a
high initial density (Stiemer et al., 1989).
The erythemal and tanning responses of human skin are genetically determined. Res-
ponses to a first seasonal exposure of about 30 min to the midday sun have been used as part
of the basis for a skin type classification for white-skinned people ranging from Celtic to
Mediterranean (Morison, 1983a; Pathak et al., 1987):
Skin type I Always bum, never tan
Skin type II Usually burn, tan less than average (with difficulty)
Skin type III
Skin type IV
OTHER RELEVANT DATA
Sometimes mild bum, tan about average
Rarely bum, tan more than average (with ease)
169
UV A radiation produces immediate changes in melanocytes in white-skinned people. In
individuals with type-II skin, multiple pinocytotic vesicles, larger vacuoles, swelling and
partial-to-total dissolution of the itmer membranes of mitochondria and numerous small
vesicles associated with an enlarged Golgi apparatus were seen with doses that did not
produce immediate pigment darkening (Beitner & Wennersten, 1983). In those with
type-III skin, similar changes occurred but only with doses that produced immediate
pigment darkening (Beitner, 1986).
Three Japanese skin types have been described on the basis of personal reactions to the
sun (Kawada, 1986). Experimental exposure to monochromatic UVR showed that the MED
correlated well with skin type. Immediate tanning occurred but was not related to skin type.
After irradiation with the minimal dose that would produce immediate tanning, the tan faded
within 3-15 mini after greater exposures, the tan remained longer but never for more than 60
min. The action spectrum for immediate tanning had a maximum at 320 nm and decreased
gradually towards 400 nm. New pigment formation (delayed tanning) after exposure to 290
nm and 305 nm radiation began about 65 h after irradiation and increased until it reached a
maximum at 124 h (with a dose four times the MED) or 151 h (with a dose eight times the
MED). Following a dose three times the MED, some delayed tanning was still evident after
two months. The minimal melanogenic dose (producing delayed tanning) was greater than
the MED for all Japanese skin types, in contrast to findings in white Caucasians.
Parrish et al. (1981) showed that repeated dai ly exposure to doses of broad-band UVB
and UV A lower than the MED lowered the threshold for both erythema and true melano-
genesis for several subsequent days; the threshold for melanogenesis was decreased to a
greater extent than that for erythema, a separation that was more pronounced for UV A than
for UVB radiation.
(ii) Pigmented naevi
Exposure to the sun appears to stimulate the occurrence and behaviour of acquired
pigmented naevi. Kopf et al. (1985) showed, in 80 consecutive patients with dysplastic
naevus syndrome, that the concentration of naevi on areas of the thorax protected relatively
well from the sun was substantially lower than that on areas exposed to the sun. Augustsson
et al. (1990) showed that, in melanoma cases as weld as in controls, the concentration of
common naevi was higher on the sun-exposed skin of the back than on the protected skin of
the buttocks. An Australian study compared naevi excised in summer to those excised in
winter in Western Australia. Inflammation, regression, mitotic activity and lymphocytic
infiltration were significantly more prevalent in naevi excised in summer than in winter
(Holman et al., 1983b; Armstrong et al., 1984). [The Working Group noted that these
observations may be confounded by the site of the naevi.]
In an Australian cross-sectional study of 511 people, the presence of palpable naevi on
the forearm was associated with female sex, young age, not having southern European
grandparents, being born in Australia and intermediate categories of variables indicating sun
exposure (Armstrong et al., 1986).
Gallagher et al. ( 1990a,b) studied risk factors for common naevi in school children in
Vancouver, British Columbia, Canada. The number of naevi increased with age (from six to
18 years). Naevi occurred most commonly on intermittently than on constantly exposed
parts of the body and less commonly in skin that was rarely exposed. Light and freckled
skin, propensity to burn rather than tan upon exposure to the sun and a history of frequent or
severe sunburn were associated with a large number of naevi.
Green el al. (1988b) compared the prevalence of melanocytic naevi (benign pigmented
moles) in children aged 8-9 in Kiddermister, United Kingdom, and Brisbane, Australia.
Regardless of skin colour, the mean number of naevi was at least five times larger in the
Australian children than in the British children. In both populations, naevi were more
prevalent in children with fair skin.
(i ii) Ultrastructural changes
Jones, S.K. et al. (1987) and Roth et al. (1989) each described a patient who developed
many freckle-like lesions on all exposed sites following repeated exposure to high-dose
UVA from a home sunbed for tanning the skin. Biopsy showed increased numbers of large
melanocytes in the basal layers.
Rosario et al. (1979) examined the sequential histological changes produced by single
exposures to UV A, UVB and UYC radiation on untanned skin of the lower back.
Exposures were designed to cause approximately equal degrees of erythema. Following
UVB and UVC, dyskeratotic cells ('sunburn cells') were scattered throughout the malpighian
layer of the epidermis at 24 and 48 h. By 72 h and seven days, they formed a continuous
band in the upper malpighian layer or the stratum corneum. Epidermal hyperkeratosis,
parakeratosis and acanthosis appeared concurrently at 72 h. The granular layer was focally
absent at 24 and 48 h and had increased focally at 72 h and seven days. There was a
minimal-to-moderate lymphocytic infiltrate in the dennis which was most pronounced after
48-72 h. Infrequent mitotic figures were observed in keratinocytes. UV A caused fewer
dyskeratotic cells at all time intervals, and these never coalesced into a band. UV A,
however, elicited the greatest degree of inflammation at 24, 48 and 72 h in terms of both
quantity and depth of cellular infiltrate. Endothelial cell swelling, nuclear dust and
extravasation of red blood cells were generally observed together. These dermal findings
were more pronounced at 72 h. Neither epidermal hyperkeratosis, parakeratosis nor
acanthosis was observed. Intracellular oedema of moderate degree was noted with all
wavebands at all time intervals. The authors considered that the production of more
prominent dermal changes by UV A than by UVB and UVC might be related to greater
penetration of longer wavelengths. The histological changes returned to normal earliest
after UVB and latest after UV A irradiation.
Pearse et al. (1987) examined the effects of repeated irradiation with UVB (0.5, 1 and 2
times the MED three times a week for six weeks) and UV A (6 J/cm
2
[60 kJ/m
2
] three times
a week for three weeks). UVB irradiation at twice the MED led to significant increases in
epidermal thickness, stratum corneum thickness and keratinocyte height, as did UV A irra-
diation. Both UV A and UVB significantly increased glucose-6-phosphate dehydrogenase
activity and decreased succinic dehydrogenase activity throughout the epidermis. The auto-
radiographic labelling index was significantly increased following the highest dose of UVB.
The benign skin changes attributed to sunlight and seen on physical examination include
wrinkles, atrophy, cutis rhomboidalis nuchae (thick, yellow, furrowed skin, particularly on
the back of the neck), yellow papules and plaques on the face, colloid milium (firm, small,
yellow, translucent papules on the face, forearms and hands), telangiectasia, diffuse
erythema, diffuse brown pigmentation, ecchymoses in sun-damaged areas, freckles, actinic
lentigo (large, irregular, brown areas), Favre-Racouchot syndrome (yellow, thick come-
dones and follicular cysts of the periorbital, malar and nasal areas) and reticulated pig-
mented poikiloderma (reddish-brown reticulated pigmentation with telangiectasia and
atrophy and prominent hair follicles on exposed chest and neck) (Goldberg & Altman,
1984). Although most commonly seen in fair-skinned Caucasians, these changes may also
be seen in Chinese heavily exposed to the sun (Giam, 1987). A visual system using facial
photographs has been developed to enable grading of the degree of elastosis (Cameron et
al., 1988).
Holman et al. (1984a,b) made silicone rubber moulds of the microtopography of the skin
of the hands of 1216 subjects and developed a grading system to describe alterations in skin
surface characteristics observed under a low-power microscope. Using multivariate
analysis, independent risk factors for topographic evidence of actinic skin damage were:
male sex, age, tendency to bum upon exposure to the sun and outdoor occupation. Similar
results were reported by Green ( 1991 ).
Everett et al. (1970) reported ultrastructural changes in the epidermis of six elderly,
fair-skinned, freckled, blue-eyed, Caucasian male farmers with a history of multiple actinic
keratoses and skin cancers. Light microscopy showed effacement of epidermal rete ridges
and an irregular decrease in epidermal thickness in areas of skin exposed to sunlight. Three
groups of changes were apparent upon transmission electron microscopic examination:
firstly, local areas of degeneration involving groups of adjacent cells, with degenerative
changes resembling dyskeratosis in both the basal and the spinous layers of the epidermis;
secondly, disturbed cellular cohesion, with variable numbers, distribution and degrees of
maturity; and thirdly, changes in epidermal pigment- with the melanin concentration
varying from none to excessive-and melanosome complexes that were often abnormally
large.
Kligman (1969) described the changes in elastic tissue (elastic hyperplasia or actinic
elastosis) seen in the dermis of sun-exposed Caucasian facial skin. Such changes were quite
advanced before the extent of the damage became visible clinically. Some elastic
hyperplasia was seen in elderly blacks over the age of 70, but the changes were markedly
less extensive than those seen in whites.
Bouissou et al. (1988) studied elastic fibres in protected skin and skin highly exposed to
the sun from normal Caucasians of different ages, using light and electron microscopy. In
skin exposed to the sun, there was elastotic degeneration in the reticular dermis and
progressive thickening and curling of the elastic fibres in the upper dermis. Altered fibres
progressively formed thick, irregular masses, with clumps of amorphous, granular, elastotic
material and large areas of uneven staining appearing frequently thereafter. Electron
microscopy revealed that normal collagen and elastotic material were often contiguous but
never continuous.
(iv) Keratosis
The occurrence of keratosis, a benign but probably premalignant squamous neoplasm of
the skin (Marks et al., 1988), has been studied in relation to exposure to sunlight in several
cross-sectional studies.
Chronic solar damage (assessed by cutaneous microtopographs and paraocular photo-
graphs) was associated with keratosis, after adjustment for age, in a study of 1216 people in
Busselton, Australia (Holman et al., 1984a). A similar association between cutaneous
microtopography and prevalence of keratosis was observed by Green (1991) in a study of
1539 people in Nambour, Australia.
Vitasa et al. (1990) conducted a study of 808 white watermen in Maryland, USA. The
prevalence of keratosis was 25%. The risk factors for this condition were found in a
multivariate analysis to be age, individually estimated cumulative exposure to sunlight, blue
eyes, childhood freckling and a tendency to sunburn.
Marks et al. (1983) studied 2113 adults in Maryborough, Australia. The prevalence of
keratosis was 56.9%. Adjusted for age, the prevalence of keratosis was significantly
associated with being born in Australia, with a tendency to sunburn and not tan and with
blue eye colour. In another survey by these authors, of 2000 adult in-patients from a
hospital in Melbourne, Australia, the prevalence of keratosis on the light-exposed areas of
the head and neck, forearms and back of hands was 37.7%. Prevalence of keratosis was
significantly associated with age and with being born in Australia and, among men, with
outdoor occupation (Goodman et al., 1984). The Melbourne and Maryborough populations
were compared further by Marks and Selwood ( 1985), who attributed the higher prevalence
of keratosis in Maryborough to the fact that this population had a 14.2 higher erythemal
UVR level.
Foley et al. (1986) studied 766 consecutive patients with keratosis. Lesions on the hands
and forearms in men were seen more often on the right side than on the left, which the
authors attributed to the higher exposure of the right side while driving an automobile. In
women, more lesions of the head and neck were on the left side.
(v) Photosensitivity disorders
Abnormal reactions to solar radiation, termed photosensitivity disorders, occur in a
relatively small number of exposed individuals; these have been reviewed comprehensively
(Harber & Bickers, 1981; Bernhard et al., 1987). Genetic and metabolic diseases that
maybe associated with photosensitivity include xeroderma pigmentosum, phenylketonuria,
Bloom's syndrome, Cockayne's syndrome, Rothmund-Thomson syndrome, certain
porphyrins, Hartnup syndrome and pseudoporphyria cutanea tarda. The excision repair
disorders are discussed on pp. 191-194. Defects in pigmentation due to an absence of
melanocytes (vitiligo) and defective functioning of melanocytes (albinism) also confer
susceptibility to UVR because of failure to develop photoprotection through tanning
responses.
In idiopathic photodermatoses, the primary abnormality is an acquired alteration in
reaction to sunlight. The commonest form is polymorphous light eruption, in which indi-
viduals who previously tolerated sun exposure develop itchy papules, vesicles or erythe-
matous patches or plaques on exposed areas after moderate exposure to the sun (Bernhard et
al., 1987). Other photosensitivity conditions include solar urticaria (Armstrong, 1986),
hydroa vacciniforme (hydroa aestivale) (Halasz et al., 1983) and actinic reticuloid (Bernhard
et al., 1987).
Photoaggravated dermatoses are conditions that may occur in the absence of exposure to
sunlight but can be induced or exacerbated by such exposure. The commonest is
recurrences of hel pes simplex viral eruptions, usually on the upper lip; this viral infection
has been reproduced by exposure to artificial sources ofUVR (Spruance, 1985).
Other skin diseases reported to be photoaggravated include lupus erythematosus, Darier's
disease, acne vulgaris, atopic dermatitis, bullous pemphigoid, disseminated superficial
actinic porokeratosis, erythema multiforme, lichen planus, pellagra, pemphigus, pityriasis
alba, pityriasis rubra pilaris, psoriasis, acne rosacea, seborrheic dermatitis and transient
acantholytic dermatitis (Grover's disease) (Bernhard et al., 1987).
A gin et al. (1981 b) found that single exposures to UV A plus UVB caused thickening of
the whole epidermis and stratum corneum in pigmented and albino hairless mice.
Sterenborg et al. (1986) found similar changes after repeated exposures to mainly UVB in
hairless albino mice.
C57B1 mice irradiated with UVB daily for 10 days had a four-fold increase in the
number of epidermal melanocytes, with increased pigmentation and local thickening of the
epidermis (Rosdahl, 1979). A gradual, delayed, three-fold increase in the number of mela-
nocytes also occurred in shielded contralateral ears, without increased pigmentation or epi-
dermal thickening.
Generally consistent observations have been reported on chronic changes (photoageing)
in hairless mice (Bissett et al. , 1987, 1989; Kligman, 1989). Bissett et al. (1987) described
the progression of chronic UV damage to the skin in albino hairless Skh:Hr-1 mice
irradiated with UVB or UVB plus UVA three times a week for 16 weeks, with a 17-week
recovery period. UVB and a combination of UV A and UVB produced similar changes. An
early increase in transepidermal water loss was seen, with a doubling of skin thickness and
changes in the microtopography of the skin surface with visible skin wrinkling. Dose-
dependent histological changes were seen, with thickening and hyperplasia of the epidermis.
Dermal elastic fibres thickened and proliferated throughout the upper dermis, and there was
a proliferation of fibroblasts, sebaceous cysts and dermal cysts in the upper dermis. By
week 16, the skin was clearly elastotic, with thick, tangled masses of elastic fibres in the
dermis. Use of a broad-spectrum sunscreen product with a claimed SPF (skin protector
factor) of 15 retarded but did not completely prevent the effects of UVB and of UVB plus
UV A radiation. Animals exposed to UVB and then allowed to recover for 12 weeks
exhibited a zone of clearance of all abnormal elastin from the dermal-epidermal junction to
mid-way down the dermis.
Animals exposed to UVA alone for 33 weeks with a recovery period of 18 weeks
(Bissett et al., 1987) exhibited a different pattern of changes. Epidermal thickening
occurred at a slower rate, there was no increase in water loss; and sagging rather than
wrinkling of the skin occurred. There was a very gradual increase in cellularity; focal areas
of collagen damage and absence of elastic fibres were seen; the size and number of dermal
cysts increased; and there was only slight evidence of recovery after 18 weeks. UV A
appeared to accelerate several changes similar to those that occur with chronological ageing
in mice. Using a dual grating monochromator, Bissett et al. (1989) examined the action
spectra for these changes. Most were similar and occmTed in the UVB waveband:
wrinkling, glycosaminoglycan increase, collagen damage, elastosis, epidermal thickening,
dermal cellularity and dermal inflammatory cell increase. In contrast, the spectrum for skin
sagging was very broad, with a maximum near 340 nm. These results suggest that more
than one chromophore is involved in UV -induced chronic skin changes.
High doses of UVA (cumulative dose, 3000 J/cm
2
) were reported to produce severe
elastic fibre hyperplasia, but no large aggregates of elastosis or destruction of collagen, in
female Skh-hr 1 albino mice (Kligman et al., 1985; Kligman, 1989). A dose of 13 000 J/cm
2
from a filtered (50% cutoff at about :S 345 nm) UV A source, however, produced only
insignificant changes. Dose-response studies with another UV A source, filtered to remove
all radiation below 340 nm, produced some elastin thickening at a total dose of 8000 J/cm
2
as well as increased epidermal proliferation and increased and enlarged dermal cysts
(Kligman et al., 1987).
Kligman and Sayre (1991) found that the action spectrum for elastosis in albino hairless
mice was similar to that for erythema, except that longer UV A wavelengths ( > 330 nm)
were less effective for elastosis.
The chronic effect of repeated UV irradiation was also investigated in naked albino Ng/-
mice using high total doses ( > 20 000 J/cm
2
) from a predominantly UV A source (but con-
taining some UVB) administered for 16 h daily for 8.5 months (Berger et al., 1980a).
Dermal changes simjlar to those seen in human actinic elastosis were observed. There was
endothelial swelling of dilated small capillary vessels and slight perivascular infiltration.
Particularly in the upper dermis, collagen was replaced with an amorphous material that
stained faintly with haematoxylin-eosin. Mast cells and a relatively increased number of
spindleshaped fibroblasts were found in the middle and lower dermis. Large aggregates of
numerous tangled, thickened fibres with the staining properties of elastic tissue were seen.
Electron microscopy showed that elastic fibres were increased in number and size and there
was splitting of collagen fibres. Most small blood vessels were dilated, with multiple basal
lamina. The elastic tissue changes showed no signs of regression 2.5 months after
irradiation had been discontinued, although the epitheliai changes regressed over this period.
Similar changes in elastic tissue (Berger et al., 1980b) were found after exposure to a
filtered UV A source which contained no UVB, but no alteration of collagen was observed
and inflammatory changes were absent. Electron microscopy showed changes similar to
those observed in actinic elastosis.
In female, lightly pigmented, hairless Oslo/Born mice, UVB alone produced moderate
elastosis, UVB and UVA together produced a slightly reduced degree of elastosis, but UVB
followed by large doses ofUVA produced severe elastosis; UVA alone was reported to have
no effect (Poulsen et al., 1984). In Skh:Hr 1 albino hairless mice, a combination of UV A
and UVB had additive effects (Kligman et al., 1985).
(c) Comparison of humans and animals
No direct comparison has been repotied of the optical properties of whole human and
mouse epidermis; however, the available data suggest that the absorption/transmission
spectra are of a similar general shape but have marked quantitative differences. For
example, a comparison of data on a graph of effects on human epidermis not previously
exposed to UVR (Bruls et al., 1984a) with tabulated data on mouse epidermis not previously
exposed (Sterenborg & van der Leun, 1988), generated in the same laboratory, showed that
transmission in the mouse was two orders of magnitude greater in the UVC region and one
order of magnitude greater in the UVB and UV A regions than in humans. In human and
mouse epidermis, prior exposure to UVR resulted in marked decreases in UVR
transmission. No study has been reported on mouse stratum corneum.
4.2.2 Immune response
Exposure to solar radiation and UVR can alter immune function in experimental animals
and humans. This area of research is known as photoimmunology and has recently been
reviewed (Daynes et al., 1983; Parrish, 1983; Parrish et al., 1983; Bergstresser, 1986;
Roberts et al., 1986; K.rutmann & Elmets, 1988; Morison, 1989).
(a) Humans
(i) Contact hypersensitivity (allergy)
Exposure of normal subjects to radiation in a tanning solarium which emitted mainly
UV A but also UVB radiation reduced allergic reactions to 2,4-dinitrochlorobenzene (Hersey
et al., 1983a). Halprin et al. ( 1981) and Nusbaum et al. ( 1983) found that UVB radiation
partially suppressed the development of contact allergy to nitrogen mustard in patients with
mycosis fungoides and psoriasis. Exposure to UVB was begun prior to treatment with
mustard, and the field of exposure to the chemical was included in the area exposed to
radiation, so that both a local and systemic effect may have been measured. In both studies,
the proportion of patients sensitized to mustard gas was reduced by exposure to UVB
radiation, and sensitization, when it did occur, was delayed. [The Working Group noted that
the presence of diseases known to influence the immune system makes the findings difficult
to interpret.]
Response to 2,4-dinitrochlorobenzene was diminished in sun-damaged skin in subjects
previously sensitized to the allergen (Kocsard & Ofner, 1964; O'Dell et al., 1980). UVB-
induced suppression of contact allergy to nickel and other allergens (e.g., cobalt) has also
been reported (M.0rk & Austad, 1982; Sjovall & Christensen, 1986).
Studies on the possible mechanism of suppression have focused mainly on the effects on
antigen presentation in the skin. At low doses of UVB ( < 15 mJ/cm
2
) , Langerhans' cells are
the only epidermal cells to be altered morphologically (Aberer et al., 1981 ). Depletion of
Langerhans' cells after a few exposures to UVB radiation is transient (Tjemlund & Juhlin,
1982; Scheibner et al., 1986a); however, chronic exposure to sunlight appears to result in a
sustained reduction, since fewer Langerhans' cells are found in exposed than in unexposed
skin of older adults but not of young adults (Gilchrest et al., 1982; Scheibner et al., 1983;
Thiers et al., 1984; Czernielewski et al., 1988). Pigmentation does not seem to protect
Langerhans' cells, since exposure to UVB plus UVA radiation (simulating natural UVR)
produced similar degrees of depletion of these cells in dark-skimled Australian aboriginals
and in fair-skinned people of Celtic descent (Hollis & Scheibner, 1988); Langerhans' cells
were equally affected in fair-skinned and dark-skinned people after multiple exposures to
sunlight (Scheibner et al., 1986b).
The antigen-presenting function of Langerhans' cells is also diminished after irradiation
in vivo with UVB (Cooper et al., 1985; Rasanen et al., 1989). The function returns to the
epidermis within 24 h, owing to the appearance of two cell populations that are distinct and
different from Langerhans' cells (Cooper et al., 1986). Both populations have receptors for
the monoclonal OKM5 antibody; one also has receptors for the OKMI antibody and is
possibly a dendritic cell from blood, while the other is OKM1- and is related to a subset of
blood monocytes. These cells can activate T cells in the absence of exogenous antigen and
lead to the generation of T -suppressor cells which can inhibit various immune responses.
Baadsgaard et al. (1988) showed that epidermal cells from UVB-irradiated skin can stimu-
late suppressor/cytotoxic lymphocytes. This may occur via at least two pathways: activation
of T-suppressorlinducer cells or induction of interleukin-2 production. These observations
suggest that UV -induced immune suppression is more closely related to the appearance of
OKM5+ cells in the epidermis than to the disappearance ofLangerhans' cells.
Systemic suppression of contact allergy may also result from exposure to UVR.
Granstein and Sauder (1987) exposed subjects to a MED of mainly UVB radiation and
measured levels of serum interleukin-1 activity that peaked 1-4 h after exposure and
returned to baseline by 8 h. This activity may originate from the skin. in which increased
levels have been detected after UVB irradiation (Kupper et al. , 1987; Oxholm et al., 1988;
Rasanen et al., 1989).
A recent study (Yoshikawa et al., 1990) showed that suppression of UVB-induced
contact allergy may be a risk factor for nonmelanocytic skin cancer. Approximately 60% of
normal subjects were sensitized by application of 2,4-dinitrochlorobenzene to UVB-irra-
diated skin compared to 8% of patients with a history of skin cancer. Many skin cancer
patients were also immunologically tolerant to this allergen; this was not observed in normal
subjects.
Pigmentation does not protect against UV-induced immunosuppression, since it occurs
in the same proportion of black and white people (Vermeer et al., 1991).
(ii) Lymphocytes
A single, whole-body exposure to UVB radiation which produced painful erythema
produced a transient decrease in the proportion of circulating E rosette-forming cells and in
the response of lymphocytes to a mitogen (Morison et al., 1979a). McGrath el al. (1986)
found a decrease in the proportion of circulating suppressor cells following exposure to half
the MED of UVB, although the total number ofT lymphocytes was not altered. Exposure
of normal subjects to sunlight daily for two weeks, however, produced different effects: The
total proportion of T lymphocytes was diminished owing to a pronounced drop in the
proportion of helper/inducer cells associated with an increase in the proportion of suppressor
cells in the peripheral blood (Hersey et al., 1983b). Similar changes occurred after exposure
of normal subjects to UVAplus UVB radiation (Hersey et al., 1983a). When UVB radiation
was removed by a Mylar filter (Flersey et al., 1988) or a sunscreen (Flersey et al., 1987),
most of the effect was removed. The numbers of circulating T cells and helper-T cells were
significantly reduced by exposure of normal subjects to solar lamps containing UV A (with
minimal UVB) and to fluorescent tubes emitting mainly visible light, which contained small
quantities of UVB, but the number ofT-suppressor cells was only slightly reduced. These
effects were considered to be due to the UVB radiation (Rivers et at., 1989).
(iii) Infectious diseases
Recurrent infections to herpes simplex virus types l and 2 can be induced by exposure to
UVB radiation (Wheeler, 1975; Spruance, 1985; Klein & Linnemann, 1986; Perna et al.,
1987). Presumably, local alterations of immunity, associated with extensive UV-induced
tissue damage, are responsible for this reactivation.
(iv) Photosensitive disease
An interaction between solar radiation and the immune system was first postulated on
the basis of observations that the pathogenesis of several diseases is characterized by photo-
sensitivity. Solar urticaria, photoallergy and lupus erythematosus are the main examples
(for reviews, see Morison, 1983b,c; Morison & Kochevar, 1983).
(i) Contact hypersensitivity
The first report of UV -induced suppression of contact hypersensitivity was in
guinea-pigs that received applications of a sensitizing chemical through UV -irradiated skin
(Haniszko & Suskind, 1963). This effect has since been termed local suppression of contact
hypersensitivity. Later, in studies of UV -induced tumour susceptibility in mice, it was
found that UVR could also induce systemic suppression of contact hypersensitivity when
the sensitizer is applied through unexposed skin only (Kripke et al., 1977). This occurred
during chronic treatment of mice, was transient and appeared to be due to failure of an
ejector mechanism (efferent block) of the immune response (Jessup et al., 1978). These two
phenomena, local and systemic suppression of contact hypersensitivity, are probably
mediated by different mechanisms.
Local suppression of contact hypersensitivity: Pretreatment of mice with low doses of
UVB radiation ( 100-700 J/m
2
fluorescent sunlamp radiation daily for four days) suppressed
the development of contact hypersensitivity to sensitizing chemicals (e.g., 2,4-dinitrotluoro-
benzene) applied subsequently to irradiated skin (Toews eta/., 1980; Elmets et al., 1983).
This effect was associated with generation of hapten-specific LyT -1 + T cells which
suppress the induction phase of the immune response (Elmets et al., 1983). The most
effective wavelengths are < 300 nm (Elmets et at., 1985). Local suppression of contact
hypersensitivity by UVB radiation also occurs in hamsters (Streilein & Bergstresser, 1981 ).
Several hypotheses have been explored to explain the mechanism of local suppression.
Multiple exposures to sunlight result in a striking reduction in the number of Langerhans'
cells in guinea-pigs, as detected by ultrastructural examination (Fan et al., 1959).
UV -induced alterations occur in Ia + Langerhans' cells (Streilein et al., 1980; Perry &
Greene, 1982; Gurish et al., 1983; Stingl et at., 1983), but alterations in other cells may be
involved.
Thy-1 + dendritic epidermal cells (identified by antibodies to surface markers on lym-
phocytes), found in mouse but not reported in human skin, are bone marrow-derived
lymphocytes which down-regulate contact hypersensitivity. They are not affected by low-
dose UVR, and hap/en-conjugated Thy- 1 + dendritic epidermal cells can induce tolerance
on subcutaneous injection into the footpad or after intravenous injection (Welsh & Kripke,
1990). This finding is supported by the observations (Okamoto & Kripke, 1987) that (i) the
draining lymph nodes of mice treated with low doses of UVR contained these hapten-
conjugated cells after exposure to a contact sensitizer, (ii) injection of these cells into other
syngeneic mice resulted in the generation of suppressor cells, and (iii) removal of these cells
from the lymph node cells abolished the suppression.
1-Y, Thy-1-, Ia- antigen-presenting cells, which are also resistant to low doses ofUVB
radiation and preferentially generate a suppressor cell pathway, may also be involved in
local suppression (Granstein et at., 1984; Granstein, 1985; Granstein et al., 1987; Okamoto
& Kripke, 1987).
Keratinocytes may also be involved through the production of epidermal cell-derived
thymocyte-activating factor (ETAF), which is functionally and biochemically very similar
to interleukin-1, a nonspecific helper factor necessary for activation ofT cells by antigen.
lnterleukin-1 can reduce expression of contact hypersensitivity in mice (Robertson et al.,
1987). Studies by several workers have suggested that exposure to UVR inhibits the
production of ETAF (Saucer et at., 1983) or decreases its activity (Sting! et al., 1983).
When antigen-presenting cells are exposed to UVR, their ability to activate T cells is
markedly inhibited (Tominaga et al., 1983). UV irradiation of mice induces the release of a
specific interleukin-1 inhibitor, keratinocyte-derived, EC-contra IL 1 (Schwarz et al., 1988).
Other workers (Ansel et al., 1983; Gahring et al., 1984) have found increased production of
ET AF. [The Working Group noted that differences in the radiation sources and model
systems could explain the discrepancies between the results of these studies.]
Systemic suppression of contact hypersensitivity: Systemic suppression of contact hyper-
sensitivity in mice requires a higher exposure dose ( 40-50 KJ/m
2
) than local suppression
(Kripke & Morison, 1986a)_ A dose of 8.2 kJ/m
2
at 320 nm produced nearly 50% systemic
suppression, and I 00 kJ/m
2
produced 80% suppression (Noonan et al., 1984). Like local
suppression, systemic suppression is associated with the generation of suppressor Lyt-1 + T
lymphocytes (Noonan et al., 198la; Ullrich & Kripke, 1984). The pathways leading to the
appearance of these lymphocytes are, however, probably different. Systemic suppression
has also been induced in guinea-pigs (Morison & Kripke, 1984) and in the South American
opossum, Monodetphis domestica (Applegate et al., 1989). Artificial sources of UVB
radiation and sunlight, but not UV A, induce systemic suppression of contact allergy in mice
and guinea-pigs (Morison et al., 1985).
Determination of an action spectrum for systemic suppression of contact hypersensitivity
in mice revealed peak activity in the 260-270 nm region, which is consistent with a
superficial location of the cbromophore in the epidermis (De Fabo & Noonan, 1983;
Noonan & De Fabo, 1985). Two candidate molecules, urocanic acid and DNA, have been
suggested.
Several lines of evidence indicate that abnormalities in Langerhans' cells are not in-
volved in systemic suppression, in contrast to local suppression (Lynch et al., 1983;
Morison et al., 1984; Noonan et at., 1984), and that a defect of antigen presentation is not an
initial step (Kripke & McClendon, 1986). Soluble mediators are released from irradiated
skin and may generate suppressor cells in a distant organ. Serum collected from
UV -exposed mice and epidermal cells exposed to UVR in vitro contain factors that can
induce systemic suppression (Schwarz et al., 1986). The situation is far from
straightforward, however, since a recent study indicated that multiple suppressive factors,
with different immunosuppressive properties, may be released by different wavelengths of
UVR (Kim et al., 1990). Indomethacin blocks the development of suppression (Chung et
al., 1986; Jun et al., 1988), indicating that prostaglandins may also be involved in the
pathway.
Several properties of the suppressor cells have been defmed: (i) they suppress primary
proliferative responses but not a secondary response in vitro (this is consistent with the idea
that they suppress induction of sensitization but not with the proposal that they elicit a
response in a previously sensitized animal) (Ullrich, 1985); (ii) their action is limited to
T -dependent antigens (Ullrich, 1987); and (iii) they can modulate other immunological
pathways, such as formation of anti-hapten antibodies and cytotoxic-T lymphocytes (Ullrich
et al., 1986a).
(ii) Delayed hypersensitivity injected antigens
Systemic suppression of delayed hypersensitivity was induced by UVB irradiation of
mice following injection of 2,4-dinitrochlorobenzene into the footpad (Jessup et al., 1978),
of hapten-coupled spleen cells into the footpad (Greene et al., 1979) or the ear (Noonan et
al., 1981 b) or of erythrocytes and soluble protein antigens into the footpad (Ullrich et al.,
1986b) and is associated with the generation of antigen-specific T lymphocytes. This
suppression differs from the suppression of contact hypersensitivity to topically applied
allergens because delayed hypersensitivity can be restored in UV -inadiated mice by
injection of hapten coupled spleen cells from normal mice (Noonan et al., 1981b; Kripke &
Mmison, 1985, 1986b). Furthermore, systemic injection of methylprednisolone before
immunization prevented suppression of delayed hypersensitivity but had no effect on the
suppression of contact hypersensitivity (Kripke & Morison, 1986b ).
Systemic depression of splenic antigen-presenting cell function was demonstrated in
UVB-exposed mice (Letvin et al., 1980a,b; Gurish et al., 1982). Two explanations have
been advanced: a transient redistribution of antigen-presenting cells to peripheral lymphoid
tissues in response to UV -induced inflammation (Gurish et al., 1982; Spangrude et al.,
1983) or direct damage to blood monocytes or other precursors of splenic antigen-presenting
cells as they circulate through the skin (Spangrude et al., 1983). The latter theory is
supported by the observation that immunization with hapten-conjugated splenic
antigen-presenting cells or epidermal cells exposed in vitro to UVR can induce
hapten-specific T-suppressor cells (Fox et al., 1981; Sauder et al., 1981).
The role of one of the proposed chromophores, urocanic acid, has been explored. UV-
irradiated urocanic acid (containing 74% cis-urocanic acid after 4 h) suppresses delayed
hypersensitivity to HSV -1 when injected subcutaneously or applied to the skin of mice
(Ross et al., 1986), and is thus similar to UVB radiation (Ross et al., 1987). In both
instances, phenotypically similar suppressor cells were induced (Howie et al., 1986a; Ross
et al., 1987). In addition, intravenous administration of cis-urocanic acid impairs
antigen-presenting cell function in splenic dendritic cells. These observations suggest that
trans-urocanic acid is the photoreceptor for UVB-induced systemic suppression of delayed
hypersensitivity and that cis-urocanic acid acts as an immunomodulator (Noonan et al.,
1988).
(iii) Immunology of ultraviolet-induced skin cancer
Most UV -induced turnouts in mice are highly antigenic and are rejected upon transplan-
tation into normal syngeneic recipients; however, they grow progressively in immuno-
suppressed recipients (Kripke, 1974). The specific immunological rejection of these trans-
planted tumours is mediated by cytolytic-T lymphocytes aided by natural killer and
cytotoxic-T cells (Fortner & Kripke, 1977; Fortner & Lill, 1985; Streeter & Fortner,
1988a,b ). Tumours grow in UV -irradiated recipients or primary hosts because T -suppressor
lymphocytes induced by the exposure to UVR block the normal immunological surveillance
system (Fisher & Kripke, 1977; Spellman et al. , 1977; Fisher & Kripke, 1978; Spellman &
Daynes, 1978). The function of these suppressor cells is specific in that, whereas they
prevent development of UVR-induced tumours, they do not alter the growth of chemically
induced tumours or skin allografts (Kripke & Fisher, 1976; Fisher & Kripke, 1978).
The phenotype of the suppressor cells is LyT 1 + 2-, Ia- (antibodies to surface markers on
lymphocytes), similar to that of other UV-induced suppressor cells (Ullrich & Kripke,
1984). These suppressor cells are important in the development of primary neoplasms. de
Gruijl and van der Leun (1982b, 1983) found accelerated development of UVR-induced
tumours in hairless mice that had been exposed previously to UVR at a separate site. Fisher
and Kripke ( 1982) observed that, if suppressor cells were present from the time of
commencement of exposure to UVR, the latent period for development of turnouts was
shortened and the tumour yield was increased. Thus, photocarcinogenesis in mice appears
to involve at least two UVR-inducecd alterations: (i) an alteration in DNA leading to
transformation of cells (see pp. 188-189) and (ii) a specific systemic immunological
alteration that permits expression of the turnout (Fisher & Krifke, 1977).
Suppressor cells can be induced by doses of 40-50 kJ/m of radiation from fluorescent
sunlamps (see Fig. 9c, p. 64) (Kripke & Morison, 1986a), and susceptibility to transplanted
tumours is evident long before the de-novo appearance of tumours (Fisher & Kripke, 1977).
Suppressor cells can be induced by exposure to UVC (from low-pressure mercury discharge
lamps) (Lill, 1983), UVB (De Fabo & Kripke, 1980), large doses of OVA (Morison, 1986)
and sunlight (Morison & Kelley, 1985). Wiskemann et al.. (1986) described an effect of
neutral white fluorescent bulbs. [The Working Group considered that this effect may have
been due to low levels ofUVB from this source.]
(iv) Transplantation immunity
The immune responses in graft rejection and graft-versus-host disease are complex and
directed against class I antigens of the major histocompatibility complex which are
expressed on all nucleated cells and class II Ia antigens which are expressed normally on
lymphocytes and macrophages. Lindahl-Kiessling and Safwenberg (1971) demonstrated
that UV irradiation of stimulator cells could abrogate the proliferation of responder cells in a
mixed lymphocyte reaction. Subsequent studies (Alter et al. , 1973; Bach et al. , 1977)
indicated that this effect was due to alteration of class II Ia antigens on the cells bearing
them. These initial observations have been extended to various systems.
Pre-transplant, donor-specific blood transfusions have been used to reduce the need for
post-transplant immunosuppression, with varying success. The basis for this effect is
thought to be generation of donor-specific T -suppressor lymphocytes in the host. Lau et al.
(1983) found that exposure of the blood to UVB radiation prior to transfusion greatly
enhanced this effect and permitted long-term survival of allografts of islets of Langerhans
across a major histocompatibility barrier in rats. The effect was shown to be due to
inactivation of lymphocytes by radiation) resulting in cancellation of a signal from Ia
antigen-positive cells and permitting the generation of donor-specific T-suppressor cells. A
simi lar effect was demonstrated with rat heart allografts (Balshi et al., 1985).
Deletion of Ia antigens or inactivation of cells bearing them may explain prolonged graft
survival in other systems. Exposure of mouse tail skin to UVB radiation in vitro prolonged
its survival as a graft when !-region differences only were present, but UVB had no effect in
the case of complete H-2 differences (Claas et al., 1985). Similarly, mouse corneal allograft
survival was prolonged by exposure to UVB radiation in vitro (Ray-Keil & Chandler, 1986).
Prolonged survival as grafts of r at islets of Langerhans exposed to UVB radiation in vitro
was apparently due to inactivation of dendritic cells bearing Ia antigens (Lau et al., 1984).
The model of UVR-induced systemic suppression of delayed hypersensitivity has been
extended to transplantation studies, because of the considerable potential for manipulating
the immune system in transplantation. Sensitization of mice with allogeneic spleen cells
after a single exposure to UVB radiation suppressed the delayed hypersensitivity response to
these cells and proliferation oflymphocytes from the irradiated mice in a mixed-lymphocyte
reaction; these effects are due to generation of suppressor cells specific for donor antigens
(Ullrich, 1986). Interestingly, exposure of the mice to radiation need not precede exposure
to the antigen but can be delayed up to five days after first contact with the antigen, unlike
other forms of suppression of delayed hypersensitivity (Magee et al., 1989a). Similar
observations have been made in rats, but suppressor cells were not demonstrated in the
spleen (Magee et al., 1989b). Subcutaneous injection of epidetmal cells that have been
exposed to UVB radiation in vitro can similarly cancel a delayed hypersensitivity response
in mice; this effect is associated with prolongation of skin allograft survival (Tamaki &
lijima, 1989).
Graft-versus-host disease can also be reversed by UVR. Two rat models have been
studied. Pretreatment of donor bone marrow with UVB radiation did not increase the failure
of grafts, but it prevented graft-versus-host disease in most instances (Pepino et al., 1989).
Pre-irradiation of rat skin with UVB prevented subsequent development of cutaneous
graft-versus-hcyst disease at the site of exposure (Glazier et al., 1984). In both of these
studies, an alteration ofla-bearing cells was postulated as the mechanism.
(v) Infectious diseases
Classic delayed hypersensitivity to complex protein antigens (correlated with resistance
to a number of infections) can be suppressed by exposure to UVB radiation (Ullrich et al.,
1986b).
Exposure of mice to low doses (1.3-3.4 kJ/m
2
) of UVB (less than a human MED) at the
site of intradermal infection with herpes simplex type 2 virus increased the severity of the
disease. Unirradiated mice developed only a single vesicle at the site of inoculation,
whereas irradiated mice developed zosterifonn lesions which heal ed slowly and. at the
highest dose of radiation, were lethal. At doses that increased the severity of the infections,
systemic suppression of delayed hypersensitivity to the virus due to generation of
antigen-specific T -suppressor lymphocytes was observed (Yasumoto et al., 1987). In-vitro
assays showed UVB-induced impairment of antigen presentation, which may have been due
to the presence of suppressor factors in the supernatant (Hayashi & Aurelian, 1986). Similar
results were found in a model of herpes simplex virus type 1 infections in mice (Howie et
al., 1986a,b,c; Otani & Mori, 1987). [The Working Group considered that these
experiments have not demonstrated clearly that the effect of radiation on the induction of
immunity is local , since the possibility of an indirect systemic effect has not been explored.]
Exposure to low doses of UVB radiation prevented the development of delayed hyper-
sensitivity to the protozoan, leishmania, and reduced the number and severity of skin lesions
when leishmania was inoculated at the site of exposure. Exposure to radiation did not,
however, alter the viability of the organisms or the degree of their dissemination to distant
sites-the spleen, lymph nodes and skin. Furthermore, the irradiated mice reacted to a
second, distant inoculation as if it were a primary infection, presumably because they lacked
the cell-mediated immunity that would be needed to control this second attack of the
organism (Giannini, 1986).
Exposure of mice to UVB radiation also caused systemic suppression of delayed
hypersensitivity to the yeast Candida albicans (Denkins et al., 1989), through two possible
mechanisms: one mediated by suppressor cells (detected in the spleen) triggered by
exposure to radiation prior to contact with the antigen and another which did not involve
splenic suppressor cells and was triggered by exposure to radiation following exposure to
the antigen.
(vi) Human lymphocytes in vitro
Lymphocytes are highly sensitive to low doses of UVR. UVC was approximately 10
times more effective than UVB and I 0
5
times more effective than UV A on mononuclear
peripheral blood cells in vitro (Morison et al., 1979b). Cripps et al. (1978) found that UVC
was preferentially toxic to T lymphocytes, but that T and B lymphocytes were similarly
susceptible to UVB. UV A did not appear to kill T orB cells. Exposure of mononuclear
peripheral blood cells to UVB radiation inhibited both natural killer cell activity and the
response of these cells to stimulation by a mitogen (phytohaemagglutinin) (Schacter- et al.,
1983), in the absence of any apparent change in viability. The effect on natural killer cell
activity occurred selectively at the post-binding stage of lysis (Elmets et al., 1987) and could
be virtually reversed by the addition of interleukin-2 and superoxide dismutase (Toda et al.,
1986).
Firstly, most observations have been made in experimental systems and few studies have
involved humans, and it can be only assumed that results of studies in mice can be
extrapolated to humans. Furthermore, in no instance have parallel studies in an
experimental system and in humans been performed to test this assumption. Secondly,
while most investigations of photoimmunology have focused on the effects of 'UVB'
radiation, in most studies this term refers to the emission spectrum of a fluorescent sunlamp
(see Fig. 9c, p. 64) which contains both UVC and UV A, as well as UVB radiation, besides
having little in common with the spectrum of sunlight. Fortunately, in the few studies in
which the effects of fluorescent sunlamps and sunlight have been compared in experimental
systems. similar alterations in immunity have been observed. Finally, with few exceptions,
the effect of exposure to UVR is to suppress immunity highly selectively, at least in
experimental animals. Thus, in mice, certain cell-mediated immune responses are
suppressed by UVR, whereas humoral immunity is largely unaffected. The selective nature
of UVR-induced immunosuppression has not been established in humans, but no evidence
exists to suggest that it does not apply. The importance of such selectivity is that it differs
from the forms of immunosuppression seen most commonly in humans, namely viral and
drug-induced suppression, which affect most functions of the immune system. Exposure of
humans to UVR is unlikely to cause paralysis of immune function but probably selectively
negates a few immune responses.
4.2.3 Eye
(a) Humans
(i) Anterior eye (cornea, conjunctiva)
The cornea absorbs UVC and UVB radiation (Sliney & Wolbarsht, 1980). Sunlight has
been implicated as causing nodular band keratinopathies (spheroidal degeneration and
climatic droplet keratopathy), pinguecula, pterygium, photokeratitis and photokerato-
conjunctivitis (Wittenberg, 1986). Artificial sources of UVR, including welding arcs and
germicidal lamps, cause photokeratoconjunctivitis and photokeratitis (Sliney, 1986). A
study by Taylor et al. (1989) of the association between exposure to broad-band UVR and
corneal disease in 838 fishermen in Chesapeake Bay, Maryland, USA, reported a significant
association with pterygium and climatic droplet keratopathy but a weak association with
pinguecula.
(ii) Lens
The lens absorbs radiation between 305 and 400 nm (Wittenberg, 1986). UVR produces
substantial photodamage to both the structural proteins and key enzymes of the lens (for
review, see Andley, 1987).
Taylor et al. (1988) studied the two major types of senile cataract (nuclear and cortical
cataracts) in 838 Maryland fishermen for each of whom mean annual and cumulative UVB
exposure had been assessed. High cumulative exposure to UVB and high annual exposure
to UVB were both associated with increased risk of cortical cataract, but no association was
seen with nuclear cataracts. The association between exposure to solar radiation and
cataract is also supported by studies of cataract in northern India and China and in
aborigines in Australia and by an analysis of data from the US National Health and
Nutritional Examination Survey. These studies were reviewed by Wittenberg (1986).
It has been claimed that the presence of low levels of photosensitizing compounds in
lens tissue may contribute to cataractogenesis (Lerman, 1988).
(iii) Posterior eye
The posterior eye is composed of the vitreous humour and the retina (German, 1980). In
the normal eye. solar radiation in the visible and near infrared regions (400-1400 nm)
reaches these structures. Refraction of this waveband by the cornea and lens greatly
increases the irradiance between the surface of the cornea and the retina (Sliney &
Wolbarsht, 1980).
Permanent retinal damage was observed after direct viewing of the sun and viewing of
solar eclipses and in aircraft spotters during the Second World War, but no epidemiological
study has associated retinal pathology with routine environmental exposure to sunlight
(Wittenberg, 1986). The suggestion that senile macular degeneration is related to solar
exposure was not supported by a large study of fishermen in Maryland (West et al., 1989).
(i) Anterior eye
Pitts et al. (1977) and Cullen (1980) studied the effects of exposure to UVR at 295 nm
on the corneas of pigmented rabbit eyes. The threshold dose for corneal damage was 0.05
J/cm
2
. Changes observed with a slit lamp biomicroscope included discharge, corneal debris,
haziness, granular change, epithelial exfoliation, stromal opacities and stromal haze.
Applegate and Ley (1991) showed that UVR-induced corneal opacification and neo-
vascularization of the cornea of the South American opossum M. domestica was due to
DNA damage, as these effects could be delayed by subsequent illumination with
photoreactivation light, which specifically monomerizes pyrimidine dimers.
(ii) Lens
Cataracts have been produced in pigmented rabbit eyes by exposure to UVB radiation
(Pitts et al., 1977). Cataracts were produced in young albino mice 60 weeks after irradiation
with a black light (predominantly UVA) (Zigman & Vaughan, 1974; Zigman et al., 1974).
Albino mice developed anterior lens opacities after daily exposure for one to two months to
a UVB plus UV A source (290-400 nm), but not after the source was filtered to remove
radiation < 320 nm (Jose & Pitts, 1985).
(iii) Posterior eye
The effects of solar radiation on the posterior eye have been reviewed (Wittenberg,
1986, Andley, 1987). Irradiation ol' cali: vitreous humour in vitro with visible radiation in
the presence of photosensitizers resulted in partial liquefaction, suggesting that
photogenerated active species of oxygen may damage the vitreous structure. In rabbits in
vivo, however, little liquefaction was seen, suggesting a protective mechanism in the intact
organ (Pitts et al., 1977).
Damage to the retina by exposure to sunlight may also be due to thennal effects at high
irradiances or to photochemical effects at lower irradiances. In various animals, continuous
exposure to sunlight produces a photochemical lesion involving the entire retina and
affecting both rods and cones (Young, 1988). The photopigment, rhodopsin, is the chromo-
phore for damage to the rods, while the three cone pigments are the chromophores for
cones. In monkeys, blue-light damage caused by exposure to the 400-500 nm waveband
affected the macular or paramacular region of the retinal pigment epithelium. The
chromophore involved has been postulated to be melanin; active species of oxygen appear
to act as mediators of the photochemistry (Lerman, 1980; And ley, 1987).
The limited data available indicate that the optical properties of the components of
human and animal eye are broadly similar.
OTHER RELEVANT DATA
4.3 Photoproduct formation
4.3.1 DNA photoproducts
185
A multitude of photoproducts are formed in cellular DNA by solar UVR, many of which
were first recognized after their induction by non-solar radiation at a wavelength of 254 nm.
The ratio of the different photoproducts changes markedly with wavelength. A brief
description of the photoproducts is given below, together with a note on the wavelength
dependence of formation and susceptibility to repair. Substantial information on biological
consequences is available only for cycl.obutane-type pyrimidine dimers and pyrimidine-
pyrimidone (6-4) photoproducts.
(a) Cyclobutane-type pyrimidine dimers
Shortly after the observation that thymine compounds irradiated with UVC in the frozen
state rapidly lose their absorption (Beukers et al., 1958), a dimer of thymine was shown to
be responsible for this effect, the two molecules being linked by a cyclobutane ring
involving the 5 and 6 carbon atoms (Beukers & Berends, 1960, Wulff & Fraenkel, 1961 ).
Continued irradiation leads to a wavelength-dependent equilibrium between dimer
formation and dimer splitting to reform the monomer. Dimer formation is favoured when
the ratio of dimer to monomer absorbance is relatively small (wavelengths > 260 nm),
whereas monomerization is favoured at shorter wavelengths (around 240 nm), when the
ratio is larger (Johns et al., 1962). Although several isomers of the cyclobutane-type
thymidine dimer have been isolated from irradiated thymine oligomers, only the cis-syn
isomer appears to predominate in biological systems (Ben-Hur & Ben-lshai, 1968; Varghese
& Patrick, 1969; Banerjee et al., 1988).
Cytosine-thymine ( cyt+--+thy), thymine+--+thymine (thy+-+ thy) and cytosine-cytosine
(cyt+--+cyt) cyclobutane-type dimers are also formed in irradiated Escherichia coli DNA but
dearninate to uracil+--+thymine (ura+--+thy) and uracil-uracil dimers after the acid hydrolysis
usually used in chromatographic analysis (Setlow & Carrier, 1966). Cytosine moieties in
dimers are also deaminated at a slower rate under physiological conditions that produce
uracil residues (Fix, 1986), and recent evidence obtained in bacteria suggests that the rate
may be more significant than was previously thought (Tessman & Kennedy, 1991). After
treatment at 254 nm, thy+--+thy, cyt+--+thy and cyt+--+cyt appear in irradiated DNA at a ratio of
2:1:1 (Unrau et al., 1973), but this ratio changes quite markedly at longer wavelengths, e.g.,
to 5:4:1 at 265 nm (Setlow & Carrier, 1966). At 254 nm, the relative proportion of
cyclobutane dimers was: 5'-thy+--+thy, 0.68; 5'-cyt+--+thy, 0.17; 5'-cyt+--+thy, 0.08; and
5'-cyt+--+cyt, 0.07 (Kraemer et al., 1988). Ellison and Childs (1981) showed in E. coli that
the ratio of cyt+--+thy:thy+--+thy increases from 0.75 at 254 nm to 1.5 at 313 nm then
decreases to 0.8 at 320 nm, the longest wavelength tested. At 365 nm, the longest
wavelength at which dimers have been detected, the ratio of thy+--+thy:ura+--+thy was 5-6: I
(Tyrrell, 1973). The proportion of cyt+--+cyt:thy+--+thy increased up to 300 nm, but cyt+--+cyt
was undetectable at longer wavelengths (Ellison & Childs, 1981). On the basis of these
data, the latter authors argued that the predominant dimer species formed in E. coli by
exposure to sunlight are likely to be mixed dimers of cyt+--+thy rather than thy+--+thy
( cyt+--+thy:thy+--+thy, 1.2:1 ). The ratio of formation of thy+--+thy:ura+--+thy dimers in bacterial
DNA at 254 and 365 nm is approximately 7 x 10
5
nm (Tyrrell, 1973). A similar ratio of
total dimer product formation was found in cultured human skin fibroblasts irradiated at
254-265 nm (Enninga et al., 1986)
Fisher and Johns (1976) described the photochemistry and mechanism of formation of
cyclobutane-type pyrimidine dimers in considerable detail. The mechanism of dimer
formation in the UVB region almost certainly involves direct absorption, since the action
spectrum for induction closely resembles that for the appropriate monomer for wavelengths
as long as 313 nm (Ellison & Childs, 1981 ). The mechanism of formation by longer wave-
lengths (e.g., 365 nm) has not been clarified.
Cyclobutane-type dimers can be removed from the DNA of both prokaryotic and
eukaryotic cells by the powerful excision repair mechanism that is deficient in cells from
most sun-sensitive, skin cancer-prone patients with the hereditary disease, xeroderma
pigmentosum (see Friedburg, 1984; Cleaver & Kraemer, 1989). Photoreactivation is
specific for pyr+--+pyr (pyrimidine dimers) and monomerizes them in situ via a photolyase.
Many microorganisms and higher eukaryotes contain a photolyase, but the proteins and
light-activation spectra differ from species to species. The specificity of this process has
proved a powerful tool in analysing the role of pyr+--+pyr in biological effects. For example,
the potential photoreactivation of pyr+--+pyr has been studied in a set of experiments to
demonstrate that the presence of UVC-induced pyr+--+pyr in fish can be a precarcinogenic
lesion (Setlow, 1975). More recently, the small opossum, M. domestica, has been used by
Ley and coworkers as an animal model in studies on the effects of UVR, predominantly
UVB, mainly because cells of the skin of this animal, unlike that of the mouse, contain a
photoreactivating enzyme(s). They showed that several biological effects, including
decreased hair growth, erythema and tumour formation, were suppressed by exposure to
longer wavelengths (photoreactivating light) (Ley & Applegate, 1989; Ley et al., 1991).
Considerable evidence, including the fact that photoreactivation prevents formation of
the majority of mutations induced in bacteria by UVC, shows that the argument that
pyr+--+pyr is a major premutagenic lesion is overwhelming (Doudney, 1976). Recognition
that UV induced mutagenesis in bacteria is an inducible process (see Witkin, 1976),
however, complicates this argument, since, assuming that a structure involving pyr+--+pyr
constitutes the inducing event, its elimination by photoreactivation would preclude
error-prone repair at the site of any premutagenic lesion. When all inducible functions
relevant to mutagenesis are turned on, the photoreversibility of UVC mutagenesis at several
pyr+--+pyr sites disappears (Bridges & Brownn, 1992); e.g., UV -induced mutagenesis to his+
in certain recA441 lexA5l bacteria was not photoreversible, indicating that pyrimidine
dimers are not target lesions (Ruiz-Rubio et al., 1986). This suggests that
non-photoreversible photoproducts (such as the pyrimidine-pyrimidone 6-4 photoproduct)
are the principal premutagenic lesions at dithymine sequences and that cyclobutane-type
thymine dimers are weakly mutagenic. This conclusion is consistent with the results of
other studies with single-stranded vector DNA containing cyclobutane-type (6-4) thy+--+thy
photo products at specific sites (Banerjee et al. , 1988, 1990; LeClerc et al. , 1991 ).
(b) Pyrimidine-pyrimidone ( 6-4) photoproducts
The most extensively studied non-dimer photoproduct is that formed from thymine and
cytosine. Indirect evidence (Varghese & Patrick, 1969) suggests that this structure is the
in-vivo precursor of the compound 6-4'-[pyrimidin-2'-one]thymine (thy(6-4)pyo), originally
found in acid hydrolysates of UV-inadiated DNA (Varghese & Wang, 1967; Wang &
Varghese, 1967). Some years later, a type of UV -induced photoproduct, the pyrimidine
nucleoside-cytidine lesion, was recognized in highly reiterated sequences of human DNA
(Lippke et al., 1981); this is also probably a precursor of the thy(6-4)pyo product (Brash &
Haseltine, 1982; Franklin et al., 1982). Using DNA sequencing analysis, UV photoproducts
were more frequent at the 3' end of pyrimidine runs. Although the overall ratio of 6-4
photoproducts to dimers was 15% at certain sequences, 6-4 photoproducts
occurred at approximately the same frequency as that of the cyclobutane dimer (Kraemer et
at., 1988).
Patrick ( 1977) originally reported that the action spectrum for ( 6-4) photo product
formation resembles that for cyclobutane dimer formation, although the quantum yields are
two and ten times lower than that of and formation, respectively. Using
inadiation at wavelengths as long as 334 nm, Chan et al. (1986) found that the action
spectrum for induction of hot alkali sites (presumably the thy(6-4)pyo hydrolysis product)
was also similar to that for formation. The action spectra for the induction of
thymine dimers and (6-4) photoproducts were similar from 180 to 300 nm, whereas the
action spectrum values for thymine dimer induction were about nine and 1.4 times higher or
more than the values for (6-4) photoproduct induction below 160 nm and above 313 nm,
respectively (Matsunaga et al., 1991 ).
Most xeroderma pigmentosum patients are defective in the excision of (6-4) photo-
products (Mitchell et al., 1985) and cyclobutane pyrimidine dimers (Cleaver & Kraemer,
1989). ln addition, a group of patients with trichothiodystrophy (type 3) showed a marked
reduction in the repair of (6-4) photoproducts (Broughton et al., 1990).
Glickman et al. (1986) demonstrated in E. coli that the cytosine-cytosine pyrimidine-
pyrimidone (6-4) photoproduct is highly mutagenic; however, in other studies (e.g.,
Flutchinson et al., 1988), cyclobutane dimers were shown to be responsible for the majority
of observed mutations. Assessment of the relative contributions to mutagenesis of all
dipyrimidine photoproducts will require comprehensive studies in different biological
systems with specifically designed sequences containing the appropriate photoproducts.
Both pyrimidine dimers and pyrimidine-pyrimidone ( 6-4) photoproducts appear to be
impotiant in inducing cytotoxic and mutagenic lesions in human cells, although the relative
contributions of each type remain controversial (Mitchell, 1988).
(c) Thymine glycols
A group of monomeric ring-saturated lesions of the 5,6-dihydroxydihydrothymine type
(thymine glycols) have been detected by alkaline-acid degradation in the DNA of UV-
inadiated human cells (Hariharan & Cerutti, 197 6, 1977). Alkaline-acid degradation (see
Cerutti, 1981) can be used to detect a class of structural ly related lesions rather than a single
lesion, with a yield that has been estimated to be approximately 20% of the total of ring-
saturated thymine products (tsat).
Two aspects of this class of UV photoproduct are of particular interest: frrstly, they are
closely related to a class of ionizing radiation products and are believed to arise through a
similar mechanism, i.e., indirectly via the action of hydroxyl radicals; secondly, their yield
(relative to that of other UV-induced base damage) increases with exposures in the UVB
region. Measurements in BeLa cells showed that at 265 nm the ratio of h y ~ thy to t sat was
21, whereas at 313 run the ratio decreased to 1.3 (Cerutti & Netrawali. ]979). The saturated
thymine damage induced by UV A and UVB radiation may thus be due to the effects of
active oxygen species generated via endogenous cell components. There is little evidence
pertaining to the lethal or other biological consequences of such lesions in mammalian cells,
although a glycosylase capable of repairing these lesions has been isolated from human cells
(Higgins et al. 1987).
(d) Cytosine damage
The photochemical induction of pyrimidine hydrates has been reviewed (Fisher & Johns,
1976). Significant levels of hydrates are probably formed initially by UVR; however, their
instability hampers measurement of their induction and removal in cells, and it has not been
possible to establish a cause-and-effect relationship between photohydrate induction and
biological effects in viva. Using sequencing techniques, Gallagher et al. (1989) observed
incision by human endonucleases of unidentified cytosine photoproducts that were neither
cyclobutane-type nor (6-4) pyrimidine dimers. The frequency of these two photoproducts
was two orders of magnitude lower than that of pyrimidine dimers, and the optimal wave-
lengths for induction were between 270 and 295 nm.
(e) Purine damage
Purine damage has been studied less frequently than pyrimidine damage, since the
quantum yields are at least one order of magnitude lower; however, the development of
sequencing techniques has made their detection easier (Kumar et al., 1991). Incisions
(endonuclease V) are detected at unidentified purine or purine-pyrimidine moieties after
broad-spectrum UV irradiation (Gallagher & Duker, 1986). Such damage appears to be
induced maximally in the wavelength region of 260-300 nm (Gallagher & Duker, 1989).
Although the overall yield is much lower than that of y r ~ p y r similar yields occur at
certain loci.
(f) DNA strand breaks
UVC radiation induces a lower proportion of single-strand breaks than of other photo-
products. In contrast, strand breaks are the commonest initial lesion induced by ionizing
radiation. Although strand breaks form only a minority of lesions after irradiation at wave-
lengths up to 365 nm, they become increasingly important at longer wavelengths in the solar
UV region (290-400 nm). At 313 nm, the ratio of DNA strand breakage to p y r ~ p y r induc-
tion in intact E. coli was 1:44 (Miguel & Tyrrell, 1983), whereas at 365 nm one strand break
was formed for approximately every two pyrimidine dimers (Tyrrell et al., 1974). An action
spectrum for break induction in Bacillus subtilis DNA in vivo is available (Peak & Peak,
1982). More recently, an action spectrum for single-strand breaks in human skin cells has
been determined which shows that irradiation in the presence of deuterium (which enhances
singlet oxygen lifetime) increases the number of strand breaks observed at 365 and 405 nm.
At wavelengths of 405 nm and longer, strand breaks and DNA-protein cross-links are the
only forms of photochemical damage that have been determined (Peak et al., 1987).
Between 10 and 20% of the breaks induced at 365 nm are not frank breaks but rather
alkali-labile bonds which presumably include apurinic and apyrimidinic sites (Ley et al.,
1978; Peak & Peak, 1982). The formation of breaks is strongly dependent upon oxygen at
both 313 (Miguel & Tyrrell, 1983) and 365 nm (Tyrrell et al., 1974; Peak & Peak, 1982).
Their formation in vitro at 365 nm is also quenched by free-radical scavengers. Strand
breaks are repaired rapidly by a variety of cellular mechanisms in both prokaryotes and
eukaryotes. The role of these lesions in the biological action of solar radiation is not well
understood (Tyrrell et al., 1974).
(g) DNA-protein cross-links
The photochemical addition of nucleic acids to amino acids and proteins both in vitro
and in vivo has been the subject of several reviews (Smith, 1976; Shetlar, 1980). Of the 22
common amino acids, 11 undergo photochemical addition to labelled uracil, the most
reactive of which is cysteine, and several heterophotoproducts involving cysteine have been
isolated and characterized.
Several prokaryotic and eukaryotic proteins have been cross-linked photochemically to
DNA in vitro, including DNA polymerase, RNA polymerase, helix destabilizing protein and
mixtures of proteins (Shetlar, 1980).
There is evidence that DNA-protein cross-links are formed in mammalian cells in
significant yields by wavelengths longer than 345 nm (Bradley et al., 1979; Peak & Peak,
1991). Action spectra for the formation of DNA-protein cross-links in human cells have
now been obtained. Two peaks of induction are observed: one at 254-290 nm,
corresponding to the peak of DNA absorption, and a second at 405 nm, presumably
resulting from a photosensitization reaction (Peak et al., 1985). [The Working Group noted
that DNA-protein cross-links are likely to have important consequences for cells, but no
data are available to allow evaluation of their effects in eukaryotic cells.]
4.3.2 Other chromophores and targets
In addition to DNA, many other cellular components absorb and/or are damaged by solar
UVR and may influence the biological outcome of exposure. Both informational and
transfer RNA molecules are susceptible to photomodification. Studies in insects indicate
that damage to messenger RNA may be relevant to embryonic development, but the rele-
vance of these results to mammalian systems is unclear (Kalthoff & Jackie, 1982). Detailed
results of bacterial studies on the photolability of certain components of transfer RNA
(Jagger, 1981) are almost certainly not relevant to mammalian cells. Damage to proteins
could lead to modification of the level of persistent primary damage in DNA, such that
cellular DNA repair and antioxidant pathways are compromised (Tyrrell, 1991). There is
also evidence that components of electron transport and oxidative phosphorylation, as well
as membranes and membrane transport systems, can be damaged by solar wavelengths
(Jagger, 1985). Non-DNA chromophores and targets become particularly relevant at longer
wavelengths.
(a) Chromophores
Both nucleic acids and proteins weakly absorb UV A, and, although direct photochemical
events may occur, it appears likely that the initial event in the biological effects of UV A
radiation is absorption by a non-DNA chromophore which results in generation of active
oxygen species or energy transfer to the critical target molecules. As a consequence, at long
UV wavelengths, the range of targets is extended to all critical molecules that are
susceptible to active intermediates generated by chromophores.
Most of the knowledge on relevant chromophores has been obtained from in-vitro
experiments or from studies in bacteria (Eisenstark, 1987). Indirect evidence indicates that
porphyrins play a role in the inactivation of Propionibacterium acnes by UV A (Kjeldstad &
Johnsson, 1986). It has also been shown that E. coli mutants defective in the synthesis of
8-aminolaevulinic acid are resistant to inactivation by UV A (Tuveson & Sammartano,
1986), which strongly suggests that porphyrin components of the respiratory chain act as
endogenous photosensitizers. This conclusion is supported by the finding that strains that
overproduce cytochrome were sensitive to broad-band UV A radiation (Sammartano &
Tuveson, 1987). Porphyrins are also essential to human cellular metabolism, and
overproduction of iron-free porphyrins in erythropoietic or hepatic tissues is the underlying
cause of the photodestruction of the skin seen in the group of diseases known as porphyrias.
Although direct evidence is lacking, free porphyrins and proteins containing haem (such as
catalase, peroxidases and cytochromes) are also potentially important chromophores in skin
cells from normal individuals. Many other cellular compounds which contain unsaturated
bonds, such as flavins, steroids and quinones, should also be considered potential
chromophores. Although n01mal levels of catalase (which contains haem) and alkyl
hydroperoxide reductase (which contains FAD) would be expected to exert a protective role
in bacteria (see below), overproduction of these enzymes is correlated with an increase in
sensitivity to UV A radiation in bacteria (Kramer & Ames, 1987).
Porphyrins are an important class of photodynamic sensitizers which are believed to
exert their biological action via the generation of singlet oxygen. Recent experiments have
shown that deuterium oxide (which prolongs the lifetime of singlet oxygen) sensitizes
human fibroblast cell populations to the lethal action of UV A radiation, while sodium azide
(which destroys singlet oxygen) protects them (Tyrrell & Pidoux, 1989). Although this
finding is consistent with the involvement of porphyrins in the lethality of UV A, other
cellular compounds may also generate singlet oxygen. It is also important to consider active
oxygen species that may be generated intracellularly. Not only can hydrogen peroxide be
generated by UV A irradiation of tryptophan (McCormick et at., 1976), but both superoxide
anion and hydrogen peroxide can be generated by photo-oxidation of NADH and NADPH
(Czochralska et al., 1984; Cunningham et al., 1985).
The presence of chromophores (such as psoralens) in the diet may also influence
susceptibility to damage, but this reaction is clearly subject to enormous individual varia-
bility. Accidental and deliberate application of chemical agents (such as sunscreens and
drugs) to the skin may also introduce potentially damaging chromophores.
(b) Membranes
The lipid membrane is readily susceptible to attack by active oxygen intermediates.
Many reports (e.g. , Desai et al., 1964; Roshchupkin et al., 1975; Putvinsky et al., 1979;
Azizova et al., 1980) have shown that UVR can induce peroxidation of membrane lipids.
In-vitro studies with lecithin microvesicles have shown UVR-induced changes in the
microviscosity of membrane bilayers (Dearden et al. , 1981) which are correlated with the
degree of unsaturation of fatty acid chains (Dearden et al., 1985). UVC and UV A radiation
and sunlight have been shown to cause lipid peroxidation in the liposomal membrane
(Mandai & Chatterjee, 1980). Haem proteins such as cytochrome c and catalase are known
to catalyse lipid peroxidation and peroxidative breakdown of membranes (e.g., Brown &
Wuthrich, 1977; Gofii et al., 1985; Szebeni & Toll in, 1988). A dose-dependent, linear
increase in lipid peroxidation of liposomal membranes was induced by UV A radiation,
which was inhibited to a large extent by butylated hydroxytoluene, a nonspecific scavenger
of lipid-free radicals. Since both sodium azide and L-histidine (quenchers of singlet
oxygen) led to 40-50% inhibition of peroxidation, the authors suggested that singlet oxygen
is involved in initiation of the reaction (Bose et al., 1989).
UV A irradiation of liposomes leads to lipid peroxidation in the absence of photo-
sensitizer molecules, so that singlet oxygen may arise through direct stimulation of
molecular oxygen (Bose et al., 1989). Biological membranes are, however, rich in
endogenous photosensitizer molecules, such as those involved in electron transport, and
these may contribute to the peroxidation of lipids observed in biological systems (see
Jagger, 1985). Membrane damage has long been implicated in the lethality of UV A in
bacteria (Hollaender, 1943) and almost certainly contributes to the sensitivity of
UV A-treated populations plated on minimal medium- a phenomenon which is highly
dependent on oxygen (Moss & Smith, 1981 ). Sensitivity to UV A has been related to levels
of unsaturated fat in membranes (Klamen & Tuveson, 1982; Chamberlain & Moss, 1987).
Furthermore, the presence of deuterium oxide enhances the levels of membrane damage,
sensitivity to UVA and lipid peroxidation (Chamberlain & Moss, 1987), suggesting that
singlet oxygen plays a role in all three processes. Leakage experiments have also been used
to assess UVA-induced membrane damage in yeast: again, changes in permeability
correlated well with lethality and were highly oxygen dependent (Ito & Ito, 1983). UVA
irradiation of cultured human and mouse fibroblasts led to the release of arachidonate
metabolites from the membrane in a dose-dependent fashion. The release was also
dependent on the presence of both oxygen and calcium ion and may be related to the
induction of cutaneous erythema, which is also oxygen dependent (Hanson & DeLeo, 1989).
Studies of the effects of UVR on membrane transport have been undertaken in prokaryotes
(Jagger, 1985), but no infmmation was available on the effects of UVR on eukaryotic
membrane transport.
4.4 Human excision repair disorders
4.4.1 Xeroderma pigmentosum
The commonest, most characteristic photoproducts produced in DNA by UVB and UVC
radiation involve adjacent pyrimidines. Evidence summarized above argues strongly that
these products give rise to a wide variety of alterations in DNA sequence and gene
expression. Like many other types of DNA damage, these photoproducts may be excised,
and the resulting gap in one strand can be resynthesized accurately using the undamaged
strand as a template. How this is accomplished is best understood in the bacterium E. coli,
in which a multi protein complex including the products of the uvr A, B and C genes excises
an oligonucleotide 12 or 13 bases in length containing the photoproduct. The resulting gap
is filled by a DNA polymerase (usually III), and the final ligase link to the adjacent DNA is
effected by polynucleotide ligase (Bridges et al. , 1987; Bridges, 1988; Bridges & Bates,
1990). Other gene products are involved in the process, and a more comprehensive
discussion is given by Sancar and Rupp (1983). Bacteria that have defects in the uvr A orB
genes cannot excise UV photoproducts and are 10-20 times more sensitive to killing and the
induction of mutations by UVC. They are also more sensitive to UVB and (under certain
conditions) UV A (Webb, 1977). It can be concluded that the function of excision repair is
to minimize the deleterious consequences of DNA damage, such as the persistence of UV
photoproducts.
A similar process takes place in humans. Although much less is known about the
mechanism, many genes have been shown to be involved, and these are being cloned and
the role of their products is being elucidated (Hoeijmakers & Bootsma, 1990; Bootsma &
Hoeijmakers, 1991 ). Like bacteria, humans can also be deficient in aspects of excision
repair. The prototypic example is the genetic disorder xeroderma pigmentosum, which is
actually a complex of disorders comprising at least 10 different forms of DNA repair defect
(nine excision defective complementation groups and one excision repair proficient variant
group) (Kraemer et al., 1987; Cleaver & Kraemer, 1989). The sensitivity of fibroblasts and
lymphocytes from excision-defective individuals with xeroderma pigmentosum to mutation
and lethality by UVC is up to 10 times greater than that of cells from normal individuals
(Arlett et al., 1992) and for UVR from a solar simulator (Patton et al., 1984). The
pigmentary abnormalities are confined to sun-exposed portions of the skin.
The incidences of turnouts of the skin, anterior eye and tip of the tongue in these
individuals are much higher than those in unaffected populations (Kraemer et at. , 1987), and
the median age of patients at onset of skin cancers appears to be much younger than that of
the general population. Multiple primary skin cancers are common which arise predo-
minantly on sunlight-exposed areas of the body (Kraemer et al., 1987); there is anecdotal
information that they are largely prevented if protection against exposure to sunlight is
afforded early in life (Kraemer & Slor, 1984). Studies of patients with excision-defective
xeroderma pigmentosum provide the strongest evidence that sunlight-induced photoproducts
can result (in the absence of repair) in the genesis of basal-cell carcinomas, squamous-cell
carcinomas and melanomas and strongly support the contention that they can also do so in
normal individuals in whom repair is more efficient (although probably never complete).
The photoproducts that fail to be excised in xeroderma patients are known to be produced in
human skin, not only by UVC (used in most laboratory experiments with cells) but also by
UVB, particularly by wavelengths around 300 nm (Bridges, 1990; Athas et al., 1991).
Action spectra show that the difference in the cytotoxic action of UVB on cultured cells
from normal and xeroderma pigmentosum patients is similar to that of UVC, whereas the
differences in the response to UVA are only slight (Keyse et al., 1983). The studies on
xeroderma pigmentosum illustrate that DNA repair is a major defence of the human skin
against the carcinogenic action of sunlight.
4.4.2 Trichothiodystrophy
The conclusions derived from studies of xeroderma pigmentosum have become more
complex with the availability of information on two related excision disorders. Trichothio-
dystrophy is a rare disease in which patients generally have skin judged to be sun-sensitive
by erythemal response but no indication of the pronounced freckling or elevated incidence
of early skin tumours associated with xeroderma pigmentosum (Bridges, 1990). In the
majority of cases studied, trichothiodystrophy is associated with a deficiency in the ability to
repair UV -induced damage in cellular DNA.
Three categories of response to UVR have been identified. In type 1, the response is
completely normal, whereas type-2 cells are deficient in excision repair, with properties
indistinguishable from those of xeroderma pigmentosum complementation group D. Type-3
cells survive normally after UV irradiation, and the rates of removal of cyclobutane pyri-
midine dimer sites are also normal (Broughton et al., 1990). In xeroderma pigmentosum
diploid fibroblast lines, catalase activity was decreased on average by a factor of five as
compared to controls, while heterozygotic lines exhibited intermediary responses. All
trichothiodystrophy lines tested were deficient in UV -induced lesion repair and exhibited a
high level of catalase activity; however, molecular analysis of catalase transcription showed
no difference between normal, xeroderma and trichothiodystrophy cell lines. UV irradiation
induces five times more hydrogen peroxide production in xeroderma lines than in trichothio-
dystrophy lines and three times more than in controls. These striking differences indicate
that UVR, directly or indirectly, together with defective oxidative metabolism may increase
the initiation and/or the progression steps in patients with xeroderma pigmentosum to a
greater degree than in people with trichothiodystrophy, which may partly explain the
different tumoral phenotypes in the two diseases (Vuillaume et al., 1992).
Five patients with trichothiodystrophy type 2 appeared to be in one of the xeroderma
pigmentosum complementation groups: Fibroblasts from these individuals were indistin-
guishable from xeroderma fibroblasts in the same complementation group and were equally
sensitive to the lethal and mutagenic effects ofUVC (Stefanini et al., 1986; Lehmann et al.,
1988). Two other trichothiodystrophy patients (type 3) had cells markedly defective in the
removal of ( 6-4) pyrimidine photoproducts but not cyclobutane-type dimers (Broughton et
al., 1990).
4.4.3 Cockayne's syndrome
A third sun-sensitive excision repair disorder is Cockayne's syndrome. Patients with this
condition have fibroblasts which undergo normal excision repair in the overall genome but
which are defective in the excision of dimers from DNA strands undergoing active
transcription (Mayne et al., 1988). Cockayne's syndrome cells are sensitive to both killing
and mutation induction by UVC (Arlett & Harcourt, 1983) and have reduced repair of
cyclobutane dimers; they show, however, normal repair of non-dimer photoproducts in a
UV treated shuttle vector plasmid. Like patients with trichothiodystrophy, those with
Cockayne's syndrome do not have pronounced freckling or enhanced early incidence of skin
cancers (Barrett et at., 1991).
4.4.4 Role of immunosuppression
If it is assumed that UV -induced DNA damage sustained by patients with trichothio-
dystrophy type 2 results in the same photo-induced mutations in their skin cells (including
mutations associated with the initiation of cancer) as is seen in xeroderma pigmentosum
patients of the same complementation group (D) (Bridges, 1990; Broughton et al., 1990),
something other than unrepaired DNA damage and an elevated frequency of mutations must
be needed to trigger initiated cells into clonal expansion and early tumours, as is seen in
xeroderma pigmentosum. The assumed latency of initiated cells in such trichothiodystrophy
patients may be related to the latency seen in epidemiological studies of skin cancer in the
normal population (see section 2).
The nature of the circumstances that allow initiated skin cells to develop into tumours in
xeroderma pigmentosum patients, and perhaps later in life in other individuals, is unclear.
Burnet (1971) first suggested that individuals with this disorder might be deficient in some
immunosurveillance step. Bridges (1990) proposed that theywere also hypersensitive to both
the immunosuppressive and the mutagenic action of UVR, so that the elevated skin cancer
rate in individuals with xeroderma pigmentosum would not accurately reflect the actual
increase in mutation frequency in exposed skin but would exaggerate it greatly.
4.5 Genetic and related effects
Any cell that is UV -irradiated can be expected to sustain DNA damage. The nature of
this damage is wavelength-dependent, and the major photoproducts of short-wavelength UV
irradiation are various types of dipyrimidine photoproducts, while DNA strand breakage and
DNA-protein cross-linkage occur relatively more frequently after irradiation with long-
wavelength UVR. As the wavelength is increased above 290 nm, the efficiency of
formation of pyrimidine dimers and other DNA photoproducts decreases greatly. This
wavelength dependency of response presents a fundamental problem for the quantitative
interpretation of the genetic activities of different regions ofthe UV spectrum. In most
experimental studies with UV A and UVB irradiation and, of course, simulated solar
radiation, monochromatic radiation was not used. Also, the characteristics of the radiation
emitted from the source are variable over time and from source to source. Because of these
practical considerations, comparisons of the effects seen in different studies in terms of dose
are commonly invalid: Photoproduct yield is dependent on the energy contributions from the
different wavelengths within the spectrum used, but incident doses (fluences) are measured
only as energy fluxes over the whole spectrum emitted from the source. The problem of
dosimetry within experimental systems is compounded by the fact that absorbed dose is
determined by the geometry of the system and the position of the target within it: absorption
by one layer (e.g. , the medium or a layer of cells) will affect the fluence incident upon the
layer beneath. The fluence absorbed may thus differ substantially from the incident fluency
of the system. For these reasons, it was considered inappropriate to compile quantitative
genetic profiles as is customary in these monographs.
Given the generally significant responses in many different tests for the genetic activity
of UVR in a wide range of organisms and cultured cells, the simple qualitative questions
appear to have been answered in abundance. The main issues of outstanding interest are:
identification of the types of damage induced by the valious portions of the UV spectrum;
the mechanisms by which damage is translated into mutation or other genetic changes; and
the dose characteristics of these responses.
4.5. 1 Humans
The portions of the body that receive most exposure to UVR are the skin, anterior eye
and lip. Because dermal capillaries approach the skin surface, it can be anticipated that
Blood will be exposed to the portion ofUVR (see Kraemer & Weinstein, 1977; Morison et
al., 1979a; Larcom et al., 1991) that penetrates the dermis. The biological consequences of
this exposure are unknown.
DNA damage in skin cells has been studied using three methods that are sensitive
enough to detect DNA damage after exposure to doses of UVR too low to induce erythema:
(i) use of antibodies specific for UV-altered DNA, followed by immunofluorescence.
This method can be used with immunoperoxidase staining and a secondary antibody
(Eggset et al., 1983, 1986) or without them (Tan & Stoughton, 1969);
(ii) autoradiography after tritiated thymidine incorporation (Epstein et al. , 1969, 1970;
Honigsmann et al., 1987; Wolf et al., 1988); and
(iii) treatment of extracted DNA with Micrococcus luteus cyclobutyl pyrimidine dimer
site-specific endonuclease, followed by alkaline agarose gel electrophoresis of the
single-stranded DNA fragmented at the dimer sites (Sutherland et al. , 1980;
D'Ambrosio et al., 1981; Gange et al. , 1985; Freeman et al., 1986, 1987, 1989;
Alcalay et al., 1990). This method suffers the disadvantage that damage cannot be
localized to particular layers of the skin, but dimer yield can be calculated. Methods
for the study of resolved genetic damage have not been pursued.
(a) Epidemis
(i) Broad-spectrum ultraviolet radiation, including solar simulation
Effects on DNA synthesis were demonstrated in human skin in vivo which had been
exposed to three times the MED ofUVR (< 320 nm; mercury arc lamp [Fig. 9a, p. 64]) and
then injected intradermally with tritiated thymidine (8-41 x 10
6
ergs/cm
2
[8-41 kJ/m
2
]) in the
irradiated area immediately and at 0.25, 3, 5 and 24 h subsequently. S Phase was
suppressed in cells of the basal layer at 3-h and 5-h sampling times, but not at 24 h.
Sparsely labelled cells (indicating DNA repair) occurred in greatly variable proportions
from person to person in the basal, malpighian and granular layers at 0, 0.25. 3 and 5 h, but
not at 24 h, indicating that repair was complete by 24 h (Epstein et al., 1969). DNA repair
was also reduced in the skin cells of three patients with xeroderma pigmentosum in
comparison to eight normal controls (Epstein et al., 1970).
Sutherland et al. (1980) demonstrated a dose-related response for the induction of pyri-
midine dimers after exposure to a Westinghouse sun lamp (Fig. 9c, p. 64), with 50%, energy
< 320 nm, at 0, 970, 1940 and 3880 J/m
2
. In one subject. 0.5 of the MED of sun-lamp
exposure resulted in about 6 0.6 dimers per 10
8
Da.
D'Ambrosio et al. (1981) reported that approximately 12.8 and 23.6 dimers per 10
8
Da
were induced in skin DNA in vivo following irradiation with a mercury arc lamp (200-450
nm) at 150 and 300 J/m
2
, respectively. Repair or removal of dimers was measured 0-24 h
following exposure. About 50% of the dimers were lost 58 min after irradiation, and less
than 10% remained at 24 h. In an experiment with patients with lupus erythematosus,
D'Ambrosio et al. (1983) obtained results similar to those found in the skin of nonnal
individuals.
Strickland et al. (1988) measured the induction of cyclobutane dithymidine photo-
products in human skin samples after exposure to simulated solar radiation. Tissue samples
from three non-pigmented (white) individuals were exposed to 18 or 36 kJ/m
2
UVR (0.5-1
MED), and those from three constitutively pigmented (black) individuals were exposed to
72 and 144 kJ/m
2
. Constitutively pigmented skin required doses of UVR two to four times
higher than non-pigmented skin to produce roughly equivalent levels of thymine dimers.
[The Working Group noted the small number of people studied.]
(ii) UVA radiation
Freeman et al. (1987) showed in two subjects that similar pyrimidine dimer yields were
produced in skin by a broad-band UV A source (UV ASUN 2000), by broadband UV A
filtered to remove all light of wavelengths < 340 nm and by narrow-band radiation centred
at 365 nm (xenon-mercury compact arc), indicating that UV A radiation and not stray shorter
wavelength radiation was responsible. Dimer production was observed following exposures
to 5 x 10
5
J/m
2
. Since exposure to a UV A-emitting tanning lamp results in a dose of about 5
x 10
5
J/m
2
, UV A exposure for cosmetic purposes could result in measurable levels of DNA
damage.
(iii) UVB radiation
The efficiency of UV A- and UVB-induced tans in protecting against erythema and the
formation of dimers induced by UVB was studied in five subjects by Gange et al. (1985).
The radiation sources were a UV ASUN 2000 lamp (UV A; Fig. 8d, p. 61) and an FS36 Elder
fluorescent sunlamp (UVB). UVB-induced tanning protected against erythema produced by
subsequent UVB exposure two to three times better than UVA-induced taming; however,
tanning with either UV A or UVB was associated with a similar reduction in yield of
endonuclease-sensitive sites in epidermal DNA (about 50%).
Eggset et al. (1983) observed DNA damage in both epidermis and dermis following
exposure to a Westinghouse FS-20 sunlamp(Fig. 9c, p. 64) at 0.5-2 MED (2 MED , 900
J/m
2
). The outer layers were more heavi ly damaged after smal l doses than the basal layer,
which may be better protected by its deeper location and shielding by melanin. The authors
claimed that DNA repair was well under way after 4-5 h and was apparently nearly
complete at 24 h, as judged by immunofluorescence and immunoperoxidase staining.
Repair was faster in the presence of visible light than when irradiated skin was shielded with
thick black plastic. [The Working Group noted the absence of quantitative data.]
In a study of two volunteers (Eggset et al., 1986), tanning was shown to protect against
DNA damage in skin (induced in a UVB solarium), but the conclusions were based solely
on observations of immunofluorescence. [The Working Group noted the absence of quanti-
tative data.]
Freeman et al. (1986) measured UVB-induced DNA damage in the skin of seven indivi-
duals with different sensitivities to UVB irradiation, as measured by the MED, with irra-
diation from an FS36 Elder fluorescent sunlamp (280-320 nm). The production of dimers
was correlated inversely with the MED. The slopes of the dose-response curves for the
most UVB-sensitive individual (MED, 240 J/m
2
) and for the least sensitive individual
(MED, 1460 J/m
2
) were 11.5 X 1 o-
4
and 2.6 X I o-
4
dimer sites per 1 000 bases per mJ/cm
2
[ 10
J/m
2
], respectively.
Honigsmann et at. (1987) studied unscheduled DNA synthesis in epidermal cells in the
skin of25 male volunteers (four with skin type nand 21 with skin type III; see pp. 168-169)
after exposure to doses of UVB of 0.06-6 MED, from a 6-k W xenon arc lamp (292-304
nm). The MED values ranged from 140 to 550 11m
2
The dose-response curve showed a
significant increase in unscheduled DNA synthesis between 0.06 and 1 MED but no
difference between 1 and 6 MED, suggesting a saturation of excision repair in vivo.
Freeman (1988) studied interindividual variability in 17 healthy volunteers in the repair
of pyrimidine dimers induced following exposure to 0.25-1.5 MED from a Westinghouse
FS-40 sunlamp (see Fig. 9c, p. 64). Removal of dimers was detected within 6 h of
irradiation. The average half-time for removal of dimers was 11.0 4.3 (SD) h (range,
5.5-21.1 h). [The Working Group noted that the spectra and doses used in this study were
different from those used by D'Ambrosio et al. (1981). It is not clear if the interindividual
variability is greater than the experimental error.]
Interindividual variability in the repair of UVB-induced pyrimidine dimers was also
studied by Alcalay et al. (1990) in 22 patients aged 31-84 with at least one basal-cell
carcinoma. The control group consisted of 19 cancer-free volunteers aged 25-61. Both
groups were given one MED of radiation from a 150-W xenon arc solar UV -simulated lamp
equipped with a 50-cm liquid light guide and a filter eliminating wavelengths below 295
nm. Dimers were measured immediately and after 6 h. The two groups were similar at time
0, but after 6 h, 22 4% (range about 8-64) of the dimers were removed in the cancer group
compared to 33 4% (range about 4-64) in the control group. Of the cancer patients, 23%
had repaired more than 30%, of the DNA damage, compared to 53% of the control group.
[The Working Group noted that it is not clear if the interindividual variability is greater than
the experimental error.]
Wolf et al. ( 1988) observed measurable amounts of unscheduled DNA synthesis in the
skin of 23 volunteers exposed to 0.5 MED UVB irradiation from a high-pressure mercury
lamp [spectral emission not given]. Administration of carotenoids (to reduce light
sensitivity in patients with erythropoietic protoporphyria) at a dose of 150 mg per day for 30
days did not significantly alter the amount of unscheduled DNA synthesis (6 1.2
grains/cell before and 8 2 grains/cell after carotenoid treatment; seven subjects). The
same investigation showed no significant protection by carotenoids against UV A-, UVB- or
PUV A-induced erythema, on the basis of pre- and post-carotenoid MED or minimal
phototoxic dose.
In 30 volunteers, it was demonstrated that the action spectrum for the frequency of
pyrimidine dimer formation in human skin DNA for a given fluence (incident dose) has its
maximum near 300 nm and decreases sharply on either side of this wavelength (Fig.l2).
The decrease at < 300 nm is probably due to absorption in the upper layers of skin. These
data were used to estimate that, at a solar angle of 40 , a reduction in the thickness of the
stratospheric ozone layer from 0.32 em down to 0.16 em would be expected to result in a
2.5-fold increase in dimer formation (Freeman et al., 1989).
A dose-response for the fmmation of thymine dimers in epidermal cells isolated from
human skin irradiated with UVB in vitro was determined by Roza et al. (1988) using a
monoclonal antibody.
(iv) UVC radiation
Exposure of human skin, from which the stratum corneum had been removed, to either a
germicidal (UVC) or a Hanovia hot quartz lamp in vivo resulted in DNA damage
demonstrable by immunofluorescence (Tan & Stoughton, 1969). When the stratum
corneum was intact, DNA damage was detected only after exposure to the germicidal lamp.
Fig. 12. Action spectrum for pyrimidine dimer formation in human skin () and
solar spectra at the surface of the Earth for stratospheric ozone levels of 0.32 em
(dotted line) and 0.16 em (solid line). Each point in the action spectrum represents
the slope of the dose-response line (dimer yields at three exposures) for one
volunteer at one wavelength, obtained from triplicate independent determinations.
Thirty points occur at 302 nm, although some points overlie other values; five
points occur at each other wavelength: points at 290 and 334 nm are circled to
indicate that identical dimer yields were recorded for two volunteers. ph, photon;
ESS, endonuclease-sensitive site
Action spectrum (ESS/kb/oh/cm
2
x 10
19
)
-
Q.:.
...
~
-0,
...
-~ -0.!.
< ~ ~ 7 7 . ~ ~ ~ ~ ~ ~ ~ ~
< ... ~ --------
m g -------------._
<
CD w
- ...
" 0
:J
(0 ~
-o
::::T

g
>
-o.
Solar flux (ph/cm
2
/s/nm X 1 0"
13
)
From Freeman et al. (1989)
-o_
[The Working Group noted that more sensitive analytical techniques for DNA damage are
now available.]
(b) Lymphocytes
(i) Broad-spectrum ultraviolet radiation
In addition to cells of the skin, white blood cells are also subject to exposure to UVB and
UV A, partly because some are temporarily resident in the skin and partly because it has
been estimated that the equivalent of the total blood volume circulates through the dermal
capillaries approximately every 11 min (Kraemer & Weinstein, 1977). Detecting effects,
e.g., on lymphocytes, is likely to be extremely difficult owing to the fact that they are
continually moving between the blood and other tissues; indeed, 90% of the lymphocyte
population at any given time is resident outside the blood. Thus, the concentration in the
blood of any lymphocytes irradiated while passing through the skin may fall substantially
over time after irradiation ends as they are diluted in the whole body lymphocyte pool.
Extravascular lymphocytes resident in the skin may also receive higher doses of UVR.
Nevertheless, studies have been reported of genetic or related effects on lymphocytes
sampled from peripheral blood.
Larcom et al. ( 1991) examined the capacity for DNA synthesis of lymphocytes from
eight subjects exposed in two commercial tanning salons. Blood was taken immediately
before tanning and again 24 h after tanning. System I used a sunlamp with a UVB:UVR
ratio of 0.02% for 280-300 nm and 1.4% for 300-315 nm; the output of system II (Solana
Voltarc lamp) was not indicated. There was a 24-84% (average, 53%) decrease in
phytohaemagglutinin-induced DNA synthesis with system I and a 8-58% (average, 30%)
decrease with system II.
(ii) UVA radiation
Seven of 13 psoriasis patients recetvmg oral 8-methoxypsoralen and high-intensity,
long-wave UV A radiation had reduced leukocyte DNA synthesis; this did not occur in any
of 10 controls (Kraemer & Weinstein, 1977). These results indicate that UV A reduces the
incorporation of tritiated thymidine in lymphocytes circulating through the skin.
(iii) UVP radiation
In normal, fair-skinned subjects given whole-body exposure to 1.5-3 x MED doses of
UVB from a sunlamp (280-380 nm), a dose-dependent decrease was seen in the incorpo-
ration of tritiated thymidine into DNA following stimulation by photohaemagglutinin; the
proportion of circulating lymphocytes was decreased and the proportion of null cells was
increased (Morison et al., 1979a).
These studies indicate that leukocytes should be included in any inventory of human
cells potentially exposed to solar radiation or artificial UVR.
4.5.2 Experimental systems [see Tables 32-35, in which exposures are separated according
to type ofUVR]
(a) DNA damage
Inhibition of DNA synthesis has been induced in hairless albino mouse epidermis at
wavelengths of 260-320 nm, with a maximal effect at 290 nm. Inhibition was not detected
at 335 nm (Kaidbey, 1988). The action spectrum was similar to that for formation of cyclo-
butane-type pyrimidine dimers (Cooke & Johnson, 1978; Ley et al., 1983) and pyrimidine-
pyrimidone (6-4) photoproducts in mouse skin (Olsen et al., 1989). Pyrimidine dimers
(measured as endonuclease-sensitive sites) have been measured in the corneal DNA of the
marsupial, M. domestica. following exposure to a sunlamp (280-400 nm) (Ley et al. 1988).
While DNA is the main photochromophore for UVC, there is evidence that active
oxygen intermediates are involved in the production of DNAdamage by UV A (Tyrrell,
1991 ). The production oi several types of photo lesions is oxygen dependent (Tyrrell,
1484, 1991 ). In addition, the irradiation lethality of both cultured bacterial (Webb, 1977)
and mammalian (Danpure & Tyrrell, 1976) cells is dependent on the presence of oxygen;
this observation was later linked with the production of singlet oxygen (Tyrrell & Pidoux,
1989). It has also been observed that irradiation of cultured human skin cells with UVB
(302 nm, 313 nm), UV A (334 nm, 365 nm) and visible ( 405 nm) radiation is strongly
enhanced in glutathione-depleted cells (Tyrrell & Pidoux, 1986, 1988). This apparent
protection by glutathione appears to be due to its radical scavenging properties at the stated
wavelength but may be due to induction of a more specific pathway (such as its essential
role as a hydrogen donor for glutathione peroxidase) at longer wavelengths. Francis and
Giannelli (1991) found that the abnormally high yield of single-stranded DNA breaks
produced by UV A in six UV A-sensitive human fibroblasts (three from actinic reticuloid
patients, two from sisters with familial actinic keratoses and internal malignancies and one
from a patient with an abnormally high incidence of basal-cell carcinomas) could be
reduced if sensitive cells were co-cultivated with normal fibroblasts or with radical
scavengers. They suggested that the UV A-sensitive cells had deficits of
small-molecular-weight scavengers of active oxygen species and that intercellular
cooperation allows the transfer of these substances from resistant to sensitive cells. The
presence of non-DNA chromophores that generate active oxygen species can also occur
with UVC. Melanin, normally regarded as a solar screen, has also been associated with the
fmmation of oxidative DNA damage, such as thymine glycols in mouse cells that vary in
melanin content (Huselton & Hill, 1990). A slight increase in pyrimidine dimer yield was
seen in human melanocytes as compared to keratinocytes following exposure to UVR at
254, 297, 302 and 312 nm but was significant only at 297 nm (Schothorst et al., 1991 ).
(b) Mutagenicity
Numerous reports show that sunlight or solar-simulated radiation induces mutations in
bacteria, plants, Chinese hamster ovary (CHO) and lung (V79) cells, mouse lymphoma cells
and human skin fibroblasts.
Studies in bacteria exposed to radiation throughout the solar UV spectrum (reviewed by
Webb, 1977) demonstrate mutagenic activity unambiguously. The effects of sunlight on
mammalian cells have been reviewed (Kantor, 1985). UVA (320-400 nm) is mutagenic to
yeast and cultured mammalian cells, UVB (290-320 nm) to bacteria and cultured mam-
malian cells and UVC (200-290 nm) to bacteria, fungi, plants, cultured mammalian cells,
including CHO and V79 cells, and human lymphoblasts, lymphocytes and fibroblasts.
Since wavelengths in the UVC range do not reach the surface of the Earth, they are of no
significance as a source of damage in natural sunlight.
A characteristic of all of these studies is that UV A appears to be relatively inefficient as
a mutagen in comparison with UVB and UVC when activity is expressed per unit of energy
fluence, but not necessarily so when expressed per DNA photoproduct (see Tyrrell, 1984).
Webb (1977) compiled action spectra for the introduction of mutations in bacteria, as did
Coohill et al. (1987) for mutagenesis in human epithelial cells. In both Salmonella and
human cells, wavelengths > 320 nm were at least 103 times less effective than those
between 270 and 290 nm.
A comparison of the mutagenicity of various UV -containing light sources towards a set
of S. typhimurium strains was reported by De Flora et al. (1990). The approach did not
involve measurement of cytotoxicity, and mutagenicity was compared at roughly equitoxic
doses rather than as a function of fluence. Halogen lamps were as mutagenic as 254-nm
UVC and more mutagenic than fluorescent sunlamps or sunlight. The mutagenicity of
halogen lamps was attributed to their UVC component, in contrast to sunlight which
produced mutagenic effects over a wide UV spectrum. The mutagenicity of halogen lamps,
fluorescent lamps and sunlight was partially inhibited by catalase, suggesting that peroxides
may be involved in this in-vitro system. It is also relevant that pretreatment of E. coli with
hydrogen peroxide results in an increase in both UV A resistance and hydrogen peroxide
scavenging ability (Moss, S.H., quoted by Tyrrell, 1985; Sammartano & Tuveson, 1985;
Tyrrell, 1985).
Further evidence for the complexity of responses to the UVR region comes from
Schothorst et al. ( 1987b ), who examined the mutational response of human skin fibroblasts
to 12 lamps differing widely in their emission characteristics. Surprisingly, they found that,
whatever the light source, mutation induction perMED was similar with UVC, UVB and
solar radiation; with UV A (only one data point), mutation induction per MED was much
greater. The authors emphasized that these conclusions hold only if it is valid to calculate
the mutagenicity of a light source by adding the effects of the contributing wavelengths;
however, the data ofCoohill et al. (1987) argue against this assumption.
The inevitable consequence of the absorption spectrum maximum of DNA is that there is
a considerable body of data on mutagenicity toward microorganisms of UVC, which is
usually delivered by radiation from germicidal lamps with more than 90% of their output at
254 nm. The types of mutations that are induced by UVC and the mechanisms of their
induction have been reviewed (Witkin, 1976; Hall & Mount, 1981; Walker, 1984;
Hutchinson & Wood, 1986; Bridges et al., 1987; Hutchinson, 1987). Specific cellular
proteins, including the products of recA and umuC genes, together with a cleaved derivative
of the umuD gene product, must be present for mutations to result from most types of DNA
damage. These proteins are themselves part of an inducible response to DNA damage, and
their intracellular level increases dramatically when photoproducts or other lesions are
detected in DNA. It is not yet clear to what extent inducible systems are involved in UV
mutagenesis in higher eukaryotes.
Current evidence suggests that all photoproducts are likely to be potentially mutagenic,
although with greatly different specificities and potencies. The major UV photoproducts,
cyclobutane-type thymine-thymine dimers, are, for example, relatively weakly mutagenic
(Banerjee et al., 1988, 1990), owing in part to the propensity of polvmerases to insert
adenine when the template instruction is unclear or missing (Sagher & Strauss, 1983;
Schaaper et al., 1983; Kunkel, 1984). The relatively minor (6-4) thymine-thymine
photoproduct is, in contrast, highly mutagenic, the dominant mutation being a 3' T
transition (LeClerc et al., 1991 ). By far the most frequent UVC-induced change in human
cells is the transition from G:C to A:T (Bredberg et al., 1986; Seetharam et al., 1987; Hsia
et al., 1989; Dorado et al., 1991). A number of investigators have noted the production of
tandem transitions from G:C,G:C to A:T,A:T. Although this is not the most frequent
change, it seems to be particularly characteristic for UVC mutagenesis in human cells. The
frequency of mutation per lethal event at the hprt locus (which detects a broad spectrum of
mutations) is approximately the same at 254 nm and 313 nm in human 1ymphoblastoid cells:
however, the mutation frequency per lethal event at the Na+/K+ ATPase locus (which detects
point mutations) is considerably higher at 313 nm. This finding may indicate a difference in
types of premutagenic lesions and/or rates of mutation between the two wavelength regions
(Tyrrell, 1984). Two bacterial studies provide positive evidence for the mutagenic activity
of fluorescent lamps. De Flora et al. (1990) employed Sylvania 36 W cool white tubes with
E. coli and Salmonella strains. [The Working Group had difficulty in evaluating these data
because they are presented in a highly transformed format.] Hartman et al. (1991) used
General Electric F15T8CW lamps; a lowest effective dose of 5500 J/m
2
can be estimated
from the results with Salmonella tester strains. Filters that block wavelengths < 370 nm
effectively eliminated mutagenesis, while radical scavengers such as superoxide dismutase
or catalase stimulated mutagenesis.
Hsie et al. (1977) irradiated the hprt CHO system with Westinghouse white light
F40CW lamps. The minimal effective dose was 3.96 x 106 J/m
2
. Putting lids on the petri
dishes reduced mutant frequency by 30%. [The Working Group noted that the results were
based on a single dose point in a single experiment.] Jacobson et al. (1978) exposed mouse
lymphoma L5178Y tk+l- cells to Sylvania F l8T8 cool white lamps. The estimated lowest
effective dose was 2 x 10
4
J/m
2
. [The Working Group noted that the selective agent used,
BUdR, is regarded as inefficient and has been superseded by trichlorothymidine, so these
results require confirmation.]
(c) Chromosomal effects
Sunlamps have been shown to produce sister chromatic exchange in amphibian cells
(Chao & Rosenstein, 1985) and in human fibroblasts (Bielfe1d et al., 1989; Roser et al.,
1989). Fibroblasts from a panel of cutaneous malignant melanoma patients (Roser et al. ,
1989) and heterozygotes of xerodetma pigmentosum (Bielfeld et al., 1989) were more
susceptible to the induction of both sister chromatic exchange and micronuclei than those
from normal donors. Micronuclei were also induced in mouse splenocytes by exposure to
sunlamps in vitro (Dreosti et at., 1990).
A study with CHO cells provided evidence for a dose-related increase in the induction of
sister chromatic exchange by UV A, but the increased induction of chromosomal aberrations
showed no dose-response relationship (Lundgren & Wulf, 1988).
UVB induced sister chromatic exchange in CHO cells (Rasmussen et al. , 1989) and
chromosomal aberrations in frog ICR 2A cells (Rosenstein & Rosenstein, 1985). In the
latter study, photoreactivation reduced the number of chromosomal aberrations more
effectively at 265, 289 and 302 than at 313 nm, suggesting that non-cyclobutane dimer
photoproducts are more important primary lesions at the higher wavelength.
For UVC, more extensive data are available. Sister chromatic exchange was induced in
Chinese hamster V79 (Nishi et al. , 1984) and CHO (Rasmussen et al., 1989) cells.
Chromatid exchange was also recorded in cultured fetal fibroblasts from New Zealand black
mice, which proved to be more sensitive than BALB/c cells (Ready et al., 1978). The
induction of chromosomal aberrations in Chinese hamster cells has been reported on a
number of occasions (Chu, 1965a,b; Trosko & Brewen, 1967; Bender et al., 1973; Griggs &
Bender, 1973; Ikushima & Wolff, 1974).
Exposure of frog ICR 2A cells to 254 or 265 nm radiation induced both sister chromatic
exchange (Chao & Rosenstein, 1985) and chromosomal abenations, while photoreactivating
light significantly reduced the frequency of chromosomal aberrations, which implies a role
for pyrimidine dimers in their genesis (Rosenstein & Rosenstein, 1985). Chromosomal
aberrations were also seen with Xenopus cell cultures (Griggs & Bender, 1973). The
frequencies of sister chromatic exchange and chromosomal aberrations induced by UVC
were reduced by photoreactivating light in chicken embryo fibroblasts (Natarajan et al.,
1980), lending further support to the concept that the cyclobutane pyrimidine dimer
represents a primary lesion in these two end-points.
Farshad et al. (1980a) reported the induction of chromosomal damage in human IMR-90
fibroblasts following treatment with 4.6 W/m
2
over 20 h (331 kJ/m
2
) from F 15T8-CW tubes.
Shielding and radical scavengers reduced the level of damage.
Extensive data are available on the induction of sister chromatic exchange in fibroblasts
from patients with Bloom's syndrome (Krepinsky et at., 1980), xeroderma pigmentosum (De
Weerd-Kastelein et al., 1977; Fujiwara et al., 1981) or Cockayne's syndrome (Marshall et
al., 1980; Fujiwara et al., 1981), as well as from normal individuals. In comparison with
normal individuals, more sister chromatic exchanges were induced per lethal lesion in
fibroblasts from excision-competent Bloom's syndrome (Kurihara et al., 1987) and
Cockayne's syndrome (Marshall et al., 1980) patients. No such increase in sister chromatic
exchange was seen in fibroblasts from excision-defective xeroderma pigmentosum patients
or from an individual defective in the ligation step of repair (Henderson et al., 1985).
The induction of sister chromatic exchange by UV irradiation has also been studied in
human lymphocytes, with conflicting results. In one study, they were reported to be less
responsive than either human fibroblasts or CHO cells (Ferticone et al., 1986), while another
report, in which chromosomal aberrations were also studied, suggested that lymphocytes
were more sensitive than fibroblasts in their response at both end-points (Murthy et al.,
1982). These results may have implications for the interpretation of the effect of UV on the
immune system.
Fibroblasts from xeroderma pigmentosum patients are more sensitive to the induction of
chromosomal aberrations than cells from normal donors (Farrington et al., 1971; Farrington,
1972; Marshall & Scott, 1976). Seguin et al. (1988) showed that lymphoblastoid cells from
five Cockayne's syndrome patients were similarly hypersensitive to UVC-induced
chromosomal aberrations. The induction of micronuclei in two normal and three Bloom's
syndrome-derived fibroblast cell cultures was reported by Krepinsky et al. (1980). One
culture from a Bloom's syndrome patient, GM1492, proved to be exceptionally sensitive to
the induction of micronuclei; the other two were indistinguishable from n01mal cells. This
result emphasizes the potential importance of heterogeneity in response among patients with
rare genetic syndromes.
(d) Tramformation
Morphological transformation of mammalian cells has been induced by solar radiation,
unshielded fluorescent tubes, solar simulators, UV A, UVB and, most extensively, UVC.
There is weak evidence (Baturay et al., 1985) for the induction of transformation by predo-
minantly UV A radiation (20T12BLB bulbs) in BALB/c 3T3 cells. In the same report, UV A
was shown to have promoting activity following initiation with 13-propiolactone. The most
effective wavelength for Syrian hamster embryo cells (Doniger et al., 1981) and human em-
bryonic fibroblasts (Sutherland et al., 1981) appears to be in the UVC range at about 265
nm. Transformation of human cells can be enhanced by delivering the dose on a number of
separate occasions (Sutherland et al., 1988). It has also been reported that excision repair-
defective xeroderma pigmentosum cells can be transformed to the anchorage-independent
phenotype at lower doses than those required for cells from normal individuals (Maher et
al., 1982). Fisher and Cifone (1981) showed enhanced metastatic potential of mouse fibro-
sarcoma cells. Plasmids containing the human N-ras gene which were irradiated with UVR
(254 nm) in vitro acquired the ability to transform cultured rat-2 cells after transfection;
photoreactivation of irradiated plasm ids eliminated their transforming ability (van der Lubbe
et al., 1988). In another study, UVB irradiation activated the human Ha-ras gene on a
plasmid in a transformation assay with mouse NIH-3T3 cells (Pierceall & Ananthaswamy
(1991).
An investigation of chromosomal breaks and malignant transformation in embryonic
mouse cells (Sanford et al. , 1979; Parshad et al., 1980b) revealed that exposure of cultured
cells to fluorescent lamps induced malignant transformation, as measured by tumour
formation following implantation into syngeneic hosts. The potential importance of active
oxygen species was revealed by experiments in which the partial pressure of oxygen in
cultures was increased, resulting in increased malignant transformation and correlated chro-
mosomal breakage.
Kennedy et al. (1980) reported induction of transformation in C3H 10Tl/2 mouse
embryonic cell cultures by light from General Electric P18T8 lamps. The lowest effective
dose was estimated at 2 x 10
5
J/m
2
, and use of petri dish lids was effective in reducing
transformation.
(e) Effects on cellular and viral gene expression
A number of cellular oncogenes and other genes involved in the regulation of growth are
implicated in the process of carcinogenesis, as they are subject to both gene mutation and
alteration in expression due to chromosomal rearrangement. Many of these genes also show
transient alterations in expression following DNA damage, which has led to the suspicion
that such transient changes are involved, either directly or indirectly, in the carcinogenic
process.
UVC radiation was found to increase transiently the expression of various cellular genes,
including those that code for collagenase (Stein et at., 1989), the fos protein (Hollander &
Pomace, 1989; Stein et al., 1989), the jun protein (Ronai et al., 1990), metallothioneins I
and II (Pomace et al., 1988) and human plasminogen activator (Miskin & Ben-Ishai, 1981 ).
UV A radiation enhanced expression of the genes that code for the fos protein (Hollander &
Pornace, 1989), and UVB radiation increased the level of ornithine decarboxylase (Verma et
al., 1979). Different levels of cytotoxicity were seen in these experiments. UV A radiation
at doses that inactivate a small fraction of the fibroblast cell population induced expression
of the haem oxygenase gene (Keyse & Tyrrell, 1989) by a transient enhancement in
transcription rate (Keyse et al., 1990). cis-Acting enhancer elements have been shown to be
involved in activation of the collagenase and c-fos, as well as human immunodeficiency
promoter (Stein et al., 1989). In both rat fibroblasts and human keratinocyte cell lines,
exposure to UVR increased the levels of c-fos RNA within 1 0 min and of c-myc RNA after
about 1 h. The levels peaked at 30 min and 7 h and returned to normal within 1 h and 24 h,
respectively. The order of effectiveness was UVC > UVB > UV A (Ronai et al. , 1990).
Elevated levels of p53 protein were observed in mouse cells treated with UVR; the increase
was due to post-translation activation or stabilization (Maltzman & Czyzyk, 1984). In
human keratinocytes exposed to UV A, increased levels of human epidermal growth factor
receptor RNA (HER-1) were found (Yang et al., 1988).
The mechanisms that mediate these transient and immediate inducible responses are
largely unknown. Some of them, however, overlap with those seen in response to tumour
promoters, and it is significant that natural sunlight has been reported to enhance the
expression of protein kinase C in cultured human epithelial P3 cells (Peak et al., 1991a).
For reviews of this general area, see Ananthaswamy and Pierceall (1990) and Ronai et al.
(1990).
Other transient responses to UVR have been noted at somewhat later times (12-48 h).
Methotrexate resistance due to gene amplification was reported in 3T6 mouse cells (Tlsty et
al., 1984). Another selective DNA amplification response is induction by UVR of viral
DNA synthesis, e.g .. of polyoma virus in rat fibroblasts. UVC was more effective than
UVB, and UV A was ineffective (Ronai et al., 1987). In Chinese hamster embryo cells,
UVC irradiation increased DNA binding to the early domain of the SV40 minimal origin,
resulting in SV 40 DNA amplification (Lucke-Ruhle et al., 1989). The induction of
asynchronous viral replication is mediated by cellular proteins that bind to specific
sequences in the DNA of polyoma (Ronai & Weinstein, 1988) and SV40 viruses
(Lucke-Ruhle et al., 1989).
Exposure to UVR can activate viruses. This phenomenon has been known for herpes
simplex virus for a long time (for a recent report, see Rooney et al., 1991). It was reported
recently that UVC can activate the gene promoters of the human immunodeficiency virus
(HIV) (Valerie et al., 1988) and Moloney murine sarcoma virus (Lin et al., 1990).
Furthermore, activation of complete HIV grown in cells pre-exposed to UVC radiation was
observed (Valerie et al., 1988). HIV activation may contribute to faster development of
AIDS, which in tum may facilitate development of malignancies. Further studies showed
that the HIV promoter and HIV are activated by UVC and UVB, but not UVA radiation
even at very high exposures (Stanley et al., 1989; Beer et al., 1991 [abstract]; Lightfoote et
al., 1992). There are indications that pyrimidine dimers (Stein et al., 1989) or chromatin
damage (Valerie & Rosenberg, 1990) play a role in the initiation of HIV activation by UVR.
The in-vitro observations have been verified for UVC, UVB and UV A in experiments with
transgenic mice carrying the HIV promoter/reporter gene constructs (Cavard et al., 1990;
Frucht et al., 1991; Vogel et al., 1992). For reviews on the activation HIV by UVR, see
Zmudzka and Beer (1990) and Beer and Zmudzka (1991).
Table 32. Genetic and related effects of solar, simulated solar and sunlamp (UV A and UVB) irradiation
Test system
BS?, Bacillus subtilis, mutation
SSB, Saccharomyces cerevisiae D7, DNA damage
PLM, Wheat mutation
DIA, DNA damage, TCR 2A frog cells in vitro
DIA, DNA damage, ICR 2A frog cells in vitro
DIA, DNA strand breaks, Chinese hamster V79 cells in vitro
DIA, DNA damage, Chinese hamster V79 cells in vitro
DIA, DNA damage, C3H lOTl/2 mouse cells in vitro
GCO, Gene mutation, Chinese hamster ovary cells in vitro
G9H, Gene mutation, Chinese hamster V79lung cells in vitro, 6-TGr
G5T, Gene mutation, mouse lymphoma L5178Y cells in vitro
G9H, Gene mutation, Chinese hamster V79lung cells in vitro, 6-TGr
G9H, Gene mutation, Chinese hamster V79 lung cells in vitro, 6-TGr
SIA, Sister chromatid exchange, ICR 2A frog cells in vitro
MIA, Micronucleus test, mouse splenocytes in vitro
TBM, Cell transfonnation, BALB/c 3T3 mouse cells in vitro
TBM, Cell transformati on, BALB/c mouse epidermal cells in vitro
TCM, Cell transformation, C3H lOTl/2 mouse embryo cells in vitro
TCM, Cell transformation, C3H 1 OTI/2 mouse cells in vitro
TCL, Cell transformation, mouse fibrosarcoma cells in vitro
TCL, Cell transformation, 1 OTl/2 mouse skin fibroblasts in vitro
DIA, DNA damage, fish in vitro
DIH, DNA damage, human skin fibroblasts in vitro
DIH, DNA damage, human HeLa cells in vitro
GIH, Gene mutation, human xeroderma pigmentosum fibroblasts in vitro
SHF, Sister chromatid exchange, humanb fibroblasts in vitro
Resulta
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Reference
Munakata (1989)
Hannan et al. (1984)
Morgan et al. (1988)
Chao & Rosenstein (1986)
Rosenstein et al. ( 1989)
Elkind & Han (1978)
Suzuki et at. (1981)
Suzuki et al. (1981)
Hsie et a/. ( 1977)
Zolzer et al. (1988)
Jacobson et al. ( 1978)
Bradley & Sharkey (1977)
Burki & Lam (1978)
Suzuki et al. (1981)
Dreosti et al. (1990)
Withrow et al. (1980)
Ananthaswamy & Kripke (1981)
Kennedy et al. (1980)
Suzuki etal. (1981)
Fisher & Cifone (1981)
Ananthaswamy (1984a)
Applegate & Ley (1988)
Rosenstein et al. (1985)
Chao & Rosenstein ( 1986)
Rosenstein (1988)
Rosenstein & Mitchell ( 1991)
Elkind & Han ( 1978)
Patton eta/. (1984)
Knees-Matzen et al. (1991)
OTHER RELEVANT OAT A
Table 32 ( contd)
Test system
SfH, Sister chromatid exchange, human xeroderma pigmentosum
fibroblasts
SIH, Sister chromatid exchange, human malignant melanoma cells
MIH, Micronucleus test, human xeroderma pigmentosum fibroblasts
MIH, Micronucleus test, human malignant melanoma cells
DV A, DNA damage, BALB/c mouse skin cells in vivo
DV A, DNA damage, marsupial corneal cells in vivo
DV A, DNA damage, marsupial corneal cells in vivo
TVI, Cell transformation, lOTl/2 mouse skin fibroblasts treated in vivo
scored in vitro
DVH, DNA damage, human skin cells in vivo
DVH, DNA damage, human skin cells in vivo
a+, positive
bFirst-degree relatives of melanoma patients
Resulta
+
+
+
+
+
+
+
+
+
+
Reference
Bielfeld et al. ( 1989)
Roser et al. (1989)
Bielfeld eta/. (1989)
Roser et al. ( 1989)
Ananthaswamy & Fisher (1981)
Freeman eta/. (1988a)
Ley et al. (1988)
Ananthaswamy (1984b)
Eggset et al. ( 1983)
Freeman et al. (1988b)
207
Table 33. Genetic and related effects of predominantly UV A irradiation (near UV)
Test system Resuie
SA9, Salmonella typhimurium TA98, reverse mutation +
ECW, Escherchia coli WP2 uvrA, reverse mutation +
EC2, Escherichia coli WP2 her-, reverse mutation +
ECR, Escherichia coli B/r/l, trp, reverse mutation +
ECR, Escherichia coli WP2 recA, reverse mutation +
ECR, Escherichia coli WP2 uvrA recA, reverse mutation +
ECR, Escherichia coli Blr uvrA trp thy, reverse mutation +
ECR, Escherichia coli wild type, reverse mutation +
ECR, Escherichia coli, mutation +
SSB, Saccharomyces cerevisiae wild type, DNA damage +
SSB, Saccharomyces cerevisiae excision-deficient, DNA damage +
SSB, Saccharomyces cerevisiae D7, DNA damage +
DIA, DNA damage, Chinese hamster ovary cells in vitro +
DIA, DNA strand breaks, Chinese hamster ovary cells in vitro +
GCO, Gene mutation, Chinese hamster ovary cells in vitro +
G9H, Gene mutation, Chinese hamster lung V79 cells, hprt locus +
G90, Gene mutation, Chinese hamster lung V79 cells, 6-TGr +
G5T, Gene mutation, mouse lymphoma L5178Y cells, tk locus +
SIC, Sister chromatid exchange, Chinese hamster ovary cells in vitro +
CIC, Chromosomal abberations, Chinese hamster ovary cells in vitro (+)
TCL, Cell transformation, Syrian hamster embryo cells in vitro (neoplastic transformation) +
TCL, Cell transformation, Syrian hamster embryo cells in vitro (morphological transformation)
DIH, DNA strand breaks, human fibroblasts in vitro +
DIH, DNA strand breaks, human teratoma cells in vitro +
DIH, DNA double strand breaks, human teratocarcinoma cells in vitro +
DIH, DNA stand breaks, human fibroblasts in vitro +
DIH, DNA-protein cross-links, human teratocarcinoma cells in vitro +
DIH, DNA strand breaks, human epithelial P3 cells in vitro +
DIH, Pyrimidine dimer fonnation, human skin fibroblasts in vitro +
Reference
Calkins et al. ( 1987)
Tyrrell ( 1982)
Kubitschek ( 1967)
Webb & Malina (1970)
Tyrell ( 1982)
Tyrell (1982)
Tyrell ( 1982)
Tyrell ( 1982)
Wood et al. (1984)
Zolzer & Kiefer (1983)
Zolzer & Kiefer ( 1983)
Hannan et al. (1984)
Zelle et al. ( 1980)
Churchill et al. (1991)
Zelle et al. (1980)
Singh & Gupta (1982)
Lundgren & W u1f (1988)
Wells & Han ( 1984)
Wells & Han (1984)
Hitchins et al. (1987)
Lundgren & Wulf (1988)
Lundgren & Wulf ( 1988)
Barrett et al. (1978)
Barrett et al. (1978)
Rosenstein & Ducore (1983)
Peak et al. (1987)
Peak & Peak (1990)
Francis & Giannelli ( 1991)
Peak & Peak (1991)
Peak et al. (1991b)
Enninga et al. (1986)
Table 33 ( contd)
Test system
DIH, Pyrimidine dimer formation, human skin fibroblasts in vitro
GIH, Gene mutation, human lymphoblastoid cell line in vitro
GIH, Gene mutation, human skin fibroblasts in vitro
GIH, Gene mutation, human epithelial cells in vitro
DVH, Pyrimidine dimer formation, human skin in vivo
a+, positive;(+), weakly positive;-, negative
bPositive result with 365 nm but not with 334 nm at same fluence
OTHER RELEVANT DATA
ResuJe Reference
+ Rosenstein & Mitchell
(1987)
Tyrrell ( 1984)
+ Enninga eta!. (1986)
+b Jones et al. ( 1987)
+ Freeman et al. ( 1989)
209
Table 34. Genetic and related effects of predominantly UVB irradiation
Test system
SA9, Salmonella typhimurium TA98, reverse mutation
EC2, Escherichia coli WP2, reverse mutation
TSC, Tradescantia, chromosomal aberrations
DIA, DNA damage, Chinese hamster ovary cells in vitro
DIA, DNA strand breaks, Chinese hamster V79 cells
DTA, DNA-protein cross-links, Chinese hamster V79 cells
G9H, Gene mutation, Chinese hamster V79 lung cells, hprt locus
G90, Gene mutation, Chinese hamster V79 lung cells, ouabainr
G9H, Gene mutation, Chinese hamster V79 lung cells in vitro, 6TGr
G51, Gene mutation, mouse lymphoma L5178Y cells in vitro
SIC, Sister chromatid exchange, Chinese hamster ovary cells in vitro
CIA, Chromosomal aberrations, ICR 2A frog cells in vitro
TCS, Cell transfonnation, Syrian hamster embryo cells in vitro
DIH, DNA strand breaks, human skin fibroblasts in vitro
DIH, DNA strand breaks, human teratoma in vitro
DIH, DNA double strand breaks, human teratocarcinoma in vitro
DIH, DNA-protein cross-links, human teratocarcinoma in vitro
DIH, Pyrimidine dimer formation in human skin keratinocytes in vitro
DIH, Thymine dimer formation, human fibroblasts in vitro
GIH, Gene mutation, human lymphoblastoid cell1ine in vitro
Result
3
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
GIH, Gene mutation, human skin fibroblasts in vitro +
GIH, Gene mutation, human epithelial cells in vitro +
TIH, Cell transformation, human fibroblasts in vitro +
OVA, Cyclobutane dimers in SV40 plasmid DNA in human fibroblasts in vivo +
OVA, Cytosine photohydrates in SV40 plasmid DNA in human fibroblasts in vivo +
Reference
Calkins et al. (1987)
Peak et al. ( 1984)
Kirby-Smith & Craig (1957)
Zelle et at. ( 1980)
Matsumoto et at. ( 1991)
Matsumoto et al. (1991)
Zelle et al. ( 1980)
Rasmussen et al. (1989)
Wells & Han (1984)
Wells & Han (1984)
Colella et al. (1986)
Jacobson et al. (1981)
Rasmussen et at. ( 1989)
Rosenstein & Rosenstein (1985)
Doniger eta/. (1981)
Rosenstein & Mitchell ( 1987)
Peak eta/. (1987)
Peak & Peak (1990)
Peak & Peak (1991)
Schothorst et al. (1991)
Roza et al. (1988)
Tyrrell (1984)
Enninga et al. ( 1986)
Jones, C.A. et al. (1987)
Sutherland et al. (1981)
Mitchell eta!. (1991)
Mitchell et al. ( 1991)
Table 34 ( contd)
Test system
DV A, Pyrimidine dimer induction, mouse skin in vivo
DV A, Pyrimidine dimer formation, mouse skin in vivo
DVA, (6-4) Photoproduct formation, mouse epidermis in vivo
DVH, Pyridime dimer formation, human skin in vivo
UVH, Unscheduled DNA synthesis, human cornea in vivob
a+, positive;-, negative
bFrom people who had been dead for 15 min
OTHER RELEVANT DATA
Result
8
+
+
+
+
+
Reference
Cooke & Johnson (1978)
Ley et al. (1983)
Olsen et al. (1989)
Freeman et al. ( 1989)
Grabner & Brenner ( 1981)
211
Table 35. Genetic and related effects of UVC irradiation
Test system
ECB, Escherichia coli, thymine dimer formation
ECB, Escherichia coli, photoproduct formation
ECB, Escherichia coli, thymine photoadduct formation
ECB, Escherichia coli, pyrimidine dimers
ECB, Escherichia coli, (6-4) photoproducts
ECF, Escherichia coli, miscellaneous strains, forward mutation
ECR, Escherichia coli, mutation
SSB, Saccharomyces cerevisiae, pyrimidine dimer formation
SSB, Saccharomyces cerevisiae, pyrimidine dimer formation
SCN, Saccharomyces cerevisiae, aneuploidy
SCF, Saccharomyces cerevisiae, forward mutation
SCR, Saccharomyces cerevisiae, reverse mutation
PLU, Plants, DNA damage
PLU, Nicotiana tabacum, unscheduled DNA synthesis
PLU, Chlamydomonas reinhardtii, pyrimidine dimer formation
PLM, Chlamydomonas reinhardtii, mutation
TSC, Tradescantia, chromosomal aberrations
DM?, Drosophila melanogaster embryo cells in vitro, DNA damage
DTA, DNA damage, ICR 2A frog cells in vitro
DTA, DNA strand breaks, Chinese hamster V79 cells
DTA, DNA damage, Chinese hamster ovary cells in vitro
G9H, Gene mutation, Chinese hamster V79 lung cells in vitro
G90, Gene mutation, Chinese hamster V79 lung cells, ouabain'
Resulta
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Reference
Setlow eta/. ( 1963)
Setlow ( 1968)
Smith (1964)
Brash & Haseltine (1982)
Brash & Haseltine ( 1982)
Miller ( 1985)
Witkin (1976
Walker (1984)
Franklin & Haseltine (1986)
Bridges er al. ( 1987)
Schaaper et al. (1987)
Wheatcroft et al. (1975)
Resnick et al. (1987)
Parry et al. (1979)
Lee et al. ( 1988)
Siede & Eckardt (1986)
McLennan ( 1987)
Cieminis et al. ( 1987)
Vlcek eta/. ( 1987)
Vlcek eta!. ( 1987)
Kirby-Smith & Craig (1957)
Koval ( 1987)
Elkin & Han (1978)
Zelle et a/. ( 1980)
Zelle et al. (1980)
Rasmussen eta/. ( 1989)
Drobetsky & Glickman (1990)
Colella eta/. (1986)
Suzuki eta/. (1981)
Suzuki et at. (1981)
OTHER RELEVANT DATA
Table 35 ( contd)
Test system
G51, Gene mutation, mouse lymphoma L5178Y cells in vitro
SIC, Sister chromatid exchange, Chinese hamster V79 cells in vitro
SIC, Sister chromatid exchange, Chinese hamster ovary cells in vitro
SIA, Sister chromatid exchange, ICR 2A frog cells in vitro
SIA, Sister chromatid exchange, chick embryo fibroblasts in vitro
CIC, Chromosomal aberrations, Chinese hamster fibroblasts in vitro
CIC, Chromosomal aberrations, Chinese hamster fibroblasts in vitro
CIC, Chromosomal aberrations, Chinese hamster V79 cells in vitro
CIC, Chromosomal aberrations, Chinese hamster V79 cells in vitro
CIC, Chromosomal aberrations, Chinese hamster ovary cells in vitro
CIC, Chromosomal aberrations, Chinese hamster CHEF-125 cells in vitro
CIA, Chromosomal aberration, chick embryo fibroblasts in vitro
CIA, Chromosomal aberrations, A8W243 Xenopus cells in vitro
CIA, Chromosomal aberrations, ICR 2A frog cells in vitro
CIA, Chromosomal aberrations, New Zealand black mouse fetal fibroblasts
TBM, Cell transformation, BALB/c 3T3 mouse cells
TCM, Cell transformation, C3H 10TI/2 mouse cells
TCM, Cell transformation, C3H 10T1/2 mouse cells
TCM, Cell transformation, C3H 10TI/2 mouse cells
TCM, Cell transformation, C3H lOTI /2 mouse cells
TCM, Cell transformation, C3H IOTI /2 mouse cells
TCS, Cell transformation, Syrian hamster embryo cells
TCS, Cell transfonnation, Syrian hamster embryo cells
TCS, Cell transfonnation, Syrian hamster embryo cells
TEV, Cell transfom1ation, SV-40/BALB/c 3T3 mouse cells
DIH, DNA strand breaks, human skin fibroblasts in vitro
DIH, DNA strand breaks, human teratoma cells in vitro
DIH, Thymine dimer formation, human skin fibroblasts in vitro
Resulta
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Reference
Jacobson eta!. (1981)
Nishi et al. (1984)
Rasmussen et al. ( 1989)
Natarajan et a/. (1980)
Chu (1965a)
Chu (1965b)
Bender et al. (1973)
Griggs & Bender (1973)
Ikushima & Wolff (1974)
Trosko & Brewen (I 967)
Natarajan eta/. (1980)
Griggs & Bender (1973)
Rosenstein & Rosenstein (1985)
Reddy et al. (1978)
Withrow eta/. ( 1980)
Chan & Little (1976)
Monda! & Heidelberger (1976)
Chan & Little (1979)
Suzuki eta!. ( 1981)
Borek eta/. ( 1989)
DiPaolo & Donovan (1976)
Do niger et al. ( 1981)
Borek eta/. ( 1989)
Withrow eta!. ( 1980)
Rosenstein et al. (1985)
Rosenstein & Mitchell (1987)
Peak et al. (1987)
Roza et al. (1988)
213
Table 35 ( contd)
Test system
DIH, DNA strand breaks, human fibroblasts in vitro
DIH, DNA-protein cross-links, human fibroblasts in vitro
DIH, DNA double strand breaks, human teratocarcinoma cells in vitro
DIH, DNA-protein cross-links, human teratocarcinoma cells in vitro
DIH, Pyrimidine dimer formation, human skin keratinocytes and
melanocytes in vitro
GlH, Gene mutation, human fibroblasts in vitro
GlH, Gene mutation, human fibroblasts in vitro
GIH, Gene mutation, human lymphocytes in vitro
GIH, Gene mutation, human lymphoblastoid cell line in vitro
GIH, Gene mutation, human skin fibroblasts in vitro
GIH, Gene mutation, human epithelial cells in vitro
GIH, Gene mutation, human HeLa cells in vitro
GIH, Gene mutation, human lymphocytes in vitro
GIH, Gene mutation, human fibroblasts in vitro
GIH, Gene mutation, human fibroblasts in vitro
GIH, Gene mutation, human melanoma cells in vitro
SHF, Sister chromatid exchange, human fibroblasts in vitro
SHF, Sister chromatid exchange, human fibroblasts in vitro
SHL, Sister chromatid exchange, human lymphocytes in vitro
SHL, Sister chromatid exchange, human lymphocytes in vitro
SHF, Sister chromatid exchange, human skin fibroblasts
MIH, Micronucleus test, human skin fibroblasts in vitro
CHF, Chromosomal aberrations, human fibroblasts in vitro
CHF, Chromosomal aberrations, human skin fibroblasts
CHL, Chromosomal aberrations, human lymphocytes in vitro
CHL, Chromosome exchanges, human lymphocytes in vitro
TIH, Cell transformation, human fibroblasts in vitro
Resulta
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
Reference
Lai & Rosenstein ( 1990)
Lai & Rosenstein ( 1990)
Peak & Peak ( 1990)
Peak & Peak (1991)
Schothorst eta/. (1991)
Maher eta!. (1979)
Myhr eta/. (1979)
Sanderson et al. (1984)
Tyrrell (1984)
Enninga et al.(1986)
Jones, C.A. et al. (1987)
Musket a/. (1989)
Norimura eta/. ( 1990)
Dorado et al. ( 1991)
McGregor et a/. ( 1991)
Musk eta/. (1989)
Fujiwara eta!. ( 1981)
Kurihara eta/. ( 1987)
Murthy et al. (1982)
Perticone eta/. (1986)
De Weerd-Kastelein et al. (1977)
Krepinsky et al. (1980)
Marshall et al. (1980)
Henderson et al. ( 1985)
Krepinsky et al. (1980)
Farrington (1972)
Farrington et a/. (1971)
Marshall & Scott (197 6)
Murthy et al. (1982)
Holmberg & Gamauskas (1990)
Maher et al. (1982)
OTHER RELEVANT DATA
Table 35 ( contd)
Test system
???, Cyclobutane dimers in SV40 plasmid DNA in human skin fibroblasts in vitro and in vivo
???,Cytosine photohydrates in SV40 plasmid DNA in human skin fibroblasts in vitro and in vivo
DV A, Pyrimidine dimer formation, mouse skin in vivo
a +, positive
Result"
+
+
+
+
Reference
Mitchell et al. (1991)
Mitchell et al. (1991)
Bowden et at. (1975)
215
5. Summary of Data Reported and Evaluation
5.1 Exposure data
Terrestrial life is dependent on radiant energy from the sun. Approximately 5% of solar
terrestrial radiation is ultraviolet radiation (UVR), and solar radiation is the major source of
human exposure to UVR. Before the beginning of this century, the sun was essentially the
only source of UVR, but with the advent of artificial sources the opportunity for additional
exposure has increased.
UVR spans the wavelengths from l 00 to 400 nm. The biological effects of UVR vary
enormously with wavelength; by convention, the ultraviolet spectrum has been further
subdivided into three regions: UVC (100-280 nm), UVB (280-315 nm) and UVA (315-400
nm).
Solar UVR that reaches the Earth's surface comprises approximately 95% UV Aand 5%
UVB: UVC is completely filtered out by the Earth' s atmosphere. The amount of solar UVR
measured at the Earth' s surface depends upon a number of factors, which include solar
zenith angle (time of day, season and geographical latitude), stratospheric ozone,
atmospheric pollutants, weather, ground reflectance and altitude.
Exposed skin surface is irradiated differently depending on cultural and social behaviour,
clothing, the position of the sun in the sky and the relative position of the body. Exposure to
UVB of the most exposed skin surfaces, such as nose, tops of the ears and forehead, relative
to that of the lesser exposed areas, such as underneath the chin, normally ranges over an
order of magnitude. Ground reflectance plays a major role in exposure to UVB of the eye
and shaded skin surfaces, particularly with highly reflective surfaces such as snow.
In cutaneous photobiology, radiant exposme is frequently expressed as 'exposure dose' in
units of J/cm
2
(or J/m
2
). 'Biologically effective dose', derived from radiant exposure
weighted by an action spectrum, is expressed in units of J/cm
2
(effective) or as multiples of
'minimal erythema dose' (MED). In cellular photobiology, the term 'fluence' is often used
incorrectly as equivalent to radiant exposure.
The cumulative annual exposure dose of solar UVR varies widely among individuals in
a given population, depending to a large extent on occupation and extent of outdoor
activities. For example, it has been estimated that indoor workers in mid-latitudes ( 40-60
0
N) receive an annual exposure dose of solar UVR to the face of about 40-160 times the
MED, depending upon propensity for outdoor activities, whereas the annual solar exposure
dose for outdoor workers is typically around 250 times the MED. Because few actual
measurements have been reported of personal exposures, these estimates should be
considered to be very approximate and subject to differences in cultural and social
behaviour, clothing, occupation and outdoor activities.
-217-
Cumulative annual outdoor exposures may be augmented by exposures to articial
sources of UVR. For example, the use of cosmetic tanning appliances increased in
popularity in the 1980s. The majority of users are young women, and the median annual
exposure dose is probably 20-30 times the MED. Currently used appliances emit primarily
UV A radiation; prior to the 1980s, tanning lamps emitted higher proportions of UVB and
uvc.
UVR has been used for several decades to treat skin diseases, notably psoriasis. A
variety of sources of UVR are employed, and nearly all emit a broad spectrum of radiation.
A typical dose in a single course ofUVB phototherapy might lie between 200 and 300 times
the MED.
UVR is used in many different industries, yet there is a paucity of data concerning
human exposure from these applications, probably because in normal practice sources are
well contained and exposure doses are expected to be low. Acute reactions to overexposure
are common among electric arc welders. Staff in hospitals who work with unenclosed
phototherapy equipment are at potential risk of overexposure unless protective measures are
taken. Individuals exposed to lighting from fluorescent lamps may typically receive annual
exposure doses of UVR ranging from 0 to 30 times the MED, depending on illuminance
levels and whether or not the lamps are housed behind plastic diffusers. There is increasing
use of tungsten-halogen lamps, which also emit UVR, for general lighting.
5.2 Human carcinogenicity data
5.2.1 Solar radiation
Subjects with the inherited condition xeroderma pigmentosum appear to have fre-
quencies of nonmelanocytic skin cancer and melanoma that are much higher than expected.
Some evidence suggests that the greatest excess occurs on the head and neck.
(a) Nonmelanocytic skin cancer
The results of descriptive epidemiological studies suggest that exposure to sunlight
increases the risk of nonmelanocytic skin cancer. These tumours occur predominantly on
the skin of the face and neck, which is most commonly exposed to sunlight, although the
distribution of basal-cell carcinomas is not as closely related to the distribution of exposure
to the sun as is that of squamous-cell carcinomas. There is a strong inverse relationship
between latitude and incidence of or mortality from skin cancer and, conversely, a positive
relationship between incidence or mortality and measured or estimated ambient UVR.
Migrants to Australia from the British Isles have lower incidence of and mortality from non-
melanocytic skin cancer than the Australian-born population. People who work primarily
outdoors have higher mortality from these cancers, and there is some evidence that outdoor
workers have higher incidence.
In several cross-sectional studies, positive associations have been seen between mea-
sures of solar skin damage and the prevalence of basal- and squamous-cell carcinomas.
Measures of actual exposure to the sun have been less strongly associated with these
cancers, possibly because of errors in measurement and inadequate control for potential
confounding variables. In a study of US fishermen, estimates of individual annual and
cumulative exposure to UVB were positively associated with the occurrence of
SUMMARY OF DATA REPORTED AND EVALUATION 219
squamous-cell carcinoma but not with the occurrence of basal-cell carcinoma. Only two
population-based case-control studies have been conducted. In one of these, from Canada,
the response rate was low and the measures of exposure were crude. In the other study,
from Australia, facial telangiectasia and solar elastosis of the neck were strongly associated
with the risk for squamous-cell carcinoma, and cutaneous microtopography and solar
elastosis of the neck were strongly associated with risk for basal-cell carcinoma. Migrants
to Australia had a lower risk of squamous-cell carcinoma than did native-born Australians,
and migrants who arrived after childhood had a lower risk for basal-cell carcinoma.
The hospital-based case-control studies that have been conducted suffer from methodo-
logical deficiencies, including choice of controls, measurement of exposure and
confounding by reaction to sunlight, and are therefore difficult to interpret.
In a cohort study of nurses in the USA, those who spent more than 8 h per week outside
without sunscreens had a simi lar incidence rate of basal-cell carcinoma to those who spent
fewer than 8 h per week outdoors. In a cohort study from Victoria, Australia, the rates of
both types of skin cancer were increased in outdoor workers, but the effect was not
significant after adjustment for reaction to sunlight.
(b) Cancer of the lip
Cancer of the lip has been related to outdoor occupation in a number of descriptive
studies. Migrants to Australia and Israel have lower risks than native-born residents.
Three case-control studies provide useful information about the association between
outdoor work, taken as a proxy measure for exposure to UVR, and cancer of the lip. All of
them showed a significantly increased risk, although potential confounding by tobacco use
was not controlled adequately in any of the studies.
Assessment of the carcinogenicity of solar radiation for the lip is complicated by the fact
that carcinoma of the lip as actually diagnosed is a mixture of cancers of the external lip and
cancers of the buccal membranes. Use of alcohol and tobacco are known causes of the latter
tumours.
(c) Malignant melanoma of the skin
Descriptive studies in whites in North America, Australia and several other countries
show a positive association between incidence of and mortality from melanoma and resi-
dence at lower latitudes. Studies of migrants suggest that the risk of melanoma is related to
solar radiant exposure at the place of residence in early life. The body site distribution of
melanoma shows lower rates per unit area on sites usually unexposed to the sun than on
usually or regularly exposed sites.
A large number of case-control studies are pertinent to the relationship between
melanoma and exposure to the sun. These include large, carefully conducted population-
based studies carried out in Western Australia, Queensland, western Canada and Denmark.
Their results are generally consistent with positive associations with residence in sunny
environments throughout life, in early life and even for short periods in early adult life.
Positive associations are generally seen between measurements of cumulative sun damage
expressed biologically as microtopographical changes or history of keratoses or nonmelano-
cytic skin cancer.
In contrast, the associations with total exposure to the sun over a lifetime or in recent
years, as assessed by questionnaire, are inconsistent. This inconsistency may be due to
differences in the effects of chronic and intermittent exposure. Chronic exposure, as
assessed through occupational exposure, appeared to reduce melanoma risk in three of the
large studies, particularly in men; this observation is consistent with the descriptive
epidemiology of the condition, which shows lower risks in groups that work outdoors.
Several other studies, which were generally smaller or had less detailed methods of
exposure assessment, show either no effect or an increased risk associated with occupational
exposures.
Assessment of intermittent exposure is complex; nonetheless, most studies show positive
associations with measure of intermittent exposure, such as particular sun-intensive acti-
vities, outdoor recreation or vacations.
Most studies show positive associations with a history of sunburn; however, this asso-
ciation cannot be easily interpreted, because while it might accurately reflect sunburn it
could just as well reflect either the tendency to sunburn, if exposed, or intermittent exposure
more generally.
(d) Melanoma of the eye
There is no latitude gradient among white populations of the incidence of ocular neo-
plasms, some 80% of which are likely to be ocular melanomas. No effect of southern US
birthplace was seen in the two descriptive studies in the USA that examined this aspect.
Four case-control studies, from western Canada and from Philadelphia, San Francisco
and Boston, USA, provided information on the association between exposure to solar radia-
tion and ocular melanoma. All of these studies demonstrate an increased risk of ocular
melanoma in people with light skin, light eye colour or light hair colour. Two of the studies
compared effect of southern US birthplace with birth elsewhere in the USA; a significant
difference was seen in the Philadelphia study.
Past residence south of 40 N latitude was positively associated with ocular melanoma in
the Boston study but was not significant in the Philadelphia study after control for southern
birthplace. Although several outdoor activities, such as gardening and sunbathing, were
associated in the Philadelphia study with ocular melanoma, participation in outdoor acti-
vities did not increase risk significantly in Boston or San Francisco.
The lack of consistency of the results of these studies makes their interpretation difficult.
(e) Other cancers
No adequate study was available to evaluate the role of solar radiation in cancers at other
body sites.
5.2.2 Artificial sources of ultraviolet radiation
No adequate study was available on nonmelanocytic skin cancer in relation to exposure
to artificial sources ofUVR.
Two case-control studies, one from Scotland and one from Ontario, with detailed infor-
mation on use of sunbeds and sunlamps showed positive relationships between duration of
use and risk of melanoma of the skin. Several other studies with limited information
showed no association.
One case-control study from Sydney, Australia, showed a positive relationship between
melanoma of the skin and exposure to fluorescent lights at work among women, but the
measurement of exposure was crude and among exposed cases there was a relative excess of
melanoma on the trunk, a site likely to be covered at work. A more detailed study from
Australia showed no consistent association between cumulative exposure or rate of exposure
to fluorescent lights and melanoma. Two other studies had detailed information on
exposure. One, from Scotland, showed no such association, while the other, from England,
had inconsistent effects depending on the method of ascertainment of information. Another
study, from New York, with limited information also showed inconsistent effects depending
on the source of information.
Two case-control studies, from Boston and Philadelphia, USA, showed significant posi-
tive associations between use of sunlamps and melanoma of the eye. Another case-control
study, from San Francisco, showed an increased risk for exposure to 'UV or black light',
although the nature of the exposure was not specified.
Two studies, from Philadelphia and Montreal, showed significant positive associations
between welding and melanoma of the eye.
5.2.3 Molecular genetics of human skin cancers
Base substitutions in a tumour suppressor gene, p53, found in human squamous-cell skin
carcinomas that had developed at sites exposed to the sun were similar to those found in
experimental systems exposed to UVR, and especially to UVB.
5.3 Carcinogenicity in experimental animals
Solar radiation was tested for carcinogenicity in a series of exceptional studies in mice
and rats. Large numbers of animals were studied, and well-characterized benign and mali-
gnant skin turnouts developed in most of the surviving animals. Although the reports are
deficient in quantitative details, the results provide convincing evidence that sunlight is
carcinogenic for the skin of animals.
Broad-spectrum UVR (solar-simulated radiation and ultraviolet lamps emitting mainly
UVB) was tested for carcinogenicity in many studies in mice, to a lesser extent in rats and in
a few experiments in hamsters, guinea-pigs, opossums and fish. Benign and malignant skin
tumours were induced in all of these species except guinea-pigs, and tumours of the cornea
and conjunctive were induced in rats, mice and hamsters.
The predominant type of tumours induced by UVR in mice is squamous-cell carcinoma.
Basal-cell carcinomas have been observed occasionally in athymic nude mice and rats
exposed to UVR. Melanocytic neoplasms of the skin were shown to develop following
exposure of opossums and hybrid fish to broad-spectrum UVR.
Studies in hairless mice demonstrated the carcinogenicity of exposures to UVR in the
wavelength ranges 315-400 run (UV A), 280-315 nm (UVB) and :S 280 nm (UVC), UVB
radiation being the most effective, followed by UVC and UVA. UVB radiation is three to
four orders of magnitude more effective than UV A. Both short-wavelength UVA (315-340
nm) and long-wavelength UVA (340-400 nm) induced skin cancer in hairless mice. The
carcinogenic effectiveness of the latter waveband is known only as an average value over
the entire range; the uncertainty of this average is about one order of magnitude. In none of
the experiments involving UVC was it possible to exclude completely a contribution of
UVB, but the size of the effects observed indicate that they cannot be due to UVB alone.
No experimental data were available on the carcinogenicity to animals of radiation from
general lighting fixtures, including fluorescent and quartz halogen lamps.
UVR has been studied in protocols involving two-stage chemical carcinogenesis (substi-
tuting UVR for the chemical initiator or for the chemical promoter or giving it in addition to
both). UVR has been reported to exert many effects on the carcinogenic process, including
initiation, promotion, cocarcinogenicity and even tumour inhibition. Chemical immuno-
suppressive agents have been shown to enhance the probability of developing UVR-induced
tumours m mice.
5.4 Other relevant data
5.4.1 Transmission and absorption
Studies of transmission in whole human and mouse epidermis and human stratum
corneum in vitro show that these tissues attenuate radiation in the solar UVR range. This
attenuation, which is more pronounced for the UVB than for the UV A wavebands, affords
some protection from solar UVR to dividing cells in the basal layer.
The different components of the human eye act as optical filters for the UVR range.
Consequently, little or no UVR reaches the retina in the normal eye.
5.4.2 Effects on the skin
UVR produces erythema, melanin pigmentation and acute and chronic cellular and
histological changes in humans. Generally consistent changes are seen in experimental
species, including the hairless mouse.
The action spectra for erythema and tanning in humans and for oedema in hairless mice
are similar. UVB is three to four times more effective than UV A in producing erythema. In
humans, pigmentation protects against erythema and histopathological changes. People
with a poor ability to tan, who burn easily and have light eye and hair colour are at a higher
risk of developing melanoma, basal-cell and squamous-cell carcinomas (see section 5.2).
In humans, acquired pigmented naevi and solar keratoses, indicators of melanomas and
squamous-cell carcinomas, respectively are induced by exposure to the sun.
Xeroderma pigmentosum patients have a high frequency of pigmentary abnormalities
and skin cancers on sun-exposed skin. These patients also have defective DNA repair.
5.4.3 Effects on the immune response
Relatively few investigations have been reported of the effects of UVR on immunity in
humans, but changes do occur. There is evidence that contact all ergy is suppressed by
exposure to UVB and possibly to UV A radiation. The number of Langerhans' cells in the
epidermis is decreased by exposure to UVR and sunlight, and the morphological loss of
these cells is associated with changes in antigen-presenting cell function in the direction of
suppression; this change may be due not only to simple loss of function but also to active
migration of other antigen-presenting cells into the skin. A reduction in natural killer cell
activity also occurs, which can be produced by UV A radiation. These changes are
short-lived, and their functional significance is unknown. Pigmentation of the skin may not
protect against some UVR-induced alterations of immune function.
Several immune responses are suppressed by UVR in mice and other rodents. Sup-
pression of contact hypersensitivity has received most attention, and this response may be
impaired locally, at the site of exposure to radiation, or systemically, at a distant, unexposed
site. The two forms of suppression have different dose dependencies- systemic suppression
requiring much higher doses- and their mechanisms appear to differ, but the efferent limb
of each involves generation of hap/en-specific T -suppressor cells that block induction but
not elicitation of contact hypersensitivity. Systemic suppression of delayed hypersensitivity
to injected antigens can also be produced by exposure to UVB radiation, and several obser-
vations suggest that the mechanism of this suppression differs from that of systemic
suppression of contact hypersensitivity.
Alterations in immune function induced by exposure to UVR play a central role in
photocarcinogenesis in mice. UVR-induced T-suppressor cells block a normal irnmuno-
surveillance system that prevents the growth of highly antigenic UVR-induced tumours. It
is not known whether this mechanism operates in humans.
5.4.4 DNA photoproducts
Solar UVR induces a variety of photoproducts in DNA, including cyclobutane-type
pyrimidine dimers. pyrimidine-pyrimidine (6-4) photoproducts, thymine glycols, cytosine
damage, purine damage, DNA strand breaks and DNA-protein cross-links. Substantial
information on biological consequences is avai lable only for the first two classes. Both are
potentially cytotoxic and can lead to mutations in cultured cells, and there is evidence that
cyclobutane-type pyrimidine dimers may be precarcinogenic lesions. The relative and
absolute levels of each type of lesion vary with wavelength. Substantial levels of thymidine
glycols, strand breaks and DNA-protein cross-links are induced by solar UVA and UVB
radiation, but not by UVC radiation. The ratio of strand breaks to cyclobutane-type dirner
lesions increases as a function of increasing wavelength. In narrow band-width studies, the
longest wavelength at which cyclobutane-type pyrimidine dimers have been observed is 365
nm, whereas the induction of strand breaks and DNA-protein cross-links has been observed
at wavelengths in the UVB, UV A and visible ranges. Non-DNA chromophores such as
porphyrins, which absorb solar UVR, appeared to be important in generating active
intermediates that can lead to damage. Solar UVR also induces membrane damage.
5.4.5 Genetic and related effects
Measurable DNA damage is induced in human skin cells in vivo after exposures to
UV A, UVB and UVC radiation, including doses in the range commonly experienced by
humans. Most of the DNA damage after a single exposure is repaired within 24 h. The
importance of these wavelength ranges depends on several factors. UVB is the most
effective, UVC being somewhat less effective and UV A being much less effective, when
compared on a per photon basis, probably owing to a combination of the biological
effectiveness of the different wavebands and of their absorption in the outer layers of the
skin.
Summary table of genetic and related effects of ultraviolet A radiation
Nonmammalian systems Mammalian systems
Proka- Lower Plants Insects In vitro In vivo
ryotes eukaryotes
o i G oiR I G IA oiGic Rj GlcJ A
Animal cell I Human cells Animals I Humans
D I G I S l Ml Cl AI T IJ IDI Gl s I M I Cl AI Tl I I D G I S I M I c I DL I A I D I s I ML c I A
+ + + +
A, aneuploidy; C, chromosomal aberrations; D, DNA damage; DL, dominant lethal mutations; G, gene mutation; I, inihibition of intercellular communication; M, micronuclei; R,
mitotic recombination and gene conversion; S, sister chromatid exchange; T, cell transformation
In completing the tables, the following symbols indicate the consensus of the Working Group with regard to the results for each endpoint:
+ considered to be positive for the specific endpoint and level of biological complexity
+
1
considered to be positive, but only one valid study was available to the Working Group; sperm abnormality, mouse
considered to be negative
considered to be negative, but only one valid study was available to the Working Group
? considered to be equivocal or inconclusive (e.g., there were contradictory results from different laboratories; there were confounding exposures; the results were
equivocal)
Summary table of genetic and related of ultraviolet B radiation
Proka- Lower Plants Insects In vitro l n vivo
ryotes eukar votes
Animal cells I Humans cells Animals I Humans
o i G o i R Gj A o lolc Rj o1 c iA Dl GJ Sj Ml Cl Aj TJ I J D j GJ S I Ml Cj AI TJ I D I G I s I M J c I DL J AI D] s I Ml C j A
+ +
I
+ + +
I
+ + + +
I
+ +
A, aneuploidy; C, chromosomal aberrati ons; D, DNA damage; DL, dominant lethal mutati on; G, gene mutation; I, inhibition of intercellular communication; M,
mi cronuclei; R, mitotic recombination and gene conversion; S, sister chromatid exchange; T, cell transformation
+
1
considered to be negative, but only one valid study was available to the Working Group
? considered to be equivocal or inconclusive (e.g., there were contradictory results from the di fferent laboratories; there were confOlmding exposures; the
results were equivocal)
Summary table of genetic and related effects of ultraviolet C radiation
Proka- Lower Plants Tnsects In vitro In vivo
ryotes eukar otes
Animals cells I Human Animals I Humans
D I G DIR dA nlo lc R I G _l c I A D l G LS I Ml C j A IT IT ID IG l s LMI Cl AIT I I D J G J s J M L c I DL I A _I DJ s L Ml c I A
+ + + + +
l l
+ +
l l
+ + + + +
l
+ + + + + + +
l
+
l
A, aneuploidy; C, chromosomal aberrations; D, DNA damage; DL, dominant lethal mutati on; G, gene mutation; r, inhibition of intercellular communication; M,
micronuclei; R, mitotic recombination and gene conversion; S, sister chromatid exchange; T, cell transformation
+
1
considered to be negative, by only one valid study was avai lable to the Working Group
? considered to be equivocal or inconclusive (e.g., there were contradictory results from di fferent laboratories; there were confounding exposures; the
results were equivocal)
Solar and 'solar-simulated' radiation and radiation from sunlamps (UV A and UVB) are
mutagenic to prokaryotes and plants, induce DNA damage in fish and in amphibian cells in
vitro, are mutagenic to and induce sister chromatic exchange in amphibian cells, induce
micronucleus formation and transformation in mammalian cells in vitro are mutagenic to
and induce DNA damage and sister chromatic exchange in human cells in vitro and induce
DNA damage in mammalian skin cells irradiated in vivo.
UV A radiation is mutagenic to prokaryotes and induces DNA damage in fungi. It is
mutagenic to and induces DNA damage, chromosomal aberrations and sister chromatic
exchange in mammalian cells and induces DNA damage and mutation in human cells in
vitro.
UVB radiation is mutagenic to prokaryotes and induces chromosomal aberrations in
plants. It is mutagenic to and induces DNA damage, sister chromatic exchange and transfor-
mation in mammalian cells, is mutagenic and induces DNA damage and transformation in
human cells in vitro and induces DNA damage in mammalian skin cells irradiated in vivo.
UVC radiation induces DNA damage in and is mutagenic to prokaryotes, fungi and plants
and induces DNA damage in insects and aneuploidy in yeast. It induces sister chromatic
exchange in amphibian and avian cells in vitro; it is mutagenic to and induces DNA damage,
chromosomal aberrations, sister chromatic exchange and transformation in mammalian and
human cells in vitro; and it induces DNA damage in mammalian skin cells irradiated in vivo.
UVR in the three wavelength ranges can induce or enhance cellular and viral gene
expressiOn.
5.5 Evaluations
1
There is sufficient evidence in humans for the carcinogenicity of solar radiation. Solar
radiation causes cutaneous malignant melanoma and nonmelanocytic skin cancer.
There is limited evidence in humans for the carcinogenicity of exposure to ultraviolet
radiation from sunlamps and sunbeds.
There is inadequate evidence in humans for the carcinogenicity of exposure to fluo-
rescent lighting.
There is inadequate evidence in humans for the carcinogenicity of other sources of arti-
ficial ultraviolet radiation.
There is sufficient evidence for the carcinogenicity of solar radiation in experimental
animals.
There is sufficient evidence for the carcinogenicity of broad-spectrum ultraviolet radia-
tion in experimental animals.
There is sufficient evidence for the carcinogenicity of ultraviolet A radiation in experi-
mental animals.
1
For definition of the italicized terms, see Preamble.
There is sufficient evidence for the carcinogenicity of ultraviolet B radiation in experi-
mental animals.
There is sufficient evidence for the carcinogenicity of ultraviolet C radiation in experi-
mental animals.
Overall evaluation
Solar radiation is carcinogenic to humans (Group 1 ).
Ultraviolet A radiation is probably carcinogenic to humans (Group 2A).
Ultraviolet B radiation is probably carcinogenic to humans (Group 2A).
Ultraviolet C radiation is probably carcinogenic to humans (Group 2A).
Use of sunlamps and sun beds entails exposures that are probably carcinogenic to humans
(Group 2A).
Exposure to fluorescent lighting is not classifiable as to its carcinogenicity to humans
(Group 3)
CONTINUED
6. References
Aberer, W., Schuler. G., Stingl, G., Honigsmann, H. & Wolff, K. (1981) Ultraviolet light depeletes
surface markers ofLangerhans cells. J. invest. Dermatol., 16, 202-210
Adam, S.A., Sheaves, J.K., Wright, N.H., Mosser, G., Harris, R.W. & Vessey, M.P (1981) A case-
control study of the possible association between oral contraceptives and malignant melanoma.
Br. J. Cancer, 44, 45-50
Agin, P.P. , Rose, A.R, ill, Lane, C.C .. Akin, F.J. & Sayre, R.M. (1981a) Changes in epidermal
forward scattering absorption after UVA or UVA-UVB irradiation. J. invest. Dermatol., 16, 17-
177
Agin, P.P, Lane. C.C. & Sayre, R.M. (1981b) Ultraviolet irradiation induces optical and structural
changes in the skin of hairless mice. Photochem. Photobiophys., 3, 185-194
Alcalay, J. , Freeman, S.E. , Goldberg, L.H. & Wolf, .T.E. (1990) Excision repair of pyrimidine
dimers induced by simulated solar radiation in the skin of patients with basal cell carcinoma. J.
invest. Dermatol., 95, 506-509
Alter, B.J., Schendel, D.J., Bach, M.L., Bach, F.H., Klein, J. & Stimpfling, J.H. (1973)
Cell-mediated lympholysis. Importance of serologically defined H-2 regions. J. exp. Med. , 137,
1303-1309
Ambach, W & Rehwald, W. (1983) Measurements of the annual variation of the erythema dose of
global radiation. Radiat. environ. Biophys., 21, 295-303
American Conference of Governmental Industrial Hygienists ( 1991) 1991-1992 Threshold Limit
Values for C'hernical Substances and Physical Agents and Biological Exposure Indices,
Cincinnati, OH, pp. 123-128
Ananthaswamy, H.N. (1984a) Lethality and transformation of 10T1/2 mouse embryo fibroblast cell
line by various wavelengths of ultraviolet radiation. Photodermatology, 1, 265-176
Ananthaswamy, H.N. (1984b) Neoplastic transformation of neonatal mouse skin fibroblasts in
culture after a single exposure to ultraviolet radiation in vivo. Photodermatology, 1,10-113
Ananthaswamy, H.N. & Fisher, M.S. (1981) Photoreactivation of ultraviolet radiation-induced pyri-
midine dimers in neonatal BALB/c mouse skin. Cancer Res., 41, 1829-1833
Ananthaswamy, H.N. & Kripke, M.L. (1981) In vitro transformation of primary cultures of neonatal
BALB/c mouse epidermal cells with ultraviolet-B radiation. Cancer Res., 41, 2882-2890
Ananthaswamy, H.N. & Pierceall, W.E. (1990) Molecular mechanisms of ultraviolet radiation
carcinogenesis. Photochem. Photobiol., 52, 1119-1 136
Ananthaswamy, H.N., Price, J. E. , Goldberg, L.H. & Bales, E.S. (1988) Detection and identification
of activated oncogenes in human skin cancers occurring on sun-exposed bodysites. CancerRes.,
48, 3341-3346
Anderson, D.E. (1963) Effect of pigment on bovine ocular squamous carcinoma. Ann. N. Y.Acad.
Sci ., 100, 436-446
Anderson, R.R. & Parrish. J .A. ( 1981) The optics of human skin. J. in vest. Derrnatol., 77, 13-19
Anderson, R.R. & PatTish, J.A. (1982) Optical properties of human skin. In: Regan, J.D. & Parrish,
J.A., eds, The Science of Photo-medicine, New York, Plenum, pp. 147-194
-229-
Anderson. L.M. & Rice, J.M. (1987) Tumorigenesis in athymic nude mouse skin by chemical
carcinogenesis and ultraviolet light. J. natl Cancer lnst., 78, 125-134
Andley, U.P. (1987) Yearly review. Photodamage to the eye. Photochem. Photobiol., 46, 1057-1066
Anon. (1979) Sun-lamps, Solariums, Sunbeds and Health Lamps, London, Which? Consumer Asso-
ciation, pp. 268-272
Anon. (1987) The Truth About Tanning, London, Which? Consumer Association, pp. 21-216
Ansel, J.C., Luger, T.A. & Green, I. (1983) The effect of in vitro and in vivo UV irradiation on the
production of ET AF [ epidennal cell derived thymocyte-activating factor] activity by human and
murine keratinocytes. J. invest. Dermatol., 81, 519-523
Applegate, L.A. & Ley, R.D. ( 1988) Ultraviolet radiation-induced lethality and repair of pyrimidine
dimers in fish embryos. Mutat. Res., 198, 85-92
Applegate, L.A. & Ley. R.D. (1991) DNA damage is involved in the induction of opacification and
neovascularization of the cornea by ultraviolet radiation. Exp. Eye Res., 52, 493-497
Applegate, L.A., Ley, R.D., Alcalay, J. & Kripke, M.L. (1989) Identification of the molecular target
for the suppression of contact hypersensitivity by ultraviolet radiation. J. exp. Med. , 170,
1117-1131
Arlett, C.F. & Harcourt, S.A. ( 1983) Variation in response to mutagens amongst normal and repair-
defective human cells. In: Lawrence, C.W., ed., Induced Mutagenesis: Molecular Mechanisms
and their Implications for Environmental Protection, New York, Plenum, pp. 249-267
Arlett, C.F., Harcourt, S.A., Cole, J., Green, M.H. & Anstey, A.V (1992) A comparison of the
response of unstimulated and stimulated T -lymphocytes and fibroblasts from normal, xeroderma
pigmentosum and trichothiodystrophy donors to the lethal action of UV-C. Mutat. Res., 273,
127-135
Armstrong, B.K. (1984) Melanoma ofthe skin. Br. med. Bull., 40, 346-350
Armstrong, R. B. (1986) Solar urticaria. Dermatol. Clin., 4, 253-259
Armstrong, B.K. (1988) Epidemiology of malignant melanoma: intermittent or total accumulated
exposure to the sun? J. Dermatol. surg. Oneal. , 14, 835-849
Armstrong, B.K., Woodings, T.L. , Stenhouse, N.S. & McCall, M.G. (1983) Mortality from Cancer
in Migrants to Australia 1962 to 1971, Perth, University of Western Australia, pp. 21, 81-83
Armstrong, B.K., Heenan, P.J., Caruso, V., Glancy, R.J. & Flolman, C.D.J. (1984) Seasonal
variation in the junctional component of pigmented naevi (Letter to the Editor). Int. J. Cancer,
34,441-442
Armstrong, B.K., de Klerk, N.H. & Holman, C.D.J. (1986) Etiology of common acquired
melanocytic nevi: constitutional variables, sun exposure, and diet. J. natl Cancer lnst., 77,
329-335
Athas, W.F., Hedayati, M.A.. Matanoski, G.M., Fanner, E.R. & Grossman, L. (1991) Development
and field-test validation of an assay for DNA repair in circulating human lymphocytes. Cancer
Res., 51, 5786-5793
Atkin, M., Penning, J., Heady, J.A., Kennaway. E.L. & Kennaway, N.M. (1949) The mortality from
cancer of the skin and lip in certain occupations. Br. J. Cancer, 3, 1-15
Aubry, F. & MacGibbon, B. (1985) Risk factors of squamous cell carcinoma of the skin. A case-
control study in the Montreal region. Cancer, 55, 907-911
Auerbach, H. (1961) Geographic variation in incidence of skin cancer in the United States. Public
Health Rep., 76, 345-348
Augustsson, A., Stierner, U., Rosdahl, I. & Suurkiila, M. (19903 Melanocytic naevi in sun-exposed
and protected skin in melanoma patients and controls. Acta derm. venereal., 71, 512-517
REFERENCES 23 1
Azizova, A.O., lslomov. A.T., Roshchupkin, D.J. , Predvoditelev, D.A., Remizov, A.N. &
Vladimirov, Y.A. ( 1980) Free radicals formed on ultraviolet irradiation of the lipids of
biological membranes. Biophysics, 24, 407-414
Baadsgaard, 0., Fox, D.A. & Cooper, K.D. (1988) Human epidermal cells from ultraviolet light-
exposed skin preferentially activate autoreactive CD4 + 2H4 + suppressor-inducer lymphocytes
and CDS+ suppressor/cytotoxic lymphocytes. J. lmmunol., 140, 1738-1744
Bach, F.H., Grillot-Courvalin, C., Kuperman, O.J. , Sollinger, H.W., Hayes, C. , Sondel, RM., Alter,
B.J. & Bach, M. L. ( 1977) Antigenic requirements for triggering of cytotoxic T lymphocytes .
Immunol. Rev., 35, 76-96
Bachem, A. (1956) Ophthalmic ultraviolet action spectra. Ann. J. C)phthalmol., 41, 969-975
Baden, H.P & Pathak, M.A. ( 1967) The metabolism and function of urocanic acid in skin. J. invest.
Dermatol., 48, 11-17
Ba1shi, J.D., Francfort, J.W. & Perloff, L.J. (1985) The influence of ultraviolet irradiation on the
blood transfusion effect. Surgery, 98, 243-249
Banerjee, S.K., Christensen. R.B., Lawrence, C.W. & LeClerc, J.E. (1988) Frequency and spectrum
of mutations produced by a single cis-syn thymine-thymine cyclobutane dimer in a
single-stranded vector. Proc. nat/ Acad. Sci. USA, 85, 8141-8145
Banerjee, S.K., Borden, A. , Christensen, R.B. , LeClerc, J.E. & Lawrence, C.W. (1990)
SOS-Dependent replication past a single trans-syn T-T cyclobutane dimer gives a different
mutation spectrum and increased error rate compared to replication past this lesion in uninduced
cells. J. Bacteriol., 172, 2105-21 12
Barrett, J.C., Tsutsui, T. & Ts'o, P.O.P (1978) Neoplastic tranforn1ation induced by a direct pertur-
bation of DNA. Nature, 274, 229-232
Barrett, S.F., Robbins. J. H., Tarone, R.E. & Kraemer, K.H. (1991) Evidence fist defective repair of
cyclobutane pyrimidine dimers with normal repair and other DNA photoproducts in a trans-
criptionally active gene transfectcd into Cockayne syndrome cells. Mutat. Res., 255, 281-291
Barth, C., Knuschke, P. & Barth, J. (1990) Examination of welders exposed to UV radiation (Ger.).
Z. ges. Hyg. , 36, 654-55
Baturay, N.Z. , Targovnik, H.S., Reynolds, R.J. & Kennedy, A.R. (1985) Induction of in vitro
transformation by near-uv light and its interaction with 13-propiolactone. Carcinogenesis, 6, 465-
468
Beadle, PC. & Burton, J.L. (1981) Absorption of ultraviolet radiation by skin surface lipid. Br J.
Dermatol., 104, 549-551
Beardmore, G.L. (1972) The epidemiology of malignant melanoma in Australia. In: McCarthy,
W.H., ed., Melanoma anal Skin Cancer, Sydney, V.C.N. Blight (Goverment Printer). pp. 39-4
Bech-Thomsen. N., Wulf, H.C., Poulsen, T. & Lundgren, K. (1988a) Pretreatment with long-wave
ultraviolet light inhibits ultraviolet-induced skin tumor development in hairless mice. Arch.
Dermatol., 124, 1215-1218
Bech-Thomsen, N., Wulf, H.C. & Lundgren, K. (1988b) Pre-treatment with UVA influences broad-
spectrum UV photocarcinogencsis in hairless mice (Abstract). In: Rikles, E., ed., Proceedings of
the lOth International Congress on Photobiology Jerusalem, Jerusalem, International
Association of Biology, p. 34
Beer, J.Z. & Zmudzka, B.Z. (1991) Activation of human immunodeficiency virus by radiation. In:
Seymour. C.B. & Mothersill. C., eds, New Developments in Fundamental and Applied
Radiology, London. Taylor & Francis, pp. 113-123
Beer, J.Z., Olvey, K.M., Miller, S.A. & Zmudzka, B.Z. (1991) Studies on the human
immunodeficiency virus promoter induction by UV A and UVB radiation (Abstract).
Photochem. Photobiol., 53, 25S
Beitner, H. (1986) The effect of high dose long-wave ultraviolet radiation (UVA) on epidermal
melanocytes in human skin: a transmission electron microscopic study. Photodermatology, 3,
133-139
Beitner, H. & Wennersten, G. (1983) The immediate action of long-wave ultraviolet radiation
(UV A) on suprabasal melanocytes in human skin: a transmission electron microscopical study.
Acta denmatol. venereol., 63, 328-334
Beitner, H., Norell, S.E., Ringborg, U., Wennersten, G. & Mattson, B. (1990) Malignant melanoma:
aetiologieal importance of individual pigmentation and sun exposure. Br. 1. Dermatol., 122,
43-51
Bender, M.A., Griggs, H.G. & Walker, P.L. (1973) Mechanisms of chromosomal aberration
production. I. Aberration induction by ultraviolet light. Mutat. Res., 20, 387-402
Ben-Hur, E. & Ben-lshai, R. (1968) Trans-syn thymine dimers in ultraviolet-irradiated denatured
DNA: identification and photoreactivability. Biochim. biophys. Acta, 166, 9-15
Beral, V. & Robinson, N. (1981) The relationship of malignant melanoma, basal and squamous skin
cancers to indoor and outdoor work. Br J. Cancer, 44, 886-891
Beral, V., Evans, S., Shaw, H. & Milton, G. (1982) Malignant melanoma and exposure to
fluorescent lighting at work. Lancet, ii, 290-293
Berg, M. (1987) Ultraviolet radiation in operation rooms. In: Passchier, WF. & Bosnjakovic,
B.F.M., eds, Human exposure to Ultraviolet Radiation: Risk and Regulations, Amsterdam,
Elsevier, pp. 359-363
Berger, D.S. (1976) The sunburning ultraviolet meter: design and performance. Photochem. Photo-
bioi., 24, 587-593
Berger, H. & Kaase, H. (1983) Risk for carcinoma from sources of UV A radiation (Ger.). Z.
Hautkr., 58, 63
Berger, D.S. & Urbach. F. (1982) A climatology of sunburning ultraviolet radiation. Photochem.
Photobiol., 35, 187-192
Berger, H., Tsambaos, D. & Mahrle, G. (1980a) Experimental elastosis induced by chronic
ultraviolet exposure. Light and electron-rnicrosccypic study. Arch. dermatol. Res., 269, 39-49
Berger, H., Tsambaos, D. & Kaase, H. (1980b) Experimental actinic elastosis by chronic exposure
to filtered UV A radiation (Ger.). Z. Hautkr, 55, 1510-1527
Bergstresser, P.R. (1986) Ultraviolet B radiation induces 'local immunosuppression'. CurrenlProbl.
Dermatol., 15, 205-218
Bernhard, J.D., Pathak, M.A., Kochevar, I.E. & Parrish, J.A. (1987) Abnorinal reactions to
ultraviolet radiation. In: Fitzpatrick, T.B. , Eisen, A.Z., Wolff, K., Freedberg. I.M. & Austen,
K.F. , eds, Dermatology in General Medicine, 3rd ed., New York, McGraw-Hill, pp. 1481-1507
Beukers, R. & Berends, W. ( 1960) Isolation and identification of the irradiation product of thymine.
Biochim. biophys. Acta, 41, 550-551
Beukers, R. , Ylstra, J. & Berends, W. (1958) The effect of ultraviolet I ight on some components of
the nucleic acids. II. In rapidly frozen solutions. Reel. Trav chim. Pays-Bas, 77, 729-732
Bielfeld, V., Weichenthal, M., Roser, M., Breithart, E., Berger, J. , Seemanova, E. & Rudiger. H.W.
(1989) Ultraviolet-induced chromosomal instability in cultured fibroblasts of heterozygote
carriers for xeroderma pigmentosum. Cancer Genet. Cytogenet ., 43, 219-226
Bissett, D.L., Flannon. D.R. & Orr, TV (1987) An animal model of solar-aged skin: histological,
physical, and visible changes in UV -irradiated hairless mouse skin. Photochem. Photobiol. , 46,
367-378
REFERENCES 233
Bissett, D.L., Hannon, D.P & Orr, TV. (1989) Wavelength dependence of histological, physical,
and visible changes in chronically UV-irradiated hairless mouse skin. Photochem. Photobiol.,
50,763-769
Blum, H.F. (1948) Sunlight as a causal factor in cancer of the skin of man. J. natllnst. Cancer, 9,
247-258
Blum, H. F. (1959) Carcinogenesis by Ultraviolet Light, Princeton, NJ, Princeton University Press
Blum, H.F. & Lippincott, S.W. (1942) Carcinogenic effectiveness of ultraviolet radiation of
wavelength 2537 A. J. natl Cancer lnst., 3, 211-216
Blum, H.F., Butler, E.G., Dailey, T.H., Daube, J.R., Mawe, R.C. & Soffen, G.A. (1959) Irradiation
of mouse skin with single doses of ultraviolet light. J. natl Cancer /nst., 22, 979-993
Blumthaler, M., Canaval, H. & Ambach, W. (1983) Measurement of erythemally active doses of
solar UVB radiation (Ger.). Z. Med.-Meteorol., 2,14-17
Blumthaler, M., Ambach, W. Ac Daxecker, F. (1985a) Computation of radiation doses from
different solar UVB radiations producing photoelectric keratosis (Ger.). Klin. Mbl. Augenheilk.,
186, 275-278
Blumthaler, M., Ambach, W. & Canaval, H. (1985b) Seasonal variation of solar UV-radiation at a
high mountain station. Photochem. Photobiol. , 42, 147-152
Boettner, E.A. & Wolter, J.R. (1962) Transmission of the ocular media. Invest. Ophthalmol., I,
776-783
Bootsma, D. & Hoeijmakers, J.H.J. (1991) The genetic basis of xeroderma pigmentosum. Ann.
Genet., 34, 143-150
Borek, C., Ong, A. & Mason, H. (1989) Ozone and ultraviolet light act as additive cocarcinogens to
induce in vitro neoplastic transformation. Teratog. Carcinog. Mutag., 9, 71-74
Bos, A.J.J. & de Haas, M.P (1987) On the safe use of a high power ultraviolet laser. In: Passchier,
W.F. & Bosnjakovic, B.F.M., eds, Human Exposure to Ultraviolet Radiation: Risks and
Regulations, Amsterdam, Elsevier, pp. 377-382
Bose, B., Agarwal, S. & Chatterjee, S.N. (1989) UVA induced lipid peroxidation in liposomal
membrane. Radiat. environ. Biophys., 28, 59-65
Bouissou, H., Pieraggi, M.-T., Julian, M. & Savit, T. (1988) The elastic tissue of the skin. A
comparison of spontaneous and actinic (solar) aging. Int. J. Dermatol., 27, 327-335
Bowden, G.T., Trosko, J.E., Shapas, B.G. & Boutwell, R.K. (1975) Excision of pyrimidine dimers
from epidem1al DNA and nonsemiconservative epidermal DNA synthesis following ultraviolet
irradiation of mouse skin. Cancer Res., 35, 3599-3607
Bowker, K.W. & Longford, A.R (1987) Ultraviolet radiation hazards from the use of solaria. In:
Passchier, W.F. & Bosnjakovic, B.F.M., eds, Human Exposure to Ultraviolet Radiation: Risks
and Regulations, Amsterdam, Elsevier, pp. 365-369
Boyd, R.W. (1983) Radiometry and the Detection of Optical Radiation, New York, John Wiley &
Sons
Boyle, J., MacKie, R.M. , Briggs, J.D. , Junor, B.J.R. & Aitchison, T.C. (1984) Cancer, warts and
sunshine in renal transplant patients. A case-control study. Lancet, i, 702-705
Bradley, M.O. & Sharkey, N.A. (1977) Mutagenicity and toxicity of visible fluorescent light to
cultured mammalian cells. Nature, 266, 724-726
Bradley, M.O., Hsu, I.C. & Harris, C.C. (1979) Relationships between sister chromatic exchange
and mutagenicity, toxicity and DNA damage. Nature, 282, 318-320
Brash, D.E. & Haseltine, W.A. (1982) UV-Induced mutation hotspots occur at DNA damage hot-
spots. Nature, 298, 189-192
Brash, D.E., Rudolph. J.A., Simon, J.A., Lin, A., McKenna, G.J., Baden. H.P, Halperin, A.J. &
Ponten, J. (1991) A role for sunlight in skin cancer: UV-induced p53 mutations in squamous cell
carcinoma. Proc. natlAcad. Sci. USA, 88, 10124-10128
Bredberg. A., Kraemer, K.H. & Seidman, M.M. (1986) Restricted ultraviolet mutational spectrum
in a shuttle vector propagated in xeroderma pigmentosum cells. Proc. nat/ Acad. Sci. USA, 83,
8273-8277
Bridges, B.A. (1988) Mutagenic DNA repair in Escherichia coli. XVI. Mutagenesis by ultraviolet
light plus delayed photoreversal in recA strains. Mutat. Res. , 198, 343-350
Bridges, B.A. (1990) Sunlight, DNA damage and skin cancer: a new perspective. Jpn J. Cancer
Res., 81, 105-107
Bridges, B.A. & Bates, H. (1990) Mutagenic DNA repair in Escherichia coli. VIII. Involvement of
DNA polymerase III a-subunit (DnaE protein) in mutagenesis alter exposure to UV light,
Mutagenesis, 5, 35-38
Bridges, B.A. & Brown. G.M. (1992) Mutagenic repair in Escherichia coli. XXI. A stable SOS-
inducing signal persisting after excision repair of ultraviolet damage. Mutat. Res. (in press)
Bridges, B.A., Woodgate, R., Ruiz-Rubio, M., Sharif, F., Sedgwick, S.G. & Hiibscher. U. (1987)
Current understanding of UV-induced base pair substitution mutation in E. coli with particular
reference to the DNA polymerase III complex. Mutat. Res., 181, 219-226
Broughton, B.C., Lenmann. A.R., Harcourt, S.A., Arlett, C.F., Sarasin, A., Kleijer W.J., Beemer,
F.A., Nairn, R. & Mitchell, D. L. (1990) Relationship between pyrimidine dimers, 6-4
photoproducts, repair synthesis and cell survival: studies using cells from patients with
trichothiodystrophy. Mutut. Res., 235, 33-40
Brown, L.R. & Wuthrich, K. (1977) A spin label study of lipid oxidation catalyzed by heme
proteins. Biochim. biophys. Acta, 464, 356-369
Brown, J., Kopf, A.W., Rigel, D.S. & Fridmai1, R.J. (1984) Malignant melanoma in World War ll
veterans. Int. J. Dermatol., 23, 661-663
Bruggers, J.H.A., de Jong, W.E., Bosnjakovic. B.F.M. & Passchier, W.F. (lti87) Use of artificial
tanning equipment in the Netherlands. In: Passchier, WE & Bosnjakovic, B.F.M., eds, Human
Exposure to Ultraviolet Radiation: Risk and Regulations, Amsterdam, Elsevier, pp. 235-239
Bruls, W.A.G., Slaper. H., van der Leun, J.C. & Berrens, L. (1984a) Transmission of human
epidermis and stratum corneum as a function of thickness in the ultraviolet and visible
wavelengths. Photochem. Photohiol. , 40, 485-494
Bruls, W.A.G., van Weelden, H. & van der Leun, J.C. (1984b) Transmission of UV-radiation
through human epidermal layers as a factor influencing the minimal erythema dose. Photochem.
Photobiol., 39, 63-67
Bruyneel-Rapp, F., Dorsey, S.B. & Guin, J.D. (1988) The tanning salon: an area survey of
equipment, procedures, and practices. J. Am. Acad. Dern1atol, 18, 1030-1038
Burki, H.J. & Lam, C.K. (1978) Comparison of the lethal and mutagenic effects of gold and white
fluorescent lights on cultured mammalian cells. Mutat. Res., 54, 373-377
Burnet, F.M. (1971) Implications of immunological surveillance for cancer therapy. /sr. J. med.
Sci., 7, 9-16
Calkins, J., Selby, C. & Enoch, H.G. (1987) Comparison of UV action spectra for lethality and
mutation in Salmonella typhimurium using a broad band source and monochromatic radiations.
Photochem. Photobiol., 45, 631-636
REFERENCES 235
Cameron, L.L., Vitasa, B.C., Lewis, P.G., Taylor, H.R. & Emmett, E.A. (1988) Visual assessment
of facial elastosis using photographs as a measure of cumulative ultraviolet exposure. Photo-
dermatology, 5, 277-282
Caplan, R.M. (1967) Medical uses ofthe Wood's lamp. J. Am. med. Assoc., 202, 1035-1038
Cavard, C., Zider, A. , Vernet, M., Bennoun, M., Saragosti, S., Grimber, B. & Briand, P. (1990) In
vivo activation by ultraviolet rays of the human immunodeficiency virus type 1 long terminal
repeat. J. clin. Invest., 36, 1369-1374
Cerutti, PA. (1981) Measurement of thymine damage induced by oxygen radical species. In:
Friedberg, E.C. & Hanawalt, PC., eds, DNA Repair. A laboratory, Manual of Research
Procedures. Vol. 1, Part A, New York, Marcel Dekker, pp. 57-68
Cerutti, PA. & Netrawali, M. (1979) Formation and repair of DNA damage induced by indirect
action of ultraviolet light in normal and xerodenna pigmentosum skin fibroblasts. Radiat. Res.,
Suppl., 423-432
Cesarini, J.-P & Muel, B. (1989) Erythema induced by quartz-halogen sources. Photodermatology,
6, 222-227
Challoner, A.V.J., Corless, D. , Davis, A., Deane, G.H.W., Diffey, B.L., Gupta, S.P. & Magnus, I.A.
(1976) Personnel monitoring of exposure to ultraviolet radiation. Clin. exp. Dermatol. , 1,
175-179
Challoner, A.V.J., Corbett, M.F., Davis, A., Diffey, B.L. , Leach, J.F. & Magnus, I.A. (1978) Des-
cription and application of a personal ultraviolet dosimeter: a review of preliminary studies. Natl
Cancer lnst. Monogr., 50, 97-100
Chamberlain, J. & Moss, S.H. (1987) Lipid pcroxidation and other membrane damage produced in
Eschenchia cold EC1060 by near-UV radiation and deuterium oxide. Photochem. Photobiol. ,
45, 625-630
Chan, G.L. & Little, J.B. (1976) Induction of oncogenic transformation in vitro by ultraviolet light.
Nature, 264, 442-444
Chan, G.L. & Little, J .B. (1979) Correlation of in vitro transformation with in vivo tumorigenicity
in 10TI/2 mouse cells exposed to UV light. Br. J. Cancer, 39, 590-593
Chan, G.L., Peak, M.J., Peak, J.G. & Haseltine, W.A. (1986) Action spectrum for the formation of
endonuclease-sensitive sites and (6-4) photoproducts induced in a DNA fragment by ultraviolet
radiation. Int. J. Radiat. Bioi. , 50, 641-648
Chao, C.C.-K. & Rosenstein, B.S. (1985) Induction of sister-chromatic exchanges in ICR 2A frog
cells exposed to 254 nm and solar UV wavelengths. Photochem. Photobiol., 41, 625-627
Chao, C.C.-K. & Rosenstein, B.S. (1986) Use of metabolic inhibitors to investigate the excision
repair of pyrimidine dimers and non-dimer DNA damages induced in human and ICR 2A frog
cells by solar ultraviolet radiation. Photochem. Photobiol. , 43, 165-170
Chu, E.H.Y. (1965a) Effects of ultraviolet radiation on mammalian cells. I. Induction of
chromosome aberrations. Mutat. Res., 2, 75-94
Chu, E.H. Y. ( 1965b) Effects of ultraviolet radiation on mammalian cells. II. Differential UV and
x-ray sensitivity of chromosomes to breakage in 5-aminouracil synchronized cell populations.
Genetics, 52, 1279-1294
Chung, H.-T., Burnham, D.K., Robertson, B., Roberts, L.K. & Daynes, R.A. (1986) Involvement of
prostaglandins in the humane alterations caused by the exposure of mice to ultraviolet radiation.
J. lmmunol., 137, 2478-2484
Churchill, M.E., Peak, J.G. & Peak, M.J. (1991) Correlation between cell survival and DNA single-
strand break repair proficiency in the Chinese hamster ovary cell lines AA8 and EM9 irradiated
with 365-nm ultraviolet A radiation. Photochem. Photobiol., 53, 229-236
Cieminis, K.G.K., Ranceliene, V.M., Prijalgauskiene, A.J., Tiunaitiene, N.V., Rudzianskaite, A.M.
& Jancys, Z .. J. (1987) Chromosome and DNA damage and their repair in higher plants
irradiated with short-wave ultraviolet light. Mutat. Res., 181, 9-16
Claas, F.H.J. , Rothert, M., Havinga, I., Schothorst, A.A. , Vermeer, B.-J. & van Rood, J.J. (1985)
Influence of ultraviolet radiation treatment on the survival of heterotopic skin grafts in the
mouse. J. invest. Dermatol. , 84, 31-32
Clark, J.H. ( 1964) The effect of long ultraviolet radiation in the development of tumors induced by
20-methylcholanthrene. Cancer Res., 24, 207-211
Cleaver, J.E. (1968) Defective repair replication of DNA in xeroderma pigmentosuin. Nature, 218,
652-656
Cleaver, J.E. & Kraemer, K.H. (1989) Xeroderma pigmentosum. In: Scriver. C.R., Beaudet, A.L.,
Sly, W.S. & Valle, D., eds, The Metabolic Basis of Inherited Disease, New York, McGraw-Hil1,
pp. 2949-2971
Clemmesen, J. (1965) Statistical Studies in theAetiology of Malignant Neoplasms. I. Review and
Results, Copenhagen, Munksgaard, p. 409
Cole, C.A., Forbes. P.D. , Davies, R.E. & Urbach, F. (1985) Effect of indoor lighting on normal
skin. Ann. N.Y. Acad. Sci., 453, 305-3 16
Cole, C.A., Forbes, P.D. & Davies, R.E. (1986) An action spectrum for UV photocarcinogenesis.
Photochem. Photobiol., 43,275-284
Colella, C.M., Bogani, P. Agati, G. & Fusi, E (1986) Genetic effects ofUVB: mutagenicity of 308
nm light in Chinese hamster V79 cells. Photochem. Photobiol., 43,437-442
Commission lnternationale de l'Eclairage (International Cornn1ission on Illumination) (1987) Voca-
bulaire International de l'Eclairage [International Lighting Vocabulary! (CIE Publication No.
17.4), 4th ed., Geneva, Bureau Central de Ia Commission E1ectrotechique lntemationale, p. 3
Coohill, T.P., Peak, M.J. & Peak, J.G. (1987) The effects of the ultraviolet wavelengths of radiation
present in sunlight on human cells in vitro. Photochem. Photobiol.., 46, 1043-1050
Cooke, K.R. & Fraser, J. (1985) Migration and death from malignant melanoma. Int. J. Cancer, 36,
175-178
Cooke, A. & Johnson, B.E. (1978) Dose response, wavelength dependence and rate of excision of
ultraviolet radiation induced pyrimidine dimers in mouse skin DNA. Biochim. biophys. Acta,
517, 24-30
Cooke, K.R., Skegg, D.C.G. & Fraser, J. (19S4) Socio-economic status, indoor and outdoor work,
and malignant melanoma. Int. J. Cancer, 34, 7-62
Cooper, K.D., Fox, P., Neises, G. & Katz, S.I. (1955) Effects of ultraviolet radiation on human
epidermal cell alloantigen presentation: initial depression of Langerhans cell-dependent
function is followed by the appearance of T6-Dr+ cells that enllance epidermal alloantigen
presentation. J.lmmunol., 134, 129-137
Cooper, K.D., Neises, G.R. & Katz. S.I. (1986) Antigen-presenting OKM5 + melanophages appear
in human epidermis after ultraviolet radiation. J. invest. Dermatol., 86, 363-370
Corominas. M., Karnino. H. , L.eon, J. & Pellicer, A. (1989) Oncogene activation in human benign
tumors of the skin (keratoacanthon1as): Is H-ras involved in dii'ferentiation as well as proli-
feration? Proc. natl Acad. Sci. USA, 86,6372-6376
Cox, C.W.J. (1987) Ultraviolet irradiance levels in welding processes. In: Passchier. W.F-. &
Bosnjakovic. B.FM., eds. Human Exposure to Ultraviolet Racticition: Risks und Regulations,
Amsterdam, Elsevier, pp. 383-386
REFERENCES 237
Cripps, D.J. , Horowitz, S. & Hong, R. (1978) Spectrum of ultraviolet radiation on human B and T
lymphocyte viability. Clin. exp. Dermatol., 3, 43-50
Cristofolini, M., Franceschi, S., Tasin, L. , Zumiani, G., Piscioli, F., Talamini, R. & LaVecchia, C.
( 1987) Risk factors for cutaneous malignant melanoma in a northern Italian population. Int. J.
Cancer, 39, 150-154
Crombie, I.K. (1981) Di stribution of malignant melanoma on the body surface. Br. J. Cancer, 43,
842-849
Cullen, A.P (1980) Additive effects of ultraviolet radiation. Am. J. Optometry physiol. Optics. 57,
808-814
Cunningham, M.L., Johnson, J.S., Giovanazzi, S.M. & Peak, M.J. (1985) Photosensitized
production of superoxide anion by monochromatic (290-405 nm) ultraviolet irradiation of
NADH and NADPH coenzymes. Photochem. Photobiol., 42, 125-128
Czernielewski, J.M., Masouye, 1. , Pisani, A., Ferracin, J., Auvolat, D. & Ortonne, J.-P ( 1988) Effect
of chronic sun exposure on human Langerhans cell densities. Photodermatology 5, 116-120
Czochralska, B., Kawczynski, W. , Bartosz, G. & Shugar, D. (1984) Oxidation of excited-state
NADH and NAD dimer in aqueous medium- involvement of 0
2
- as a mediator in the presence
of oxygen. Biochim. biophys. Acta, 801, 403-409
D'Ambrosio, S.M., Slazinski. L. , Whetstone, J.W. & Lowney, E. (1981) Excision repair of
UV-induced pyrimidine dimers in human skin in vivo. J. invest. Dermatol., 77, 3l1-313
D'Ambrosio, S.M. , Bisaccia, E., Whetstone, J.W., Scarborough, D.A. & Lowney, E. (1953) DNA
repair in skin of lupus erythematosus following in vivo exposure to ultraviolet radiation J.
invest. Dermatol. , 81, 452-454
Danpure, H.J. & Tyrrell, R.M. (1976) Oxygen-dependence of near-UV (365 nm) lethality and the
interaction of near-UV and x-rays in two mammalian cell lines. Photochetn Photohiol., 23,
171-177
Dardanoni, L.. Gafa. L.. Paterno, R. & Pavone, G. (1984) A case-control study on lip cancer risk
factors in Ragusa (Sicily). Int. J. Cancer, 34, 335-337
Davies, R.E. & Forbes. P. D. ( 1988) Retinoids and photocarcinogenesis: a review. J. Toxicol.
cutaneous ocul. Toxicol., 7, 241-253
Davies, R.E., Dodge, H.A. & DeShields, L.H. (1972) Alteration of the carcinogenic activity of
7,12-dirnethylbenz(a)anthracene (DMBA) by light (Abstract no. 54). Proc. Am. Assoc. Cancer
Res., 13, 14
Davis, A. & Sims. D. (1983) Weathering of Polymers, London, Applied Science
Daynes, R.A., Spellman, C.W, Woodward, J.G. & Stewart, D.A. (1977) Studies into the
transplantation biology of ultraviolet light-induced tumors. Transplantation, 23, 343-348
Daynes, R.A., Spikes. J.D. & Krueger, G. (1983) Experimental and Clinical Photoimmunology,
Vols I and II, Boca Raton, FL, CRC Press
Dearden, S.J., Hunter, T.F. & Philp, J. (1981) Bilayer microviscosity changes due to 0
2
1
~ g )
peroxidation in lipid vesicles. Chern. Phys. Lett., 81, 606-609
Dearden, S.J.. Hunter, TE & Philp, J. (1985) Fatty acid analysis as a function of photo-oxidation in
egg yolk lecithin vesicles. Photochetn. Photobiol., 41, 213-215
De Fabo, E. C. & Kripke, M.L. (1980) Wavelength dependence and dose-rate independence of UV
radiation-induced immunologic unresponsiveness of mice to a UV- induced fibrosarcoma.
Photochem. Photobiol .. 32, 183-188
De Fabo, E. C. & Noonan, F.P. (1983) Mechanism of immune suppression by ultraviolet irradiation
in vivo. I. Evidence for the existence of a unique photoreceptot in skin and its role in photo-
immunology. J. exp. Med., 157, 84-98
De Flora, S., Camoirano. A., Izzotti , A. & Bennicelli, C. (1990) Potent genotoxicity of halogen
lamps, compared to fluorescent light and sunlight. Carciriogenesis, 11, 2171-2177
Denkins, Y., Fidler, I.J. & Kripke, M.L. ( 1989) Exposure of mice to UVB radiation suppresses
delayed hypersensitivity to Candida albicans. Photochem. Photobiol. , 49, 615-619
Desai , I.D., Sawant, P.L. & Tappe!, A.L. (1964) Peroxidative and radiation damage to isolated Iyso-
somes. Biochim. biophys. Acta, 86, 277-285
De Weerd-Kastelein, E.A., Keijzer, W., Rainaldi, G. & Bootsma, D. (1977) Induction of sister chro-
matid exchanges in xeroderma pigmentosum cells after exposure to ultraviolet light. Mutat.
Res., 45, 253-261
Diffey, B.L. ( 1982) The consistency of studies of ultraviolet erythema in normal human skin. Phys.
me d. Biol., 27, 715-720
Diffey, B.L. (1986) Use ofUV A sunbeds for cosmetic tanning. Br. J. Derrnatol., 115, 67-76
Diffey, B. (1987a) A comparison of dosimeters used for solar ultraviolet radiometry. Photochem.
Photobiol., 46, 55-60
Diffey, B.L. ( 1987b) Analysis of the risk of skin cancer from sunlight and solaria in subjects living
in northern Europe. Photodermatology, 4, 118-126
Diffey, B.L. (1987c) Cosmetic and medical applications of ultraviolet radiation: risk evaluation and
protection techniques. In: Passchier, W.P. & Bosnjakovic, B.F.M., eds, Human Exposure to
Ultraviolet Radiation: Risks and Regulations, Amsterdam, Elsevier, pp. 305-315
Diffey, B.L. (1988) The risk of skin cancer from occupational exposure to ultraviolet radiation in
hospitals. Phys. me d. Bioi. , 33, I 187-1193
Diffey, B.L. (1989a) Ultraviolet radiation and skin cancer. Are physiotherapists at risk? Physio-
therapy, 75, 615-616
Diffey, B.L. (1989b) Ultraviolet radiation dosimetry with polysulphone film. In: Diffey. B.L., ed.,
Radiation Measurement in Photobiology, London, Academic Press, pp. 135-159
Diffey, B.L. (1990a) Human exposure to ultraviolet radiation. Semin.. Dermatol., 9. 2-10
Diffey, B.L. (1990b) Ultraviolet radiation lainps for the phototherapy and photochemotherapy of
skin disease. In: Champion. R.H. & Pye, R.J., eds, Recent Advances in Dermatology, Vol. 8.
Edinburgh, Churchill Livingstone, pp. 21-40
Diffey, B.L. ( 1991) Solar ultraviolet radiation effects on biological systems. Phys. med. Bioi., 36,
299-328
Diffey, B.L. & Parr, PM. (1987) An appraisal of ultraviolet lamps used for the photoytherapy of
psoriasis. Br. J. Dermatol..l17, 49-56
Diffey, B.L. & Parr, P.M. (1989) The normal range in diagnostic phototesting. Br J. Dermatol., 120,
517-524
Diffey, B.L. & Farr, PM. (1991a) Quantitative aspects of ultraviolet erythema. Clin. phys. physiol.
Meas., 12,311 -325
Diffey, B.L . & Parr, P.M. (1991b) Tanning with UVB orUVA: an appraisal ofrisks. J. Photochem.
Photobiol. B. Bioi. , 8, 219-223
Diffey, B.L. & Langley, F.C. (1986) Evaluation of Ultraviolet Radiation Hazards in Hospitals
(Report No. 49), London, institute ofPhysica1 Sciences in Medicine
Diffey, B.L. & McKinlay, A.E (1983) The UVB content of UVA fluorescent lamps' and its
erythemal effectiveness in human skin. Phys. med. Biol. , 28, 351-358
Diffey, B.L., Kerwin, M. & Davis, A. (1977) The anatomical distribution of sunlight. Br. J.
Dermatol., 97, 407-410
REFERENCES 239
Diffey, B.L., Tate, T.J. & Davis, A. (1979) Solar dosimetry of the face: the relationship of natural
ultraviolet radiation exposure to basal cell carcinoma localization. Phys. med. Bioi., 24, 931-939
Diffey, B.L., Challoner, A.V.J. & Key, P.J. (1980) A survey of the ultraviolet radiation emissions of
photochemotherapy units. Br. J. Dermatol., 102, 301-306
Diffey. B.L., Larko, 0., Meding, B., Edeland, H.G. & Wester, U .. (l986) Personal monitoring of
exposure to ultraviolet radiation in the car manufacturing industry. Ann. occup. Hyg., 30,
163-170
Diffey, B.L., Farr, PM. & Oakley, A.M. (1987) Quantitative studies on UV A-induced erythema in
human skin. Br. J. Dermatol., 117, 57-66
DiPaolo, J.A. & Donovan, P.J. (1976) In vitro morphologic transtonnation of Syrian hamster cells
by UV-irradiation is enhanced by X-irradiation and unaffected by chemical carcinogens. Int. J.
radiat. Bioi., 30,41-53
Doda, D.D. & Green. A.E.S. ('I 98()) Surface reflectance measurements in the UV A from an
airborne platform. Part 1. Appl. Optics, 19, 2140-2145
Doll, R. (1991) Urban and rural factors in the aetiology of cancer. Int. J. Cancer, 47, 803-810
Doll, R., Muir, C. & Waterhouse, J., eds (1970) Cancer Incidence in Five Continents, Vol. II,
Berlin, Springer
Doniger, J. , Jacobson, E.D., Krell, K. & DiPaolo, J.A. (1981) Ultraviolet light action spectra for
neoplastic transfonnation and lethality of Syrian hamster embryo cells correlate with spectrum
for pyrimidime dimer formation in cellular DNA. Proc. natl Acad. Sci. USA, 78., 2378-2382
Dorado, G., Steingrimsdottir, H. , Arlett, C.F. & Lehmann, A.R. ( 1991) Molecular analysis of ultra-
violet-induced mutations in a xeroderma pigmentosum cell line. J. mol. Bioi., 217, 217-222
Dorn, H.F. ( 1944a) Illness from cancer in the United States. Public Health Rep., 59, 33-48
Dorn, H.F. (1944b) Illness front cancer in the United States: IV Illness from cancer of specific sites
classed in broad groups. Public Health Rep., 59, 65-77
Dorn, H.F. (1944c) Illness from cancer in the United States: VI. Regional differences in illness from
cancer. Public Health Rep., 59, 97-ll4
Dorn, C.R., Taylor, D.O.N. & Schneider, R. (1971) Sunlight exposure and risk of developing
cutaneous and oral squamous cell carcinomas in white cats. J. natl Cancer Inst., 46, I 073-1078
Doudney, C.O. (1976) Mutation in ultraviolet light-damaged microorganisms. In: Wang, S.Y., ed.,
Photochemistry and Photobiology of Nucleic Acids, Vol. 2, New York, Academic Press, pp.
309-374
Dougherty, M.A., McDermott, R.J. & Hawkins, M.J. (1988) A profile of users of commercial
tanning salons. Health Values, 12, 21-28
Dreosti. I.E., Baghurst, PA., Partick, E.J. & Turner, J. (1990) Induction of micronuclei in cultured
marine splenocytes exposed to elevated levels of ferrous ions, hydrogen peroxide and ultraviolet
irradiation. Mutat. Res., 244, 337-343
Drobetsky, E.A. & Glickman, B.W. (1990) The nature of ultraviolet light-induced mutations at the
heterozygous aprt locus in Chinese hamster ovary cells. Mutat. Res., 232, 281-289
Dubin, N., Moseson, M. & Pasternack, B.S. (1986) Epidemiology of malignant melanoma:
pigmentary traits, ultraviolet radiation, and the identify cation of high-risk populations. Recent
Results Cancer Res., 102, 56-75
Dubin, N., Moseson, M. & Pasternack, B.S. (1989) Sun exposure and malignant melanoma among
susceptible individuals. Environ. Health Perspectives, 81, 139-151
Eaton, G.J., Custer, R.P & Crane, A.R. (1978) Effects of ultraviolet light on nude mice. Cutaneous
carcinogenesis and possible leukemogenesis. Cancer, 42, 182- 188
Ebbesen, P. (1981) Enhanced lymphoma incidence in BALB/c mice after ultraviolet light treatment.
J. natl Cancer Inst., 67, 1077-1078
Eggset, G., Volden. G. & Krokan, H. (1983) UV-Induced DNA damage and its repair in human skin
in vivo studied by sensitive immunohistochemical methods. Carcinogenesis, 4, 745-750
Eggset, G., Krokan, H. & Volden, G. (1986) UV-Induced DNA damage and its repair in tanned and
untanned human skin in vivo. Photobiochem. Photobiophys. , 10, 181-187
Eisenstark, A. (1987) Mutagenic and lethal effects of near-ultraviolet radiation (290-400 nm) on
bacteria and phage. Environ. mol. Mutag., 10, 317-337
Eklund, G. & Malec, E. (1978) Sunlight and incidence of cutaneous malignant melanoma. Effect of
latitude and domicile in Sweden. Scand. J. plast. reconstr. Surg., 12, 231-241
Elkind, M.M. & Han, A. (1978) DNA single-strand lesions due to 'sunlight' end UV light: a compa-
rison of their induction in Chinese hamster and human cells, and their fate in Chinese hamster
cells. Photochem. Photobiol., 27, 717-724
Ellison, M.J. & Childs, J.D. (1981) Pyrimidine dimers induced in Escherichia coli DNA by
ultraviolet radiation present in sunlight. Photochem. Photobiol. , 34, 465-469
Elmets, C.A., Bergstresser, P.R., Tigelaar, R.E., Wood, P.J. & Streilein, J.W. (1983) Analysis of the
mechanism of unresponsiveness produced by haptens painted on skin exposed to low dose ultra-
violet radiation. J. exp. Med., 158, 781-794
Elmets, C.A., LeVine, M.J. & Bickers, D.R. (1985) Action spectrum studies for induction of
immunologic unresponsiveness to dinitrofluorobenzene following in vivo low dose ultraviolet
radiation. Photochem. Photobiol., 42, 391-397
Elmets, C.A., Larson, K., Urda, G.A. & Schacter, B. (1987) Inhibition of postbinding target cell
lysis and of lymphokine-induced enhancement of human natural killer cell activity by in vitro
exposure to ultraviolet B radiation. Cell. lmmunol., 104, 47-58
Elwood, J.M. (1989a) Epidemiology and control of melalloma in white populations and in Japan. J.
invest. Dennatol. , 92, 214S-221S
Elwood, J.M. (1989b) Epidemiology of melanoma: its relationship to ultraviolet radiation and ozone
depletion. In: Jones, R.R. & Wigley, T., eds, Ozone Depletion. Health and Environmental
Consequences, New York, John Wiley & Sons, pp. 169-189
Elwood, J.M. & Gallagher, R.P (1983) Site distribution of malignant melanoma. Can. med. Assoc.
]. , 128, 1400-1404
Elwood, J.M., Lee, J.A.H., Walter, S.D., Mo, T. & Green, A.E.S. (1974) Relationship of melanoma
and other skin cancer mortality to latitude and ultraviolet radiation in the United States and
Canada. Int. J. Epidemiol. , 3, 325-332
Elwood, J.M., Gallagher, R.P, Hill, G.B., Spinelli, J.J., Pearson, J.C.G. & Threlfall, W. (1984)
Pigmentation and skin reaction to sun as risk factors for cutaneous melanoma: Western Canada
Melanoma Study. Br. med. 1., 288, 99-102
Elwood, J.M., Gallagher, R.P, Davison, J. & Hill, G.B. (1985a) Sunburn, suntan and the risk of
cutaneous malignant melanoma: the Western Canada Melanoma Study. Br. J. Cancer, 51,
543-549
Elwood, J.M., Gallagher, R.P, Hill, G.B. & Pearson, J.C.G. (1985b) Cutaneous melanoma in
relation to intermittent and constant sun exposure: the Western Canada Melanoma Study. Int. J.
Cancer, 35, 427-433
Elwood, J.M., Williamson, C. & Stapleton, P.J. (1986) Malignant melanoma in relation to moles,
pigmentation, and exposure to fluorescent and other lighting sources. Br J. Cancer, 53, 65-74
Elwood, J.M., Gallagher, R.P, Worth, A.J. , Wood, W.S. & Pearson. J.C.G. (1987) Etiological
differences between subtypes of cutaneous malignant melanoma: Western Canada Melanoma
Study. J. nat! Cancer lnst., 18, 37-44
REFERENCES 241
Elwood, J.M., Whitehead, S.M., Davison, J., Stewart, M. & Galt, M. (1990) Malignant melanoma in
England: risk associated with naevi, freckles, social class, hair colour, and sunburn. Int. J.
Epidemlol., 19, 801-810
Emmett, E.A. (1973) Ultraviolet radiation as a cause ofskin tumors. Crit. Rev. Toxicol., 2, 211-255
Emmett, E.A. & Horstman, S. W (1976) Factors influencing the output of ultraviolet radiation
during welding. J. occup. Med., 18,41-44
Engel, A., Johnson, M.-L. & Haynes, S.G. (1988) Health effects of sunlight exposure in the United
States. Results from the first National Health and Nutrition Examination Survey, 1971-1974.
Arch. Derrnatol. , 124, 72-79
English, D.R., Rouse, I.L., Xu, Z., Watt, J.D., Holman, C.D.J., Heenan. P.J. & Armstrong, B.K.
(1985) Cutaneous malignant melanoma and fluorescent lighting. J. natl Cancer Inst., 74,
1191-1197
Enninga, I.C., Groenendijk, R.T.L., Filon, A.R., van Zeeland, A.A. & Simons. J.W.I.M. (1986) The
wavelength dependence of UV-induced pyrimidine dimer formation, cell killing and mutation
induction in human diploid skin fibroblasts. Carcinogenesis, 7, 1899-1836
Epstein, J.H. (1965) Comparison of the carcinogenic and cocarcinogenic effects ofultraviolet light
on hairless mice. J. nat/ Cancer Inst., 34, 741-745
Epstein, J.H. (1985) Animal models for studying photocarcinogenesis. In: Maibach, H. & Lowe, N.,
eds, Models in Dermatology. Vol.2, Basel, Karger, pp. 303-312
Epstein, J.H. (1988) Photocarcinogenesis promotion studies with benzoyl peroxide (BPO) and
croton oil. J. invest. Dermatol., 91, 114-116
Epstein, J.H. (1990) Induction of melanomas in experimental animals. Photodermatol. Photo-
immunol. Photomed., 7, 95-97
Epstein, J.H. & Epstein, W.L. (1962) Cocarcinogenic effect of ultraviolet light on DMBA tumor
initiation in albino mice. J. invest. Dermutol. , 39, 455-460
Epstein, J.H. & Epstein, W.L. (1963) A study of tumor types produced by ultraviolet light in
hairless mice and hairy mice. J. invest. Dermatol., 41, 463-73
Epstein, J.H. & Roth, H.L. (1968) Experimental ultraviolet light catcinogenesis. A study of croton
oil promoting effects. J invest. Dermatol. , 50, 387-389
Epstein, J.H., Epstein. W.L. & Nakai, T. (1967) Production of melanomas from DMBA-induced
'blue nevi' in hairless mice with ultraviolet light. J. natl Cancer lnst., 38, 19-30
Epstein, W.L. , Fukuyama, K. & Epstein, J.H. (1969) Early effects of ultraviolet light on DNA
synthesis in human skin in vivo. Arch. Dermatol., 100, 84-89
Epstein, J.H., Fukuyama. K. , Reed, W.B. & Epstein, W.L. (1970) Defect in DNA synthesis in skin
of patients with xeroderma pigmentosum demonstrated in vivo. Science, 168, 1477-1478
Eriksen, P. (1987) Occupational applications of ultraviolet radiation: risk evaluation and protection
techniques. In: Passchier, W.F. & Bosnjakovic, B.F.M., eds, Human Exposure to Ultraviolet
Radiation. Risks and Regulations, Amsterdam, Elsevier, pp. 317-331
Eriksen, P., Moscato, P.M., Franks, J.K. & Sliney, D.H. (1987) Optical hazard evaluation of dental
curing lights. Community Dent. oral Epidemiol., 15, 197-201
Everett, M.A., Yeargers, E., Sayre, R.M. & Olson, R.L. (1966) Penetration of epidermis by
ultraviolet rays. Photochem. Photobiol., 5, 533-542
Everett, M.A., Nordquist, J. & Olson, R.L. (1970) Ultrastructure of human epidermis following
chronic sun exposure. Br. J. Dermatol., 84, 248-257
Fan, J., Schoenfeld, R.J. & Hunter, R. (1959) A study of the epidennal clear cells with special
reference to their relationship to the cells ofLangerhans. J. invest. Dermatol. , 32, 445-50
Farr, PM. & Diffey, B.L. (1985) The erythemal response of human skin to ultraviolet radiation. Br.
J. Dermatol., 113, 65-76
Fears, T.R., Scotto, J. & Schneiderman, M.A. (1976) Skin cancer, melanoma, and sunlight. Am. J.
public Health, 66, 461-464
Fears, T.R., Scotto, J. & Schneiderman, M.A. (1977) Mathematical models of age and ultraviolet
effects on the incidence of skin cancer among whites in the United States. Am. J. Epidemiol.,
105, 42-427
Fincham, S. & Hill, G.B. (1989) Risk factors for non-melanocytic skin cancer in Alberta males
(Abstract). Am. J. Epidemiol., 130, 841
Findlay, G.M. (1928) Ultra-violet light and skin cancer. Lancet, ii, 1070-1073
Findlay, C.M. (1930) Cutaneous papillomata in the rat following exposure to ultra-violet light.
Lancet, i, 1229-1231
Fisher, M.S. & Cifone, M.A. (1981) Enhanced metastatic potential of murine fibrosarcomas treated
in vitro with ultraviolet radiation. Cancer Res., 41, 3018-3023
Fisher, G.J. & Johns, H.E. (1976) Pyrimidine photohydrates. In: Wang, S., ed., Photochemistry and
Photobiology of Nucleic Acids, Vol. 1, New York, Academic Press. pp. 169-224
Fisher, M.S. & Kripke, M.L. (1977) Systemic alteration induced in mice by ultraviolet light
irradiation and its relationship to ultraviolet carcinogenesis. Proc. natl Acad. Sci. USA, 74,
1688-1692
Fisher, M.S. & Kripke, M.L. (1978) Further studies on the tumor-specific suppressor cells induced
by ultraviolet radiation. J. Immunol. , 121, 1139-1144
Fisher, M.S. & Kripke, M.L. ( 1982) Suppressor T lymphocytes control the development of primary
skin cancers in ultraviolet-irradiated mice. Science, 216, 1133-1134
Fix, D.F. (19S6) Thermal resistance of UV -mutagenesis to photoreactivation in E. coli Blr uvrA
ung: estimates of activation energy and further analysis. Mol. gen. Genet. , 204, 452-456
Foley, P., Lanzer, D. & Marks, R. (1986) Are solar keratoses more common on the driver's side?
Br. med. J., 293, 18
Forbes, P.D. & Urbach, F. (1975) Experimental modification ofphotocarcinogenesis. I. Fluorescent
whitening agents and short-wave UVR. Food Cosmet. Toxicol., 13, 335-337
Forbes, P.D., Davies, R.E. & Urbach, E. (1978) Experimental ultraviolet photocarcinogenesis:
wavelength interactions and time-dose relationships. Nat! Cancer lnst. Monogr. , 50, 31-38
Forbes, P.D., Blum, H.F. & Davies, R.E. (1981) Photocarcinogenesis in hairless mice:
dose-response and the influence of dose-delivery. Photochem. Photobiol., 34, 361-365
Forbes, P.D., Davies, R.E., Urbach, F., Berger, D. & Cole, C. (1982) Simulated stratospheric ozone
depletion and increased ultraviolet radiation: effects on photocarcinogenesis in hairless mice.
Cancer Res., 42, 2796-2803
Forbes, P.D., Davies, R.E., Urbach, E. & Dunnick, J.K. (1990) Long-term toxicity of oral
8-methoxypsoralen plus ultraviolet radiation in mice. J. Toxicol. cutaneous ocul. Toxicol., 9,
237-250
Fomace, A.J., Jr, Schalch, H. & Alamo, I., Jr (1988) Coordinate induction ofmetallothioneins I and
II in rodent cells by UV irradiation. Mol. cell. Biol., 8, 4716-4720
F'ortner, G.W. & Kripke, M.L. (1977) In vitro reactivity of splenic lymphocytes from normal and
UV irradiated mice against syngeneic UV-induced tumors. J. Immunol., 118, 1483-1487
Fortner, G.W. & Lill, P.H. (1985) Immune response to ultraviolet-induced tumours. I.
Transplantation immunity developing in syngeneic mice in response to progressor
ultraviolet-induced tumors. Transplantation, 39, 44-49
Foukal, P. (1990) SolarAstrophysics, New York, John Wiley & Sons, pp. 6-75
REFERENCES 243
Fox, T.J., Sy, M.-S., Benacerraf, B. & Greene, M.T. (1981) Impairment of antigen-presenting cell
function by ultraviolet radiation. II. Effect of in vitro ultraviolet irradiation on antigenpresenting
cells. Transplantation, 31, 262-265
Francis, A.J. & Giannelli, F. (1991) Cooperation between human cells sensitive to USA radiations:
a clue to the mechanism of cellular hypersensitivity associated with different clinical conditions.
Exp. Cell Res., 195, 47-52
Franklin, W.A. & Haseltine, W.A. (1986) The role of the (6-4) photoproduct in ultraviolet
light-induced transition mutations in E. coli. Mutat. Res., 165, 1-7
Franklin, S.A., Lo, K.M. & Haseltine, W.A. (1982) Alkaline lability of fluorescent photoproducts
produced in ultraviolet light-irradiated DNA. J. bioi. Chern., 257, 13535-13543
Frederick, J.E. (1990) Trends in atmospheric ozone and ultraviolet radiation: mechanisms and
observations for the northern hemisphere. Photochem. Photobiol. , 51,757-763
Freeman, R.G. (1975) Data on the action spectrum for ultraviolet carcinogenesis. J. natl Cancer
Inst. , 55, 1119-1122
Freeman, R.G. (1978) Action spectrum for ultraviolet carcinogenesis. Natl Cancer Jnst. Monogr. ,
50, 27-29
Freeman, S.E. (1988) Variations in excision repair of UVB-induced pyrimidine dimers in DNA of
human skin in situ. J. invest. Dermatol., 90, 814-817
Freeman, R.G. & Knox, J.M. (1964) Ultraviolet-induced corneal tumors in different species and
strains of animals. J. invest. Dermatol. , 43, 431-436
Freeman, S.E., Gange, R.W. , Matzinger, E.A. & Sutherland, B.M. (1986) Higher pyrimidine dimer
yields in skin of normal humans with higher UVB sensitivity. J. invest. Dermutol. , 86, 34-36
Freeman, S.E., Gange. R.W., Sutherland, J.C., Matzinger, E.A. & Sutherland, B.M. (1987)
Production of pyrimidine dimers in DNA of human skin exposed in situ to UV A radiation. J.
Freeman, S.E., Applegate, L.A. & Ley, R.D. (1988a) Excision repair of UVR-induced pyrimidine
dimers in corneal DNA. Photochem. Photobiol. , 47, 159-163
Freeman, S.E., Ley, R.D. & Ley, K.D. (1988b) Sunscreen protection against UV-induced
pyrimidine dimers in DNA of human skin in situ. Photodermatology, 5, 243-247
Freeman, S.E., Hacham, H., Gange, R.W., Maytum, D.J., Sutherland, J.C. & Sutherland, B.M.
(1989) Wavelength dependence of pyrimidine dimer formation in DNA of human skin
irradiated in situ with ultraviolet light. Proc. natl Acad. Sci. USA, 86, 5605-5609
Friedburg, E. (1984) DNA Repair, New York, W.H. Freeman & Co.
Frucht, D.M., Lamperth, L., Vicenzi, E., Belcher, J.H. & Martin, M.A. (1991) Ultraviolet radiation
increases HIV-long terminal repeat-directed expression in transgenic mice. AIDS Res. hum.
Retroviruses, 7, 729-733
Fujiwara, Y. , Ichihashi, M., Kano, Y., Goto, K. & Shimizu, K. (1981) A new human photosensitive
subject with a defect in the recovery of DNA synthesis after ultraviolet-light irradiation. J.
Gafa, L., Filippazzo, M.G., Tumino. R., Dardanoni, G., Lanzarone, F. & Dardanoni, L. (1991) Risk
factors of nonmelanoma skin cancer in Ragusa, Sicily: a case-control study. Cancer Causes
Control, 2, 395-399
Gahring, L., Baltz, M., Pepys, M.B. & Daynes, R. (1984) Effect of ultraviolet radiation on
production of epidermal cell thymocyte-activating factor/interleukUl 1 in vivo and in vitro.
Immunology, 81, 1198-1202
Gallagher, R.P. (1988) Ocular melanoma in fanners (Letter to the Editor). Am. J. ind. Med., 13,
523-525
Gallagher, P.E. & Duker, N.J. (1986) Detection ofUV purine photoproducts in a defined sequence
ofhuman DNA. Mol. cell. Bioi., 6, 707-709
Gallagher, P.E. & Duker, N.J. (1989) Fonnation of purine photoproducts in a defined human DNA
sequence. Photochem. Photobiol., 49, 599-605
Gallagher, C.H., Path, F.R.C., Canfield, P.J., Greenoak, G.E. & Reeve, VE. (1984a)
Characterization and histogenesis of tumors in the hairless mouse produced by low-dosage
incremental ultraviolet radiation. J. invest. Dermatol., 83, 169-174
Gallagher, C.H., Greenoak, G.E., Reeve, V.E. , Canfield, P.J. , Baker, R.S.U. & Bonin, A.M. (1984b)
Ultraviolet carcinogenesis in the hairless mouse skin. Influence of the sunscreen 2-ethylhexyl-p-
methoxycinnamate. Aust. J. exp. Bioi. med. Sci., 62, 577-588
Gallagher, R.P, ThrelfaU, W.J., Jeffries, E., Band, P.R., Spinelli, J. & Coldman, A.J. (1984) Cancer
and aplastic anemia in British Columbia farmers. J. natl Cancer lnst. , 72, 1311-1315
Gallagher, R.P, Elwood, J.M., Rootman, J., Spinelli, J.J., Hill, G.B., Threlfall, W.J. & Birdsell, J.M.
(1985) Risk factors for ocular melanoma: Western Canada Melanoma Study. J. natl Cancer
Inst., 74, 775-778
Gallagher, R.P, Elwood, J.M. & Hill , G.B. (1986) Risk factors for cutaneous malignant melanoma:
the Western Canada Melanoma Study. Recent Results Cancer Res., 102, 38-55
Gallagher, R.R, Elwood, J.M., Threlfall, W.J., Spinelli, J.J. , Fincham, S. & Hill, G.B. (1987) Socio-
economic status, sunlight exposure, and risk of malignant melanoma: the Western Canada Mela-
noma Study. J. natl Cancer lnst., 79, 647-652
Gallagher, P.E., Weiss, R. B., Brent, T.P. & Duker, N.J. (1989) Wavelength dependence of DNA
incision by a human ultraviolet endonuclease. Photochem. Photobiol., 49, 363-367
Gallagher, R.P, McLean, D.l., Yang, C.P, Coldman, A.J., Silver, H.K.B., Spinelli, J.J. & Beagrie,
M. ( 1990a) Anatomic distribution of acquired melanocytic nevi in white children. A
comparison with melanoma: the Vancouver mole study. Arch. Dermatol. , 126, 466-471
Gallagher, R.P, McLean, D.l., Yang, C.P., Coldman, A.J., Silver, H.K.B., Spinelli, J.J. & Beagrie,
M. (1990b) Suntan, sunburn and pigmentation factors and the frequency of acquired
melanocytic nevi in children. Similarities to melanoma: the Vancouver mole study. Arch.
Dermatol., 126, 770-776
Gange, R.W. (1987) Acute effects of ultraviolet radiation in the skin. In: Fitzpatrick, T.B., Eisen.
A.Z., Wolff, K. , Freedberg, I.M. & Austen, K.F., eds, Dermatology in General Medicine, 3rd
ed., New York, McGraw-Hill, pp. 1451-1457
Gange, R.W., Blackett, A.D. , Matzinger, E.A., Sutherland, B.M. & Kochevar, I.E. (1985)
Comparative protection efficiency of UV A- and UVB-induced tans against erythema and
formation of endonuclease-sensitive sites in DNA by UVB in human skin. J. invest Dermatol. ,
85, 362-364
Garbe, C., Kriiger, S., Stadler, R., Guggenmoos-Holzmann, I. & Orfanos, C.E. (1989) Markers and
relative risk in a German population for developing malignant melanoma. Int. J. Dermatol., 28,
517-523
Gardiner, B.G. & Kirsch, P.J. (1991) European ln.tercomparison of Ultraviolet Spectrometers,
Panorama, Greece, 3-12 July 1991 (Report to the Commission of the European Communities
[STEP Project 76], Science and Technology for Enviromnental Protection), Luxembourg,
Commission ofthe European Communities
Garland, F.C., White, M.R .. Garland, C.F. , Shaw, E. & Gorham, E.D. (1990) Occupational sunlight
exposure and melanoma in the US Navy. Arch. environ. Health. 45, 261-267
REFERENCES 245
Garrison, L.M., Murray, L.E., Doda, D.D. & Green, A.E.S. (1978) Diffuse-direct ultraviolet ratios
with a compact double monochromator. Appl. Optics, 17, 827-836
Gellin, G.A. , Kopf, A.W. & Garfinkel, L. ( 1965) Basal cell epithelioma. A controlled study of asso-
ciated factors. Arch. Dermatol. , 91, 38-45
Gellin, G.A., Kopf, A.W. & Garfinkel, L. (1969) Malignant melanoma. A controlled study of
possibly associated factors. Arch. Dermatol. , 99, 43-48
Gensler, H.L. ( 1988a) Enhancement of chemical carcinogenesis in mice by systemic effects of
ultraviolet radiation. Cancer Res., 48, 620-623
Gensler, H.L. (1988b) Prevention of chemically induced two-stage skin carcinogenesis in mice by
systemic effects ofultraviolet irradiation. Carcinogenesis, 9, 767-769
Gensler, H.L. & Bowden, G.T. (1987) UVB-lnduced modulation of mouse skin tumor induction by
benzo[a]pyrene (Abstract no. 546). Proc. Am. Assoc. Cancer Res., 28, 137
Gensler, H.L. & Welch, K. (1992) Prevalence of tumor prevention rather than tumor enhancement
when repetitive UV radiation treatments precede initiation and promotion. Carcinogenesis, 13,
9-13
Giam, Y.C. (1987) Cutaneous gerontology: solar degeneration of the skin. Ann. Acad. Med.
Singapore, 16, 94-97
Giannini, M.S.H. (1986) Suppression of pathogenesis in cutaneous leishmaniasis by UV irradiation.
Infect. lmmun., 51, 838-843
Gibbs, N.K., Young, A.R. & Magnus, I.A. (1985) Failure of UVR dose reciprocity for skin tumori-
genesis in hairless mice treated with 8-methoxypsoralen. Photochem. Photobiol., 42, 39-42
Gibson, P. & Diffey, B.L. (1989) Techniques for spectroradiometry and broadband radiometry. In:
Diffey, B.L., ed., Radiation Measurement in Photobiology, London, Academic Press, pp.
135-159
Gies, P., Roy, C. & Elliott, G. (1988) The anatomical distribution of solar UVR with emphasis on
the eye. In: Radiation Protection Practice, Vol. 1. Sydney, Pergamon Press, pp. 341-344
Gilchrest, B.A., Soter, N.A., Stoff, J.S. & Mihm, M.C., Jr (1981) The human sunburn reaction:
histologic and biochemical studies. J. Am. Acad. Dermatol. , 5, 411-422
Gilchrest, B.A., Murphy, G.F. & Soter, N.A. (1982) Effect of chronologie aging and ultraviolet irra-
diation on Langerhans cells in human epidermis. J. invest. Dermatol. , 19, 85-88
Giles, G.G., Marks, R. & Foley, P. (1988) Incidence of non-melanocytic skin cancer treated in
Australia. Br med. 1. , 296, 13-17
Glazier, A., Morison, W.L., Bucana, C., Hess, A.D. & Tutschka, P.J. (1984) Suppression of
epidermal graft-versus-host disease with ultraviolet radiation. Transplantation, 37, 211-213
Glickman, B.W., Schaaper, R.M., Haseltine, W.A., Dunn, R.L. & Brash, D.E. (1986) The C-C (6-4)
UV photoproduct is mutagenic in Escherichia coli. Proc. natl Acad. Sci. USA, 83, 6945-6949
Goldberg, L.H. & Altman, A. (1984) Benign skin changes associated with chronic sunlight
exposure. Cutis, 34, 33-39
Gofii, F.M., Ondarroa, M., Azpiazu, I. & Macarulla, J.M. ( 1985) Phospholipid oxidation catalysed
by cytochrome c in liposomes. Biochim. biophys. Acta, 835, 549-556
Goodman, G.J. , Marks, R., Selwood, T.S., Ponsford, M.W. & Pakes, W. (1984) Non-melanocytie
skin cancer and solar keratoses in Victoria-clinical studies II. Aust. J. Derm., 25, 103-106
Gordon, D. & Silverstone, H. (1976) Worldwide epidemiology of premalignant and malignant cuta-
neous lesions. In: Andrade, R., Gumport, S.L., Popkin, G.L. & Rees, RD., eds, Cancer of the
Skin, Vol. 1, London, W.B. Saunders, pp. 405-434
Gorlin, R.J. (1987) Nevoid basal-cell carcinoma syndrome. Medicine, 66,98-113
Grabner, G. & Brenner, W. (1981) Unscheduled DNA repair in the human cornea following solar
simulating radiation. Acta ophthalmol., 59, 847-853
Grady, H.G., Blum, H.F. & Kirby-Smith, J.S. (1941) Pathology of tumors of the external ear in
mice induced by ultraviolet radiation. J. natl Cancer Inst. , 2, 269-276
Grady, H.G., Blum, H.F. & Kirby-Smith, J.S. (1943) Types of tumor induced by ultraviolet
radiation and factors influencing their relative incidence. J. natl Cancer Inst., 3, 371-378
Graham, S., Marshall, J. , Haughey, B., Stoll, H., Zielezny, M., Brasure, J. & West, D. (1985) An
inquiry into the epidemiology of melanoma. Am. J. Epidemiol., 122, 606-619
Granstein, R.D. (1985) Epidermal 1-J-bearing cells are responsible for transferable suppressor cell
generation after immunization of mice with ultraviolet radiation-treated epidermal cells. J.
Granstein, R.D. & Sauder, D.N. (1987) Whole-body exposure to ultraviolet radiation results in
increased serum interleukin-1 activity in humans. Lymphokine Res., 6, 187-193
Granstein, R.D. , Lowy, A. & Greene, M.I. (1984) Epidermal antigen-presenting cells in activation
of suppression: identification of a new functional type of ultraviolet radiation-resistant
epidermal cell. J. Immunol., 132, 563-565
Granstein, R.D., Askari, M., Whitaker, D. & Murphy, G.F. (1987) Epidermal cells in activation of
suppressor lymphocytes: further characterization. J. Immunol., 138, 4055-4062
Greaves, M.W., Hensby, C.N., Black, A, K., Plummer, N.A., Fincham, N., Warin, A.P. & Camp, R.
(1978) Inflammatory reactions induced by ultraviolet irradiation. Bull. Cancer, 65, 299-304
Green, A.C. (1984) Sun exposure and the risk of melanoma. Aust. J. Dermatol., 25,99-102
Green, A.C. (1991) Premature ageing of the ski n in a Queensland population. Med. J. Aust., 155,
473-478
Green, A.C. & Battistutta, D. (1990) Incidence and detenninants of skin cancer in a high-risk
Australian population. Int. J. Cancer, 46, 356-361
Green, A. & O'Rourke, M.G.E. (1985) Cutaneous malignant melanoma in association with other
skin cancers. J. natl Cancer /nst., 74, 977-980
Green, A. & Siskind, V. (1983) Geographical distribution of cutaneous melanoma in Queensland.
Med. J. Aust., 1, 407-410
Green, A.E.S., Findley, G.B., Jr, Klenk, K.F., Wilson, W.M. & Mo, T. (1976) The ultraviolet dose
dependence of non-melanoma skin cancer incidence. Photochem. Photobiol. , 24, 353-362
Green, A.C., Siskind, V., Bain, C. & Alexander, J. (1985a) Sunburn and malignant melanoma. Br J.
Cancer, 51, 393-397
Green, A.C., McLennan, R. & Siskind, V. (1985b) Common acquired naevi and the risk of
malignant melanoma. Int. J. Cancer, 35, 297-300
Green, A. C. , Bain, C., McLennan, R. & Siskind, V. ( 1986) Risk factors for cutaneous melanoma in
Queensland. Recent Results Cancer Res., 102, 76-97
Green, A.C., Beardmore, G., Hart, V., Leslie, D., Marks, R. & Staines, D. (1988a) Skin cancer in a
Queensland population. J. Am. Acad. Dermatol., 19, 1045-1052
Green, A.C., Sorahan, T., Pope, D., Siskind, V., Hansen, M., Hanson, L., Leech, P. , Ball, P.M. &
Grimley, R.P. (1988b) Moles in Australian and British schoolchildren (Letter to the Editor).
Lancet, ii, 1497
Greene, M.I., Sy, M.S., Kripke, M. & Benacerraf, B. (1979) Impairment of antigen-presenting cell
function by ultraviolet radiation. Proc. natl Acad. Sci. USA, 76, 6591-6595
Griggs, H.G. & Bender, M.A. (1973) Photoreactivation of ultraviolet-induced chromosomal aberra-
tions. Science, 179, 86-88
REFERENCES 247
Grob, J.J., Gouvernet, J., Aymar, D., Mostaque, A., Romano, M.H., Collet, A.M., Noe, M.C.,
Diconstanzo, M.P. & Bonerandi, J.J. (1990) Count of benign melanocytic nevi as a major indi-
cator of risk for nonfamilial nodular and superficial spreading mclarloma. Cancer, 66,387-395
de Gruijl, F.R. & van der Leun, J.C. (1982a) Effect of chronic UV exposure on epidermal
transmission in mice. Photochem. Photobiol., 36, 433-438
de Gruijl, F.R. & van der Leun. J.C. (1982b) Systemic influence of pre-irradiatiorl of a limited skin
area on UV-tumorigenesis. Photochem. Photobiol., 35, 379-383
de Gruijl, F.R. & van der Leun, J.C. (1983) Follow up on systemic influence of partial
pre-irradiatior on UV -tumorigenesis. Photochern. Photobiol., 38, 381-383
de Gruijl, F.R., van der Meer, J.B. & van der Leun, J.C. (1983) Dose-time dependency of tumor
formation by chronic UV exposure. Photochem. Photobiol. , 37, 53-62
Gupta, A.K. & Anderson, T.F. (1987) Psoralen photochemotherapy. J. Am. Acad. Dermatol. , 17,
703-734
Gurish, M.F., Lynch, D.H. & Daynes, R.A. (1982) Changes in antigen-presenting cell function in
the spleen and lymph nodes of ultraviolet-irradiated mice. Transplantation, 33, 280-284
Gurish, M.F, lynch, D.H., Yowell, R. & Daynes, R.A. (1983) Abrogation of epidermal antigen-
presenting cell function by ultraviolet radiation administered in vivo. Transplantation, 36, 304-
309
Haenszel, W. (1963) Variations in skin cancer incidence within the United States. Nat! Cancer Inst.
Monogr. , 10, 225-243
Halasz, C.L.G., Leach, E.E., Walther, R.R. & Poh-Fitzpatrick, M.B. (1983) Hydroa vacciniforme:
induction oflesions with ultraviolet A. J. Am. Acad. Dermatol., 8, 171-176
Hall, J.D. & Mount, D.W. (1981) Mechanisms of DNA replication and mutagerlesis in ultraviolet-
irradiated bacteria and man1malian cells. Prog. nucleic Acid Res. mol. Biol., 25, 53-126
Halprin, K.M., Comerford, M., Presser, S.E. & Taylor, J.R. (1981) Ultraviolet light treatment delays
contact sensitization to nitrogen mustard. Br. J. Dermatol., 105, 71-76
Haniszko, J. & Suskind, R.R. (1963) The effect of ultraviolet radiation on experimental cutaneous
sensitization in guinea pigs. J. invest. Dermatol., 40, 183-191
Hannan, M.A., Paul , M., Amer, M.H. & AI-Watban, F.H. (1984) Study of ultraviolet radiation and
genotoxic effects of natural sunlight in relation to skin cancer in Saudi Arabia. Cancer Res., 44,
2192-2197
Hanson, D.L. & DeLeo, VA. (1989) Long wave ultraviolet radiation stimulates arachidonic acid
release and cyclooxygenase activity in mammalian cells in culture. Photochem. Photobiol., 49,
423-430
Harber, L.C. & Bickers, D.R. (1981) Photosensitivity Diseases. Principles of Diagnosis and
Treatment, Philadelphia, P A, W .B. Saunders
Hardie, I.R., Strong, R.W, Hartley, L.C.J., Woodruff, P.W.H. & Clunie, C.J.A. (1980) Skin cancer
in Caucasian renal allograft recipients living in a subtropical climate. Surgery, 87, 177-183
Hariharan, P.V. & Cerutti, P.A. (1976) Excision of ultraviolet and gamma ray products of the
5,6-dihydroxy-dihydrothymine type by nuclear preparations of xeroderma pigmentosum cells.
Biochim. biophys. Acta, 447,375-378
Hariharan, P.V & Cerutti, P.A. (1977) Formation of products of the 5,6-dihydroxydibydrothymine
type by ultraviolet light in HeLa cells. Biochemistry, 16, 2791-2795
Hartman, Z., Hartman, P.E. & McDermott, W.L. (1991) Mutagenicity of coolwhite fluorescent light
for Salmonella. Mutat. Res., 260, 25-38
Hawk, J.L.M. & Parrish, J.A. (1982) Responses of normal skin to ultraviolet radiation. In: Reegan,
J.D. & Parrish, J.A., eds, The Science of Photomedicine, New York, Plenum, pp. 219-260
Hayashi, Y. & Aurelian, L. (1986) Immunity to herpes simple virus type 2: viral antigen-presenting
capacity of epidermal cells and its impairment by ultraviolet irradiation. J. lmmunol., 136,
1087-1092
Health Council of the Netherlands (1986) UV Radiation: Human Exposure to Ultraviolet Radiation
(Report 1986/93), The Hague
Hefferren, J.J., Cooley, R.O., Hall, J.B., Olsen, N.H. & Lyon, H.W. (1971) Use of ultraviolet illumi-
nation in oral diagnosis. J. Am. dent. Assoc., 82, 1353-1360
Henderson, S.T ( 1970) Daylight and Its Spectrum, New York, American Elsevier
Henderson, L.M., Arlett, C.F., Harcourt, S.A., Lehmann, A.R. & Broughton, B.C. (1985) Cells from
an immunodeficient patient (46BR) with a defect in DNA ligation are hypomutable but hyper-
sensitive to the induction of sister chromatic exchanges. Proc. natl Acad. Sci. USA, 82, 2044-
2048
Henseler, T., Wolff, K., Honigsmann, H. & Christophers, E. (1981) Oral 8- methoxypsoralen photo-
chemotherapy of psoriasis. The European PUV A study: a cooperative study among 18 European
centres. Lancet, i, 853-857
Herity, B., O'Loughlin, G., Moriarty, M.J. & Conroy, R. (1989) Risk factors for non-melanoma skin
cancer. Irish med. J. , 82, 151-152
Hersey, P., Bradley, M., Hasic, E., Haran, G., Edwards, A. & McCarthy, W.H. (1983a)
Immunological effects of solarium exposure. Lancet, i, 545-548
Hersey, P., Haran, G., Hasic, E. & Edwards, A. (1983b) Alteration ofT cell subsets and induction
of suppressor T cell activity in nonnal subjects after exposure to sunlight. J. lmmunol., 31,
171-174
Hersey, P. MacDonald, M., Burns, C., Schibeci , S., Matthews, H. & Wilkinson, F.J. (J 987)
Analysis of the effect of a sunscreen agent on the suppression of natural killer cell activity
induced in human subj ects by radiation from solarium lamps. J. invest Dermatol., 88, 271-276
Hersey, P., MacDonald, M., Henderson, C., Schibeci, S., D'Alessandro, G., Pryor, M. & Wilkinson,
E.J. (1988) Suppression of natural killer cell activity in humans by radiation from solarium
lamps depleted ofUVB. J. invest. Dermatol., 90, 305-310
Hietanen, M. (1990) Spectroradiometic and Photometric Studies on Ocular Exposure to Ultraviolet
and Visible Radiation from Light Sources, PhD Thesis, Helsinki, University of Helsinki,
Department of Physics
Higgins, S.A., Frenkel, K., Cummings, A. & Teebor, G.W. (1987) Definitive characterisation of
human thymine glycol N-glycosylase activity. Biochemistry, 26, 1683-1688
Hinds, M.W. & Kolonel, L.N. (1980) Malignant melanoma of the skin in Hawaii, 1960-1977.
Cancer, 45,811-817
Hitchins, V.M., Beer, J.Z., Strickland, A.G. & Olvey, K.M. (1987) Lethal and mutagenic effects of
UVA radiation in six strains of L5178Y mouse lymphoma cells. In: Passchier, W.F. &
Bosnjakovic, B.F.M., eds, Humun Exposure to Ultraviolet Radiation: Risk and Regulations,
Hoeijmakers, J.H.J. & Bootsma, D. (1990) Molecular genetics of eukaryotic DNA excision repair.
Cancer Cells, 2, 311-320
Hoffman, J.S. (1987) Ultraviolet Radiation and Melanoma (with a Special Focus on Assessing the
Risks of Stratospheric Ozone Depletion) (EPA 400/ 1-87-001 D), Washington DC, Office of Air
and Radiation. US Environmental Protection Agency
REFERENCES 249
Hogan, D.J., To, T., Gran, L., Wong, D. & Lane, P.R. (1989) Risk factors for basal cell carcinoma.
Int. J. Dermatol., 28, 591-594
Holick, M.F., MacLaughlin, J.A., Clark, M.B., Holick, S.A., Potts, J.T., Jr, Anderson, R.R., Blank,
I.H., Parrish, J.A. & Elias, P. (1980) Photosynthesis of previtarnin D3 in human skin and the
physiologic consequences. Science, 210, 203-205
Hollaender, A. (1943) Effect of long ultraviolet and sh01t visible radiation (3500 to 4900 A) on
Eschenchia coli. J. Bacterial., 46, 531-541
Hollander, M.C. & Fomace, A.J., Jr (1989) Induction of fos RNA by DNA-damaging agents.
Cancer Res., 49, 1687-1692
Hollis, D.E. & Scheibner, A. (1988) Ultrastructural changes in epidermal Langerhans cells and
melanoeytes in response to ultraviolet irradiation, in Australians of aboriginal and Celtic
descent. Br. J. Dermatol., 119, 21-31
Holly, E.A., Kelly, J.W. , Shpall, S.N. & Chiu, S.-H. (1987) Number ofmelanocytic nevi as a major
risk factor for malignant melanoma. J. Am. Acad. Dermatol., 17, 459-468
Holly, E.A., Aston, D.A., Char, D.H., Kristiansen, J.J. & Ahn, D.K. (1990) Uveal melanoma in rela-
tion to ultraviolet light exposure and host factors. Cancer Res., 50, 5773-5777
Holly, E.A., Aston, D.A., Ahn, D.K., Kristiansen, J.J. & Char, D.H. (1991) No excess prior cancer
in patients with uveal melanoma. Ophthalmology, 98, 608-611
Holman, C.D.J. & Armstrong, B.K. (1984a) Pigmentary traits, ethnic origin, benign nevi, and
family history as risk factors for cutaneous malignant melanoma. J. nat/ Cancer Inst., 72,
257-266
Holman, C.D.J. & Armstrong, B.K. (1984b) Cutaneous malignant melanoma and indicators of total
accumulated exposure to the sun: an analysis separating histogenetic types. J. natl Cancer Inst.,
73, 75-82
Holman, C.D.J., Gibson, I.M., Stephenson, M. & Armstrong, B.K. (1983a) Ultraviolet irradiation of
human body sites in relation to occupation and outdoor activity: field study using personal UVR
dosimeters. Clin. exp. Dermatol., 8, 269-277
Holman, C.D.J., Heenan, P.J., Caruso, V. , Glancy, R.J. & Armstrong, B.K. (1983b) Seasonal
variation in the junctional component of pigmented naevi . Int. J. Cancer, 31, 213-215
Holman, C.D.J., Armstrong, B.K., Evans, P.R., Lumsden, G.J., Dallimore, K.J., Meehan, C.J.,
Beagley, J. & Gibson, I.M. (1984a) Relationship of solar keratosis and history of skin cancer to
obj ective measures of actinic skin damage. Br. J. Dermatol., 110, 129-138
Holman, C. D., Evans, P.R., Lumsden, G.J. & Armstrong, B.K. ( 1984b) The determinants of actinic
skin damage: problems of confounding among environmental and constitutional variables. Am.
J. Epidemiol., 120, 414-422
Hohnan, C.D.J., Armstrong, B.K. & Heenan, P.J. (1986a) Relationship of cutaneous malignant
melanoma to individual sunlight-exposure habits. J. natl Cancer Into., 76, 403-414
Holman, C.D.J., Armstrong, B.K., Heenan, P.I., Blackwell, J.B., Cumming, F.J., English, D.R.,
Holland, S., Kelsall, G.R.H., Matz, L.R., Rouse, I.L. , Singh, A., Ten Seldam, R.E.J., Watt, J.D.
& Xu, Z. (1986b) The causes of malignant melanoma: results froth the West Australian Lions
Melanoma Research Project. Recent Results Cancer Res., 102, 18-37
Holmberg, M. & Gumauskas, E. (1990) Chromosome-type exchange aberrations are induced by
inhibiting repair of UVC-induced DNA lesions in quiescent human lymphocytes. Mutat. Res.,
232, 261-266
Honigsmann, H., Brenner, W. Tanew, A. & Ortel, B. (1987) UV-lnduced unscheduled DNA
synthesis in human skin: dose response, correlation with erythema, time course and split dose
exposure in vivo. J. Photochem. Photobiol. B Bioi., 1, 33-43
Hoover, T.L., Morison, W.L. & Kripke, M.L. (1987) Ultraviolet carcinogenesis in athymic nude
mice. Transplantation, 44, 693-695
Howie, S.E.M., Norval, N., Maingay, J. & Ross, J.A. (1986a) Two phenotypically distant T cells
(Ly1 +2" and Lyr2+) are involved in ultraviolet-B light-induced suppression of the efferent DTH
response to HSV-1 in vivo. Immunology, 58,653-658
Howie, S.E.M., Norval, M. & Maingay, J.P. (1986b) Alterations in epidermal handling of HSV-1
antigens in vitro induced by in vivo exposure to UVB light. Immunology, 57, 225-230
Howie, S., Norval, M. & Maingay, J. (1986c) Exposure to low-dose ultraviolet radiation suppresses
delayed-type hypersensitivity to herpes simplex virus in mice. J. invest. Dermatol., 86, 125-128
Hsia, H.C., Lebkowski, J.S. , Leong, P.-M., Calos, M.P. & Mi ller, J.H. (1989) Comparison of
ultraviolet inadiation-induced mutagenesis of the lad gene in Escherichia coli and in human
293 cells. J. mol. Biol., 205, 103-113
Hsie, A. W ., Li, A.P & Machanoff, R. (1977) A fluence response study of lethality and mutagenicity
ofwhite, black, and blue fluorescent light, sunlamp, and sunlight irradiation in Chinese hamster
ovary cells. Mutat. Res., 45, 333-342
Hsu, J. , Forbes, P.D., Harber, L.C. & Lakow, E. (1975) Induction of skin tumors in hairless mice by
a single exposure to UV radiation. Photochem. Photobiol. , 21, 185-188
Hueper, W.C. (1942) Mmphological aspects of experimental actinic and arsenic carcinomas in the
skin of rats. Cancer Res., 2, 551-559
Huldschinsky, K. (1933) Eye sarcoma in rats caused by abnormally long UV (Ger.). Dtsch. med.
Wochenschr., 59, 530-531
Hunter, D.J., Colditz, G.A., Stampfer, M.J., Rosner, B., Willett, W.C. & Speizer, F.E. (1990) Risk
factors for basal cell carcinoma in a prospective cohort of women. Ann. Epidemiol., 1, 13-23
Husain, Z., Yang, Q. & Biswas, D.K. (1990) cHa-ras Proto oncogene. Amplification and over-
expression in UVB-induced mouse skin papillomas and carcinomas. Arch. Dermatol., 126, 324-
330
Husain, Z., Pathak, M.A., Flotte, T. & Wick, M.M. (1991) Role of ultraviolet radiation in the
induction of melanocytic tumors in hairless mice following 7,12-dimethylbenz(a)antbracene
application and ultraviolet irradiation. Cancer Res., 51, 4964-4970
Huselton, C.A. & Hill, H.Z. ( 1990) Melanin photosensitizes ultraviolet light (UVC) DNA damage
in pigmented cells. Environ. mol. Mutag. , 16, 37-43
Hutchinson, E. (1987) A review of some topics concerning mutagenesis by ultraviolet light. Photo-
chem. Photobiol. , 45, 897-903
Hutchinson, F. & Wood, R.D. (1986) Mechanisms of mutagenesis of E. coli by ultraviolet light.
Basic Life Sci., 38, 377-383
Hutchinson, E. Yamamoto, K., Stein, J. & Wood, R.D. (1988) Effect of photoreactivation on muta-
genesis oflambda phage by ultraviolet light. J. mol. Bioi., 202, 593-601
IARC (1980) !ARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to
Humans, Vol. 24, Some Pharmaceuticul Drugs, Lyon, pp. 101-124
IARC (1985) !ARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to
Humans, Vol. 37, Tobacco Habits Other than Smoking; Betel Quid and Areca-nut Chewing;
and Some Related Nitrosamines, Lyon. pp. 37-136
IARC (1986a) /ARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to
Humans, Vol. 40, Some Naturally Occurring und Synthetic Food Compounds, Furocoumarins
and Ultraviolet Radiation. Lyon, pp. 317-371
IARC (1986b) /ARC Monographs on the Evalaution of the Carcinogenic Risk of Chemicals to
Humans. Vol. 38, Tobacco Smoking, Lyon
REFERENCES 251
IARC (1987a) /ARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Suppl. 7,
Overall Evaluations of Carcinogenicity: An Updating of IARC Monographs Volumes 1 to 42,
Lyon, pp. 175-176
IARC (1987b) /ARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Suppl. 7,
Overall Evaluations of Carcinogenicity: An Updating of IARC Monographs Volumes I to 42,
Lyon, pp. 242-245
IARC (1988) /ARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Vol. 44,
Alcohol Drinking, Lyon
IARC (1990) /ARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Vol. 49,
Chromium, Nickel and Welding, Lyon, pp. 447-525
Ikushima, T. & Wolff, S. (1974) UV-Induced chromatic aberrations in cultured Chinese hamster
cells after one, two, or three rounds of DNA replication. Mutat. Res. , 22, 193-201
International Electrotechnical Commission (1987) Safety of Household and Similar Electrical
Appliances, Part 2, Particular Requirements for Ultra-violet and Infra-red Radiation Skin
Treatment Appliances for Household Use (Publication 335-2-27: 1987), (Geneva, Bureau
Central de la Commission Electrotechnique lntemationale
International Electrotechnical Commission (1989) Amendment, Geneva, Bureau Central de la Com-
mission Electrotechnique Internationale
International Non-ionizing Radiation Committee of the International Radiation Protection Asso-
ciation (1985) Guidelines on limits of exposure to ultraviolet radiation of wavelengths between
180 nm and 400 nm (incoherent optical radiation). Health Phys. , 49, 331-340
International Non-ionizing Radiation Committee of the International Radiation Protection Asso-
ciation (1989) Proposed change to the IRP A 1985 guidelines on limits of exposure to ultraviolet
radiation. Health Phys., 56, 971-972
Iqbal, M. (1983) An Introduction to Solar Radiation, New York, Academic Press, pp. 335-373
Ito, M. ( 1966) The photodynamic effect of acridine orange and ultraviolet on the mouse skin
carcinogenesis by 20-methylcholanthrene. Nagoya med. 1., 12, 247-259
Ito, A. & Ito, T. (1983) Possible involvement of membrane damage in the inactivation by
broad-band near-UV radiation in Saccharomyces cerevisiae cells. Photochem. Photobiol. , 37,
395-401
Iversen, O.H. (1986) Carcinogenesis studies with benzoyl peroxide (Panoxyl gel 5%). J. invest.
Dermatol., 86, 442-448
Iversen, O.H. (1988) Skin tumorigenesis and carcinogenesis studies with 7,12-dimethylbenz[a]-
anthracene, ultraviolet light, benzoyl peroxide (Panoxyl gel 5%) and ointment gel.
Carcinogenesis, 9, 803-809
Jacobson, E.D., Krell, K., Dempsey, M.J., Lugo, M.H., Ellingson, 0. & Hench, C.W., II (1978)
Toxicity and mutagenicity of radiation from fluoreceent lamps and a sunlamp in L5178Y mouse
lymphoma cells. Mutat. Res., 51, 61-75
Jacobson, E.D., Krell, K. & Dempsey, M.J. (1981) The wavelength dependence ofultraviolet light-
induced cell killing and mutagenesis in L5178Y mouse lymphoma cells. Photochem. Photobiol.,
33, 257-260
Jagger, J. (1981) Near-UV radiation effects on microorganisms. Photochem. Photobiol., 34,
761-768
Jagger, J. (1985) Solar-UV Actions on Living Cells, New York, Praeger
Jessup, J.M., Hanna, N., Palaszynski, E. & Kripke, M.L. (1978) Mechanisms of depressed reactivity
to dinitrochlorobenzene and ultraviolet-induced tumors during ultraviolet carcinogenesis in
BALB/c mice. Cell. Immunol., 38, l 05-115
Johns, H.E., Rapaport, S.A. & Delbriick, M. (1962) Photochemistry of thymine dimers.J. mol. Bioi. ,
4, 104-114
Jones, C.A., Huberman, E., Cunningham, M.L. & Peak, M.J. (1987) Mutagenesis and cytotoxicity
in human epithelial cells by far- and near-ultraviolet radiations: action spectra. Radiat. Res.,
110, 244-254
Jones, S.K., Moseley, H. & Mackie, R. (1987) UV A-Induced melanocytic lesions. Br J. Dermatol.,
117, 111-115
Jose, J.G. & Pitts, D.G. (1985) Wavelength dependency of cataracts in albino mice following
chronic exposure. Exp. Eye Res., 41, 545-563
Joshi, S.D., Shrikhande, S.S. & Ranade, S.S. (1984) Studies on UV-induced skin tumors in experi-
mental animals. Photobiochem. Photobiophys., 7, 235-243
Joshi, S.D., Shrikhande, S.S., Naik, S.N. & Ranade, S.S. (1986) Multiplicity of tumours induced in
the laboratory animals by ultraviolet (UVC) irradiation. Indian J. anim. Sci., 56, 736-743
Jun. B.-D., Roberts, L.K., Cho, B.-H., Robertson. B. & Daynes, R.A. (1988) Parallel recovery of
epidermal antigen-presenting cell activity and contact hypersensitivity responses in mice
exposed to ultraviolet irradiation: the role of a prostaglandin-dependent mechanism. J. invest.
Dermatol., 90,311-316
Kaase, H., Zidowitz, B. & Berger, H. (1984) Spectral radiation measurements of UV-irradiation
apparatus and experiments to estimate the risk of photocarcinogenesis (Ger.). Lichtforschung, 1,
43-47
Kaidbey, K.H. (1988) Wavelength dependence for DNA synthesis inhibition in hairless mouse epi-
dermis. Photodermatology, 5, 65-70
Kaidbey, K.H. , Agin, P.P, Sayre, R.M. & Kligman, A.M. (1979) Photoprotection by melanin- a
comparison of black and Caucasian skin. J. Am. Acad Dermatol., 1, 249-260
Kalthoff, K. & Jackie, H. (1982) Photoreactivation of pyrimidine dimers generated by a photo-
sensitized reaction in RNA of insect embryos (Smittia spec.). In: Helene, C., Charlier, M.,
Montenay-Garestier, T. & Laustriat, G., eds, Trends in Photobiology, New York, Plenum,
pp.173-188
Kantor, G.J. (1985) Effects of sunlight on mammalian cells. Photochem. Photobiol. , 41,741-746
Kawada, A. ( 1986) UVB-Induced erythema, delayed tanning, and UV A-induced immediate tanning
in Japanese skin. Photodermatology, 3, 327-333
Keijzer, W. , Mulder, M.P., Langeveld, J.C.M., Smit, E.M.E., Bos, J.L., Bootsma, D. &
Hoeijmakers, J.H.J. (1989) Establishment and characterization of a melanoma cell line from a
xeroderma pigmentosum patient: activation of N-ras at a potential pyrimidine dimer site.
Cancer Res., 49, 1229-1235
Kelfkens, G., de Gruijl, F.R. & van der Leun, J.C. (1991a) Tumorigenesis by short-wave ultraviolet
A: papillomas versus squamous cell carcinomas. Carcinogenesis, 12, 1377-1382
Kelfkens, G., van Weelden, H., de Gruijl, F.R. & van der Leun, J.C. (199lb) Influence of dose rate
in ultraviolet tumorigenesis. J. Photochem. Photobiol. B Bioi., 10, 41-50
Keller, A.Z. (1970) Cellular types, survival, race, nativity, occupations, habits and associated
diseases in the pathogenesis of lip cancers. Am. J. Epidemiol. , 91, 486-499
Kelly, G.E., Meikle, W.D. & Sheil, A.G.R. (1987) Effects of immunosuppressive therapy on the
induction of skin tumors by ultraviolet irradiation in hairless mice. Transplantation, 44, 429-434
Kelly, G.E., Meikle, W.D. & Moore, D.E. (1989) Enhancement ofUV-induced skin carcinogenesis
by azathioprine: role of photochemical sensitisation. Photochem. Photobiol., 49, 59-65
REFERENCES 253
Kennedy, A.R., Ritter, M.A. & Little, J.B. (1980) Fluorescent light induces malignant
transformation in mouse embryo cell cultures. Science, 207, 1209-1211
Keyse, S.M. & Tyrrell, R.M. (1989) Heme oxygenase is the major 32-kDa stress protein induced in
human skin fibroblasts by UV A radiation, hydrogen peroxide and sodium arsenite. Proc. natl
A cad. Sci USA, 86, 99-103
Keyse, S.M., Moss, S.H. & Davies, D.J.G. (1983) Action spectra for inactivation of normal and
xeroderma pigmentosum human skin fibroblasts by ultraviolet radiations. Photochem.
Photobiol., 37, 307-312
Keyse, S.M., Applegate, L.A., Tromvoukis, Y. & Tyrrell, R.M. (1990) Oxidant stress leads to trans-
criptional activation of the human heme oxygenase gene in cultured skin fibroblasts. Mol. cell.
Bioi., 10,4967-4969
Khlat, M., Vail, A., Parkin, M. & Green, A. (1992) Mortality from melanoma in migrants to
Australia: variation by age at arrival and duration of stay. Am. J. Epidemiol., 135, 1103-1113
Kim, T.-Y., Kripke, M.L. & Ullrich, S. E. ( 1990) Immunosuppression by factors released from
UVirradiated epidermal cells: selective effects on the generation of contact and delayed hyper-
sensitivity after exposure to UV A or UVB radiation. J. invest. Dermatol., 94, 26-32
Kinlen, L.J., Sheil, A.G.R., Peto, J. & Doll, R. (1979) Collaborative United Kingdom-Australasian
study of cancer in patients treated with immunosuppressive drugs. Br. med. J., ii, 1461-1466
Kinsey, V.E. (1948) Spectral transmission of the eye to ultraviolet radiatioris. Arch. Opthalmol., 39,
508-5 13
Kirby-Smith, J.S. & Craig, D.L. (1957) The induction of chromosome aberrations in Tradescantia
by ultraviolet radiation. Genetics, 42, 176-187
Kirkpatrick, C.S., Lee, J.A.H. & White, E. (1990) Melanoma risk by age and socio-economic status.
Int. J. Cancer, 46, 1-4
Kjeldstad, B. & Johnsson, A. ( 1986) An action spectrum for blue and near ultraviolet inactivation of
Propionibacterium acnes; with emphasis on a possible porphyrin photosensitization.
Klamen, D.L. & Tuveson, R.W. (1982) The effect of membrane fatty acid composition on the
near-UV (300-400 nm) sensitivity of Eschenchia coli Kl 060. Photochem. Photobiol., 35,
167-173
Klein, K.L. & Linnemann, C. C., Jr (1986) Induction of recurrent genital herpes simplex virus type 2
infection by ultraviolet light (Letter to the Editor). Lancet, i, 796-797
Klepp, 0. & Magnus, K. (1979) Some environmental and bodily characteristics of melanoma
patients. A case-control study. Int. J. Cancer, 23, 482-486
Kligman, A.M. (1969) Early destructive effect of sunlight on human skin. J. Am. med. Assoc., 210,
2377-2380
Kligman, L.H. (1988) UV A enhances low dose UVB tumorigenesis (Abstract). Photochem.
Photobiol., 47, 8S
Kligman, L.H. (1989) The ultraviolet-irradiated hairless mouse: a model for photoaging. J. Am.
Acad. Dermatol., 21, 623-631
Kligman, L.H. & Kligman, A.M. ( 1981) Histogenesis and progression of ultraviolet light-induced
tumors in hairless mice. J. natl Cancer Inst. , 67, 1289-1297
Kligman, L.H. & Sayre, R.M. (1991) An action spectrum for ultraviolet induced elastosis in hairless
mice: quantification of elastosis by image analysis. Photochem. Photobiol., 53, 237-242
Kligman, L.H., Akin, F.J. & Kligman, A.M. (1985) The contributions of UVA and UVB to
connective tissue damage in hairless mice. J. invest. Dermatol., 84, 272-276
Kligman, L.H., Kaidbey, K.H., Hitchins, V.M. & Miller, S.A. (1987) Long wavelength(> 340 nm)
ultraviolet-A induced skin damage in hairless mice is dose dependent. Im Passchier, W.F. &
Bosnjakovic, B.F.M., eds, Human Exposure to Ultraviolet Radiation. Risks and Regulations,
Kligman, L.H., Crosby, M.J. , Miller, S.A., Hitchins, V.M. & Beer, J.Z. (1990) Skin cancer
induction in hairless mice by long-wavelength UV A radiation: a progress report (Abstract).
Photochem. Photobiol., 51, 18S-19S
Kligman, L.H., Crosby, M.J., Miller, S.A., Hitchins, V.M. & Beer, J.Z. (1992) Carcinogenesis by
long wavelength UVA and failure of reciprocity. In: Urbach, F., ed., Biological Responses to
Ultraviolet A Radiation, Overland Park, KS, Valdenmar, pp. 301-308
Knees-Matzen, S., Roser, M., Reimers, U., Ehlert, U., Weichenthal, M., Breitbart, E.W. & Rudiger,
H.W. ( 1991) Increased UV -induced sister-chromatic exchange in cultured fibroblasts of first-
degree relatives of melanoma patients. Cancer Genet. Cytogenet., 53, 265-270
Kocsard, E. & Ofner, F. (1964) Contact eczematous sensitization and sensitivity of the solar
elastotic skin. Aust. J. Dermatol. , 7, 203-205
Kolari . P.J., Lauharanta, J. & Hoikkala, M. (1986) Midsummer solar UV-radiation in Finland
compared with the UV -radiation from phototherapeutic devices measured by different
techniques. Photodermatology, 3, 340-345
Kopecky, K.E., Pugh, G.W., Jr, Hughes, D.E., Booth, G.D. & Cheville, N.F. (1979) Biological
effect of ultraviolet radiation on cattle: bovine ocular squamous cell carcinoma .. Am. J. vet. Res.,
40, 1783-1788
Kopf, A.W, Lindsay, A.C., Rogers, G.S., Friedman, R.J. , Rigel, D.S. & Levenstein, M. (1985)
Relationship of nevocytic nevi to sun exposure in dysplastic nevus syndrome. J. Am Acad.
Dermatol., 12, 656-662
Kostkowski, H.J., Saunders, R., Ward, J., Popenoe, C. & Green, A. ( 1982) Measurement of Solar
Terrestrial Spectral lrradiance in the Ozone (Cut-off Region, Self-studyManual on Optical
Radiation Meusurements. Part III, Applications (NBS Technical Note 910-5; PB 83-184382),
Washington DC, National Bureau of Standards
Koval, T.M. (1987) Photoreactivation of UV damage in cultured Drosophila cells. Experientia, 43,
445-446
Kraemer, K.H. (1980) Xeroderma pigmentosum. A prototype disease of enviromental-genetic inter-
action. Arch. Dermatol., 116, 541-542
Kraemer, K.H. Greene, M.H. (1985) Dysplastic nevus syndrome. Familial and sporadic precursors
of cutaneous melanoma. Dermatol. Clin., 3, 225-237
Kraemer, K.H. & Slor, H. (1984) Xeroderma pigmentosum. Clin Dermatol., 2, 33-69
Kraemer, K.H. & Weinstein, G. D. (1977) Decreased thymidine incorporation in circulating leuko-
cytes after treatment of psoriasis with psoralen and long-wave ultraviolet light. J. invest.
Dermatol. , 69,211-214
Kraemer, K.H., Lee, M.M. & Scotto, J. (1987) Xerodenna pigmentosum. Cutaneous. Ocular, and
neurologic abnormalities in 830 published cases. Arch. Dermatol., 123, 241-250
Kraemer, K.H., Seetharam, S., Protic-Sablj ic, M., Brash, D.E., Bredberg, A. & Seidman, M.M.
(1988) Defective DNA repair and mutagenesis by dimer and non-dimer photoproducts in
xerodenna pigmentosum measured with plasmid vectors. In: Friedberg, E.C. & Hanawalt, P.C.,
eds, Mechanisms and Consequences of DNA Damage Processing, New York, Alan R. Liss, pp.
325-335
Kramer, G.F. & Ames, B.N. (1987) Oxidative mechanisms of toxicity of low-intensits near- UV
light in Salmonella typhimurium. J. Bactenol., 169, 2259-2266
REFERENCES 255
Krepinsky, A.B., Rainbow, A.J. & Heddle, J.A. (1980) Studies on the ultraviolet light sensitivity of
Bloom's syndrome fibroblasts. Mutat. Res., 69, 357-368
Kricker, A., English, D.R., Randell, P.L. , Heenan, P.J. , Clay, C.D., Delaney, T.A. & Armstrong,
B.K. (1990) Skin cancer in Geraldton, Western Australia: a survey of incidence and prevalence.
Med. J. Aust., 152, 399-407
Kricker, A., Armstrong, B.K., English, D.R. & Heenan, P.J. (1991a) Pigmentary and cutaneous risk
factors for non-melanocytic skin cancer-a case-control study. Int. J. Cancer, 48, 650-662
Kricker, A., Armstrong, B.K., English, D., Heenan, P.J. & Randell, P.L. (1991b) A case-control
study of non-melanocytic skin cancer and sun exposure in Western Australia (Abstract No. III,
P2). Cancer Res. clin. Oncol., 117 (Suppl. II), S75
Kripke, M.L. (1974) Antigenicity of murine skin tumors induced by ultraviolet light. J. natl Cancer
lnst., 53, 1333-1336
Kripke, M.L. (1977) Latency, histology, and antigenicity of tumors induced by ultraviolet light in
three inbred mouse strains. Cancer Res., 37, 1395-1400
Kripke, M.L. & Fisher, M.S. (1976) Immunologic parameters of ultraviolet carcinogenesis. J. nat!
Cancer lnst., 57, 211-215
Kripke, M.L. & McClendon, E. (1986) Studies oil the role of antigen-presenting cells in the
systemic suppression of contact hypersensitivity by UVB radiation. J. lmmunol., 137, 443-447
Kripke, M. L. & Morison, W.L. (1985) Modulation of immune function by UV radiation. J. invest
Dermatol., 85, 625-665
Kripke, M.L. & Morison, W.L. (l986a) Studies on the mechanism of systemic suppression of
contact hypersensitivity by ultraviolet B radiation. Photodermatology, 3, 4-14
Kripke, M.L. & Morison, W.L. (1986b) Studies on the mechanism of systemic suppression of
contact hypersensitivity by UVB radiation. II. Differences in the suppression of delayed and
contact hypersensitivity in mice. J. invest. Dermatol., 86, 543-549
Kripke, M.L. & Sass, E.R., eds (1978) International conference on ultraviolet carcinogenesis. Nat!
Cancer Inst. Monogr. , 50
Kripke, M.L.. Lofgreen, J.S., Beard, J., Jessup, J.M. & Fisher, M.S. (1977) In vivo immune
responses of mice during carcinogenesis by ultraviolet irradiation. J. natl Cancer Inst., 59,
1227-1230
Kromberg, J.G.R., Castle, D. , Zwane, E.M. & Jenkins, T. (1989) Albinism and skin cancer in
southern Africa. Clin. Genet., 36, 43-52
Krutmann, J. & Elmets, C.A. (1988) Recent studies on mechanisms in photoimmunology.
Kubitschek, H.E. (1967) Mutagenesis of near-visible light. Science, 155, 1545-1546
Kumar, S., Joshi, P.C., Sharma, N.D., Bose, S.N., Davies, R.J.H, Takeda, N. & McCloskey, J.A.
(1991) Adenine photodimerization ill deoxyadenylate sequences: elucidation of the mechanism
through structural studies of a major d(ApA) photoproduct. Nucleic Acids Res., 19, 2841-2847
Kunkel, T.A. (1984) Mutational specificity of depurination. Proc. nat/ Acad. Sci. USA, 81,
1494-1498
Kupper, T.S., Chua, A.O., Flood, P. , McGuire, J. & Gubler, U. (1987) Interleukin 1 gene expression
in cultured human keratinocytes is augmented by ultraviolet irradiation. J. clin. Invest., 80,
430-436
Kurihara, T., Tatsumi, K., Takahashi, H. & Inoue, M. (1987) Sister-chromatid exchanges induced
by ultraviolet light in Bloom's syndrome fibroblasts. Mutat Res., 183, 197-202
Lai, L.-W & Rosenstein, B.S. (1990) Induction of DNA strand breaks and DNA-protein cross-links
in normal human skin fibroblasts following exposure to 254 nm UV radiation. J. Photochem.
Photobiol. B Bioi., 6, 395-404
Lancaster, H.O. (1956) Some geographical aspects of the mortality from melanoma in Europeans.
Med. J. Aust., 1, 1082-1087
Lancaster, H.O. & Nelson, J. (1957) Sunlight as a cause of melanoma: a clinical survey. Med. J.
Aust., 1, 452-456
Larcom, L.L., Morris, T.E. & Smith, M.E. (1991) Tanning salon exposure suppression of DNA
repair capacity and mitogen-induced DNA synthesis. Photochem. Photobiol., 53, 511-516
Larko, 0. & Diffey, B.L. (1983) Natural UV -B radiation received by people with outdoor, indoor
and mixed occupations and UV-B treatment of psoriasis. Clin. exp. Dermatol., 8, 279-285
Larko, 0. & Diffoy, B.L. (1986) Occupational exposure to ultraviolet radiation in dermatology
departments. Br. J. Dermatol. , 114, 479-484
Lau, H., Reemtsma, K. & Hardy, M.A. (1983) Pancreatic islet allograft prolongation by
donor-specific blood transfusions treated with ultraviolet irradiation. Science, 221, 754-756
Lau, H., Reemtsma, K. & Hardy, M.A. (1984) Prolongation of rat islet allograft survival by direct
ultraviolet irradiation of the graft. Science, 223, 607-609
Leach, J.F., McLeod, V.E., Pingstone, A.R., Davis, A. & Deane, C.H.W. (1978) Measurement of
the ultraviolet doses received by office workers. Clin. exp. Dermatol., 3, 77-79
LeClerc, J.E., Borden, A. & Lawrence C.W. (1991) The thymine-thymine pyrimidine-pyrirnidone
(6-4) ultraviolet light photoproduct is highly mutagenic and specifically induces 3'-thymine-to-
cytosine transitions in Escherichia coli. Proc. natl Acad. Sci. USA, 88, 9685-9689
Lee, J.A.H. (1982) Melanoma and exposure to sunlight. Epidemiol. Rev., 4, 110-136
Lee, J.A.H. & Storer, B.E. (1980) Excess of malignant melanomas in women in the British Isles.
Lancet, ii, 1337-1339
Lee, J.A.H. & Strickland, D. (1980) Malignant melanoma: social status and outdoor work. Br. J.
Cancer,41, 757-763
Lee, G.S.-F., Savage, E.A., Ritzel, R.G. & von Borstel, R.C. (1988) The base-alteration spectrum of
spontaneous and ultraviolet radiation-induced forward mutations in the URA3 locus of
Saccharomyces cerevisiae. Mol. gen. Genet. , 214, 396-404
Lehmann, A.R., Arlett, C.F., Broughton, B.C., Harcourt, S.A., Steingrimsdottir, H., Stefanini, M.,
Taylor, A.M.R., Natarajan, A.T, Green, S., King, M.D., MacKie, R.M., Stephenson, J.B.P &
Tolmie, J.L. ( 1988) Trichothiodystrophy, a human DNA repair disorder with heterogeneity in
the cellular response to ultraviolet light. Cancer Res., 48, 6090-6096
Lennan, S. (1980) Radiant Energy and the Eye, New York, MacMillan
Lerman. S. (1988) Ocular phototoxicity. New Engl. J. Med., 319, 1475-1477
Letvin, N.L., Greene, M.l., Benacerraf, B. & Germain, R.N. (1980a) Immunologic effects of whole-
body ultraviolet irradiation: selective defect in splenic adherent cell function in vitro. Proc. natl
Acad. Sci. USA, 77, 2881-2885
Letvin, N.L., Nepom, Jar, Greene, M.I., Benacerraf, B. & Germain, K.N. (1980b) Loss ofla-bearing
splenic adherent cells after whole body ultraviolet irradiation J. Immunol., 125, 2550-2554
van der Leun, J.C. (1984) Yearly review: UV-carcinogenesis. Photochem. Photohiol. , 39, 861-868
van der Leun, J.C. (1987a) Principles of risk reduction and protection. In: Passchier, W.F. &
Bosnjakovic, B.E.M., eds, Human Exposure to Ultraviolet Radiation: Risks and Regulations.
Amsterdam. Elsevier, pp. 293-303
van der Leun, J.C. (1987b) Interactions of different wavelengths in effects ofUV radiation on skin.
Photodermatology, 4, 257-264
REFERENCES 257
van der Leun, J.C. (1992) Interactions of UV A and UVB in photodermatology: what was photo-
augmentation? In: Urbach, F., ed, The Biological Responses to Ultraviolet A Radiation,
Overland Park, KS, Valdenmar, pp. 309-319
van der Leun, J.C. & van Weelden, H. (1986) UVB phototherapy: principles, radiation sources,
regimens. Curr. Probl. Dermatol., 15, 39-51
Levi, F., La Vecchia, C., Te, V.-C. & Mezanotte, G. (1988) Descriptive epidemiology of skin
cancer in the Swiss canton of Vaud. Int. J. Cancer, 42, 811-816
Lew, R.A., Sober, A.J., Cook, N., Marvell, R. & Fitzpatrick, T.B. (1983) Sun exposure habits in
patients with cutaneous melanoma: a case control study. J. Dermatol. surg. Oncol., 9, 981-986
Ley, R.D. (1985) Photoreactivation of UV-Induced pyrimidine dimers and erythema in the
marsupial Monodelphis domestica. Proc. natl Acad. Sci. USA, 82, 2409-2411
Ley, R.D. & Applegate, L.A. (1989) Photoderrnatological studies with the marsupial Monodelphis
domestica. In: Maibach, R.I. & Lowe, N.J., eds, Models in Dermatology, Vol.4, Basel, Karger,
pp. 265-275
Ley, R.D., Sedita, B.A. & Boye, E. (1978) DNA polymerase !-mediated repair of 365 nm-induced
single-strand breaks in the DNA of Escherichia coli. Photochem. Photobiol., 27, 323-327
Ley, R.D., Peak, M.J. & Lyon, L.L. (1983) Induction of pyrimidine dimers in epidennal DNA of
hairless mice by UVB: an action spectrum. J. invest. Dermatol., 80, 188-191
Ley. R.D., Applegate, L.A., Stuart, T.D. & Fry, R.J.M. (1987) UV radiation-induced skin tumors in
Monodelphis domestica. Photodermatology, 4, 144-147
Ley, R.D., Applegate, L.A. & Freeman, S.E. (1988) Photorepair of ultraviolet radiation-induced
pyrimidine dimers in corneal DNA. Mutat. Res., 194, 49-55
Ley, R.D., Appiegate, L.A., Padilla, R.S. & Stuart, T.D. (1989) Ultraviolet radiation-induced
malignant melanoma in Monodelphis domestica. Photochem. Photobiol., 50, 1-5
Ley, R.D., Applegate, L.A., Fry, R.J.M. & Sanchez, A.B. (1991) Photoreactivation of ultraviolet
radiation-induced skin and eye tumors of Monodelphis domestica. Cancer Res., 51, 6539-6542
Lightfoote, M.M., Zmundzka, B.Z., Olvey, K.M., Miller, S.A. & Beer, J.Z. (1992) Effects of UVB
radiation on human immunodeficiency virus. In: Houck, M. & Klign1an, A., eds, Biological
Effects of Light (in press)
Lill, P.H. ( 1983) Latent period and antigenicity of murine tumors induced in C3H mice by short-
wavelength ultraviolet radiation. J. invest. Dermatol., 81, 342-346
Lin, C.S., Goldthwait, D.A. & Samols, D. (1990) Induction of transcription from the long terminal
repeat of Moloney murine sarcoma provirus by UV-irradiation, x-irradiation, and phorbol ester.
Proc. natl Acad. Sci. USA, 87, 36-40
Lindahl-Kiessling, K. & Safwenberg, J. (1971) Inability ofUV-irradiated lymphocytes to stimulate
allogenic cells in mixed lymphocyte cultures. Int. Arch. Allergy, 41, 670-678
Lindqvist, C. (1979) Risk factors in lip cancer: a questionnaire survey. Am. J. Epidemiol., 109,
521-530
Lippincott, S.W. & Blum, H.F. (1943) Neoplasms and other lesions of the eye induced by
ultraviolet radiation in strain A mice. J. natl Cancer Inst., 3, 545-554
Lippke, J.A., Gordon, L.K., Brash, D.E. & Haseltine, W.A. (1981) Distribution ofUV light-induced
damage in a defined sequence of human DNA: detection of alkaline-sensitive lesions at pyri-
midine nucleoside-cytidine sequences. Proc. natl Acad. Sci USA, 78, 3388-3392
Lischko, A.M., Seddon, J.M., Gragoudas, E.S., Egan, KM. & Glynn, R.J. (1989) Evaluation of prior
primary malignancy as a determinant of uveal melanoma. A case-control study. Ophthalmology,
96, 1716-1721
Livingston, W. (1983) Landscape as viewed in the 320-nm ultraviolet. J. opt. Soc. Am., 73,
1653-1657
Logani, M.K, Sambuco, C.P., Forbes, P.D. & Davies, R.E. (1984) Skin-tumour promoting activity
of methyl ethyl ketone peroxide- a potent lipid-peroxidizing agent. Food chem. Toxicol., 22,
879-882
Luande, J., Henschke, C.l. & Mohammed, N. (1985) The Tanzanian human albino skin. Cancer, 55,
1823-1828
van der Lubbe, J.L.M., Rosdorff, H.J.M., Bos, J.L. & van der Eb, A.J. (1988) Activation of N-ras
induced by ultraviolet irradiation in vitro. Oncogene Res., 3, 9-20
Lucke-Huhle, C., Mai, S. & Herrlich, P. (1989) UV-lnduced early-domain binding factor as the
limiting component of simian virus 40 DNA amplification in rodent cells. Mol. cell. Bioi. , 9,
4812-4818
Lundgren, K & Wulf, H. C. (1988) Cytotoxicity and genotoxicity of UV A irradiation in Chinese
hamster ovary cells measured by specific locus mutations, sister chromatic exchanges and
chromosome aberrations. Photochem. Photobiol., 41, 559-563
Lunnon, R.J. (1984) Ultraviolet photography. In: Williams, R. , ed., Medical Photography Study
Guide. Lancaster, MTP Press, pp. 225-233
Lynch, D.H., Gurish, M.F. & Haynes, R.A. (1983) The effects of high-dose UV exposure on murine
Langerhans cell function at exposed and unexposed sites as assessed using in vivo and in vitro
assays. J. invest. Dermatol., 81, 336-341
Lynge, E. & Thygesen, L. (1990) Occupational cancer in Denmark. Cancer incidence in the 1970
census population. Scand. J. Work Envuon. Health, 16 (Suppl. 2), 1-35
Machta, L., Cotton, G., Hass, W. & Komhyr, W. (1975) ClAP Measurements of Solar Ultraviolet
Radiation (Final Report, Interagency Agreement DOT-A5-20082), Washington DC, US Depart-
ment of Transportation
Mack, T.M. & Floderus, B. (1991) Malignant melanoma risk by nativity, place of residence at
diagnosis, and age at migration. Cancer Causes Control, 2, 401-411
MacKie, R.M. & Aitchison, T (1982) Severe sunburn and subsequent risk of primary cutaneous
malignant melanoma in Scotland. B. J. Cancer, 40, 955-960
MacKie, R.M., Freudenberger, T. & Aitchison, T.C. (1989) Personal risk-factor chart for cutaneous
melanoma. Lancet, ii, 487-490
MacLaughlin, J. M.F. (1985) Aging decreases the capacity of human skin to produce
vitamin D3. J. clin. Invest. , 76, 1536- 1538
Madewell, B.R., Conroy, J.D. & Hodgkins, E.M. (1981) Sunlight-skin cancer association in the
dog: a report of three cases. J. cutaneous Pathol., 8, 434 -443
Magee, M.J., Kripke, M.L. & Ullrich, S.E. (1989a) Suppression of the elicitation of the immune
response to alloantigen by ultraviolet radiation. Transplantation, 41, 1008-1013
Magee, M.J., Kripke, M.L. & Ullrich, S.E. (1989b) Inhibition of the immune response to
alloantigen in the rat by exposure to ultraviolet radiation. Photochem. Photobiol. , 50, 193-199
Magnus, K. (1973) Incidence of malignant melanoma of the skin in Norway, 1955-1970. Variations
in time and space and solar radiation. Cancer, 32, 1275-1286
Magnus, K. (1991) The Nordic profile of skin cancer incidence. A comparative epidemiological
study of the three main types of skin cancer. Int. J. Cancer, 41, 12-19
Maher, V.M., Dorney, D.J., Mendrala, A.L., Konze-Thomas, B. & McCormick, J.J. (1979) DNA
excision-repair processes in human cells can eliminate the cytotoxic and mutagenic conse-
quences ofultraviolet irradiation. Mutat. Res., 62,311-323
REFERENCES 259
Maher, V.M., Rowan, L.A., Silinskas, K.C., Kateley, S.A. & McCormick, J.J. (1982) Frequency of
UV induced neoplastic transformation of diploid human fibroblasts is higher in xeroderma
pigmentosum cells than in normal cells. Proc. natl Acad. Sc. USA, 79, 2613-2617
Maltzman, W. & Czyzyk, L. (1984) UV-lrradiation stimulates levels of p53 cellular tumor antigen
in nontransformed mouse cells. Mol. cell. Bioi., 4, 1689-1694
Mandai, T.K. & Chatterjee, S.N. (1980) Ultraviolet- and sunlight-induced lipid peroxidation in lipo-
somal membrane. Radiat. Res., 83, 290-302
Mariutti, G. & Matzeu, M. (1987) Measurement of ultraviolet radiation in welding processes and
hazard evaluation. In: Passchier, W.F. & Bosnjakovic, B.F.M., eds, Human Exposure to
Ultraviolet Radiation: Risks and Regulations, Amsterdam, Elsevier, pp. 387-390
Marks, R. & Selwood, T.S. (1985) Solar keratoses. The association with erythemal ultraviolet
radiation in Australia. Cancer, 56, 2332-2336
Marks, R., Ponsford, M.W., Selwood, T.S., Goodman, G. & Mason, G. (1983) Non-melanocytic
skin cancer and solar keratoses in Victoria Med. J. Aust., 2, 619-622
Marks, R., Rennie, G. & Selwood, T.S. (1988) Malignant transformation of solar keratoses to
squamous cell carcinoma. Lancet, i, 795-797
Marks, R., Jolley, D., Dorevitch, A.P. & Selwood, T.S. (1989) The incidence of non-melanocytic
skin cancers in an Australian population: results of a five-year prospective study. Med. J. Aust.,
150, 475-478
Marshall, R.R. & Scott, D. (1976) The relationship between chromosome damage and cell killing in
UV -irradiated normal and xeroderma pigmentosum cells. Mutat. Res., 36, 397-400
Marshall, R.R., Arlett, C.F., Harcourt, S.A. & Broughton, B.A. (1980) Increased sensitivity of cell
strains from Cockayne's syndrome to sister-chromatic exchange induction and cell killing by
UV light. Mutat. Res., 69, 107-112
Matsumoto, K., Sugiyama, M. & Ogura, R. (1991) Non-dimer DNA damage in Chinese hamster
V -79 cells exposed to ultraviolet-B light. Photochem. Photobiol., 54, 389-392
Matsunaga, T., Hieda, K. & Nikaido, 0 . (1991) Wavelength dependent formation of thymine
dimers and (6-4)photoproducts in DNA by monochromatic ultraviolet light ranging from 150 to
365 nm. Photochem. Photobiol., 54, 403-410
Matsuoka, L.Y., Ide, L., Wortsman, J., MacLaughlin, J.A. & Holick, M.F. (1987) Sunscreens
suppress cutaneous vitamin D3 synthesis. J. clin. Endocnnol. Metab., 64, 1165-1168
Mayne, L.V., Mullenders, L.H.F. & van Zeeland, A.A. (1988) Cockayne's syndrome: a UV
sensitive disorder with a defect in the repair of transcribing DNA but normal overall excision
repair. In: Friedberg, E.C. & Hanawalt, PC., eds, Mechanisms and Consequences of DNA
Damage Processing, New York, Alan R. Liss, pp. 349-353
McCormick, J.P., Fisher, J.R., Pachlatko, J.P. & Eisenstark, A. (1976) Characterisation of a
cell-lethal product from the photooxidation of tryptophan: hydrogen peroxide. Science, 191,
468-669
McCredie, M. & Coates. M.S. (1989) Cancer Incidence in Migrants to New South Wales. 1972 to
1984, New South Wales Central Cancer Registry, Woolloomooloo, NSW, New South Wales
Cancer Council, pp. 22-23, 62-63
McGovern, V.J., Shaw, H.M., Milton, G.W. & Farago, G.A. (1980) Is malignant melanoma arising
in a Hutchinson's melanotic freckle a separate disease entity? Histopathology, 4, 235-242
McGrath, H., Wilson, W .A. & Scopelitis, E. (1986) Acute effects of low-fluence ultraviolet light on
human T-lymphocyte subsets. Photochem. Photobiol. , 43, 627-631
McGregor, W.G., Chen, R.-H., Lukash, L., Maher, V.M. & McCormick, J.J. (1991) Cell cycle-
dependent strand bias for UV-induced mutations in the transcribed strand of excision repair-
proficient human fibroblasts but not in repair-deficient cells. Mol. cell. Biol., 11, 1927-1934
McKinlay, A.F. & Diffey, B.L. (1987) A reference action spectrum for ultraviolet induced erythema
in human skin. CIE (Commission lnternationale de l'Eclairage) J. , 6, 17-22
McKinlay, A.F. & Whillock, M.J. (1987) Measurement of ultra-violet radiation from fluorescent
lamps used for general lighting and other purposes in the UK In: Passchier, W.F. &
Bosnjakovic, B.F.M., eds, Human Exposure to Ultraviolet Radiation: Risks and Regulations,
McKinlay, A.F. , Whillock, M.J. & Meulemans, C.C.E. (1989) Ultraviolet Raduation and Blue-light
Emissions from Spotlights Incorporating Tungsten Halogen Lamps (Report NRPB-R228),
Didcot, National Radiological Protection Board
McLennan, A. G. (1987) The repair of ultraviolet light-induced DNA damage in plant cells. Mutat.
Res., 181, 1-7
Mecherikunnel, A.T. & Richmond, J.C. (1980) Spectral Distribution of Solar Radiation (NASA
Technical Memorandum 82021), Greenbelt, MD, Goddard Space Flight Center
Melski, J.W., Tanenbaum, L., Parrish, J.A., Fitzpatrick, T.B. & Bleich, H.L. (1977) Oral
methoxsalen photochemotherapy for the treatment of psoriasis: a cooperative clinical trial. J.
invest. Dermatol, 68, 328-335
Menzies, S.W. , Greenoak, G.E., Reeve, V.E. & Gallagher, C.H. (1991) Ultraviolet
radiation-induced murine tumors produced in the absence of ultraviolet radiation-induced
systemic tumor immunosuppression. Cancer Res., 51, 2773-2779
Miguel, A.G. & Tyrrell, R.M. (1983) Induction of oxygen-dependent lethal damage by
monochromatic UVB (313 run) radiation: strand breakage, repair and cell death.
Milham, S., Jr (1983) Occupational Mortality in Washington State 1950-1979 (DHSS (NIOSH)
Publ. No. 83-116), Cincinnati, OH, National Institute for Occupational Safety and Health
Miller, J.H. (1985) Mutagenic specificity of ultraviolet light. J. mol. Bioi., 182, 45-68
Miskin, R. & Ben-Ishai, R. (1981) Induction of plasminogen activator by UV light in normal and
xeroderma pigmentosum fibroblasts. Proc. nat! Acad. Sci USA, 78, 6236-6240
Mitchell , D.L. (1988) The relative cytotoxicity of (6-4)photoproducts and cyclobutane dimers in
mammalian cells. Photochem. Photobiol., 48, 51-57
Mitchell, D.L. , Haipek, C.A. & Clarkson, J.M. (1985) (6-4)Photoproducts are removed from the
DNA of UV-irradiated mammalian cells more efficiently than cyclobutane pyrimidine dimers.
Mutat. Res., 143, 109- 112
Mitchell, D.L., Jen, J. & Cleaver, J.E. ( 1991) Relative induction of cyclobutane dimers and cytosine
photohydrates in DNA irradiated in vitro and in vivo with ultraviolet-C and ultraviolet-B light.
Moan, J., Dahlback, A., Henriksen, T & Magnus, K. (1989) Biological amplification factor for
sunlight-induced nonmelanoma skin cancer at high latitudes. Cancer Res., 49, 5207-5212
Molina, M.J. & Rowland, F.S. (1974) Stratospheric sink for chlorofluoromethanes: chlorine atom-
catalyzed destruction of ozone. Nature, 249, 810-812
Mondal , S. & Heidelberger, C. (1976) Transformation of C3H/10T1/2 CL8 mouse embryo
fibroblasts by ultraviolet irradiation and a phorbol skin. Nature, 260, 710-711
Morgun, V.V., Logvinenko, V.F., Dvemyakov, V.S. & Timoshenko, VM. (1988) Mutagenic
activity of pulsed concentrated sunlight. Sov. Genet., 24, 189-193
Morison, W.L. (1983a) Phototherapy and Photochemotherapy of Skin Disease, New York, Praeger
REFERENCES 261
Morison, W.L. (1983b) Solar urticaria. In: Parrish, J.A., Kripke, M.L. & Morison, W.L., eds,
Photoimmunology, New York, Plenum, pp. 215-226
Morison, W.L. (1983c) Autoimmune disease. In: Parrish, J.A., Kripke, M.L. & Morison, W.L. , eds,
Photoimmunology, New York, Plenum, pp. 255-266
Morison, W.L. (1986) The effects of UVA radiation on immune function. In: Urbach, F. & Grange,
R.W., eds, The Biological Effects ofUVA Radiation, New York, Praeger, pp. 202-209
Morison, W.L. (1989) Effects of ultraviolet radiation on the immune system in humans. Photochem.
Photobiol., 50, 515-524
Morison, W.L. & Kelley, S.P. (1985) Sunlight suppressing rejection of 280 to 320 nm
UV-radiationinduced skin tumors in mice. J. nat! Cancer Inst., 74, 525-527
Morison, W.L. & Kochevar, I.E. (1983) Photoallergy. In: Parrish, J.A., Kripke, M.L. & Morison,
W.L., eds, Photoimmunology, New York, Plenum, pp. 227-253
Morison, W.L. & Kripke, M.L. (1984) Systemic suppression of contact hypersensitivity by
ultraviolet B radiation or methoxsalen/ultraviolet A radiation in the guinea pig. Cell. lmmunol. ,
85, 270-277
Morison, W.L., Parrish, J.A. , Bloch, K.J. & Krugler, J.I. (1979a) In vivo effect of UVB on
lymphocyte function. Br. 1. Dermatol., 101, 513-519
Morison, W.L. , Parrish. J.A. , Anderson, R.R. & Bloch, K.J. (1979b) Sensitivity of mononuclear
cells to UV radiation. Photochem. Photobiol. , 29, I 045-1047
Morison, W.L., Bucana, C. & Kripke, M.L. (1984) Systemic suppression of contact hypersensitivity
by UVB radiation is unrelated to the UVB-induced alterations in the morphology and number of
Langerhans cells. Immunology, 52, 299-306
Morison, W.L., Pike, R.A. & Kripke, M.L. (1985) Effect of sunlight and its component wavebands
on contact hypersensitivity in mice and guinea pigs. Photodermatology, 2, 195-204
Morison, W. L. , Jerdan, M.S. , Hoover, T L. & Farmer, E.R. ( 1986) UV radiation-induced tumors
in haired mice: identification as squamous cell carcinomas. J. natl Cancer Inst. , 77, 1155-1162
M0rk, N.-J. & Austad, J. (1982) Short-wave ultraviolet light (UVB) treatment of allergic contact
detmatitis of the hands. Acta dermatovenereol. , 63, 87-89
Morrison, H. (1985) Photochemistry and photobiology of urocanic acid. Photodermatology, 2,
158-165
Moseley, H. (1988) Non-ionising Radiation: Microwaves, Ultraviolet and Laser Radiation (Medical
Physics Handbooks 18), Bristol, Adam Hilger, pp. 110-154
Moss, S.H. & Smith, K.C. (1981) Membrane damage can be a significant factor in the inactivation
of Escherichia coli by near-ultraviolet radiation. Photochem. Photobiol., 33, 203-210
Muir, C., Waterhouse, J., Mack, T. , Powell, J. & Whelan, S., eds (1987) Cancer Incidence in Five
Continents, Vol. V (IARC Scientific Publications No. 88), Lyon, IARC
Munakata, N. (1989) Genotoxic action of sunlight upon Bacillus subtilis spores: monitoring studies
at Tokyo, Japan. J. Radiat. Res., 30, 338-351
Murthy, P.B., Rahiman, M.A. & Tulpule, P.G. (1982) UV-Induced chromosome aberrations, sister-
chromatid exchanges and cell survival in cultured lymphocytes from malnourished children.
Pediatr Res., 16, 663-664
Musk, P. Campbell, R., Staples, J., Moss, D . .J. & Parsons, R.G. (1989) Solar and UVC-induced
mutation in human cells and inhibition by deoxynucleosides. Mutat. Res., 227, 25-30
Mutzhas, M.F. (1986) UVA-Emitting light sources. In: Urbach, F. & Grange, R.W., eds. The
Biological Effects of UVA Radiation, New York, Praeger, pp. 10-23
Myhr, B.C., Turnbull, D. & DiPaolo, J.A. (1979) Ultraviolet mutagenesis of normal and xeroderma
pigmentosum variant human fibroblasts. Mutat. Res., 62,341-353
Natarajan, A.T., van Zeeland, A.A., Verdegaal-Immerzeel, E.A.M. & Filon, A.R. (1980) Studies on
the influence of photoreactivation on the frequencies of UV-induced chromosomal aberrations,
sister-chronlaticl exchanges and pyrimidine dimers in chicken embryonic fibroblasts. Mutat.
Res., 69, 307-317
Nathanson, R.B., Forbes, P.D. & Urbach, F. (1976) Modification of photocarcinogenesis by two
immunosuppressive agents. Cancer Lett., 1, 243-247
Nelson, E.W., Eichwald, E.J. & Shelly, J. (1987) Increased ultraviolet radiation-induced skin
cancers in cyclosporine-treated mice. Transplant. Proc. , 19, 526-527
Nikula, K.J., Benjamin, S.A., Angleton, G.M., Saunders, W.J. & Lee, A.C. (1992) Ultraviolet
radiation, solar dermatosis, and cutaneous neoplasia in beagle dogs. Radiat. Res., 129, 11-18
Nishi, Y., Hasegawa, M.M. , Taketomi, M., Ohkawa, Y. & Inui, N. (1984) Comparison of 6-thio-
guanine-resistant mutation and sister chromatic exchanges in Chinese hamster V79 cells with
forty chemical and physical agents. Cancer Res., 44, 3270-3279
Noonan, F.P. & De Fabo, E.C. (1985) Immune suppression by ultraviolet radiation and its role in
ultraviolet radiation induced carcinogenesis in mice. Aust. J. Dermatol. , 26, 4-8
Noonan, F.P., De Fabo, E.C. & Kripke, M.L. (1981a) Suppression of contact hypersensitivity by
UV radiation and its relationship to UV-induced suppression of tumor immunity. Photochem.
Photobiol. . 34, 683 689
Noonan, F.P, Kripke, M.L., Pedersen, G.M. & Greene, M.L (1981b) Suppression of contact hyper-
sensitivity in mice by ultraviolet irradiation is associated with defective antigen presentation.
Noonan, F.P, Bucana, C., Sauder, D.N. & De Fabo, E.C. (1984) Mechanism of systemic immune
suppression by UV irradiation in vivo. II. The UV effects on number and morphology of epi-
dermal Langerhans cells and the UV -induced suppression of contact hypersensitivity have
different wavelength dependencies. J. lmmunol., 132,2408-2416
Noonan, F.P, De Fabo, E.C. & Morrison, H. (1988) cis-Urocanic acid, a product fonned by
ultraviolet B irradiation of the skin, initiates an antigen presentation defect in splenic dendritic
cells in vivo. J. invest. Dermatol., 90, 92-99
Norbury, K.C. & Kripke, M. L. ( 1978) Ultraviolet carcinogenesis in T-cell depleted mice . J. natl
Cancer lnst., 61, 917-921
Norimura, T., Maher, V.M. & McCormick, J.J. (1990) Cytotoxic and mutagenic effect of UV,
ethylnitrosourea and ()-713,8a-dihydroxy-9a, 1 Oa.-epoxy-7,8,9, 10-tetrahydrobenzo[a]pyrene in
diploid human T lymphocytes in culture: comparison with fibroblasts. Mutagenesis, 5, 447-451
Norval, M., Mclutyre, C.R., Simpson, T.J., Howie, S.E.M. & Bardshiri, E. (1988) Quantification of
urocanic acid isomers in murine skin during development and after irradiation with UVB light.
Nusbaum, B.P, Edwards, E.K., Jr, Horwitz, S.N. & Frost, P. (1983) Psoriasis therapy. The effect of
UV radiation on sensitization to mechlorethamine. Arch. Dermatol., 119, 117-121
O'Beim, S.F., Judge, P. Urbach, F., MacCon, C.F. & Martin, F. (1970) Skin cancer in County
Galway, Ireland. Proc. natl Cancer Conf, 6, 489-500
O'Dell, B.L., Jessen., R.T, Becker, L.E., Jackson, R.T. & Smith, E.B. (19801 Diminished immune
response in sun-damaged skin. Arch. Dermatol., 116, 559-561
Oettle, A.G. (1963) Skin cancer in Africa. Natl Cancer Inst. Monogr. , 10, 197-214
Office of Population Censuses and Surveys (1986) Occupationul Mortality: the Registar General's
Decennial Supplement for Great Britain 1979-80, 1982-83 (Series DS No. 6), London, Her
Majesty's Stationery Office
REFERENCES 263
Okamoto, H. & Kripke, M.L. (1987) Effector and suppressor circuits of the immune response are
activated in vivo by different mechanisms. Proc. natl Acad. Sci. USA, 84, 3841-3845
O'Loughlin, C., Moriarty, M.J., Herity, B. & Daly, L. (1985) A re-appraisal of risk factors for skin
carcinoma in Ireland: a case control study. Irish J. med. Sci., 154, 61-65
Olsen, J.H. & Jensen. O.M. ( 1987) Occupation and risk of cancer in Demnark. An analysis of 93
810 cancer cases, 1970-1979. Scand. J. Work Environ. Health, 13 (Suppl. 1), 1-91
Olsen, W.M., Huitfeldt, H.S. & Eggset, G. (1989) UVB-Induced (6-4) photoproducts in hairless
mouse epidennis studied by quantitative immunohistochemistry. Carcinogenesis, 10, 1669-1673
Onuigbo, W.l.B. (1978) Lip lesions in Nigerian lgbos. Int. J. oral Surg., 7, 73-75
0sterlind, A. (1987) Trends in incidence of ocular malignant melanoma in Denmark 1943-1982. Int.
J. Cancer, 40, 161-164
0sterlind, A. (1990) Malignant melanoma in Denmark. Occurrence and risk factors. Acta oncol.,
29, 1-22
0sterlind, A., Olsen, J.H., Lynge, E. & Ewertz, M. (1985) Second cancer following cutaneous mela-
noma and cancers of the brain, thyroid, connective tissue, bone, and eye in Denmark, 1943-80.
Nat! Cancer Inst. Monogr., 68, 361-388
0sterlind, A., Hou-Jensen, K. & Jensen, O.M. (1988a) Incidence of cutaneous malignant melanoma
in Denmark 1978-1982. Anatomic site distribution, histologic types, and comparison with non-
melanoma skin cancer. Br. J. Cancer, 58, 385-391
0sterlind, A., Tucker, M.A., Stone, B.J. & Jensen, O.M. (1988b) The Danish case-control study of
cutaneous malignant melanoma. 11. Importance of UV -light exposure. Int. J. Cancer, 42,
319-324
0sterlind, A., Tucker, M.A., Hou-Jensen, K., Stone, B.J., Engholm, G. & Jensen, O.M. (198&) The
Danish case-control study of cutaneous malignant melanoma. I. Importance of host factors. Int.
J. Cancer, 42, 200-206
Otani, T & Mori, R. ( 1987) The effects of ultraviolet irradiation of the skin on herpes simplex virus
infection: alterations in immune function mediated by epidennal cells and in the course of
infection. Arch. Virol., 96, 1-15
Oxholm, A., Oxholm, P., Staberg, B. & Bendtzen, K. (1988) Immunohistological detection of inter-
leukin 1-like molecules and tumour necrosis factor in human epidermis before and after UVB-
irradiation in vivo. Br. J. Dermatol., 118, 369-376
Paffenbarger, R.S., Jr, Wing, A.L. & Hyde, R.T. (1978) Characteristics in youth predictive of adult-
onset malignant lymphomas, melanomas, and leukemias: brief communication. J. natl Cancer
Inst., 60, 89-92
Parrington, J.M. (1972) Ultraviolet-induced chromosome aberrations and mitotic delay in human
fibroblast cells. Cytogenetics, 11, 117-131
Parrington, J.M., Delhanty, J.D.A. & Baden, H.P. (1971) Unscheduled DNA synthesis, UV-induced
chromosome aberrations and SV 40 transformation in cultured cells from xeroderma pigmen-
tosum. Ann. hum. Genet., 35, 149-160
Parrish, J.A. (1983) The Effect of Ultraviolet Radiation on the Immune System, New Brunswick, NJ,
Johnson & Johnson
Parrish, J.A. & Jaenicke, K.F. (1981) Action spectrum for phototherapy of psoriasis. J. invest.
Dermatol, 76, 359-362
Parrish, J.A. , Fitzpatrick, T.B., Tanenbaum, L. & Pathak, M.A. (1974) Photochemotherapy of
psoriasis with oral methoxsalen and longwave ultraviolet light. New Engl. J. Med., 291,
1207-1211
Parrish, J.A., Zaynoun, S. & Anderson, R.R. (1981) Cumulative effects of repeated subthreshold
doses of ultraviolet radiation. J. invest. Dermatol., 16, 356-358
Parrish, J.A., K.ripke, M.L. & Morison, W.L. , eds (1983) Photoimmunology, New York, Plenum
Parry, J.M., Sharp, D., Tippins, R.S. & Parry, E.M. (1979) Radiation-induced mitotic and meiotic
aneuploidy in the yeast Saccharomyces cerevisiae. Mutat. Res., 61, 37-55
Parshad, R., Taylor, W.G., Sanford, K.K., Camalier, R.F., Gantt, R. & Tarone, R.E. (1980a) Fluo-
rescent light-induced chromosome damage in human IMR-90 fibroblasts. Role of hydrogen
peroxide and related free radicals. Mutat. Res. , 73, 115-124
Parshad, R., Sanford, K.K. , Jones, G .. M., Tarone, R.E., Hoffman, H.A. & Grier, A.H. ( 1980b)
Susceptibility to fluorescent light-induced chromatic breaks associated with DNA repair
deficiency and malignant transfonnalion in culture. Cancer Res., 40, 4415-4419
Pathak, M.A., Fitzpatrick, T.B., Greiter, F. & Kraus, E.W. (1987) Preventive treatment of sunburn,
dermatoheliosis, and skin cancer with sun-protective agents. In: Fitzpatrick, T.B., Eisen, A.Z.,
Wolff, K. , Freedberg, I.M. & Austen, K.F., eds, Dermatology in General Medicine, 3rd ed.,
New York, McGraw-Hill, pp. 1507-1522
Patrick, M.H. ( 1977) Studies on thymine-derived UV photoproducts in DNA-I. Formation and
biological role of pyrimidine adducts in DNA. Photochem. Photobiol., 25, 357-372
Patton, J.D., Rowan, L.A., Mendrala, A.L., Howell, J.N., Maher, V.M. & McCormick, J.J. (1984)
Xeroderma pigmentosum fibroblasts including cells from XP variants are abnormally sensitive
to the mutagenic and cytotoxic action of broad spectrum simulated sunlight. Photochem.
Photobiol., 39,37-42
Pauw, H. & Meulemans, C.C.E. (1987) Occupational exposure to artificial UV-A radiation within
the Philips factories in the Netherlands. In: Passchier, W.F. & Bosnjakovic, B.F.M. , eds, Human
Exposure to Ultraviolet Radiation: Risks and Regulations, Amsterdam, Elsevier, pp. 265-268
Peak, M.J. & Peak, J.G. (1982) Single-strand breaks induced in Bacillus subtilis DNA by ultraviolet
light: action spectmm and properties. Photochem. Photobiol., 35, 675-680
Peak, J.G. & Peak, M.J. (1990) Ultraviolet light induces double-strand breaks in DNA of cultured
human P3 cells as measured by neutral filter elusion. Photochem. Photobiol. , 52,387-393
Peak, J.G. & Peak, M.J. (1991) Comparison of initial yields of DNA-to-protein crosslinks and
singlestrand breaks induced in cultured human cells by far- and near-ultraviolet light, blue light,
and x-rays. Mutat. Res., 246, 187-191
Peak, M.J., Peak, J.G., Moehring, M.P. & Webb, R.B. (1984) Ultraviolet action spectra for DNA
dimer induction, lethality, and mutagenesis in Escherichia coli with emphasis on the UVB
region. Photochem. Photobiol., 40, 613-620
Peak, J.G., Peak, M.J., Sikorski, R.S. & Jones, C.A. (1985) Induction of DNA-protein crosslinks in
human cells by ultraviolet and visible radiations: action spectrum. Photochem. Photobiol. , 41,
295-302
Peak, M.J. , Peak, J.G. & Carnes, B.A. (1987) Induction of direct and indirect single-strand breaks in
hun1an cell DNA by far- and near-ultraviolet radiations: action spectmm and mechanisms.
Peak, J.G., Woloschak, G.E. & Peak, M.J. (199la) Enhanced expression of protein kinase C gene
caused by solar radiation. Photochem. Photobiol., 53, 395-397
Peak, J.G., Pilas, B., Dudek, E.J. & Peak, M.J. (199lb) DNA breaks caused by monochromatic 365
nm ultraviolet-A radiation or hydrogen peroxide and their repair in human epithelioid and xero-
derma pigmentosum cells. Photochem. Photobiol., 54, 197-203
Pearse, A.D., Gaskell, S.A. & Marks, R. (1987) Epidermal changes in human skin following irra-
diation with either UVB or UV A. J. invest. Dermatol., 88, 83-87
Penn, I. (1980) Immunosuppression and skin cancer. Clin. plast. Surg., 1, 361-368
REFERENCES 265
Pepino, P., Hardy, M.A., Chabot, J.A., Berger, C., Marboe, C. & Wasfie, T. (1989) UVB irradiated
allogeneic bone marrow transplantation in rats: prevention of graft versus host disease without
immunosuppression. Transpl. Proc., 21, 2995-2996
Perna, J.J. , Mannix, M.L., Rooney, J.F., Notkins, A.L. & Straus, S.E. (1987) Reactivation oflatent
herpes simplex virus infection by ultraviolet light: a human model. J. Am. A cad. Dermatol, 17,
473-478
Perry, L.L. & Greene, M.I. ( 1982) Antigen presentation by epidermal Langerhans cells: loss of
function following ultraviolet (UV) irradiation in vivo. Clin. Immunol. Immunopathol, 24,
204-219
Perticone, P. Hellgren, D. & Lambert, B. (1986) Effects ofUV-irradiation on the SCE frequency in
human lymphocyte cultures. Mutat. Res., 175, 83-89
Peto, R., Pike, M.C., Day, N.E., Gray, R.G., Lee, P.N., Parish, S., Peto, J., Richards, S. &
Wahrendorf, J. (1980) Guidelines for simple sensitive significance tests for carcinogenic effects
in long-term animal experiments. In: !ARC Monographs on the Evaluation of the Carcinogenic
Risk of Chemicals to Humans, Suppl. 2, Long-term and Short-term Screening Assays for
Carcinogens: A Critical Appraisal, Lyon, pp. 311-26
Phelps, R.G., Bernstein, L.E., Harpaz, N., Gordon, R.E., Cmickshank, F.A. & Schwartz, E. (1989)
Characterization of a dermal derived malignant mesenchymal tumor arising in ultraviolet irra-
diated mice. Am. J. Pathol., 135, 149-159
Phillips, R. (1983) Sources and Applications of Ultraviolet Radiation, London, Academic Press
Pierceall, W.E. & Ananthaswamy, R.N. (1991) Transformation of NIH 3T3 cells by transfection
with UV-irradiated human c-Ha-ras-1 proto-oncogene DNA. Oncogene, 6, 2085-2091
Pierceall, W.E., Mukhopadhyay, I, Goldberg, L.H. & Ananthaswamy, H.N. (1991) Mutations in the
p53 tumor suppressor gene in human cutaneous squamous cell carcinomas. Mol. Carcinog., 4,
445-449
Pitcher, H.M. & Longstreth, J.D. (1991) Melanoma mortality and exposure to ultraviolet radiation:
an empirical relationship. Environ. ins., 17, 7-21
Pitts, D.G., Cullen, A.P. & Hacker, P.D. (1977) Ocular effects of ultraviolet radiation from 295 to
365 nm. Invest. Ophthalmol. visual Sci. , 16, 932-939
Plummer, N.A., Greaves, M.W., Hensby, C.N. & Black, A.K. (1977) Inflammation in human skin
induced by ultraviolet radiation. Postgrad. med. J., 53, 656-657
Poulsen, J.T, Staberg, B., Wulf, H. C. & Brodthagen, H. ( 1984) Dermal elastosis in hairless mice
after UV-B and UV-A applied simultaneously, separately or sequentially. Br. J. Dermatol.,
110, 531-538
Pound, A.W. (1970) Induced cell proliferation and the initiation of skin tumour formation in mice
by ultraviolet light. Pathology, 2, 269-275
Putschar, W. & Holtz, F. (1930) Production of skin tumours in rats by long duration UV irradiation
(Ger.). Z. Krebsforsch., 33, 219-260
Putvinsky, A.V., Sokolov, A.l., Roshchupkin, D.I. & Vladimirov, Y.A. (1979) Electric breakdown
of bilayer phospholipid membranes under ultraviolet irradiation-induced lipid peroxidation.
FEBS Lett., 106, 53-55
Rasanen, L., Reunala, T., Lehto, M., Jansen, C., Rantala, I. & Leinikki, P. (1989) Immediate
decrease in antigen-presenting function and delayed enhancement of interleukin-T production in
human epidermal cells after in vivo UVB irradiation. Br. J. Dermatol, 120, 589-596
Rasmussen, R.E., Hammer-Wilson, M. & Berns, M.W. (1989) Mutation and sister chromatic ex-
change induction in Chinese hamster ovary (CHO) cells by pulsed excimer laser radiation at 193
nm and 308 nm and continuous UV radiation at 254 nm. Photochem. Photobiol, 49, 413-418
Ray-Keil, L. & Chandler, J.W. (1986) Reduction in the incidence of rejection of heterotopic murine
corneal transplants by pretreatment with ultraviolet radiation. Transplantation, 42, 403-406
Reddy, A.L., Fialkow, P.J. & Salo, A. (1978) Ultraviolet radiation-induced chromosomal abnorma-
lities in fetal fibroblasts from New Zealand black mice. Science, 201, 920-922
Reeve, V.E., Greenoak, G.E., Gallagher, C.H., Canfield, P.J. & Wilkinson, F.J. (1985) Effect of
immunosuppressive agents and sunscreens on UV carcinogenesis in the hairless mouse. Aust. J.
exp. Biol. med. Sci., 63, 655-665
Resnick, M.A., Westmoreland, J. , Amaya, E. & Bloom, K (1987) UV-Induced damage and repair in
centromere DNA ofyeast. Mol. gen. Gener., 210, 16-22
Rigel , D.S. , Friedman, R.J., Levenstein. M.J. & Greenwald, D.I. (1983) Relationship of fluorescent
lights to malignant melanoma: another view. J. Dermatol. surg. Oneal., 9, 836-838
Ringvold, A. (1980) Cornea and ultraviolet radiation. Acta ophthalmol., 58, 63-68
Rivers, J.K., Norris, P.G., Murphy, G.M., Chu, A.C., Midgley, G., Morris. J., Morris, R.W., Young,
A.R. & Hawk, J.L.M. (1989) UVA sunbeds: tanning, photoprotection, acute adverse effects and
immunological changes. Br. J. Dermatol., 120, 767-777
Roberts, L.K. & Daynes, R.A. (1980) Modification of the immunogenic properties of chemically
induced tumors arising in hosts treated concomitantly with ultraviolet light. J. Immunol. , 125,
438-447
Roberts, L.K., Samlowski, W.E. & Daynes, R.A. (1986) Immunological consequences of ultraviolet
radiation exposure. Photodermatology, 3, 284-298
Robertson, D.F. (1972) Solar Ultraviolet Radiation in Relation to Human Sunburn and Skin
Cancer, PhD Thesis, University of Queensland, Australia
Robertson, B., Gahring, L., Newton, R. & Daynes, R. (1987) In vivo administration ofinterleukin 1
to normal mice depresses their capacity to elicit contact hypersensitivity responses:
prostaglandins are involved in this modification of immune function. J. invest. Dermatol, 88,
380-387
Robinson, J. K. (1987) Risk of developing another basal cell carcinoma. A 5-year prospective study.
Cancer, 60, 118-120
Roffo, A. H. (1934) Cancer and the sun: carcinomas and sarcomas caused by the action of the sun in
toto (Fr.). Bull. Assoc.fr. Etude Cancer, 23,590-616
Roffo, A.H. (1939) Physico-chemical etiology of cancer (with special emphasis on the association
with solar radiation) (Ger.). Strahlentherapie, 66, 328-350
Ronai, Z.A. & Wei nstein, LB. (1988) Identifi cation of a UV-induced trans-acting protein that
stimulates polyomavirus DNA replication. J. Viral., 62, 1057-1060
Ronai, Z.A., Lambert, M.E., Johnson, M.D., Okin, E. & Weinstein, LB. (1987) Induction of
asynchronous replication of polyoma DNA in rat cells by ultraviolet irradiation and the effects
of various inhibitors. Cancer Res., 47, 4565-4570
Ronai, Z.A., Lambert, M.E. & Weinstein, LB. (1990) Inducible cellular responses to ultraviolet
light irradiation and other mediators of DNA damage in mammalian cells. Cell Biol. Toxicol., 6,
105-126
Rooney, J.F., Bryson, Y., Mannix, M.L., Dillon, M., Wohlenberg, C.R., Banks, S., Wallington, C.J.,
Notkins, A. L. & Straus, S.E. ( 1991) Prevention of ultraviolet-light-induced herpes labialis by
sunscreen. Lancet, 338, 1419-1422
REFERENCES 267
Rosario, R., Mark, G.J., Parrish, J.A. & Mihm, M.C., Jr (1979) Histological changes produced in
skin by equally erythemogenic doses of UV A, UVB, UVC with psoralens. Br. J. Dermatol.,
101, 299-308
Rosdahl, I.K (1979) Local and systemic effects on the epidermal melanocyte population in UV-
irradiated mouse skin. J. invest. Dermatol., 73, 306-309
Rosenstein, B.S. (1988) The induction of DNA strand breaks in normal human skin fibroblasts
exposed to solar ultraviolet radiation. Radial. Res., 116, 313-319
Rosenstein, B.S. & Ducore, J.M. (1983) Induction of DNA strand breaks in normal human
fibroblasts exposed to monochromatic ultraviolet and visible wavelengths in the 240-546 nm
range. Photochem. Photobiol., 38, 51-55
Rosenstein, B.S. & Mitchell, D.L. (1987) Action spectra for the induction of pyrimidine (6-4)pyri-
midone photoproducts and cyclobutane pyrimidine dimers in normal human skin fibroblasts.
Photochem. Photobiol, 45, 775-780
Rosenstein, B.S. & Mitchell, D.L. (1991) The repair of DNA damages induced in normal human
skin fibroblasts exposed to simulated sunlight. Rudiat. Res., 126, 338-342
Rosenstein, B.S. & Rosenstein, R.B. ( 1985) Induction of chromosome aberrations in ICR 2A frog
cells exposed to 265-313 nm monochromatic ultraviolet wavelengths and photoreactivating
light. Photochem. Photobiol. , 41, 57-61
Rosenstein, B.S. , Murphy, J.T. & Ducore, J.M. (1985) Use of a highly sensitive assay to analyze the
excision repair of dimer and non-dimer DNA damages induced in human skin fibroblasts by 254
nm and solar ultraviolet radiation. Cancer Res., 45, 5526-5531
Rosenstein, B.S., Lai, L.-W., Ducore, J.M. & Rosenstein, R.B. (1989) DNA-protein crosslinking
with nonnal and solar UV -sensitive ICR 2A frog cell lines exposed to solar UV -radiation.
Mutat. Res., 217, 219-226
Rosenthal, F.S., Phoon, C., Bakalian, A.E. & Taylor, H.R. (1988) The ocular dose of ultraviolet
radiation to outdoor workers. Invest. Ophthalmol. visual Sci., 29, 649-656
Rosenthal, F.S., Lew, R.A., Rouleau, L.J. & Thomson, M. (1990) Ultraviolet exposure to children
from sunlight: a study using personal dosimetry. Photodermatol. Photoimmunol. Photomed., 7,
77-81
Rosenthal, F.S., West, S.K., Munoz, B., Emmett, E.A., Strickland, R.T. & Taylor, H.R. (1991)
Ocular and facial skin exposure to ultraviolet radiation in sunlight: a personal exposure model
with application to a worker population. Health Phys., 61, 77-86
Roser, M., Bohm, A., Oldigs, M., Weichenthal, M., Reimers, U., Schmidt-Preuss, U., Breitbart,
E.W. & Rudiger, H.W. (1989) Ultraviolet-induced formation of micronuclei and sister
chromatic exchange in cultured fibroblasts of patients with cutaneous malignant melanoma.
Cancer Genet. Cytogenet., 41, I 29-137
Roshchupkin, D.I., Pelenitsyn, A.B., Potapenko, A.Y., Talitsky, V.V. & Vladimirov, Y.A. (1975)
Study of the effects of ultraviolet light on biomembranes-IV. The effect of oxygen on
UV -induced hemolysis and lipid photoperoxidation in rat erythrocytes and liposomes.
Ross, J.A. , Howie, S.E.M., Norval, M., Maingay, J. & Simpson, T.J. (1986) Ultraviolet-irradiated
urocanic acid suppresses delayed-type hypersensitivity to herpes simplex virus in mice. J.
Ross, J.A., Howie, S.E.M., Norval, M. & Maingay, J. (1987) Two phenotypically distinct T cell s
are involved in ultraviolet-irradiated urocanic acid-induced suppression of the efferent
delayed-type hypersensitivity response to herpes simplex virus, type 1 in vivo. J. invest
Dermatol., 89, 230-233
Roth, D.E., Hodge, S.J. & Callen, J.P. (1989) Possible ultraviolet A-induced lentigines: a side effect
of chronic tanning salon usage. J. Am. Acad. Dermatol. , 20, 950-954
Roza, L., van der Wulp, K.J.M. , MacFarlane, S.J., Lohman, P.H.M. & Baan, R.A. (1988) Detection
of cyclobutane thymine dimers in DNA of human cells with monoclonal antibodies raised
against a thymine dimer-containing tetranucleotide. Photocherm. Photobiol, 48, 627-633
Ruiz-Rubio, M., Woodgate, R., Bridges, B.A., Herrera, G. & Blanco, M. (1986) New role for
photoreversible pyrimidine dimers in induction of prototrophic mutations in excision-deficient
Escherichia coli by ultraviolet light. J. Bactenol., 166, 1141-1143
Rusch, H.P ., Kline, B.E. & Baumann, C.A. ( 1941) Carcinogenesis by ultraviolet rays with reference
to wavelength and energy. Arch. Pathol., 31, 135-146
Russell, W.O., Wynne, E.S., Loquvam, G.S. & Mehl, D.A. (1956) Studies on bovine ocular
squamous carcinoma ('cancer eye'). I. Pathological anatomy and historical review. Cancer, 9,
1-52
Saftlas, A.F. , Blair, A., Cantor, K.P. , Hanrahan, L. & Anderson, H.A. (1987) Cancer and other
causes of death among Wisconsin farmers. Am. J. ind. Med., 11, 119-129
Sagher, D. & Strauss, B. (1983) Insertion of nucleotides opposite apurinic/apyrimidinic sites in
deoxyribonucleic acid during in vitro synthesis: uniqueness of adenine nucleotides.
Biochemistry, 22, 4518-4526
Sammartano, L.J. & Tuveson, R.W. (1985) Hydrogen peroxide induced resistance to
broad-spectrum near-ultraviolet light (300 400 nm) inactivation in Escherichia coli. Photochem.
Photobiol. , 41, 367-370
Sammartano, L.J. & Tuveson, R.W. (1987) Escherichia cold strains carrying the cloned cytochrome
d tenninal oxidase complex are sensitive to near-UV inactivation. J. Bactenol. , 169, 5304-5307
Sancar, A. & Rupp, W.D. (1983) A novel repair enzyme: UVRABC excision nuclease of
Escherichia coli cuts a DNA strand on both sides of the damaged region. Cell, 33, 249-260
Sanderson, B.J.S., Dempsey, J.L. & Morley, A.A. (1984) Mutations in human lymphocytes: effects
ofx- and UV-irradiation. Mutat. Res. , 140, 223-227
Sanford, K.K., Parshad, R., Jones, G. , Randleman, S., Garrison, C. & Price, F. (1979) Role of
photosensitization and oxygen in chromosome stability and 'spontaneous' mal ignant
transformation in culture. J. nat! Cancer lnst., 63, 1245-1255
Santamaria, L., Giordano, G.G., Alfisi, M. & Cascione, F. (19GG) Effects of light on
3,4-benzpyrene carcinogenesis. Nature, 210, 824-825
Santamaria, L. , Bianch, A., Arnaboldi, A., Daffara, P. & Andreoni , L. ( 1985) Photocarcinogenesis
by methoxypsoralen, neutral red, proflavine, and long UV radiation. CancerDetect. Prev. , 8,
447-454
Sauder, D.N., Tamaki, K., Meshell, A.N., Fujiwara, H. & Katz, S.l. (1981) Induction of tolerance to
topically applied TNCB using TNP-conjugated ultraviolet light-irradiated epidermal cells. J.
l mmunol., 127, 261-265
Sauder, D.N., Noonan, F.P., DeFabo, E.C. & Katz, S.l. (1983) Ultraviolet radiation inhibits
alloantigen presentation by epidermal cells: partial reversal by the soluble epidermal cell
product, epidermal cell-derived thymocyte-activating factor (ETAF). J. invest. Dermatol., 80,
485-489
Sayre, R.M. & Kligman, L.H. (1992) Discrepancies in the measurement of spectral sources. Photo-
chem. Photobiol. , 55, 141-143
Schaaper, R.M., Kunkel, T.A. & Loeb, L.A. (1983) Infidelity of DNA synthesis associated with
bypass of apurinic sites. Proc. natl A cad. Sci. USA, 80, 487-491
REFERENCES 269
Schaaper, R. M., Dunn, R. L. & Glickman, B.W. (1987) Mechanisms of ultraviolet-induced
mutation. Mutational spectra in the Escherichia coli lacl gene for a wild-type and an
excision-repair deficient strain. J. mol. Bioi., 198, 187-202
Schacter, B., Lederman, M.M., LeVine, M.J. & Ellner, J.J. (1983) Ultraviolet radiation inhibits
human natural killer activity and lymphocyte proliferation.]. Immunol., 130, 2484-2487
Scheibner, A., McCarthy, W.H., Milton, G.W. & Nordlund, J.J. (1983) Langerhans cell and
melanocyte distribution in 'nonnal' human epidermis. Aust. J. Dermatol. , 24,9-16
Scheibner, A., Hollis, D.E., McCarthy, W.H. & Milton, G.W. (1986a) Effects of exposure to
ultraviolet light in a commercial solarium on Langerhans cells and melanocytes in human
epidennis. Aust. J. Dermatol., 21, 35-41
Scheibner, A., Hollis, D.E., McCarthy, W.H. & Milton, G.W. (1986b) Effects of sunlight exposure
on Langerhans cells and melanocytes in human epidennis. Photodermatology, 3, 15-25
Schothorst, A.A., Slaper, H., Telgt, D., Alhadi, B. & Suurmond, D. (1987a) Amounts of ultraviolet
B (UVB) received from sunlight or artificial UV sources by various population groups in the
Netherlands. In: Passchier, W.F. & Bosnjakovic, B.F.M. , eds, Human Exposure to Ultraviolet
Radiation: Risks and Regulations, Amsterdam, Elsevier, pp. 269-273
Schothorst, A.A., Enninga, E.C. & Simons, J.W.I.M. (1987b) Mutagenic effects per erythemal dose
of rutificial and natural sources of ultraviolet light. In: Passchier. W.F. & Bosnjakovic, B.E.M.,
eds, Human Exposure to Ultraviolet Radiation: Risks and Regulutions, Amsterdam, Elsevier,
pp. 103-107
Schothorst, A.A., Evers, L.M., Noz, K.C., Pilon, R. & van Zeeland, A.A. (19'31) Pyrimidine dimer
induction and repair in cultured human skin keratinocytes or melanocytes after irradiation with
monochromatic ultraviolet radiation. J. invest. Dermatol., 96, 916-920
van der Schroeff, J.G., Evers, L.M., Boot, A.J.M. & Bos, J.L. (1990) ras Oncogene mutations in
basal cell carcinomas and squamous cell carcinomas of human skin. J. invest. Dermatol., 94,
423-425
Schulze, R. (1962) On the climate of the Earth (Ger.). Strahlentherapie, II9, 321-348
Schulze, R. (1970) Global radiation climate (Ger.). Wiss. Forschungsber, 12, 1-220
Schulze, R. & Grafe, K. (1969) Consideration of sky ultraviolet radiation in the measurement of
solar ultraviolet radiation. In: Urbach, F., ed., The Biological Effects of Ultraviolet Radiation
(Proceedings of the First International Conference), Oxford, Pergamon Press, pp. 359-373
Schwartz, S.M. & Weiss, N.S. (1988) Place of birth and incidence of ocular melanoma in the
United States. Int.]. Cancer, 41, 174-177
Schwarz, T., Urbanska, A., Gschnait, E & Luger, T.A. ( 1986) Inhibition of the induction of contact
hypersensitivity by a UV -mediated epidennal cytokine. J. invest. Dermatol. , 87, 289-291
Schwarz, T., Urbanski, A., Kirnbauer, R., Kock, A., Gschnait, F. & Luger, TA. (1988) Detection of
a specific inhibitor of interleukin 1 in sera of UVB-treated mice. J. invest. Dermatol., 91,
536-540
Scotto, J. & Fears, TR. (1987) The association of solar ultraviolet and skin melanoma incidence
among Caucasians in the United States. Cancer Invest., 5, 275-283
Scotto, J., Kopf, A.W. & Urbach, F. (1974) Non-melanoma skin cancer among Caucasians in four
areas ofthe United States. Cancer, 34, 1333-1338
Scotto, J. , Fears, T. & Gori, G. (1976) Measurements of Ultraviolet Radiations in the United States
and Comparisons with Skin Cancer Data (NIH Publ. No. 76-1029). Washington DC, US
Department of Health, Education, and Welfare
Scotto, J., Fears, T.R. & Fraumeni, J.F., Jr (1982) Solar radiation. In: Schottenfeld, D. & Fraumeni,
J.F., Jr, eds, Cancer Epidemiology and Prevention, Philadelphia, W.B. Saunders, pp. 254-276
Scotto, J., Fears, T.R. & Fraumeni, J.F., Jr (1983) Incidence of Nonmelanoma Skin Cancer in the
United States (NIH Publ. No. 83-2433), Bethesda, MD, National Cancer Institute
Seddon, J.M., Gragoudas, E.S., Glynn, R.J., Egan, K.M., Albert, D.M. & Blitzer, P.H. (1990) Host
factors, UV radiation and risk of uveal melanoma. A case-control study. Arch. Ophthalmol.,
108, 1274-1280
Seetharam, S., Protic-Sabljic. M., Seidman, M.M. & Kraemer, K.H. (1987) Abnormal ultraviolet
mutagenic spectrum in plasmid DNA replicated in cultured fibroblasts from a patient with the
skin cancer-prone disease, xeroderma pigmentosum. J. clin. Invest., 80, 1613-1617
Seguin, L.R., Tarone, R.E., Liao, K. & Robbins, J.H. (1988) Ultraviolet light-induced chromosomal
aberrations in cultured cells from Cockayne syndrome and complementation group C xeroderma
pigmentosum patients: lack of correlation with cancer susceptibility. Am. J. hum. Genet., 42,
468-475
Sekiya, T., Fushimi, M., Hori, H., Hirohashi, S., Nishimura, S. & Sugimura, T. (1984) Molecular
cloning and the total nucleotide sequence of the human c-Ha-ras-1 gene activated in a
melanoma from a Japanese patient. Proc. natl Acad. Sci. USA, 81, 4771-4775
Servilla, K.S., Bumham, D.K. & Daynes, R.A. (1987) Ability of cyclosporine to promote the
growth of transplanted ultraviolet radiation-induced tumors in mice. Transplantation, 44,
291-295
Setlow, R.B. (1968) Photoproducts in DNA irradiated in vivo. Photochem. Photobiol., 7, 643-649
Setlow, R.B. (1975) The relevance of photobiological repair. An. Acad. Bras. Cienc., 45 (Suppl.
215), 215-220
Setlow, R.B. & Carrier, W.L. (1966) Pyrimidine dimers in ultraviolet-irradiated DNA's. J. mol.
Bioi., 17, 237-254
Setlow, R.B., Swenson, P.A. & Carrier, W.L. (1963) Thymine dimers and inhibition of DNA
synthesis by ultraviolet irradiation of cells. Science, 142, 1464-1466
Setlow, R.B., Woodhead, A.D. & Grist, E. (1989) Animal model for ultraviolet radiation-induced
melanoma: platyfish-swordtail hybrid. Proc. natl Acad. Sci. USA, 86, 8922-8926
Shabad. L.M. & Litvinova, S.N. (1972) The combined effect ofUV irradiation and 7,12-DMBA on
the skin of hairless mice (Russ.). Biull. eksp. Bioi. Med., 73, 84-85
Shanmugaratnam, K., Lee, H.R & Day, N.E., eds (1983) Cancer Incidence in Singapore 1968-1977
(IARC Scientific Publications No. 47), Lyon, IARC, p. 73
Shetlar, M.D. (1980) Cross-linking of proteins to nucleic acids by ultraviolet light. Photochem.
Photobiol. Rev., 5, I 05-197
Shukla, V.K., Hughes, D.C., Hughes, L.E., McCormick, F. & Padua, R.A. (1989) ras Mutations in
human melanotic lesions: K-ras activation is a frequent and early event in melanoma deve-
lopment. Oncogene Res. , 5, 121-127
Siede, W. & Eckardt, F. (1986) Analysis of mutagenic DNA repair in a thermoconditional mutant of
Saccharomyces cerevisiae. III. Dose-response pattern of mutation induction in UV -irradiated
rev2's cells. Mol. gen. Genet., 202, 68-74
Sierniatycki, J. (1991) Risk Factors for Cancer in the Workplace, Boca Raton, FL, CRC Press
Silverstone, H. & Gordon, D. (1966) Regional studies .in skin cancer. Second report: wet tropical
and subtropical coasts of Queensland. Med. J. Aust., 2, 733-740
Silverstone, H. & Searle, J.H.A. (1970) The epidemiology of skin cancer in Queensland: the
influence of phenotype and environment. Br. J. Cancer, 24, 235-252
Singh, B. & Gupta, R.S. (1982) Mutagenic response to ultraviolet light and X-rays at five
independent genetic loci in Chinese hamster ovary cells. Environ. Mutag., 4, 543-551
Sisson, T.R.C. & Vogl, T.P. (1982) Phototherapy ofhyperbilirubinemia. In: Regan, J.D. & Parrish,
J.A., eds, The Science of Photomedicine, New York, Plenum, pp. 477-509
REFERENCES 271
Sjovall, P. & Christensen, O.B. (1986) Local and systemic effect of ultraviolet irradiation (UVB and
UVA) on human allergic contact dermatitis. Acta dermatol. venereol., 66, 290-294
Slaga, T.J., Klein-Szanto, A.J.P., Triplett, L.L., Yotti, L.P. & Trosko, J.E. (1981) Skin
tumor-promoting activity of benzoyl peroxide, a widely used free radical-generating compound.
Science, 213, 1023-1025
Slaper, H. (1987) Skin Cancer and UV Exposure: Investigations on the Estimation of Risks, PhD
Thesis, University of Utrecht
Sliney, D.H. (1986) Physical factors in cataractogenesis: ambient ultraviolet radiation and tempe-
rature. Invest. Ophthalmol. visual Sci., 27, 781-790
Sliney, D.H. (1990) The IRP A/INIRC guidelines or limits of exposure to laser radiation. In:
Grandolfo, M., Rindi, S. & Sliney, D.H., eds, Light, Lasers, and Synchrotron Radiation. A
Health Risk Assessment, New York, Plenum, pp. 341-346
Sliney, D.H. & Wolbarsht, M. (1980) Safety with Lasers and Other Optical Sources: A
Comprehensive Handbook, New York, Plenum
Sliney, D., Wood, R.L., Jr, Moscato, P.M. , Marshall, W.J. & Eriksen, P. (1990) Ultraviolet
exposure in the outdoor environment: measurement of ambient ultraviolet exposure levels at
large zenith angles. In: Grandolfo, M., Rindi, S. & Sliney, D.H., eds, Light, Lasers, and
Synchrotron Radiation. A Health Risk Assessment, New York, Plenum, pp. 169-180
Smith, K.C. (1964) The photochemistry of thymine and bromouracil in vivo. Photochem.
Photobiol., 3, 1-10
Smith, K.C. (1976) The radiation-induced addition of proteins and other molecules to nucleic acids.
In: Wang, S.Y., ed., Photochemistry and Photobiology of Nucleic Acids, Vol. 2, New York,
Academic Press, pp. 187-218
Smith, K.C., ed. (1989) The Science of Photobiology, 2nd ed., New York, Plenum, pp. 47-53
Sorahan, T. & Grimley, R.P (1985) The aetiological significance of sunlight and fluorescent
lighting in malignant melanoma: a case-control study. Br. J. Cancer, 52, 765-769
Spangrude, G.J., Bernhard, E.J., Ajioka, R.S. & Daynes, R.A. (1983) Alterations in lymphocyte
homing patterns within mice exposed to ultraviolet radiation. J. immunol., 130, 2974-2981
Spellman, C.W. & Daynes, R.A. (1978) Properties of ultraviolet light-induced suppressor
lymphocytes within a syngeneic tumor system. Cell. lmmunol., 36, 383-387
Spellman, C.W., Woodward, J.G. & Daynes, R.A. (1977) Modification of immunological potential
by ultraviolet radiation. I. Immune status of short-term UV -irradiated mice. Transplantation, 24,
112-119
Spikes, J.D., Kripke, M.L. , Connor, R.J. & Eichwald, E.J. (1977) Time of appearance and histology
of tumors induced in the dorsal skin of C3Hf mice by ultraviolet radiation from a mercury arc
lamp. J. nat! Cancer Inst., 59, 1637-1643
Spitzer, W.O., Hill, G.B., Chambers, L.W., Helliwell, B.E. & Murphy H.B (1975) The occupation
of fishing as a risk factor in cancer of the lip. New Engl. J. Med. , 293, 419-424
Spruance, S.L. (1985) Pathogenesis of herpes simplex labialis: experimental induction of lesions
with UV light. J. clin. Microbiol., 22, 366-368
Staberg, B., Wulf, H.C., Poulsen, T., Klemp, P. & Brodthagen, H. (1983a) Carcinogenic effect of
sequential artificial sunlight and UV A irradiation in hairless mice. Consequences for solarium
'therapy'. Arch. Dermatol., 119, 641-643
Staberg, B., Wulf, H.C., Klemp, P. Poulsen, T. & Brodthagen, H. (1983b) The carcinogenic effect
ofUVA radiation. J. invest. Dermatol. , 31,517-519
Stanley, S.K, Folks, T.M. & Fauci, A.S. (1989) Induction of expression of human
immunodeficiency virus in a chronically infected promonocytic cell line by ultraviolet
irradiation. AIDS Res. hum. Retroviruses, 5, 375-384
Stefanini, M., Lagomarsini, P., Arlett, C.F., Marinoni, S., Borrone, C., Crovato, F., Trevisan, G.
Cordone, G. & Nuzo, F. (1986) Xeroderma pigmentosum (complementation group D) mutation
is present in patients affected by trichothiodystrophy with photosensitivity. Hum. Genet., 74,
107-112
Stein, B., Rahmsdorf, H.J., Steffen, A., Litfm, M. & Herrlich, P. (1989) UV-Induced DNA damage
is an intermediate step in UV-induced expression of human immunodeficiency virus type 1,
collagenase, c-fos, and metallothionein. Mol. cell. Bioi. , 9, 5169-5181
Steinitz, R., Parkin, D.M., Young, J.L., Bieber, C.A. & Katz, L., eds (1989) Cancer incidence in
Jewish Migrants to Israel, 1961-1981 (IARC Scientific Publications No. 98), Lyon, IARC, pp.
114-115, 134-135
Stenback, F. (1975a) Species-specific neoplastic progression by ultraviolet light on the skin of rats,
guinea pigs, hamsters and mice. Oncology, 31, 209-225
Stenback, F. ( 1975b) Studies on the modifying effect of ultraviolet radiation on chemical skin
carcinogenesis. J. invest. Dermatol., 64, 253-257
Stenback, F. (1975c) Ultraviolet light irradiation as initiating agent in skin tumor formation by the
two-stage method. Eur. J. Cancer, 11, 241-246
Stenback, F. (1978) Life history and histopathology of ultraviolet light-induced skin tumors. Natl
Cancer lnst. Monogr., SO, 57-70
Stenback, F. & Shubik, P. (1973) Carcinogen-induced skin tumorigenesis in mice: enhancement and
inhibition by ultraviolet light. Z. Krebsforsch., 19, 234-240
Sterenborg, H.J. C.M. & van der Leun, J. C. ( 1987) Action spectra for tumorigenesis by ultraviolet
radiation. In: Passchier, W.E. & Bosnjakovic, B.F.M., eds, Human Exposure to Ultraviolet
Radiation: Risks and Regulations, Amsterdam, Elsevier, pp. 173-191
Sterenborg, H.J.C.M. & van der Leun, J. C. (1988) Change in epidermal transmission due to UV-
induced hyperplasia in hairless mice: a first approximation of the action spectrum. Photo-
dermatology, 5, 71-82
Sterenborg, H.J.C.M. & van der Leun, J.C. (1990) Tumorigenesis by a long wavelength UVA
source. Photochem. Photobiol. , 51, 325-330
Sterenborg, H.J.C.M., de Gruijl, F.R. & van der Leun, J.C. (1986) UV A induced epidermal
hyperplasia in hairless mice. Photodermatology, 3, 206-214
Sterenborg, H.J.C.M., van der Putte, S.C.J. & van der Leun, J.C. (1988) The dose-response relation-
ship of tumorigenesis by ultraviolet radiation of254 nm. Photochem. Photobiol. , 47, 245-253
Sterenborg, H.J.C.M. , de Gruijl, F.R., Kelfkens, G. & van der Leun, J.C. (1991) Evaluation of skin
cancer risk resulting from long term occupational exposure to radiation from ultraviolet lasers in
the range from 190 to 400 nm. Photochem. Photobiol., 54, 775-780
Stierner, U., Rosdahl, 1., Augustsson, A. & Kagedal, B. (1989) UVB irradiation induces melanocyte
increase in both exposed and shielded human skin. J. invest. Dermatol., 92, 561-564
Sting!, L.A., Sauder, D.N., lij ima, M., Wolff, K., Pehamberger, H. & Sting!, G. (1983) Mechanism
of UVB induced impairment of the antigen-presenting capacity of murine epidennal cells. J.
lmmunol., 130, 1586-l591
Streeter, P.R. & Fortner, G.W. (1988a) Immune response to ultraviolet-induced tumors. ll. Effector
cells in tumor immunity. Transplantation, 46, 250-255
REFERENCES 273
Streeter, P.R. & Fortner, G.W. (1988b) Immune response to ultraviolet induced tumors. III.
Analysis of cloned lymphocyte populations exhibiting antitumor activity. Transplantation, 46,
256-260
Streilein, J.W. & Bergstresser, PR. (1981) Langerhans cell function dictates induction of contact
hypersensitivity or unresponsiveness to DNFB in Syrian hamsters. J. invest Dermatol., 77,
272-277
Streilein, J.W., Toews, G.T., Gilliam, J.N. & Bergstresser, P.R. (1980) Tolerance or
hypersensitivity to 2,4-dinitro-1-fluorobenzene: the role of Langerhans cell density within
epidennis. J. invest. Dermatol., 74, 319-322
Strickland, P .T. ( 1982) Tumor induction in Sen car mice in response to ultraviolet radiation.
Strickland, P.T. (1986) Photocarcinogenesis by near-ultraviolet (UVA) radiation in Sencar mice. J.
Strickland, P.T., Burns, F.J. & Albert, R.E. (1979) Induction of skin tumors in the rat by single
exposure to ultraviolet radiation. Photochem. Photobiol., 30, 683-688
Strickland, P.T., Creasia, D. & Kripke, M.L. (1985) Enhancement of two-stage skin carcinogenesis
by exposure of distant skin to UV radiation. J. natl Cancer lnst., 74, 1129-1134
Strickland, P.T., Bruze, M. & Creasey, J. (1988) Cyclobuta-dithymidine induction by
solar-simulating UV radiation m human skin. I. Protection by constitutive pigmentation.
Sundararaman, N., St John, D.E. & Venkateswaran, S.V. (1975) Solar Ultraviolet Radiation
Received at the Earth's Surface Under Clear and Cloudless Conditions, Springfield, VA,
National Technical Information Service
Sutherland, B.M., Harber, L.C. & Kochevar, I.E. ( 1980) Pyrimidine dimer formation and repair in
human skin. Cancer Res. , 40, 3181-3185
Sutherland, B.M., Delihas, N.C., Oliver, R.P. & Sutherland, J.C. (1981) Action spectra for
ultraviolet light-induced transformation of human cells to anchorage-independent growth.
Cancer Res., 41,2211-2214
Sutherland, B.M., Freeman, A.G. & Bennett, P.V. (1988) Human cell transformation in the study of
sunlight-induced cancers in the skin of man. Mutat. Res., 199, 425-436
Suzuki, H., Fukuyama, K. & Epstein, W.L. (1977) Changes in nuclear DNA and RNA during epi-
dermal keratinization. Cell Tissue Res., 184, 155-167
Suzuki, F., Han, A., Lankas, G.R., Utsumi, H. & Elkind, M.M. (1981) Spectral dependencies of
killing, mutation, and transformation in mammalian cells and their relevance to hazards caused
by solar ultraviolet radiation. Cancer Res., 41, 4916-4924
Swerdlow, A.J. (1979) Incidence of malignant melanoma of the skin in England and Wales and its
relationship to sunshine. Br. med. J., ii, 1324-1327
Swerdlow, A.J. (1983a) Epidemiology of melanoma of the eye in the Oxford region. 1952-78. Br J.
Cancer, 47, 311-313
Swerdlow, A.J. (1983b) Epidemiology of eye cancer in adults in England and Wales, 1962-77. Am.
J. Epidemiol., 118,294-300
Swerdlow, A.J. (1985) Seasonality of presentation of cutaneous melanoma, squamous cell cancer
and basal cell cancer in the Oxford region. Br J. Cancer, 52, 893-900
Swerdlow, A.J., English, J.S.C., MacKie, R.M., O'Doherty, C.J., Hunter, J.A.A., Clark, J. & Hole,
D.J. (1988) Fluorescent lights, ultraviolet lamps, and risk of cutaneous melanoma. Br. med. J. ,
297, 647-650
Sylvania (undated) Germicidal and Short-wave Ultraviolet Radiation, Danvers, MA, GTE Products
Corporation
Szabo, G., Gerald, A.B., Pathak, M.A. & Fitzpatrick, TB. ( 1972) The ultrastructure of racial color
differences in man. In: Riley, V., ed., Pigmentation. Its Genesis and Biologic Control, New
York, Appleton-Century-Crofts, pp. 23-41
Szebeni, J. & Tollin, G. (1988) Some relationships between ultraviolet light and
heme-protein-indueed peroxidative lipid breakdown in liposomes, as reflected by fluorescence
changes: the effect of negative surface charge. Photochem. Photobiol., 47, 475-479
Takebe, H., Nishigori, C. & Satoh, Y. (1987) Genetics and skin cancer ofxerodenna pigmentosum
in Japan. Jpn. J. Cancer Res. (Gann), 78, 1135-1 143
Talve, L., F. & Jansen, C.T. (1990) UV A irradiation increases the incidence of epithelial
tumors in UVB-irradiated hairless mice. Photodermatol. Photoimmunol Photomed., 7, 109-115
Tamaki, K. & Iijima, M. (1989) The effect of ultraviolet B irradiation on delayed-type
hypersensitivity, cytotoxic T lymphocyte activity, and skin graft rejection. Transplantation, 47,
372-376
Tan, E.M. & Stoughton, R.B. (1969) Ultraviolet light induced damage to deoxyribonucleic acid in
human skin. J. invest. Dermatol., 52, 537-542
Taylor, H.R. , West, S.K., Rosenthal, F.S., Munoz, B., Newland, H.S., Abbey, H. & Emmett, E.A.
(1988) Effect of ultraviolet radiation on cataract formation. New Engl. J. Med., 319, 1429-1433
Taylor, H.R. , West, S.K., Rosenthal, F.S., Munoz, B. , Newland, H.S. & Emmett, E.A. (1989)
Corneal changes associated with chronic UV irradiation. Arch. Ophthalmol. , 107, 1481-1484
Teppo, L., Pakkanen, M. & Hak:ulinen, T. (1978) Sunlight as a risk factor of malignant melanoma
of the skin. Cancer, 41, 2018-2027
Teppo, L., Pukkala, E., Hakama, M., Hakulinen, T. Herva, A. & Saxen, E. (1980) Way of life and
cancer incidence in Finland. A municipality-based ecological analysis. Scand. J. soc. Med.,
Suppl. 19, 5-84
Tessman, I. & Kennedy, M.A. (1991) The two-step model of UV mutagenesis reassessed:
deamination of cytosine in cyclobutane dimers as the likely source of the mutations associated
with photoreactivation. Mol. gen. Genet., 227, 144-148
Thiers, B.H., Maize, J.C., Spicer, S.S. & Cantor, A.B. (1984) The effect of aging and chronic sun
exposure on human Langerhans cell population. J. invest. Dermutol. , 82, 223-226
Tjernlund, U. & Juhlin, L. ( 1982) Effect of UV radiation on immunological and histochemical
markers of Langerhans cells in normal appearing skin of psoriatic patients. Acta dermatol., 272,
171-176
Tlsty, T.D., Brown, P.C. & Schimke, R.T. (1984) UV radiation facilitates methotrexate resistance
and amplification of the dihydrofolate reductase gene in cultured 3T6 mouse cells. Mol. cell.
Bioi. , 4, 1050-1056
Toda, K.-1., Miyachi, Y., Nesumi, N., Konishi, J. & Imamura, S. (1986) UVB/PUVA-induced sup-
pression of human natural killer activity is reduced by superoxide dismutase and/or interleukin 2
in vitro. J. invest. Dermatol., 86, 519-522
Toews, G.B., Bergstresser, P.R. & Streilein, J.W. (1980) Epidennal Langerhans cell density
determines whether contact hypersensitivity or unresponsiveness follows skin painting with
DNFB. J. Immunol., 124, 445-453
Tominaga, A., Lefort, S., Mizel, S.B., Dambrauskas, J.T, Granstein, R., Lowy, A., Benacerraf, B. &
Greene, M.l. (1983) Molecular signals in antigen presentation. I. Effects of interleukin 1 and on
radiation-treated antigen-presenting cells in vivo and in vitro. Clin. lmmunol. lmmunopathol, 29,
282-293
REFERENCES 275
Trosko, J.E. & Brewen, J.G. (1967) Inhibition of ultraviolet-induced chromosome breaks by cys-
teamine in 5-bromouracil-substituted mammalian cells. Radiat. Res., 32, 200-213
Tucker, M.A., Boice, J.D., Jr & Hoffman, D.A. (1985a) Second cancer following cutaneous
melanoma and cancers of the brain, thyroid, connective tissue, bone and eye in Connecticut,
1935-82. Nat/ Cancer Inst. Monogr., 68, 161-189
Tucker, M.A., Shields, J.A., Hartge, P., Augsberger, J., Hoover, R.N. & Fraumeni, J.F., Jr (1985b)
Sunlight exposure as risk factor for intraocular malignant melanoma. New Engl. J. Med., 313,
789-792
Turner, B.J., Siatkowski, R.M., Augsburger, J.J., Shields, J.A. , Lustbader, E. & Mastrangelo, M.J.
(1989) Other cancers in uveal melanoma patients and their families. Am. J. Ophthalmol. , 107,
601-608
Tuveson, R.W. & Sammartano, L.J. (1986) Sensitivity of HemA mutant Escherichia coli cells to
inactivation by near-UV light depends on the level of supplementation with 8-aminolevulinic
acid. Photochem. Photobiol. , 43,621-626
Tyrrell, R.M. (1973) Induction of pyrimidine dimers in bacterial DNA by 365 nm radiation.
Tyrrell, R.M. ( 1982) Cell inactivation and mutagenesis by solar ultraviolet radiation. In: Helene, C.,
Charlier, M., Montenay-Garestier, T. & Laustriat, 0., eds, Trends in Photobiology, New York,
Plenum, pp. 155-172
Tyrrell, R.M. (1984) Mutagenic action of monochromatic UV radiation in the solar range on human
cells. Mutat. Res., 129, 103-110
Tyrrell, R.M. (1985) A common pathway for protection of bacteria against damage by solar UVA
(334 nm, 365 nm) and an oxidising agent (H
2
0
2
) . Mutat. Res., 145, 129-136
Tyrrell, R.M. ( 1991) UV A (320-280 nm) radiation as an oxidative stress. In: Sir, H., ed., Oxidative
Stress: Oxidants and Antioxidants, London, Academic Press, pp. 57-83
Tyrrell, R.M. & Pidoux, M. (1986) Endogenous glutathione protects human skin fibroblasts against
the cytotoxic action of UVB, UV A and near-visible radiations. Photochem. Photobiol. , 44,
561-564
Tyrrell, R.M. & Pidoux, M. (1988) Correlation between endogenous glutathione content and
sensitivity of cultured human skin cells to radiation at defined wavelengths in the solar
ultraviolet range. Photochem. Photobiol., 47, 405-412
Tyrrell, R.M. & Pidoux, M. (1989) Singlet oxygen involvement in the inactivation of cultured
human fibroblasts by UV A (334 nm, 365 nm) and near visible (405 nm) radiations. Photochem.
Photobiol., 49, 407-412
Tyrrell, R.M., Ley, R.D. & Webb, R.B. (1974) Induction of single-strand breaks (alkali-labile
bonds) in bacterial and phage DNA by near UV (365 nm) radiation. Photochem. Photobiol. , 20,
395-398
Ullrich, S.E. (1985) Suppression of lymphoproliferation by hapten-specific suppressor T
lymphocytes from mice exposed to ultraviolet radiation. Immunology, 54, 343-352
Ullrich, S.E. (1986) Suppression of the immune response to allogeneic histocompatibility antigens
by a single exposure to ultraviolet radiation. Transplantation, 42, 287-291
Ullrich, S.E. (1987) The effect of ultraviolet radiation-induced suppressor cells on T -cell activity.
Ullrich, S.E. & Kripke, M.L. (1984) Mechanisms in the suppression of tumor rejection produced in
mice by repeated UV radiation. J. lmmunol., 133, 2786-2790
Ullrich, S.E., Yee, G.K. & Kripke, M.L. (1986a) Suppressor lymphocytes induced by epicutaneous
sensitization ofUV-irradiated mice control multiple immunological pathways. Immunology, 58,
185-190
Ullrich, S.E., Azizi, E. & Kripke, M.L. (1986b) Suppression of the induction of delayed-type
hypersensitivity reactions in mice by a single exposure to ultraviolet radiation. Photochem.
Photobiol., 43, 633-638
Ultra-Violet Products, Inc. ( 1977) Laboratory Applications for Ultraviolet Equipment, San Gabriel,
CA
United Kingdom Stratospheric Ozone Review Group (1991) Stratospheric Ozone 1991, London,
Her Majesty's Stationery Office
United Nations Enviromnent Programme (1989) Environmental Effects Panel Report, Geneva
Unrau, P., Wheatcroft, R., Cox, B. & Olive, T. (I 973) The formation of pyrimidine dimers in the
DNA of fungi and bacteria. Biochim. biophys. Acta, 312, 626-632
Urbach, F., Davies, R.E. & Forbes, P.D. (1966) Ultraviolet radiation and skin cancer in man. In:
Montagna, W. & Dobson, R.L., eds, Advances in Biology of Skin, Vol. VII, Carcinogenesis,
Oxford, Pergamon Press, pp. 195-214
Urbach, F., Epstein, J.H. & Forbes, P.D. (1974) Ultraviolet carcinogenesis: experimental, global
and genetic aspects. In: Pathak, M.A., Harber, L.C., Seiji, M. & Kukita, A., eds, Sunlight and
Man. Normal and Abnormal Photobiologic Responses, Tokyo, University of Tokyo Press, pp.
259-283
Vagero, D., Ringback, G. & Kiviranta, H. (1986) Melanoma and other tumours of the skin among
office, other indoor and outdoor workers in Sweden 1961-1979. Br J. Cancer, 53, 507-512
Vagero, D., Swerdlow, A.J. & Beral, V. (1990) Occupation and malignant melanoma: a study based
on cancer registration data in England and Wales and in Sweden. Br. J. inc. Med. , 47, 317-324
Valerie, K. & Rosenberg, M. ( 1990) Chromatin structure implicated in activation of HIV -1 gene
expression by ultraviolet light. New Biol., 2, 712-718
Valerie, K., Delers, A., Bruck, C., Thiriatt, C., Rosenberg, H., Debouck, C. & Rosenberg, M.
(1988) Activation of human immunodeficiency virus type 1 by DNA damage in human cells.
Nature, 333, 78-81
Varghese, A.J. & Patrick, M.H. (1969) Cytosine derived heteroadduct formation in ultraviolet-
irradiated DNA. Nature, 223, 299-300
Varghese, A.J. & Wang, S. Y. (1967) Ultraviolet irradiation of DNA in vitro and in vivo produces a
third thymine-derived product. Science, 156, 955-957
van't Veer, L.J., Burgering, B.M.T. , Versteeg, R., Boot, A.J .. M. , Ruiter, D.J. , Osanto, S., Schrier,
P .I. & Bos, J.L. ( 1989) N-ras Mutations in human cutaneous melanoma from sun-exposed body
sites. Mol. cell. Bioi. , 9, 3114-3116
Verma, A.K., Lowe, N.J. & Boutwell, R.K. (1979) Induction of mouse epidem1al ornithine
decarboxylase activity and DNA synthesis by ultraviolet light. Cancer Res., 39, 1035-1040
Vermeer, M., Schmieder, G.J., Yoshikawa, T., van den Berg, J.-W., Metzman, M.S., Taylor, J.R. &
Streilein, J.W. (1991) Effects of ultraviolet B light on cutaneous immune responses of humans
with deeply pigmented skin. J. invest. Dermatol., 97,729-734
Vitaliano, PR (1978) The use of logistic regression for modelling risk factors: with application to
nonmelanoma skin cancer. Am. J. Epidemiol., 108, 402-414
Vitasa, B.C., Taylor, H.R. , Strickland, P.T., Rosenthal, F.S., West, S. , Abbey, H., Ng, S.K, Munoz,
B. & Emmett, E. A. (1990) Association of nonmelanoma skin cancer and actinic keratosis with
cumulative solar ultraviolet exposure in Maryland watermen. Cancer, 65, 2811-2817
REFERENCES 277
Vlcek, D., Podstavkova, S., Miadokova, E., Adams, G.M.W. & Small, G.D. (1987) General charac-
teristics, molecular and genetic analysis of two new UV-sensitive mutants of Chlamydomonas
reinhardtii. Mutat. Res., 183, 169-175
Vogel, J., Cepeda, M., Tschachler, E., Napolitano, L.A. & Jay, G. (1992) UV activation of human
immunodeficiency virus gene expression in transgenic mice. J. Virol., 66, 1-5
Vuillawne, M., Daya-Grosjean, L., Vincens, P. Pennetier, J.L., Tarroux, P. Baret, A., Calvayrac, R. ,
Taieb, A. & Sarasin, A. ( 1992) Striking differences in cellular catalase activity between two
DNA repair-deficient diseases: xeroderma pigmentosum and trichothiodystrophy.
Walker, G.C. (1984) Mutagenesis and inducible responses to deoxyribonucleic acid damage in
Escherichia coli. Microbial. Rev., 48, 60-93
Walter, S.D., Marrett, L.D. , From, L., Hertzman, C., Shannon, H.S. & Roy, P. (1990) The
association of cutaneous malignant melanoma with the use of sunbeds and sunlamps. Am. J.
Epidemiol, 131, 232-243
Walter, S.D., Marrett, L.D., Shannon, H.S., From, L. & Hertzman, C. (1992) The association of
cutaneous malignant melanoma and fluorescent light exposure. Am. J. Epidemiol., 135, 749-762
Wang, S.-Y. & Varghese, A.J. (1967) Cytosine-thymine addition product from DNA irradiated with
ultraviolet light. Biochem. biophys. Res. Commun., 29, 543-549
Waterhouse, J., Muir, C., Correa, P. & Powell, J. , eds (1976) Cancer Incidence in Five Continents,
Vol. Ill (IARC Scientific Publications No. 15). Lyon, IARC
Waterhouse, J. , Muir, C., Shanmugaratnam, K. & Powell, J., eds (1982) Cancer Incidence in Five
Continents, Vol. IV (IARC Scientific Publications No. 42), Lyon. IARC
Weast, R.C. , ed. (1989) CRC Handbook of Chemistry and Physics, 70th ed., Boca Raton, FL, CRC
Press, p. E-209
Webb, R.B. (1977) Lethal and mutagenic effects of near-ultraviolet radiation. Photochem.
Photobiol. Rev., 2, 169-261
Webb, R.B. & Malina, M.M. (1970) Mutagenic effects of near ultraviolet and visible radiant energy
on continuous cultures of Escherichia coli. Photochem. Photobiol. , 12, 457-468
van Weelden, H. & van der Leun, J.C. (1986) Photorecovery by UVA. In: Urbach. F. & Gange.
R.W., eds, The Biological Effects of UVA Radiation, New York, Praeger, pp. 147-155
van Weelden, H., de Gruijl, F.R. & van der Leun, J.C. (1986) Carcinogenesis by UVA, with an
attempt to assess the carcinogenic risks of tanning with UVA and UVB. In: Urbach, F. &
Gange, R.W., eds, The Biological Effects ofWA Radiation, New York, Praeger, pp. 137-146
van Weelden, H. , de Gruij l, F.R. , van der Putte, S.C.J. , Toonstra, J. & van der Leun, J.C. (1988) The
carcinogenic risks of modern tanning equipment: is UVA safer than UVB? Arch. dermatol. Res.,
280, 300-307
van Weelden, H. , van der Putte, S.C.J., Toonstra, J. & van der Leun, J.C. (1990a) UV A-Induced
tumours in pigmented hairless mice and the carcinogenic risks of tanning with UV A. Arch.
dermatol. Res., 282, 289-294
van Weelden, H., van der Putte, S.C.J., Toonstra, J. & van der Leun, J.C. (1990b) Ultraviolet
B-induced tumors in pigmented hairless mice, with an unsuccessful attempt to induce cutaneous
melanoma. Photodermatol. Photoimmunol. Photomed., 7, 68-72
Weinstock, M.A., Colditz, G.A., Willett, W.C., Stampfer, M . .J., Bronstein, B.R., Mihm, M.C., Jr &
Speizer, F.E. (1989) Nonfamilial cutaneous melanoma incidence in women associated with sun
exposure before 20 years of age. Pediatrics, 84, 199-204
Weinstock, M.A., Colditz, G.A., Willett, W.C., Stampfer, M.J., Bronstein, B.R., Mihm, M.C., Jr &
Speizer, F.E. (1991a) Melanoma and the sun: the effect of swimsuits and a 'healthy' ten on the
risk ofnonfami1ial malignant melanoma in women. Am. J. Epidemiol., 134, 462-470
Weinstock, M.A., Stryker, W.S., Stampfer, M.J., Lew, R.A. , Willett, W.C. & Sober, A.J. (1991b)
Sunlight and dysplastic nevus risk. Results of a clinic-based case-control study. Cancer, 67,
1701-1706
Weiss, J., Garbe, C., Bertz, J., Biltz, H. , Burg, G., Hennes, B., Jung, E.G., Kreysel, H.-W., Orfanos,
C.E., Petzold, D., Schwermann, M., Stadler, R., Tilgen, W., Tronnier, H. & Volkers, W. (1990)
Risk factors for the development of malignant melanoma in Germany (Ger.). Hautartz, 41,
309-313
Wells, R.L. & Han, A. (1984) Action spectra for killing and mutation of Chinese hamster cells
exposed to mid- and near-ultraviolet monochromatic light. Mutat. Res., 129, 251-258
Welsh, E.A. & Kripke, M.L. (1990) Murine Thy-1+ dendritic epidermal cells induce immunologic
tolerance in vivo. J. lmmunol., 144, 883-891
West, S.K., Rosenthal, F.S., Bressler, N.M., Bressler, S.B., Munoz, B., Fine, S.L. & Taylor, H.R.
(1989) Exposure to sunlight and other risk factors for age-related macular degeneration. Arch.
ophthalmol, 107, 875-879
Wheatcroft, R., Cox, B.S. & Haynes, R.H. (1975) Repair of UV-induced damage and survival in
yeast. l.Dimer excision. Mutat. Res., 30, 209-218
Wheeler, C.E., Jr (1975) Pathogenesis of recurrent herpes simplex infections. J. invest. Dermatol.,
65,341-346
Whelan, S.L., Parkin, D.M. & Masuyer, E., eds (1990) Patterns of Cancer in Five Continents
(IARC Scientific Publications No. 102), Lyon, IARC, p. 29
Whitaker, C.J., Lee, W.R. & Downes, J.E. (1979) Squamous cell skin cancer in the north-west of
England, 1967-69, and its relation to occupation. Br. J. ind. Me d., 36, 43-51
WHO (1977) Manual of the International Statistical Classification of Diseases, Injuries, and
Causes of Death. International of Diseases, 1975 rev., Vol. 1, Geneva, p. 102
WHO (1979) Ultraviolet Radiation (Environmental Health Criteria 14), Geneva
Willis, 1., Menter, J.M. & Whyte, H.J. ( 1981) The rapid induction of cancers in the hairless mouse
utilizing the principle of photoaugmentation. J. invest. Dermatol., 76, 404-408
Willis, I., Menter, J.M., de Fabo, E., Margolin, R. & Fisher, M. (1986) Photoaugmentation of UVB
effects by UVA. In: Urbach, F. & Gange, R.W., eds, The Biological Effects of UVA Radiation,
New York, Praeger, pp. 187-201
Winkelmann, R.K., Baldes, E.J. & Zollman, P.K. (1960) Squamous cell tumors induced in hairless
mice with ultraviolet light. J. invest. Dermatol., 34, 131-138
Winkelmann, R.K., Zollman, P.E. & Baldes, E.J. (1963) Squamous cell carcinoma produced by
ultraviolet light in hairless mice. J. invest. Dermatol., 40, 217-224
Wiskemann, A., Sturm, E. & Klehr, N.W. (1986) Fluorescent lighting enhances chemically induced
papilloma fonnation and increases susceptibility to tumor challenge in mice. J. Cancer Res. clin.
Oncol., 112, 141- 143
Withrow, T.J., Lugo, M.H. & Dempsey, M.J. (1980) Transformation ofBalb 3T3 cells exposed to a
germicidal UV lamp and a sunlamp. Photochem. Photobiol., 31, 135-141
Witkin, E.M. (1976) Ultraviolet mutagenesis and inducible DNA repair in Escherichia coli.
Bactenol. Rev., 40, 869-907
Wittenberg, S. ( 1986) Solar radiation and the eye: a review of knowledge relevant to eye care. Am.
J. Optom. physiol. Opt., 63, 676-689
REFERENCES 279
Wolf, C., Steiner, A. & Honigsmann, H. (1988) Do oral carotenoids protect human skin against
ultraviolet erythema, psoralen phototoxicity, and ultraviolet-induced DNA damage? J. invest.
Dermatol., 90, 55-57
Wood, R.D., Skopek, TR. & Hutchinson, F. (1984) Changes in DNA base sequence induced by
targeted mutagenesis of lambda phage by ultraviolet light. J. mol. Bioi., 173, 273-291
Wulff, D.L. & Fraenkel, G. (1961) On the nature of thymine photoproduct. Biochim. biophys. Acta,
51, 332-339
Yang, X.-Y., Ronai, Z.A., Santella, R.M. & Weinstein, LB. (1988) Effect of8-methoxypsoralen and
ultraviolet light A on EGF receptor (HER-1) expression. Biochem. biophys. Res. Commun.,
157, 590-596
Yasumoto, S., Hayashi, Y. & Aurelian, L. (1987) Immunity to herpes simplex virus type 2.
Suppression of virus-induced immune responses in ultraviolet B -irradiated mice. J. Immunol. ,
139, 2788-2793
Yoshikawa, J. , Rae, V, Bruins-Slot, W., Van den Berg, J.-W., Taylor, J.R. & Streilein, J.W. (1990)
Susceptibility to effects of UVB radiation on induction of contact hypersensitivity as a risk
factor for skin cancer in humans. J. invest. Dermatol., 95, 530-536
Young, R.W. (1988) Solar radiation and age-related macular degeneration. Surv. Ophthalmol., 32,
252-269
Young, M. & Russell, W.T (1926) An Investigation into the Statistics of Cancer in D{fferent Trades
and Professions (Medical Research Council Special Report No. 99), London, Her Majesty's
Stationery Office
Young, A.R., Walker, S.L., Kinley, J.S., Plastow, S.R. , Averbeck, D., Morliere, P. & Dubertret, L.
( 1990) Phototumorigenesis studies of 5-methoxypsoralen in bergamot oil: evaluation and
modification of risk of human use in an albino mouse skin model. J. Photochem. Photobiol. B.
Biol., 1, 231-250
Zanetti, R., Rosso, S., Faggiano, F., Roffino, R., Colonna, S. & Martina, G. (1988) A case-control
study on cutaneous malignant melanoma in the province of Torino, Italy (Fr.). Rev. Epidemiol
Sante publ, 36, 309-317
Zelle, B., Reynolds, R.J. , Kottenhagen, M.J., Schuite, A. & Lohman, PH.M. (1980)The influence of
the wavelength of ultraviolet radiation on survival, mutation induction and DNA repair in
irradiated Chinese hamster cells. Mutat. Res., 72, 491-509
Zigman, S. & Vaughan, T. (1974) Near-ultraviolet light effects on the lenses and retinas of mice.
Invest. Ophthalmol., 13, 462-465
Zigman, S., Yulo, T. & Schultz, J. (1974) Cataract induction in mice exposed to near UV light.
Ophthal. Res. , 6, 259-270
Zigman, S., Fowler, E. & Kraus, A.L. (1976) Black light induction of skin tumors in mice. J. invest.
Dermatol., 61, 723-725
Zmudzka, B.Z. & Beer, J.Z. (1990) Activation of human immunodeficiency virus by ultraviolet
radiation. Photochem. Photobiol. , 52, 1153-1162
Zolzer, F. & Kiefer, J. (1983) Wavelength dependence of inactivation and mutagenesis in haploid
yeast cells of different sensitivities. Photochem. Photobiol. , 37, 39-48
Zolzer, F. & Kiefer, J. (1984) Wavelength dependence of inactivation and mutation induction to
6-thioguanine-resistance in V79 Chinese hamster fibroblasts. Photochem. Photobiol., 40, 49-53
Zolzer, F., Kiefer, J. & Rase, S. (1988) Inactivation and mutation induction to 6-thioguanine
resistance in V79 hamster fibroblasts by simulated sunlight. Photochem. Photobiol., 41, 399-
404
Dec. 2000
Appendix B: Profile for Solar Radiation and Exposure to
Sunlamps and Sunbeds. Report on Carcinogens, Ninth
Edition (2000)
101
Dec. 2000
102
Dec. 2000
SOLAR RADIATION AND EXPOSURE TO SUNLAMPS OR SUNBEDS
First listed in the Ninth Report on Carcinogens*
CARCINOGENICITY
Solar radiation is known to be a human carcinogen, based on sufficient evidence
of carcinogenicity from studies in humans, which indicate a causal relationship between
exposure to solar radiation and cutaneous malignant melanoma and non-melanocytic skin
cancer. Some studies suggest that solar radiation may also be associated with melanoma
of the eye and non-Hodgkin's lymphoma (reviewed in IARC V.55, 1992).
Exposure to sunlamps or sunbeds is known to be a human carcinogen, based on
sufficient evidence of carcinogenicity from studies in humans, which indicate a causal
relationship between exposure to sunlamps or sunbeds and human cancer.
Epidemiological studies have shown that exposure to sunlamps or sunbeds is associated
with cutaneous malignant melanoma (Swerdlow et al., 1988; Walter eta!., 1990; Autier
et al., 1994; Westerdahl et al., 1994). Exposure-response relationships were observed for
increasing duration of exposure, and effects were especially pronounced in individuals
under 30 and those who experienced sunburn. Malignant melanoma of the eye is also
associated with use of stmlamps. In contrast, there is little support for an association of
exposure to sunlamps or sunbeds with non-melanocytic skin cancer (IARC V.55, 1992).
The evidence that solar radiation and exposure to sunlamps or sunbeds are human
carcinogens is supported by experimental studies in laboratory animals, and studies
demonstrating UV -induced DNA damage in human and animal cells. Sunlamps and
sunbeds emit radiation primarily in the ultraviolet A (UV A) and ultraviolet B (UVB)
portion of the spectrum. Numerous studies have shown that simulated solar radiation,
broad spectrum UV radiation, UV A radiation, UVB radiation, and UVC radiation are
carcinogenic in experimental animals. There is evidence for benign and malignant skin
tumors and for tumors of the comea and conjunctiva in mice, rats, and hamsters. UV
radiation also causes a wide spectrum of DNA damage resulting in mutations and other
genetic alterations in a variety of in vitro and in vivo assays for genotoxicity, including
assays using human skin cells (IARC V.55, 1992).
PROPERTIES
Solar radiation from the sun includes most of the electromagnetic spectrum
(IARC V.55, 1992). Of the bands within the optical radiation spectrum, UV light is the
most energetic and biologically damaging. UV light is divided into UV A, UVB, and
UVC. UV A is the most abundant of the three, representing 95% of the solar UV energy
to hit the equator, and UVB represents the other 5%. The short wavelength UVC rays are
* there is no separate CAS registry number assigned to solar radiation and exposure to sunlamps or
sunbeds.
103
Dec. 2000
absorbed by ozone, molecular oxygen, and water vapor in the upper atmosphere so that
measurable amounts from solar radiation do not reach the earth's sw-face (Farmer and
Naylor, 1996).
Molecules that absorb UV and visible light contain moieties called chromophoric
groups in which electrons are excited from the ground state to higher energy states. In
returning to lower energy or ground states, the molecules generally re-emit light (Dyer,
1965). Molecules sensitive to UV light absorb and emit UV light at characteristic
maximum wavelengths (A.), often expressed as Amax
Photochemical and photobiological interactions occur when photons of
optical radiation react with a photoreactive molecule, resulting in either a
photochemically altered molecule or two dissociated molecules (Phillips, 1983;
Smith, 1989; both cited by IARC V.55, 1992). To alter molecules, a sufficient
amount of energy is required to alter a photoreactive chemical bond (breaking the
original bond and/or forming new bonds).
UVB is considered to be the major cause of skin cancer despite its not penetrating
the skin as deeply as UVA or reacting with the epidermis as vigorously as UVC. UVB's
reactivity with macromolecules combined with depth of penetration make it the
biologically most potent portion of the UV spectrum, with respect to short-term and long-
term effects. UV A, while possibly not as dangerous, also induces biological damage
(Farmer and Naylor, 1996).
USE
Photobiological reactions of concern for skin cancer risk due to UV light
exposure are the reactions with the main chromophores of the epidermis-
urocanic acid, DNA, tryptophan, tyrosine and the melanins. DNA photoproducts
include pyrimidine dimers, pyrimidine-pyrimidone (6-4) photoproducts, thymine
glycols, and DNA exhibiting cytosine and purine damage and other damage such
as DNA strand breaks and cross-links and DNA-protein cross-links. The different
DNA photoproducts have varying mutagenic potential (IARC V.55, 1992).
UV -induced DNA photoproducts produce a variety of cellular responses that
contribute to skin cancer. Unrepaired DNA photoproducts may result in the
release of cytokines that contribute to tumor promotion, tumor progression,
immunosuppression, and the induction of latent viruses (Yarosh and Kripke,
1996; TARC V.55, 1992).
Aside from the many benefits of sunlight/solar radiation, artificial sources of
UVR are used for cosmetic tanning, promotion of polymerization reactions, laboratory
and medical diagnostic practices and phototherapy, and numerous other applications
(IARC V.55, 1992).
104
Dec. 2000
SOURCES
Ultraviolet light is naturally emitted by the sun and artificially from lamps such as
tungsten-halogen lamps, gas discharge, arc, fluorescent, metal halide, and electrodeless
lamps (IARC V.55, 1992) and lasers such as the 308-nm XeCl (xenon chloride) excimer
and the 193-nm ArF (argon fluoride) excimer (Sterenborg et al., 1991).
The use of sunlamps and tanning beds is as a cosmetic source. The latter chiefly
emit UV A (315-400 nm) although certain lamps that emitted considerable UVB and
UVC radiation were more common before the mid-1970s (IARC V.55, 1992). However,
UVB produces a better tan than UV A and recently, at least in the United States and
United Kingdom, use of sunlamps with more UVB radiation has become widespread
(Wright et al., 1997; cited by Swerdlow and Weinstock, 1998). Low-pressure mercury
vapor lamps, sunlamps, and black-light lamps are considered to be low-intensity UV
sources. High-intensity UV sources include high-pressure mercury vapor lamps, high-
pressure xenon arcs, xenon-mercury arcs, plasma torches, and welding arcs. Three
different UVA phosphors have been used in sunlamps sold in the United States over the
past 20 years, producing emission spectra that peak at 340 nm, 350 nm, or 366 nm. Two
modem U.S. sunlamps evaluated by the FDA emitted 99.0% and 95.7% UV A and the
rest UVB radiation (<320 nm). A new high-pressure UV A sunbed with eighteen 1600-W
filtered arc lamps emitted 99.9% UV A. An older-type sunlamp used more than 20 years
ago (UVB/FS type) emitted 48.7% UVA (Miller et al., 1998).
EXPOSURE
The greatest source of human exposure to UVR is solar radiation; however, the
exposure varies with the geographical location. With decreasing latitude or increasing
altitude, there is greater exposure; for every 1000 feet above sea level, a 4% compounded
increase in UVR exists. Decreases in the stratospheric ozone caused by chemicals
generating free radicals increase UVR exposure. Heat, wind, humidity, pollutants, cloud
cover, snow, season, and the time of day also affect UVR exposure (Consensus
Development Panel, 1991).
Although use of sunscreen is known to protect from skin damage induced by
UVR, sunscreen use has not become habitual by a large fraction of the U.S. population.
For example, Newman et al. (1996) surveyed a random sample of persons in San Diego, a
location with one of the highest incidences of skin cancer in the United States. Sunscreen
was used only about 50% of the time on both face and body by tanners, about 40% of the
time on the face, and 30% of the time on the body.
Most bulbs sold in the United States for use in sunbeds emit "substantial doses of
both UVB and UV A" (Swerdlow and Weinstock, 1998, citing "personal communication
from industry sources"). Many of the home and salon devices in the 1980s emitted both
105
Dec. 2000
UV A and UVB radiation, but current devices emit predominantly UV A (FTC, 1997;
Sikes, 1998).
FDA scientists calculated that commonly used fluorescent sunlamps would
deliver 0.3 to 1.2 times the annual UV A dose from the sun to a typical tanner requiring
20 sessions at 2 minimal erythemal doses (MED) per session. The common sunlamps
would deliver to a frequent tanner (100 sessions at 4 MED/session) 1.2 to 4.7 times the
UVA received annually from solar radiation. The frequent tanner would receive 12 times
the annual UV A from solar radiation from the recently available high-pressure sunlamps
(Miller et al., 1998).
In 1987, an American Academy of Dermatology (AAD) survey found that,
although 96% of the U.S. population surveyed knew that sun exposure causes
cancer, one-third of the adults responding develop tans. By 1987, the indoor
tanning industry was one of the fastest growing in the United States (Sikes, 1998).
Surveys of U.S. telephone book Yellow Pages found 11,000 indoor tanning
facilities in 1986 and more than 18,000 facilities in 1988. About I 1% of women
and 6% of men were frequent patrons (Research Studies-SIS, 1989). New York
State alone was estimated to have 1300 commercial tanning facilities in 1993
(Lillquist et al., 1994). By 1995, indoor tanning facilities were a $1 billion
industry serving 1 million patrons a day (Guttman, 1995). About 1 to 2 million
patrons visit tanning facilities as often as 100 times per year (Sikes, 1998).
A 1990 survey of 1,564 holders of drivers' licenses residing in New York State
outside of the New York City area, who were aged 17 to 74 years, were white, and had
never had skin cancer, found that 21.5% of the respondents had ever used sun lamps
(28.1% among those 16 to 24 years old) but that only 2.3% used sun lamps at least once a
month. Ever users were more likely to be women, younger, and never married or
divorced or separated (Lillquist et al., 1994). Surveys in the early 1990s of adolescents
who had ever used tanning devices have found about twice as many girls as boys among
the users (33% vs. 16% and 18.5% vs. 7.4%) (Banks et al., 1992; Mermelstein and
Riesenberg, 1992; both cited by Lillquist et al., 1994).
Up to 25 million persons per year in North America are currently estimated to use
sunbeds. Teenagers and young adults are prominent among users. A study of high school
students in St. Paul, Minnesota, found that 34% had used commercial sunbeds at least 4
times in the past year. Fifty-nine percent of the users reported some skin injury. A 1995
U.S. survey found that commercial tanning salon patrons included 8% aged 16 to 19
years and 42% aged 20 to 29 years; 71% were female (Hurt and Freeman, undated; cited
by Swerdlow and Weinstock, 1998).
Wisconsin dermatologists, ophthalmologists, and emergency room personnel
reported treating 372 patients with ocular and/or dermal injuries from artificial tanning
devices in a 12-month survey ca. 1990. Of these patients, 53% to 65% were exposed to
tanning beds or booths and 17 to 35% were exposed to reflector bulb lamps. In the group
of 155 emergency room patients with first or second degree skin burns from artificial
106
Dec. 2000
tanning, 58% were burned at tanning salons and 37% were burned at home (Garrett,
1990). Although FDA has mandated rules that require that tanning equipment labeling
warn about overexposure, skin cancer, possible premature skin aging, and
photosensitivity with certain cosmetics and medications, a Public Interest Research
Group survey of lOO tanning salons in 8 states and the District of Columbia found 183
tanning devices without the required warnings (Cosmetic Insiders' Report, 1991 ). Sikes
( 1998) stated, without attribution, that tanning devices caused 1,800 reported injuries in
1991, mostly in persons aged 15 to 24 years old. A survey of 31 tanning salons in 1989 in
the greater Lansing, Michigan, area, population 450,000, found that 87% of the facilities
offered their clients "tanning accelerators." Respondents of five establishments stated that
their tanning accelerators contained psoralens, but this could not be confirmed (Beyth et
al., 1991).
Workers in many occupations, e.g., agricultural, construction, and road
work laborers, spend a large component of their work day outdoors. Outdoor
workers, therefore, are the largest occupational group exposed to solar UVR.
Occupational exposure to artificial UVR occurs in industrial photo processes,
principally UV curing of polymer inks, coatings, and circuit board photoresists;
sterilization and disinfection; quality assurance in the food industry; medical and
dental practices; and welding. Welders are the largest occupational group with
artificial UVR exposure. However, only arc welding processes produce
significant levels of UVR. UVR from welding operations is produced in broad
bands whose intensities depend on factors such as electrode material, discharge
current, and gases surrounding the arc (NIOSHa, 1972). [OSHA regulations
require many protective measures to reduce UVR exposure of workers engaged in
or working in the vicinity of arc welding operations.]
A study conducted on laboratory UV lasers such as those used in cornea shaping
and coronary angioplasty showed that the relative risk may increase to a level comparable
to that of individuals with an outdoor profession (Sterenborg et al., 1991 ).
Applying a mathematical power model based on human data, Lytle et al. (1992)
suggested that there is an increased risk of squamous cell carcinoma (SCC) from
exposure to UV -emitting fluorescent lamps. The estimates of annual incidence of new
sec, for indoor workers exposed to uv light, indicated that an exposure to typical
fluorescent lighting (unfiltered by a clear acrylic prismatic diffuser) may add 3.9%
(1.6%-12%) to the potential risk from solar UVR, thus resulting in an induction of an
additiona11500 (600-4500) SCC per year in the United States. There is a small increased
risk of sec from exposure to uv -emitting fluorescent lamps, when compared to 110,000
sec caused by solar exposure.
NIOSHa (1972) estimated that 211,000 workers in the manufacturing industries
(Standard Industrial Codes [SICs] 19-39) were exposed to UVR; 49,000, in the
transportation and communication industries (SICs 40-49); 17,000, in the wholesale,
miscellaneous retail, and service stations categories (SICs 50, 59, 55); and 41,000, in the
107
Dec. 2000
services industries (SICs 70-89). The sources considered were arc welding, air purifiers,
and sanitizers.
REGULATIONS
The U.S. Food and Drug Administration (FDA) Center for Devices and
Radiological Health (CDRH) have promulgated regulations concerning sunlamp products
and UV lamps intended for use in sunlamp products. Manufacturers must notify CDRH
of product defects and repair and replacement of defects. CDRH issues written notices
and warnings in cases of noncompliance. Several performance requirements must be met
by sunlamp products (21 CFR 1040.20), including irradiance ratio limits, a timer system,
protective eyewear to be worn during product use, compatibility of lamps, and specific
labels. The label should include the statement "DANGER- Ultraviolet radiation" and
warn of the dangers of exposure and overexposure.
OSHA requires extensive UVR protective measures of employees engaged in or
working adjacent to arc welding processes. Arc welding emits broad spectrum UVR.
Workers should be protected from the UVR by screening, shields, or goggles. Employees
in the vicinity of arc welding and cutting operations should be separated from them by
shields, screens, curtains, or goggles. If possible, welders should be enclosed in
individual booths. In inert-gas metal-arc welding UVR production is 5 to 30 times more
intense than that produced by shielded metal-arc welding. OSHA-required protective
measures in shipyard employment and marine tern1inals include filter lens goggles worn
under welding helmets or hand shields and protective clothing that completely covers the
skin to prevent UVR burns and other damage (OSHA, 1998a, 1998b, 1998c).
ACGIH (1996) has set various Threshold Limit Values (TLVs) for skin and
ocular exposures. TL V s for occupational exposure are determined by these parameters:
1. "For the near UV spectral region (320 to 400 nm), total irradiance incident
upon the unprotected eye should not exceed 1.0 m W /cm
2
for periods greater
than 10
3
seconds (approximately 16 minutes) and for exposure times less than
I 0
3
seconds should not exceed l.O J/cm
2
."
2. Unprotected eye or skin exposure to UVR should not exceed 250 mJ/cm
2
( 180
nm) to l.Ox10
5
mJ/cm
2
(400nm) for an 8-hour period. The TLVs in the
wavelength range 235 to 300 nm are 3.0 (at 270 nm) to 10 mJ/cm
2
.
3. Effective irradiance for broad band sources must be determined by using a
weighting formula.
4. "For most white-light sources and all open arcs, the weighting of spectral
irradiance between 200 and 315 nm should suffice to determine the effective
irradiance. Only specialized UV sources designed to emit UV -A radiation
would normally require spectral weighting from 315 to 400 nm."
5. The pennissible ultraviolet radiation exposure for unprotected eye and skin
exposure may range from 0.1 11W/cm
2
(8 hours/day) to 30000 11W/cm
2
(0.1
sec/day).
108
Dec. 2000
6. "All of the preceding TLVs for UV energy apply to sources which subtend an
angle less than 80. Sources which subtend a greater angle need to be
measured only over an angle of80."
ACGIH ( 1996) added that even though conditioned (tanned) individuals may not
be any more protected from skin cancer, they can tolerate skin exposure in excess of the
TL V without erythemal effects. NIOSH criteria for a recommended standard for
occupational exposure to UVR are practically identical to those given in ACGIH items 1
and 2 above (NIOSHa, 1972).
The Federal Trade Commission (FTC) investigates false, misleading, and
deceptive advertising claims about sunlamps and tanning devices (FTC, 1997).
The American Medical Association passed a resolution in December 1994 that called for
a ban of the use of suntan parlor equipment for nonmedical purposes. Dermatologists
have urged the FDA to take action to discourage use of suntan parlors and suntan beds
(Blalock, 1995). Currently, the FDA Center for Devices and Radiological Health and the
Centers for Disease Control and Prevention (CDC) encourage avoidance of sunlamps and
sunbeds (AAD, 1997). Although 27 states and municipalities had promulgated some
regulations on indoor tanning facilities by late 1995, they are seldom enforced (Blalock,
1995). The American Academy of Dermatology's Tanning Parlor Initiative provides a
manual giving instructions on petitioning state, regional, and local governments on this
issue and examples of regulatory legislation (Dermatology Times, 1990). Regulations are
summarized in Volume IT, Table A-35.
109
Proc. Natl. Acad. Sci. USA
Vol. 91, pp. 360-364, January 1994
Medical Sciences
UV and skin cancer: Specific p53 gene mutation in normal skin as a
biologically relevant exposure measurement
(DNA damap/p53 mtdoa/carclaopae*)
HISAYOSHI NAKAZAWA*, DALLAS ENGLISHt , PETER L. RA.NDELLt , KEIKO NAKAZAWA:t, NICOLE MARTEL*,
BRUCE K. ARMSTRONG*, AND HIROSHI Y AMASAKI*f
International Apocy for Research oo Cancer, 150 coors Albert Thomas, 69372 Lyon cedex 08, France; f1be University of Western Australia, NedlaDds, WA
6009, Australia; and *Hospital Edouard Herriot, Lyoo, France
Communicated by Gerald N. Wogan, September 20, 1993
ABSTRACT Many human skin tumors coataln mutated
p53 genes that probably result from UV exposure. To investi
pte the liak between UV exposure ud p53 gene mutation, we
developed two metbods to deted presumptive UV -spedftc: p53
geae mutations in UV -expoaecl normal sldn. 1be metbods are
bMed oa mutaot aUeJe.spedfk PCR.s aaclllpae cbain radioDs
aacl designed to deted CC to TT mutatioas at codoas 245 and
247/248, using 10 II of DNA samples. Tbe8e spedfk muta-
tloas in the p53 gene have been reported in sldo tumors. CC to
TT mutatloas in the p53 gene were detedecl in cultured human
sldn cells ODiy after UV lrradJation, and the mutation frequency
inc:reued with lncreasiDg UV dole. Seventeea of 23 samples of
normal sldn from sua-expoaecl sites OD AustraUaa sldn
caac:er patieats coatained CC to TT mutatioas in oae or both
of codoas 245 and 247/248 of the p53 gene, aad only 1 of 20
samples from DODSUD-expoaecl sites harbored the muta
tiOD. Noae of 15 biopsies of DOI'ID8I skin from DOD-sua-espoeed
or intermJtteatiy oposed sites oa vohmteen Uvbag in France
carried such mutatioas. Our results SIIJPSl that spedfk p53
gene mutatloas auodated with human skin caac:er are induced
in normal skin by solar UV radiation. Measuremeat of these
mutatloas may be useful as a biologically rdevut measure of
UV exposure in hiUIUUIS ud as a possible predictor of risk for
sldn amcer.
One of the key problems in conducting informative epidemi-
ological studies on cancer is the lack of accurate methods for
measuring exposure to suspected carcinogens in individuals.
Mechanism-based exposure measurement methods-e.g. ,
quantification of DNA adducts in target or surrogate cells-
are being developed to overcome this problem (1). While
most such assays can estimate " recent" or " current" expo-
sure, they cannot measure cumulative exposure of target
cells, which may be the most relevant measure in cancer
epidemiology. To overcome this difficulty, the measurement
of mutation frequency-e.g. , of the HPRT gene-has been
proposed (2). We have proposed that carcinogen-specific
mutation patterns in cancer-related genes-i.e., oncogenes
and tumor suppressor be developed as biolog-
ically relevant measures of exposure to environmental car-
cinogens (3).
Molecular analysis of tumors has revealed various changes
in oncogenes and tumor suppressor genes. Since most can-
cers are considered to result from exposure to environmental
carcinogens, it is reasonable to assume that some or most
critical genetic changes are induced by such risk factors
(3-5). Animal tumors induced by specific carcinogens sup-
port this assumption: mouse skin and liver tumors and rat
mammary tumors induced by 7 ,12-dimethylbenz[a]an-
The publication costs of this article were defrayed in part by J18iC cbalJe
payment. This article must therefore be hereby marked "advertisement"
in accordance with 18 U.S.C. 1734 solely to i.ndjcate this fact.
360
thracene (DMBA) often contain an A to T mutation at codon
61 of the Ha-ras gene, while those induced by N-methyi-N-
nitrosourea (MNU) or N-methyl-N' -nitro-N-nitrosoguani-
dine (MNNG) harbor G to A mutations at codon 12 of the
same gene (reviewed in refs. 3 and 5). These findings can be
interpreted in accordance with known mechanisms of action
of the carcinogens used: DMBA is considered to form a major
adduct with adenine (6), while MNU and MNNG produce
0
6
-methylguanine, which is a promutagenic lesion for G to A
transition (7). There are other examples in which carcinogen-
specific genetic footprints can be found in animal tumors
(reviewed in refs. 3 and 5).
A few examples suggest that critical genetic changes in
human tumors may also be determined by exposure to IJU\ior
risk factors. For example, six samples of liver hemangiosar-
coma from patients who worked in vinyl chloride plants were
analyzed for the RAS mutation; five showed a G to A
transition at codon 13 of the KRAS gene (8). This mutation is
consistent with the type of vinyl chloride-DNA adducts and
mutation found in bacteria (9), and vinyl chloride is
a major etiological factor for this extremely rare cancer in
humans (10). Other examples include studies which showed
that liver tumors from areas with high exposure to aflatoxin
B 1 contained a much higher prevalence of G toT transversion
in the p53 gene than similar tumors from low-exposure areas
(11, 12). Another example has come from the study of Brash
et a/. (13), who found that p53 mutation spectra in squamous
cell carcinomas of the skin were quite different from those
found in internal tumors and were consistent with the occur-
rence of C toT or CC to TT transitions at sites of dipyrimidine
photoproducts (14, 15). Similar results have been obtained in
basal cell carcinomas (16). A very recent study by Kress et
a/. (17) has shown that UVB-induced mouse skin carcinomas
also contain UV -specific mutations in the p53 gene, including
CC to TT tandem base mutations.
While a possible link between carcinogen exposure and
genetic changes found in tumors may exist as described
above, it does not provide a means for biologically relevant
exposure estimation, nor does it prove their causal relation-
ship, unless evidence is provided that a given carcinogen
indeed induces the critical gene mutations found in tumors.
In other words, it is necessary to detect such critical gene
mutations after carcinogen exposure but before tumor oc-
currence (3). Unlike other mutation assays, which rely on
phenotypic selections of mutants, cancer-related gene muta-
tion assay has to be based on genotype analysis since
phenotypic expression of these genes is carcinogenesis per
se. We and others (18-20) have recently developed quanti-
tative and sensitive methods to detect specific RAS gene
mutations after carcinogen exposure but before cell trans
Abbreviations: AS-PCR, allele-specific PCR; AS-LCR, allele-
specific ligase chain reaction.
Ito whom reprint requests should be addressed.
Medical Sciences: Nakazawa et al.
formation or tumor appearance. These studies with model
systems indicated that genotype-based mutation assays of
cancer-related genes can be developed as a biologically
relevant method of measuring exposure to carcinogens.
To develop and validate such a method in humans, we
chose UV radiation as the model carcinogen. This model
provided us with two important advantages: UV causes skin
cancer (21) and UV -specific p53 gene mutations have been
found in human (13, 16, 22, 23) and mouse (17, 24) skin
cancers. In fact, a UV-specific CC to TT tandem base
mutation in the p53 gene has so far been found mainly in skin
cancers, while an oxygen radical may also induce the muta-
tion (25). Here we report development of two sensitive
methods to detect a UV-specific mutation (CC to TT) in the
p53 gene by using UV -exposed human skin cell cultures and
their application in normal human skin.
MATERIALS AND METHODS
CeU Culture aDd UV Irradiation in Vitro. Normal human
keratinocytes from adult donors were isolated as described
(26) and were grown in serum-free keratinocyte medium
A
wild typo
2472.
a coo===:lils
-(S2PJ
(saP) T J.
S' 6 I ljjQ ace O:::O::::O:::S::::CCICIII'
(
1
) T T
[S2P
,. p nn lA ACC k"tz?ZZZZZZIZII'
B
wild typo
247 Ul
I'=Tc;::JAAC COGc:::!IATC=S'
PCR
-
t (1) A :0 (2) ,.,(UP)

J'CZIQ'JmTTO ace t I I I I I I II'
CCtoTT-
247 241
TOO G:S:!I ATC lmmJ'
t AA
(saP),& I T !-,_
S" IIIZIO TCZZZ22TTA ACC ai:ZZZZZZZZ:r:lll'
LCA
-

no mpllflcetlon
Flo. 1. Schematic representation of strategies for detection of
CC to TT mutations at codons 247/248 of p.S3 by AS-PCR (A) and
AS-LCR (B) methods. These methods are designed to amplify only
the CC to TT mutant alleles. The same strategies were used for
detection of codon 24.S mutation. (A) Amprimers for detection of
mutation by AS-PCR: (1) for codon 24.S; 5'-AAC AGT TCC TGC
ATG GGC AA-3' (tm = 60.20C); (1) for codons 247/248, 5'-C TGC
ATG GGC GGC ATG AATT-3' (1m "' 55.30C); (2) for both codons,
5'-CAA GTG OCT CCT GAC CTG GA-3' (lm = .S5.1rC). (B)
Amprimers used for AS-LCR for codon 245: (1), .S'-TT GCC CAT
GCA GOA ACT OTT ACA C cg-3' (1m = 66.20C); (2), 32P-Iabeled
5'-cgg GAT GOO CCT CCG OTT CAT G-3' (1m = 66.10C); (3),
l2P-labeled 5'-cgg G TOT AAC AGT TCC TGC ATG GGC-3' (1m=
68.3"C); (4), 5'-AAC ATG AAC COO AGO CCC ATC CTC c-3' (1m
= 67 .30C). Amprimers used for AS-LCR for codons 247/248: (1),
5'-A ATT CAT GCC GCC CAT GCA GOA g-3' (1m= 67.1"C); (2),
lZP-Iabeled .S'-cgg GAT GOT GAG GAT GOO CCT CC-3' (1m=
67.9"C); (3),
32
P-labeled .S'-gg GT TCC TGC ATG GGC GGC ATG
AA-3' (1m= 66.00C); (4), 5'-T TOO AGO CCC ATC CTC ACC ATC
ATC-3' (1m = 65"C).
Proc. Nat/. Acad. Sci. USA 91 (1994) 361
(GIBCO) with 1.2 mM CaCh at 32C in a 90% air/10%
humidified incubator. Cells at passage 2 were exposed to
UVB radiation from a BLE-8T312 lamp (Spectronics, West-
bury, NY) (peak emission at 312 nm). Total cellular DNA was
prepared as described (18).
Skin Biopsy. Three- to 5-mm-diameter, full thickness bi-
opsies of normal skin were taken from 26 skin cancer patients
in Australia and from 17 normal volunteers in France. Paired
punch biopsies of normal skin were obtained from 17 of the
Australian skin cancer patients, one from a sun-exposed site
(shoulder) and the other from a non-sun-exposed site (but-
tock); none of these biopsies was immediately adjacent to a
skin cancer. Biopsies were immediately frozen in liquid
nitrogen and kept at -20"C to -SO"C until use.
Detedion of CC to 1T Mutation of the p53 Gene In Normal
Skin Cells. Two methods were developed, both designed to
amplify only mutant alleles, with 10 pgofDNA samples. We
chose mutations at codons 245 and 247/248 since both have
been found in skin cancers (13).
Mutant AJiele..Spedftc PCR (AS-PCR). This is a modified
nested PCR in which the entire exon 7 of the p53 gene was
initially amplified and then the mutant allele was selectively
amplified by using mutant allele-specific oligonucleotides
(Fig. 1). For the first PCR, the "amprimers" used were 5'-A
CTGGCCTCATCTTGGGCCT-3' and5'-TGTGCAGGG
TOG CAA GTG GC-3' (27) . . Each PCR mixture (100
consisted of 10 pg of genomic DNA, 1 x PCR buffer (Boeh-
ringer Mannheim), 400 dNTP, and 2 units of Taq
polymerase and amprimers. Amplification was initiated by
denaturation at 96C for 2 min and by the hot start method
(70"C) (28), followed by 40 cycles of96C, 1 min; 60"C,1 min;
72C, 30 sec. The amplified exon 7 of the p53 gene was
purified by spin column chromatography (Microcon; Ami-
con) and lyophilized, and it served as the template for the
second PCR. The mutant AS-PCR assay was performed in
the reaction mixture (100 which consisted of the ampli-
fied p53 exon 7 DNA, lx PCR buffer, 400 dNTP,
2 units of Taq polymerase, and 200 each 5'-32p-end-
labeled amprimer; amprimer sequences are given in Fig. 1.
PCR was performed by initially denaturing DNA at 96C for
2 min and by the hot start method at 75C, foUowed by 35
cycles of94C, 30 sec; 59"C for codon 245 mutation and 54C
for codons 247/248, 1 min; 72C, 30 sec. Amplified DNAs
were lyophilized and subjected to 10% PAGE. Amplified CC
to TT mutations were detected by autoradiography.
Mutant AJiele..Spedftc Ligue Cbaln Reaction (AS-LCR).
The LCR method (29) was designed with four primers to
amplify only CC to TT mutations in the p53 gene (Fig. 1).
Amprimers used are shown in Fig. 1. The primers were
designed and selected by use of the OLIGO primer analysis
program version 4.0 (MedProbe, Oslo). The LCR was per-
formed with 20 units of Ampligase (Epicentre Technologies,
Madison, WI). After initial denaturation at 96C for 2 min and
hot start at 75C, the reaction was continued for 35 cycles at
96C, 3 min; 60"C, 1 min; 72C, 30 sec. Amplified DNAs were
analyzed for the presence of the mutations as described for
the PCR method.
Each experiment contained at least two control samples
(nonexposed cultured skin cells) to detect possible PCR
(LCR) cross-contamination. Since no control showed muta-
tion bands, we concluded that cross-contamination occurs
only very rarely.
RESULTS
IDduction of p53 CC to 1T Mutations In Cultured Human
Keratlnocytes by UV Radiation: Detection by AS-PCR aad
AS-LCR Methods. To examine whether UV -specific muta-
tions in the p53 gene can be induced and detected in human
skin ceUs, we exposed early passages of human keratinocyte
362 Medical Sciences: Nakazawa et al.
A
UVB(312nm)
mJ/cm2 10 5 2 o
CCtoTT
mutation at
codon 245
UVB(312nm)
mJ/cm2 10 5 2 o
codon 24718 --
B passage N o 2 4 s
number
CCtoTT
mutation at
codon245 --
codon 24718 --
FIG. 2. Detection of UVB-induced pS3 CC to TT mutations at
codons 24S and 247/248 in cultured human normal keratinocytes by
AS-PCR and AS-LCR. (A) AS-PCR analysis: secondary cultures
were exposed to different doses ofUVB (0, 2, S, 10 mJfmZ) and DNA
was isolated after four passages. (B) AS-LCR analysis: human
keratinocytes were exposed to UVB (10 mJ/m
2
) at second passage.
DNA was isolated after different passage numbers and subjected to
AS-LCR. The CC to TT mutation-specific LCR was carried out for
codons 24S and 247/248 separately, but the products were electro-
phoresed in the same lanes. Lane N, control cells without UVB
exposure after two to four passages.
cultures to UVB and examined for the presence of CC to TT
mutations at codon 245 and codons 247/248 of the p53 gene
as described above and in Fig. 1. All experiments were
performed from samples taken from two individuals and
similar results were obtained.
Fig. 2 shows the CC to TT mutation at codons 245 and
247/248 in the p53 gene was detectable with the PCR or LCR
assay only in UVB-exposed keratinocytes. The mutations
increased in a UV dose-dependent manner (Fig. 2A) and were
not detectable until after four passages of cells following UV
irradiation (Fig. 2B). It may be that a few passages are
necessary to fix the mutations or that cells with the p53
mutations had a selective growth advantage and became
detectable after a few passages. The mutations at codons
247/248 were more frequently induced than those at codon
245. At passage 4, the mutation was detectable only when > 3
p.g of DNA (equivalent to =5 x lOS cells) was assayed,
suggesting that the frequency of this mutation in the exposed
cell population is at least 1 in 5 x lOS cells (data not shown).
Under these conditions, no mutation was observed in any
passages of cells without UV exposure (Fig. 2). These results
Proc. Nat/. Acad. Sci. USA 91 (1994)
A
shoulder
aubt-ct No. L 1 14 15 111 17 18 18 20 21 22 23 24 25 211 L2
CCtoTT
mubltlon<rt --
codon 24718
B
eubt-ct No.
CCtoTT
mutllllon 81 --
codon24718
codon 245 --
buttock
L1 14 15 16 17 18 18 20 21 22 23 24 25 26 L2

FIG. 3. Detection of CC to TT mutations in the pS3 gene in
biopsies of normal skin from sun-exposed and non-sun-exposed sites
in the same individuals. Two punch biopsies (one from shoulder and
the other from buttock) of normal skin were obtained from each of
16 Australian skin clinic patients. DNA was isolated and subjected
to AS-LCR assay to detect the CC to TT mutations at codons 24S and
247/248 of the pS3 gene as descnbed in Figs. 1 and 2. Lanes Ll and
L2, control DNA isolated from skin biopsies removed from two
volunteers living in Lyon, France; information on each subject is
given in Table 1.
indicate that both methods are sensitive to detect UV-
induced CC to TT mutations of the p53 gene.
Detectioo ofUV-Spedfk: (CC to TI) MutaUoos by PCR aod
LCR ill Sun-Exposed and Non-Sun-Exposed Normal Skin
Biopdes from the Same Individuals. To examine a possible
correlation between sun exposure and the presence of UV-
specific p53 gene mutations in human skin, we collected
normal skin biopsies from sun-exposed (shoulder) and non-
sun-exposed (buttock) sites on the same Australian skin clinic
patients. All except two had a primary diagnosis of skin
cancer. Both types of biopsies were collected from 17 pa-
tients (subjects A10-A26) and all were taken from sites
distant from the skin lesions.
More CC to TT mutations were detected by both analyses
in sun-exposed than in non-sun-exposed parts of the skin
(Fig. 3; Table 1). Thirteen of 17 sun-exposed skin biopsies
showed CC to TT mutations in at least one assay and in at
least one of two codons tested-6 at codon 245 and 10 at
codons 247/248. Only 1 of 17 non-sun-exposed skin biopsies
had the CC to TT mutation; this mutation was found at
codons 247/248 when assayed by AS-LCR but was not found
in two trials of AS-PCR. The difference in prevalence of the
mutation detected by either method in either codon between
sun-exposed (13/17) and non-sun-exposed (1/ 12) sites was
statistically highly significant (P = 0.005); when at least one
assay gave a positive response, we judged that the sample had
a mutation. While most samples gave consistent responses
when assayed by PCR and LCR methods, or twice by the
same method, some samples gave inconsistent results. This
has been expected; since presumably very few cells with the
p53 gene mutation are present in each biopsy sample, the
mutant cells cannot be equally distributed to each assay tube.
Since 10 p.g of DNA corresponding to 1.5 x 106 cells was
used for each assay, we concluded that at least 1 in 1.5 x 106
cells contained these specific mutations in positive biopsy
samples.
Detecdoo of CC to TI p53 Mutation ill Human Normal Skin
Biopsies from FraDc:e aod Austnlla. Normal skin biopsies
from 9 Australian skin cancer patients (subjects Al-A9 in
Table 2) and from 17 volunteers living in France (subjects
Medical Sciences: Nakazawa et a/. Proc. Nat/. Acad. Sci. USA 91 (1994) 363
Table 1. Detection of CC to TT mutations in sun-exposed and non-sun-exposed normal skin on the same Australian skin clinic patients
CC to TT mutation in the p53 gene
Skin lesion Shoulder Buttock
Subject Sex Age, yr Type Site Codon 245 Codons 247/248 Codon 245 Codons 247/248
AlO F 56 BCC R. nasolabial fold P- P- L- P-,L-
All M 62 BCC L. nose P- P+ L- P-,L-
Al2 F 80 BCC Behind r. ear P+ P- L- P- ,L-
Al3 F 73 BCC R. inner canthus P- ,L- P+ L- P-,L-
Al4 M 83 BCC Scapula P-,L- P-,L- L- P-,L-
AIS M 73 BCC Scapula P+,L- P+,L+ L- P-,L-
Al6 M 58 sw Ear L- P-,L- L- P- ,L-
Al7 M 49 BCC Scapula L+ P- ,L- L- P-,L-
Al8 M 59 BCC Scapula L- P+,L- L- P- ,L-
Al9 M 44 BCC Loin L- P+,L+ L- P- ,L-
A20 M 71 BCC Scapula L- P+,L+ L- P-,L-
A21 M 71 BCC Temple L+ P- ,L- L- P- ,L-
A22 M 66 AK Neck L- P- ,L- L- P-,L-
A23 M 70 sec Forehead P+,L- P+,P+,L+ P- ,L- P- ,P- ,L+
A24 M 54 BCC Nose P-,L+ P+,P- ,L+ P-,L- P-,P-,L-
A2S M S4 BCC Upper arm P+,L+ P+,P- ,L- P- ,L- P-,P-,L-
A26 M 47 BCC Loin L- P+,L+ L- P-,L-
SW, seborrheic wart; AK, actinic keratosis; BCC, basal cell carcinoma; SCC, squamous cell carcinoma; R. (orr.), right; L. ,left. The following
detection methods were used: P, allele-specific PCR; L, allele-specific LCR.
Ll-L17) were analyzed for the presence of CC to TT
mutations in the p53 gene by AS-PCR. Four samples, all from
sun-exposed sites of Australian skin cancer patients, con-
tained the mutation at codons 247/248 and one of them also
showed the mutation at codon 245 (Fig. 4). When the same
samples were subjected to AS-LCR, the same results were
obtained for codon 245 for all samples, whereas two samples
(subjects A3 and A6) showed different responses as com-
pared to the PCR (fable 2). Altogether, 5 of6 (83%) biopsies
of sun-exposed sites in Australians showed the mutation in
one or the other assay. Three biopsies of non-sun-exposed
sites from Australian cancer patients (subjects A 7-A9) and 17
biopsies from volunteers living in Lyon, France (subjects
Ll-L17), gave negative responses (Table 2). Thus, all posi-
tive biopsies were taken from sun-exposed sites of the skin
from Australians, confmning a good association between CC
Table 2. Summary of detection of CC to TT mutations in normal
skin adjacent to or distant from the skin cancer in 9 Australian
skin cancer patients (Al-A9) and from 17 volunteers living in
Lyon (Ll-L17)
Sex
Subject (age, yr)
AI M(Sl)
A2 M(69)
A3 M(64)
A4 M(62)
AS F(61)
A6 F(41)
A7 M(63)
AS F(S7)
A9 M(67)
Ll,L2 M (33,
35)
L3- M/F
L17 (1-55)
Cancer
Type Site
BCC Back
BCC Back
BCC Tmp
BCC/ Neck
sec
BCC
sec
BCC
BCC
BCC
None
None
Clav
Back
R. ear
L. tmp
R. tmp
Mutation
Biopsy Codon Codons
site 245 247/ 248
Back P- ,L- P+,L+
Back P+,L+ P+,L+
Tmp P- ,L- P- ,L+
Neck P- ,L- P+,L+
Clav P- ,L- P-,L-
Back P- ,L- P+,L-
Abd P- P-
Btk P- P-
Hip P- P-
Arm P-,L- P- ,L-
Multi P- P-
R. , right; L. , left; Tmp, temple; Clav, clavicle; Abd, abdomen;
Btk, buttock; Arm, forearm; Multi, foreskin, leg, abdomen, face, and
forearm; BCC, basal cell carcinoma; SCC, squamous cell carcinoma.
The following detection methods were used: P, allele-specific PCR;
L, allele-specific LCR.
to TT mutation of the p53 gene and sun exposure history.
While no samples from volunteers living in France were
positive, the comparison of these results with those obtained
from Australians is hindered by the fact that the former were
young normal individuals, whereas the latter were older skin
cancer patients. The difference could be due to the extent of
sun exposure but might also be due to age or poorer DNA
repair among people predisposed to skin cancer (30).
DISCUSSION
We have developed methods to detect CC to TT tandem base
mutations at codons 245 and 247/248 of the p53 gene in
apparently normal human skin cells. The presumed mutant
bands at codons 245 and 247/248 were reamplified, cloned,
and sequenced. We confirmed that they were derived from
the correct regions of the p53 gene (data not shown). We
observed dose-dependent induction of CC to TT mutations
by UVB in cultured human skin cells and analysis of normal
skin biopsies showed a higher prevalence of these mutations
in skin from sun-exposed than from non-sun-exposed sites. It
may be, therefore, that measurement of these mutations in
normal skin will be a useful measure of biologically relevant
UV exposure and predictor of risk of UV -induced skin
cancer. It is important to emphasize that the p53 gene
mutations detected represent the outcome of a series of
codon 245 codon 24718
subJect 1 2 3 l 1 4 5 6 l2
No.
-
-
F1o. 4. Detection of CC to TT mutations in codons 245 (Left) and
247/248 (Right) of the p53 gene in normal skin biopsies by AS-PCR.
Six normal skin biopsies from Australian skin cancer patients and
two from normal volunteers living in France were collected and their
total DNA was analyzed for the presence of a CC to TT mutation at
codon 245 (Left) and at codons 247/248 (Right) of the p53 gene by
AS-PCR. Biopsy sites and other information on each subject are
listed in Table 2.
364 Medical Sciences: Nakazawa et al.
UV-cell interactions, including the extent of UV exposure,
immunological status, fidelity of DNA polymerization, and
genetic background of individuals such as skin types and
difference in DNA repair capacities (30, 31).
While our study clearly shows that UV -specific p53 gene
mutations can be detected in UV -exposed skin cells in vitro
and in vivo, and our methods can detect mutated cells in a
normal cell population, they may not provide a quantitative
guide to UV exposure. Since we used the p53 gene as the target
and mutated p53 genes may provide a growth advantage, a cell
population with p53 mutations may increase relative to those
with wild-type p53 genes. In fact, we observed these mutations
in UV -exposed cultured cells only after several passases,
suggesting that the mutated cells might be selectively recruited
in a cell population. In addition, we detected
CC to TT mutations at codon 245 or 247/248 in at least 1 per
1.5 x 106 cells from skin biopsy samples. This is an extremely
high mutation frequency and suggests that selective clonal
expansion of mutant cells may have occurred in vivo. How-
ever, it is important to emphasize that we measured "cumu-
lative" mutations in populations of human skin cells that had
been exposed to the sun for decades. If a nonlethal, UV-
induced mutation occurs in stem cell populations in the skin
such as basal keratinocytes or hair follicle cells, all subsequent
cell divisions will give rise to mutated progeny cells. Continu-
ing exposure to the sun would be expected to give rise to
further mutations and increase the prevalence of mutated cells
among all basal cells. If, in addition, the mutated cells have a
growth advantage (e.g., due to mutation of the p53 gene) the
prevalence of mutated cells would increase further. Further-
more, if we assume that a series of genetic alterations is
necessary for skin carcinogenesis, as in colon carcinogenesis,
the mutation frequency of each critical gene must be higher
than we generally assume (32). Thus, the mutation frequency
we observed in the p53 gene in normal skin of 41- to
old subjects (""' 10-
6
per base) may be reasonable. Ziegler et al.
(23) recently reported that 45% of the point mutations of the
p53 gene in basal cell carcinomas are accompanied by a second
point mutation on the other allele. Since these mutations are
UV-like, the results also suggest a high mutation frequency
induced by UV (23).
The CC sequence at codons 247/248 was more frequently
mutated to TT than that at codon 245 in both UV -exposed
cultured human skin cells and in skin biopsies. The cytosine
residue in the CC sequence at codons 247/ 248 is known to be
methylated (13) and deamination of the 5-methylcytosine may
result in aCto T single base mutation (33). This characteristic,
however, does not explain the high prevalence of CC to TT
tandem base mutations at codons 247/ 248. Another difference
between the two CC sequences is location on the transcribing
strand at codons 247/ 248 and on the nontranscribing strand at
codon 245. It has been shown that strand bias exists for repair
efficiency of UV -induced DNA damage and replication fidel-
ity, but no consistent data are available to suggest strand bias
for UV -induced mutations (34). Recently, Kress et al. (17) and
Kaqjilal et al. (24) showed that C to T or CC to TT mutations
of the p53 gene in UV -induced mouse skin tumors were all
located on the nontranscribed strand. It is interesting to note,
however, that the C to T or CC to TT mutations at codons
247/ 248 (transcribing strand) are more frequently detected
than those at codon 245 (nontranscribing strand) in human skin
tumors (13, 15, 16).
Skin cancer is one of the commonest human cancers and its
incidence is increasing (35). UVB irradiance at the surface of
the earth is almost certainly increasing as a result of depletion
of stratospheric ozone; this trend will lead to increases in skin
cancer incidence and other health effects ofUV radiation (36).
Further development of the methods descnbed in this paper
will assist in accurate quantitative measurement of the rela-
tionship between UV exposure and skin cancer incidence and
Proc. Natl. A.cad. Sci. USA. 91 (1994)
will, therefore, improve our capacity to predict the effects of
and respond appropriately to UV irradiance changes.
We an: grateful to Dr. R. Newbold (Brunei University, Uxbridge,
U.K.) for his advice and help and to Ms. C. Fuchez for her excellent
secretarial help. The work was partially supported by grants from the
European Community (EV5V-CT92-0096) and from the National
Institutes of Health (ROl-CA-40534).
1. Wopn, G. N. (1992) Environ. Health Perspect. 98, 167-178.
2. Albertini, R. J. , O' Neill, J.P., Heintz, N. H.&: Kelleher, P. (1985)
Naturt (Lo!UWn) 316, 369-371.
3. Yamasaki, H., Galiana, C. &: Nakazawa, H. (1992) in Cancer
Causation: Exploring New Frontiers, ed. Iversen, 0. H. (Taylor&:
Francis, Wasbinaton), pp. 153-166.
4. Voselstein, B. &: Kinzler, K. W. (1992) Nature (London) 355,
209- 210.
S. Balmain, A. &: Brown, K. (1988) Cancer Res. 51, 147- 182.
6. Dipple, A., Pigott, M., Moscbel, R. C. &: Costantino, N. (1983)
Cancer Res. 43, 4132-4135.
7. Abbott, P. J. &: SafthiU, R. (1979) Biochim. Biophys. Acta 562, 51.
8. Marion, M. J., Froment, 0 . &: Trepo, C. (1991) Mol. Carcinogenesis
4, 450-454.
9. Barbin, A., Tenenbaum, L., Toman, Z., Radman, M. &: Bartsch, H.
(1985) Murat. Res. 52, 157- 159.
10. Creech, J. L. &:Johnson, M. N. (1974)1. Occup. Med.I6, 1S0-1S1.
11. Hsu, I. C., Metcalf, R. A., Sun, T., Welsh, J . A., Wang, N. J . &:
Harris, C. C. (1991) Nature (London) 350, 427-428.
12. OztllB, M., Bressac, B., Puisieux, A., Kew, M. , Volkmann, M. et
al. (1991) Loncet 338, 1356- 1359.
13. Brash, D. E. , Rudolph, J. A. , Simon, J. A., Lin, A., McKenna,
G. J., Baden, H. P., Halperin, A. J . &: Ponten, J. (1991) Proc. Nat/.
Acad. Sci. USA 88, 10124-10128.
14. Seidman, M. M., Bredbera, A., Seetharam, S. &: Kraemer, K. H.
(1987) Proc. Natl. Acad. Sci. USA 84, 4944-4948.
15. PierceaU, W. E., Mukhopadhyay, T. , Goldbera, L. H. &: Anan-
thaswamy, H. N. (1991) Mol. Carcinogenesis 4, 445--449.
16. Somers, K. D., Merrick, M. A., Lopez, M. E. , Incognito, L. S. ,
Schechter, G. L. &: Casey, G. (1992) Canctr Res. 52, 5997-6000.
17. Kress, S. , Sutter, C. , Strickland, P. T., Mukhtar, H., Schweizer, J .
&: Schwarz, M. (1992) Cancer Res. 52, 6400-6403.
18. Nakazawa, H., Aauelon, A.-M. &: Yamasaki , H. (1991) Mol.
Carcinog. 3, 202-209.
19. Kumar, R., Sukumar, S. &: Baroacid, M. (1990) Science 248,
1101-1104.
20. Niltitin, A. Y., Ballerina, L.A. P . Lyons, J . &: Rajewsky, M. F.
(1991) Proc. Natl. Acad. Sci. USA 88,9939-9943.
21. International Agency for Research on Cancer (1992) /ARC Mono-
graph on the Evaluatwn of Carcinogenic Risk to HumoM: Solllr and
Ultraviolet Radiation (Int. Agency Res. Cancer, Lyon, France),
Vol. SS.
22. J .-P., Moyret, C., Guillot, B., Jeanteur, P., Guilbou, J .-J.,
Tbeillet, C. & Basset-Seguin, N. (1993) Oncogene 8, 583-588.
23. Ziesler, A.-M. , Left'eU, D. J., Kunala, S., Sharma, H. W., Gailani,
M., Simon, J . A., Halperin, A. J., Baden, H. P . Shapiro, P. E.,
Bale, A. E. &: Brash, D. E. (1993) Proc. Natl. Acad. Sci. USA 90,
4216-4220.
24. Ka!Uilal, S., PierceaU, W. E., Cummings, K. K., Kripke, M. L. &:
Ananthaswamy, H. N. (1993) Cancer Res. 53, 2961-2964.
2S. Reid, T. M. &: Loeb, L. A. (1992) Cancer Res. 52, 1082-1086.
26. Hawley-Nelson, P., Sullivan, J. E., Kung, M., Hennings, H. &:
Yuspa, S. H. (1980) /. Invest. Dermotol. 173, 176-182.
27. Suzuki, Y., Sekiya, T. &: Hayashi, K. (1991) Anal. Biochem. 192,
82-84.
28. D' Aquila, R. T., Bechtel, L. J., Videler, J . A. , Eron, J. J. , Gorc-
zyca, P. &: Kaplan, J. C. (1991) Nucleic Acids Res. 19, 3749.
29. TOrmAnen, V. T. &: Pfeifer, G. P. (1992) Oncogene 1, 1729- 1736.
30. Wei, Q., Matanoski, G. M., Farmer, E. R. , Hedayati, M. A. &:
Grossman, L. (1993) Proc. Natl. Acad. Sci. USA 90, 1614--1618.
31. Yuspa, S. H. &: l'tugosz, A. A. (1991) in Physiology, Biochemistry
and Molecular Biology of the Sldn, ed. Goldsmith, L. A. (Oxford
Univ. Press, Oxford, U.K.), 2nd Ed., pp. 1365-1402.
32. Loeb, L. A. (1991) Cancer Res. 51, 3075-3079.
33. Meuth, M. (1990) Biochim. Biophys. Acta 1032, 1-17.
34. Vrieling, H., van Rooijen, M. L. , Groen, N. A., Zdzienicka, M. Z. ,
Simons, J . W. I. M., Lohman, P. H. M. &: van Zeeland, A. A.
(1989) Mol. Cell. Bioi. 9, 1277-1283.
35. Kricker, A., Enslish, D. R., RandeU, P. L., Heenan, P. J., Clay,
C. D. , Delaney, T. A. &:Armstrong, B. K. (1990) Med.J.Aust.152,
399-407.
36. Kripke, M. L., Cox, P. A., Alas, L. G. &: Yarosh, D. B. (1992)
Proc. Natl. Acad. Sci. USA 89, 7516-7520.
ILL Document De livery
NLM -- W1 CL696 (Gen); E-Journal w/ ILL access
US FEDERAL TRADE COMMISSION (FTC)
US FEDERAL TRADE COMMISSION LIBRARY
600 PENNSYLVANIA AVENUE NW
WASHINGTON. DC 20580
ATTN:
PHONE: 202-326-2395
FAX: 202-326-2732
E-MAIL: library@ftc.gov
REG Copy
SUBMITTED:
PRINTED:
REQUEST NO.
SENT VIA:
DOCLINE NO.
OPAC NO.:
J o u r na l
TITLE: CLINICAL ENDOCRINOLOGY
2009-06-18
2009-06-18
REG-14 7154 82
DOCLINE
KYB-27277338
AAAOXNN
PUBLISHER/ PLACE:
VOLUME/ ISSUE:
DATE:
Blackwell Publishing. Oxford :
2009 Feb 18; [Epub ahead of print]
2009
REG-14715482
DCUDJG
10 :09:37
15 :50:33
AUTHOR OF ARTICLE: Pilz S;Dobnig H; Nijpel s G;Heine RJ;Stehouwer
CD;Snijder MB; v
TITLE OF ARTICLE:
ISSN:
OTHER NUMBERS/LETTERS:
SOURCE:
MAX COST:
COPYRIGHT COMP.:
CALL NUMBER:
NOTES :
DELIVERY:
REPLY:
VITAMIN D AND MORTALITY IN OLDER MEN AND WOMEN .
0300-0664
Unique ID.: 0346653
KYB-27277338
19226272
PubMed
$14.00
Guidelines
W1 CL696 (Gen); E-Journal w/ILL access
CG
E-mail Post to Web: library@ftc .gov
Mail:
KEEP THIS RECEIPT TO RECONCI LE WI TH BI LLI NG STATEMENT
For problems or q uest i o ns . co n t a c t NLM a t h t t p ://wwwc f . n l m. n i h .
gov/i l l / ill_web_form.cfm or p ho ne 301-496- 5 51 1 .
Include LIBI D a nd request numbe r .
NOTE : -THIS MATERIAL MAY BE PROTECTED BY COPYRI GHT LAW ( TI TLE 17 , U.S. CODE )
NLM Col l ect i o n Ac cess Sect i o n . Be t he s da . MD
Original Article; Received Date 01-Jun-08; Returned for Revision Date 03-Nov-08; Finally
Revised Date 09-Dec-08; Accepted Date 05-Jan-09
Vitamin D and mortality in older men and women
Short title: Vitamin D and mortality
Authors: Stefan Pilz
1
, Harald Dobnig
1
, Giel Nijpels
2
3
, Robert J. Heine
2
.4, Coen D.A.
Stehouwer
5
, Marieke B. Snijder
2
6
, Rob M. van Dam
7
, Jacqueline M. Dekker
2
1
Department oflntemal Medicine, Division of Endocrinology and Nuclear Medicine, Medical
University of Graz, Austria
2
EMGO Institute, VU University Medical Center, Amsterdam, The Netherlands
3
Department of General Practice, VU University Medical Center, Amsterdam, The
Netherlands
4
Department of Endocrinology, VU University Medical Center, Amsterdam, The Netherlands
5
Department of Internal Medicine, Maastricht University Medical Centre, Maastricht, The
Netherlands
6
Institute of Health Sciences, Faculty of Earth and Life Sciences, VU University Amsterdam,
The Netherlands
7
Department ofNutrition, Harvard School of Public Health; Channing Laboratory,
Department of Medicine, Brigham and Women's Hospital, and Harvard Medical School,
Boston, MA, USA
Correspondence:
Stefan Pilz
Department of Internal Medicine, Division of Endocrinology and Nuclear Medicine, Medical
University of Graz, Austria
Auenbruggerplatz 15, 8036 Graz, Austria
Tel: +43 650 9103667; Fax: +43 316 673216
Email: stefan.pilz@chello.at
Key words: Vitamin D, 25-hydroxyitamin D, mortality, epidemiology, cardiovascular events
This is an Accepted Article that has been peer-reviewed and approved for publication in the
Clinical Endocrinology, but has yet to undergo copy-editing and proof correction. Please cite
tbis article as an "Accepted Article"; doi: 10.1111/j.1365-2265.2009.03548.x
2
SUMMARY
Objective: Vitamin D deficiency is common among the elderly and may contribute to
cardiovascular disease. The aim of our study was to elucidate whether low serum levels of25-
hydroxyvitamin D (25(0H)D] are associated with an increased risk of all-cause and
cardiovascular mortality.
Design and patients: The 1-loom Study is a prospective population-based study among older
men and women.
Measurements: Fasting serum 25(0H)D was determined in 614 study participants at the
follow-up visit in 2000-200 I , the baseline for the present analysis. To account for sex
differences and seasonal variations of25(0H)D levels we formed sex-specific quartiles,
which were calculated from the 25(0H)D values of each season.
Results: After a mean follow-up period of 6.2 years, 51 study participants died including 20
deaths due to cardiovascular causes. Unadjusted Cox proportional hazard ratios (with 95%
confidence intervals) for all-cause and cardiovascular mortality in the first when compared to
the upper three 25(0H)D quartiles were 2.24 (1.28-3.92; p=0.005) and 4.78 (l.95-1 1.69;
p=O.OOl), respectively. After adjustment for age, sex, diabetes mellitus, smoking status,
arterial hypertension, high density lipoprotein cholesterol, glomerular filtration rate and waist
to hip ratio, the hazard ratios remained sigrtificant for all-cause [1.97 (1.08-3.58; p=0.027)]
and for cardiovascular mortality [5.38 (2.02-14.34; p=0.001)].
Conclusions: Low 25(0H)D levels are associated with all-cause mortality and even more
pronounced with cardiovascular mortality but it remains unclear whether vitamin D
deficiency is a cause or a consequence of a poor health status. Therefore, intervention studies
are warranted to evaluate whether vitamin D supplementation reduces mortality and
cardiovascular diseases.
3
INTRODUCTION
Older persons are prone to low vi tamin D concentrations because the capacity of the skin to
produce vitamin D decreases with aging and the time of sun exposure is often limited.
1
.2 This
results in reduced dermal synthesis of vitamin D, which can hardly be compensated by dietary
intake of vitamin D. !.2 The estimated worldwide prevalence of vitamin D deficiency among
the elderly of almost 50% underlines that adverse health effects attributed to vitamin D
deficiency may be important for public health? It is well known that vitamin D deficiency
contributes to a higher prevalence of fractures and falls and causes muscle weakness.
1
In
addition, there is growing evidence that vitamin D deficiency may increase the risk of cancer,
autoimmune diseases and infections_2.4-7 It was recently shown that low levels of25-
hydroxyvitamin D (25(0H)D] are also independently associated with cardiovascular events in
patients with hypertension suggesting a role of vitamin D for the maintenance of
cardiovascular health.
8
This hypothesis is further supported by the abiLity of vitamin D to
suppress the renin-angiotensin system (RAS).
9
Furthermore, accumulating evidence suggests
that vitamin D deficiency may contribute to myocardial dysfunction, arterial hypertension and
diabetes mellitus. t 0-
15
We have previously shown that low levels of25(0H)D are an independent risk factor
of total and cardiovascular mortality in a large cohort of patients referred to coronary
angiography.
16
These results are in line with a recent meta-analysis, in which a significant
reduction of all-cause mortality was reported for persons receiving vitamin D supplementation
that were largely derived from trials among frail elderly people with vitamin D deficiency.
17
In the present work, we aimed to address the largely unknown association between 25(0H)D
and cardiovascular as well as all-cause mortality in a population-based study. For this purpose
we measured 25(0H)D values, which are considered to be the best indicator of vitamin D
status,
1
4
in 614 study subjects from the population-based Hoorn Study which includes older
men and women that were followed with respect to total and cardiovascular mortality.
1
8-
20
Material may be protected by copyright law (Title 17, U.s. Code)
METHODS
Study population
4
The Hoom Study is a population-based cohort study on type 2 diabetes and cardiovascular
diseases.
18
Baseline measurements ofthe white population in Hoom, a medium-sized town in
the Netherlands, were performed between 1989 and 1992 and were described elsewhere.
18
The initial study cohort was a random sample from the population register of the municipal
registry of Hoom, and encompassed 2484 participants aged from 50 to 75 years. All subjects
with type 2 diabetes at a follow-up visit between 1996 and 1998 (n = 176) and a random
samples of the remaining study cohort ( n = 898), including 193 persons with impaired and
705 persons with normal glucose metabolism, were invited for a follow-up examination
between 2000 and 2001.
19
20
In this work we present the data of the 648 subjects (60.3% of the
invited individuals) who participated in that examination between 2000-2001.
19
"
20
The reasons
for not participating in the follow-up examination were lack of interest (30%), morbidity
(23%), advanced age (7%), unwillingness to travel (6%), participation considered too time
consuming (6%) and miscellaneous reasons (15%).
19
20
The Hoom study complied with the
Declaration of Helsinki. Written informed consent was obtained from all study subjects and
the Ethics Committee of the VU University Medical Centre approved the study.
Measur ements
Blood samples were taken after an overnight fasting. Serum 25(0H)D was measured by
means of a competitive binding protein assay (DiaSorin, Stillwater, MN, USA).
19
This assay
determines both, 25-hydroxyvitamin 0
2
and 25-hydroxyvitamin 0
3
and the interassay
coefficient of variation was 10-15% with slightly lower coefficient of variation at higher
25(0H)O levels.
19
Serum parathyroid hormone (PTH) was determined with an
immunoradiometric assay (lncstar Corp., Stillwater, MN, USA). Methods for routine
laboratory measurements were described previollsly.
20
We performed a 75-g ora.l-glucose
5
tolerance test in all study participants, except in those with capillary fasting whole blood
glucose levels 2: 8 mmoVL or in patients with known diabetes mellitus, who were already
treated with oral antidiabetics and/or insulin. Diabetes mellitus and impaired glucose
metabolism (persons with impaired glucose tolerance and/or impaired fasting glucose) were
classified according to the 1999 WHO criteria.
21
Glomerular filtration rate (GFR) was
calculated according to the abbreviated MDRD forrnula?
2
Systolic and diastolic blood
pressures were determined with a random-zero sphygmomanometer (Hawksley-Gelman,
Lancing, U.K.) at the right-upper arm after five minutes at rest and the average of two
measurements was used. Arterial hypertension was defined as systolic blood pressure 2: 140,
diastolic blood pressure 2: 90 or use of antihypertensive medication. Information about
smoldng status was evaluated by a self-administered questionnaire. Percentage of whole-body
fat was determined by whole-body dual-energy X-ray absorptiometry scan using fan beam
technology (QDR-2000, software version 7.200; Hologic, Brussels, Belgium).
19
Physical
activity, expressed in hours per day, was assessed by a validated questionnaire and included
sports, bicycling, gardening, walking, doing odd jobs and household work.
23
Prior
cardiovascular disease was defined as previously described as Minnesota Code 1.1-1.3, 4.1-
4.3, 5. 1-5.3, or 7.1 on the electrocardiogram, coronary bypass operation or angioplasty, and/or
peripheral arterial bypass or amputation.
24
Information on mortality was obtained from the municipal register of the city of
Hoom. Medical records of general practitioners and local hospitals were used to determine the
causes of death. Causes of death were classified according to the ninth edition of the
Classification of Diseases (ICD-9). Cardiovascular mortal ity was classified for deceased
patients with lCD codes 390-459 (diseases of the circulatory system) or ICD code 798
(sudden death). The follow-up time for all-cause mortality is defined as the time bet\:veen the
baseline examination (2000-2001) and the date of death or the censoring date (I July 2007).
For cardiovascular mo.rtality, the follow-up time is defined as the time between the baseline
6
examination (2000-2001) and the date of death due to cardiovascular causes or the censoring
date, which is either the end of the observational period (1 July 2007) or the date of death due
to non-cardiovascular causes.
For each season of the year (summer: 21 June till 22 September, autumn: 23 September till21
December, winter: 22 December till l9March and spring: 20 March till 20 June) we formed
sex-specific vitamin D quartiles that were calculated from the 25(0H)D concentrations of the
blood samples that were drawn within the respective season. We then combined the vitamin D
quartiles of each season and of both genders into a single variable [25(0H)D quartiles]. This
was done because of the recently described sex difference in 25(0H)D levels of our study,
19
and because of the well known seasonal variations of serum 25(0H)D levels that are mainly
attributed to circannual differences in sun exposure of the sk.in.
20
.2
5
-
27
Thus, for the estimation
of the long term vitamin D status of an individual it is mandatory to consider the season of
blood draw because recent data indicate that the same person has significantly higher
25(0H)D levels when blood is drawn in summer when compared to winter?
0
25
"
27
However,
25(0H)D levels tend to be consistent when measured 12 months apart in the same person.
27
It
can thus be assumed that those individuals in a specific vitamin D quartile are likely to remain
in that quartile throughout aU seasons of a year, despite significant circannual variations in
"absolute" 25(01-I)D concentrations.
27
Differences across 25(01-I)D quartiles were calculated by X
2
-test with P for linear by
linear test for categorical and with analysis of variance (ANOVA) with P for trend for
continuous parameters. Di fferences in overall survival and survival without fatal
cardiovascular events between the lowest (first) and the upper three quartiles were graphically
displayed using Kaplan-Meier curves. Previous reports are showing that the association
between low 25(0H)D and cardiovascular events is nonHnear with a steep increase of
7
cardiovascular risk at very low 25(0H)D levels.!! This prompted us to calculate Cox
proportional hazard ratios (HRs) for all-cause and cardiovascular mortality for the lowest
when compared to the upper three 25(0H)D quartiles. The upper three quartiles were used as
the reference group.
8
In addition to unadjusted HRs we present age- and sex-adjusted HRs in
model 1. In model2 we additionally adjusted for several cardiovascular risk factors including
dichotomous categorical variables for diabetes meUitus, smokers (ex- and active smokers
versus never smokers) and arterial hypertension, and continuous variables for high density
lipoprotein cholesterol, GFR and waist to hip ratio. Alternatively, we also included percentage
of body fat instead of waist to hip ratio in the covariate list ofmodel2 because we have
recently shown that percentage of body fat is strongly associated with a poor vitamin D status
and?
7
ln addition to all covariates of model 2 we adjusted for PTH in mode13, for habitual
physical activity in model 4, for albumin in model 5 and we calculated a multivariable
adjusted model (backward LR selection method) that includes all the above mentioned
covariates. Adjustment for PTH was performed because previous study results indicated that
secondary hyperparathyroidism is an important mediator of the deleterious effects of vitamin
D deficiency?
9
30
Habitual physical activity was included as a covariate because low physical
activity is known to be associated with increased mortality and vi tamin D deficiency and
might thus be an important confounder for the association between 25(0H)D and
mortality.
12
13
16
19
.3
1
However, the reduced physical activity associated with low 25(0H)D
levels may also be regarded as a consequence of vitamin D deficiency by considering that low
levels of 25(0H)D can cause muscle weakness. It is therefore debatable whether low physical
activity is rather a mediator or confounder of harmful effects of vitamin D deficiency.
Inclusion of serum albumin in the covariate list was done to adjust for a marker of
malnutrition that is also predictive for mortality.
32
To reduce possible confounding by prior
cardiovascular disease we additionally included this variable (prior cardiovascular disease) to
the covariate list of model 2 and calculated the respective HRs for all-cause and
8
cardiovascular mortality. All statistical tests were two-sided and a P-value below 0.05 was
considered statistically significant. All our statistical analyses were performed with SPSS 15.0
statistical package (SPSS Inc., Chicago, IL, USA).
RESULTS
Serum 25(0H)D levels were available in 614 study participants. Depending on the season of
blood draw, the mean serum 25(0H)D levels standard deviations in nmoVL were 51.4
18.3 in winter, 52.1 19.2 in spring, 59.7 20.5 in summer and 56.0 20.3 in autumn.
25(0H)D concentrations were higher in men (56.5 18.8 nmol/L) than in women (50.8
19.8 nrnol/L). Baseline characteristics according to sex-spedfic 25(0H)D quartiles, which are
based on the 25(0H)D values within each season, are presented in Table 1. Low levels of
25(0H)D were significantly associated with higher age, waist to hip ratio, percentage of body
fat, parathyroid honnone concentrations, and systolic and diastolic blood pressure, whereas
high density lipoprotein cholesterol, serum albumin, GFR and physical activity were
increased in groups with higher 25(0H)D levels (Table 1). Arterial hypertension and type 2
diabetes were associated with low serum 25(0H)D concentrations (Table 1).
After a mean follow-up time of 6.2 years, 5 J study participants (34 men and 17
women) had died, including 20 deaths (12 men and.8 women) due to cardiovascular causes.
We recorded 21 deaths (13.8%) in the lowest, and 30 deaths (6.5%) in the upper three
quartiles of25(0H)D (10 deaths (6.5%] in the second, 13 (8.6%] in the third and 7 [4.5%] in
the fourth quartile). Cardiovascular deaths were classified for 12 deceased persons (7.9%) in
the lowest, and 8 (1.7%) persons in the highest three 25(0H)D quartiles (3 deaths (1.9%] in
the second, 2 [1.3%] in the third and 3 (1.9%] in the fourth quartile). Kaplan-Meier curves
that show unadjusted data for all-cause and cardiovascular mortality according to 25(0H)D
quartiles (lowest versus highest three quartiles) are graphed in Figure 1a and 1 b. Cox
proportional HRs for all-cause and cardiovascular mortality according to 25(0H)D quartiles
9
are shown in Table 2. For the lowest 25(0H)D quartile, the unadjusted risk of all-cause and
cardiovascular mortality was significantly increased when compared to the highest three
25(0H)D quartiles. Age- and sex-adjusted HRs (with 95% CI) in the first when compared to
the upper three 25(0H)D quartiles were 1.70 (0.96-3.03; p=0.069) for all cause and 3.25
(1.30-8.15; p=0.012) for cardiovascular mortality. After multivariate adjustments, low
25(0H)D levels were significantly predictive for all-cause and cardiovascular mortality (see
Table 2). Inclusion of percentage ofbody fat instead of waist to hip ration in the covariate list
ofmodel3 in Table 2 resulted in a HR for all-cause mortality of2.08 (1.12-3.86; p=0.021)
and a HR for cardiovascular mortality of 7.1 1 (2.28-22.14; p=O.OOI). After adjustment for
prior cardiovascular disease in addition to all covariates in model2 in Table 2 the HRs for all-
cause and cardiovascular mortality were 1.91 (1.02-3.58; p=0.045) and 5.90 (1.93-18.07;
p=0.002), respectively. In a multivariable adjusted model, including all covariates of Table 2,
the HRs for all-cause and cardiovascular mortality remained significant with 1.84 (1.02-3.32;
p=0.043) and 4.61 (1.76-12.01; p=0.002), respectively.
DISCUSSION
In this study we found that low senun levels of 25(0H)D are prospectively associated with
all-cause and cardiovascular mortality in a selected sample of a population based study cohort
of older men and women.
Apart from the maintenance of muscular and skeletal health vitamin D may also
protect against cancer, infections, autoimmune and vascul ar diseases suggesting that vitamin
D deficiency might contribute to a reduced life expcctancy.
1
4
"
15
Towards this, it has already
been shown that low 25(0H)D levels are associated with an increased risk of all-cause
mortality in patients with renal failure and in patients scheduled for coronary angiography.
16
33
After adjustment for age and sex, low 25(0H)D levels were also predictive for mortality in
very frail older people residing in hostels and nursing homes, but this association was no
10
longer significant after multivariate adjustments including comorbidities, nutritional status
and renal function.
28
The association between serum 25(0H)D levels and all-cause mortality
was also addressed by the Longitudinal Aging Study Amsterdam (LASA), which included
1260 communi ty-dwelling persons aged 65 years and older at baseline.
34
ln that study, low
vitamin D status was a significant predictor of mortality after adjustments for possible
confounders such as age, sex, creatinine, cognitive status, depressive symptoms,
comorbidities and lifestyle variables, but significance was lost after additional adjustments for
frailty indicators (mobility performance, low serum albumin, and low serum total cholesterol
concentrations).
34
Importantly, the investigators of the LASA concluded that the frailty
indicators, which they included as covariates in their mortality analyses might hypothetically
rather be mediators than confounders of harmful effects of vitamin D deficiency, and it is
therefore conceivable that they "overadjusted" thei r analyses.
34
Results from the Third
National Health and Nutrition Examination Survey (NHANES-Ill) which were published
during the revision of our present work, showed for the general US population that after
multivariable adjustments for possible confounders the risk of all-cause mortality was
significantly reduced by 26% in the highest versus the lowest 25(0H)D quartile.
35
Our results
concerning the association betvveen 25(0H)D and mortality are well in line with and extend
the findings of the LASA and NHANES-lli and further support the notion that vitamin D
deficiency is a risk factor for mortality.
The association between vitamin D deficiency and a higher risk of fatal cardiovascular
events was highly significant (Table 2) in our study. There was also a statistically non-
significant 20% reduction of cardiovascular mortality in the highest versus the lowest
25(0H)D quartile in NHAtmS-ill.
35
This associations between cardiovascular mortality
(diseases) and 25(0H)D levels was less significant in NHANES-1ll than in previous studies
and in our present work, which might be attributed to differences in study populations and the
way of adjustments for seasonal 25(0H)D variations.
8
16
35
Considering that other studies
11
showed a steep increase of cardiovascular risk at very low 25(0H)D concentrations,
8
16
it
might be speculated that the relatively high 25(0H)D levels of the study participants of
NHANES-III might have limited the chance to detect a more significant association of
25(0H)D levels and fatal cardiovascular events in that study. Concerning possible
explanations for the link between vitamin D deficiency and cardiovascular diseases it has
been shown that low 25(0H)D levels might contribute to arterial hypertension and diabetes
mellitus, which is consistent with the high prevalence of hypertensive and diabetic patients in
the first 25(0H)D quartile of our study (Table 1).
1
0-IS Interestingly, vitamin D deficiency
remained a significant predictor for fatal cardiovascular events in our study even after
adjustments for common cardiovascular risk factors suggesting that possible harmful effects
of low 25(0H)D levels on the cardiovascular system are independent of an involvement of
vitamin D deficiency in the pathogenesis of diabetes mellitus and arterial bypertension.
12
"
15
There exists growing evidence that low serum 25(0H)D levels contribute to heart failure and
it was shown that vitamin D treatment was associated with improved diastolic function and a
regression of myocardial hypertrophy in hemodialysis patients.
10
11
36
.3
7
Carotid intima-media
thickness was also found to be inversely and independently correlated with serum 25(0H)D
levels and recent data from NHANES-III showed that low serum 25(0H)D concentrations are
associated with a higher prevalence of peripheral arterial disease .
3
8-
39
Furthermore, results
from the Framingham Offspring Study showed that patients with 25(0H)D levels below 15
ng/ml (37 .5 nmol/L) were at increased risk of incident cardiovascular events, even after
adjustments for conventional cardiovascular risk factors.
8
We cannot prove causality for the
relationship between 25(0H)D and fatal cardiovascular events, but our results and those of
the Framingham Offspring Study point to the urgent need for interventional trials to further
evaluate whether vitamin D supplementation protects against cardiovascular diseases.
1
40
In
this context, we have to acknowledge that in the Women's Health Initiative (Will) a daily
intake of 400IU vitamin D and 1000 mg calcium carbonate did not significantly reduce the
12
risk of cardiovascular events,
41
but the results of the WID do not necessarily argue against a
possible reduction of incident cardiovascular disease by vitamin D supplementation because
the 400IU of vitamin D used in the WHJ are generally considered to be too low to adequately
treat and prevent vitamin D deficiency.
8
42
-4
4
Furthermore, we have to note, that a recent meta-
analysis in patients with chronic kidney disease failed to demonstrate a significant beneficial
effect of vitamin D (compound) treatment on biochemical markers such as PTH.
45
This
finding contrasts the current recommendations for vitamin D treatment of patients with
chronic kidney disease.
45
However, relevant data form randomised controlled trials on
outcomes such as cardiovascular events or mortality, for which observational studies strongly
suggest favourable effects of vitamin D treatment in chronic kidney disease, are still
missing.
45
-4
8
Our results are limited by the relatively low number of fatal events with subsequent
high confidence intervals of our HRs (Table 2), thus warranting further studies with larger
study cohorts to confirm and extend our findings. Despite adjustments for various potential
confounders, we cannot mle out that low serum 25(0H)D levels are only a non-specific
marker for a high risk of mortality which is confounded by other unconsidered or unmeasured
factors. lt is, however, important for us to point out that our analyses were adjusted for
physical act ivity which reduces the chance that limited mobility in persons with poor health is
a main confounder for our results. Another limitation is that om study participants had only a
single determination of25(0H)D levels and not serial measurements that would provide a
more reliable estimate of the long-term vitamin D status. Furthermore, our assumption made
for the seasonal adjustments that a person within a certain vitamin D quartile during e.g.
summer would also be within this quartile throughout the whole year is not necessarily
validated. It should also be considered that the present study cohort is a selected sample of the
initial population based cohort of the Hoom study thereby possibly limiting the
generalizability of our results.
Material may be protected by copyright law (Titl e 17, U. S. Code)
13
In summary we have shown that low serum levels of 25(0H)D arc prospectively
associated wi th all-cause and cardiovascular mortality in older men and women. Our results
provide a rationale for future studies to test whether vitamin D supplementation reduces
mortality and/or cardiovascular diseases in persons with vitamin D deficiency. These studies
are urgently needed to answer the question whether vitamin D deficiency is a cause or a
consequence of a poor health status.
Acknowledgements
This work was supported by the Netherlands Organization for Scientific Research (ZonMw
VENI grant no. 916.46.077.
Competing interests/financial disclosure
Nothing to declare.
REFERENCES
1. Mosekildc L. (2005) Vitamin D and the elderly. Clinical Endocrinology, 62, 265-281.
2. Holick MF. (2007) Vitamin D deficiency. New England Journal of Medicine, 357, 266-
281.
14
3. Nonnan AW, Bouillon R, Whiting SJ, Vieth R & Lips P. (2007) 131h Workshop consensus
for vitamin D nutritional guidelines. Journal of Steroid Biochemistry & Molecular
Biology, 103, 204-205.
4. Peterlik M & Cross HS. (2006) Dysfunct ion of the vitamin D endocrine system as
common cause for multiple malignant and other chronic diseases. Anticancer Research,
26, 2581-2588.
5. Lappe JM, Travers-Gustafson D, Recker RR & Heaney RP. (2007) Vitamin D and
calcium supplementation reduces cancer risk: results of a randomized trial. American
Journal of Clinical Nutrition, 85, 1586-1591.
6. Pilz S, Dobnig H, Winklhofer-Roob B, Riedmtiller G, Fischer JE, Seelhorst U, Wellnitz
B, Boehm BO & Marz W. (2008) Low serum levels of25-hydroxyvitamin D predict fatal
cancer in patients referred to coronary angiography. Cancer Epidemiology Biomarkers &
Prevention, 17, 1228-1233.
7. Zittermann A. (2003) Vitamin Din preventive medicine: are we ignoring the evidence?
British Journal of Nutrition, 89, 552-572.
8. Wang TJ, Pencina MJ, Booth SL, Jacques PF, Ingelsson E, Lanier K, Benjamin EJ,
D'Agostino RB, WolfM & Vasan RS. (2008) Vitamin D deficiency and risk of
cardiovascular disease. Circulation, 111, 503-511.
9. Li YC, Qiao G, Uskokovic M, Xiang W, Zheng W & Kong J. (2004) Vitamin D: a
negative regulator of the rennin-angiotensin system and blood pressure. Journal of Steroid
Biochemistry & Molecular Biology, 89-90, 87-392.
I 0. Zittermann A, Schleithoff S & Koerfer R. Vitamin D insufficiency in congestive heart
failure: Why and what to do about it? (2006) Heart Failure Reviews, 11, 25-33.
11. Pilz S, Man W, Wellnitz B, Seelhorst U, Fahrleitner-Pammer A, Dimai HP, Boehm BO
& Dobnig H. (2008) Association of vitamin D deficiency with heart failure and sudden
cardiac death in a large cross-sectional study of patients referred for coronary
angiography. J Clin Endocrinol Metab, 93, 3927-3935.
15
12. Forman JP, Giovannucci E, Holmes MD, Bischoff-Ferrari HA, Tworoger SS, Willett WC
& Curhan GC. (2007) Plasma 25-hydroxyvitamin D levels and risk of incident
hypertension. Hypertension, 49, 1063-1069.
13. Snijder MB, Lips P, Seidell JC, Visser M, Deeg DJ, Dekker JM & van Dam RM. (2007)
Vitamin D status and parathyroid hormone levels in relation to blood pressure: a
population-based study in older men and women. Journal of Internal Medicine, 261, 558-
565.
14. Pittas AG, Lau J, Hu FB & Dawson-Hughes B. (2007) The role of vitamin D and calcium
in type 2 diabetes. A review and meta-analysis. Journal of Clinical Endocrinology and
Metabolism, 92,2017-2029.
15. Need AG, O'Loughlin PD, Horowitz M & Nordin BE. (2005) Relationship between
fasting serum glucose, age, body mass index and serum 25 hydroxyvitamin D in
postmenopausal women. Clinical Endocrinology, 62, 738-741.
16. Dobnig H, Pilz S, Scharnagl H, Renner W, Seelhorst U, Wellnitz B, Kinkeldei J, Boehm
BO, Weihrauch G & Maerz W. (2008) independent association of low serum 25-
hydroxyvitamin D and 1,25-dihydroxyvitamin D levels with all-cause and cardiovascular
mortality. Arch Intern Med, 168, 1340-1349.
17. Autier P, Gandini S. (2007) Vitamin D supplementation and total mortality: a meta-
analysis of randomized controlled trials. Archives of Internal Medicine, 167, 1730-1737.
18. Mooy JM, Grootenhuis PA, de Vries H, Valkenburg HA, Bouter LM, Kostense PJ &
Heine RJ. (1995) Prevalence and determinants of glucose intolerance in a Dutch
Caucasian population. The Hoorn Study. Diabetes Care, 18, 1270-1273.
16
19. Van Dam RM, Snijder MB, Dekker JM, Stehouwer CD, Bouter LM, Heine RJ & Lips P.
(2007) Potentially modifiable determinants of vitamin D status in an older population in
the Netherlands: the Hoorn Study. American Journal of Clinical Nutrition, 85, 755-761.
20. Bos G, Dekker JM, Feskens EJ, Ocke MC, Nijpels G, Stehouwer CD, Bouter LM, Heine
RJ & Jansen H. (2005) Interactions of dietary fat intake and the hepatic lipase 4 8 0 C ~ T
polymorphism in determining hepatic lipase activity: the Hoorn Study. American Journal
of Clinical Nutrition, 81, 911 -915.
21. Gahir MM, Hanson RL, Dabela D, Imperatore G, Roumain J, Bennett PH & Knowler
WC. (2000) The 1997 American Diabetes Association and 1999 World .Health
Organization criteria for hyperglycemia and prediction of diabetes. Diabetes Care, 23,
1108-1112.
22. Levey AS, Coresh J, BalkE, Kausz AT, Levin A, Steffes MW, Hogg RJ, Perrone RD,
Lau J & Eknoyan G. (2003) Natjonal Kidney Foundation practice guidelines for kidney
disease: Evaluation, classification, and stratification. Annals of Internal Medicine, 139,
137-147.
23. Wendei-Vos GC, Schuit AJ, Saris WH & Kromhout D. (2003) Reproducibility and
reactive validity of the Short Questionnaire to Assess Health-enhancing physical activi ty.
Journal of Clinical Epidemiology, 56, 1163-1169.
24. Henry RM, Kamp 0, Kostense PJ, Spijkerman AM, Dekker JM, van Eijck R, Nijpels G,
Heine RJ, Bouter LM, Stehouwer CD. (2004) Lefi ventricular mass increases with
deteriorating g]ucose tolerance, especially in women: independence of increased arterial
stiffness or decreased flow-mediated dilation. Diabetes Care, 27, 522-529.
17
25. Bolland MJ, Grey AB, Ames RW, Mason BH, Home AM, Gamble GD & Reid lR. (2007)
The effects of seasonal variation of 25-hydroxyvitamin D and fat mass on a diagnosis of
vitamin D sufficiency. American Journal of Clinical Nutrition, 86, 959-964.
26. Levis S, Gomez A, Jimenez C, Veras L, Ma F, Shenghan L, Hollis B & Roos BA. (2005)
Vitamin D deficiency and seasonal variation in an adult South Florida population. Journal
ofClinical Endocrinology and Metabolism, 90, 1557-1562.
27. Harris SS, Dawson-Hughes B. (1998) Seasonal changes in plasma 25-hydroxyvitamin D
concentrations of young American black and white women. American Journal of Clinical
Nutrition, 61, 1232-1236.
28. Snijder MB, van Dam RM, Visser M, Deeg OJ, Dekker JM, Bouter LM, Seidell JC &
Lips P. (2005) Adiposity in relation to vitamin D status and parathyroid hormone levels: a
population-based study in older men and women. J Clin Endocrinol Metab, 90, 4119-
4123.
29. Sambrook PN, Chen SJ, March LM, Cameron lD, Cuming RG, Lord SR, Schwarz J &
Seibel MJ. (2004) Serum parathyroid hormone is associated with increased mortality
independent of25-hydroxyvitamin D status, bone mass, and renal function in the frail and
very old: a cohort study. Journal of Clinical Endocrinology and Metabolism, 89, 5477-
5481.
30. Chen SJ, Sambrook PN, March L, Cameron ID, Cumming RG, Simpson JM & Seidel MJ.
(2008) Hypovitaminosis D and parathyroid hormone response in the elderly: effects on
bone turnover and mortality. Clinical Endocrinology, 68, 290-298.
31. Monini TM, Everhart JE, Patel KV, Schaller DA, Colbert LH, Visser M, Tylavsky F,
Bauer DC, Goodpaster BH & Harris TB. (2006) Daily activity energy expenditure and
mortality among older adults. Journal of the American Medical Association, 296, 71-179.
18
32. Alberda C, Graf A & McCargar L. (2006) Malnutrition: etiology, consequences, and
assessment of a patient at risk. Best Practice & Research: Clinical Gastroenterology, 20,
419-439.
33. Wolf M, Shah A, Gutierrez 0 , Ankers E, Monroy M, Tamez I-I, SteeleD, Chang Y,
Camargo CA Jr, ToneiJi M & Thadhani R. (2007) Vitamin D levels and early mortality
among incident hemodialysis patients. Kidney international, 12, 909-911.
34. Visser M, Deeg DJ, Puts MT, Seidell JC & Lips P. (2006) Low serum concentrations of
25-hydroxyvitamin D in older persons and the risk of nursing home admission. American
Journal of Clinical Nutrition, 84, 616-622.
35. Melamed ML, Michos E, Post W, Astor B. {2008) 25-Hydroxyvitamin D levels and the
risk of mortality in the general population. Arch intern Med, 168, 1629-1637.
36. Bodyak N, Ayus JC, Achinger S, Shivalingappa V, Ke Q, Chen YS, Rigor DL, Stillman I,
Tamez H, Kroeger PE, Wu-Wong RR, Karumanchi SA, Thadhani R & Kang PM. (2007)
Activated vitamin D attenuates left ventricular abnormalities induced by dietary sodium in
Dahl salt-sensitive animals. Proceedings of the National Academy of Science, 104, 16810-
16815.
37. Kim HW, Park CW, Shin YS, Kim YS, Shin SJ, Kim YS, Choi EJ, Chang YS & Bang
BK. (2006) Calcitriol regresses cardiac hypertrophy and QT dispersion in secondary
hyperparathyroidism on hemodialysis. Nephron Clinical Practice, 102, 2 1-29.
38. Targher G, Bertolini L, Padovani R, Zenori L, Scala L, Cigolini M & Arcaro G. (2006)
Serum 25-hydroxyvitamin D3 concentrations and carotid intima-media thickness among
type 2 diabetic patients. Clinical Endocrinology, 65, 593-597.
39. Melamed ML, Muntner P, Michos ED, Uribani J, Weber C, Sharma J & Raggi P. (2008)
Serum 25-hydroxyvitamin D levels and prevalence of peripheral arterial disease. Results
from NHNAES 2001 to 2004. Arteriosclerosis, Thrombosis, and Vascular Biology, 28,
1179-85.
19
40. Hathcock JN, Shao A, Vieth R & Heaney R. (2007) Risk assessment for vitamin D.
American Journal ofCUnical Nutrition, 85, 6-18.
41. Hsia J, Heiss G, Ren H, Allison M, Dolan NC, Greenland P, Heckbert SR, Johnson KC,
Manson JE, Sidney S & Trevisan M for the Women's Health Initiative Investigators.
(2007) Calcium/vitamin D supplementation and cardiovascular events. Circulation, 115,
846-854.
42. Wallis DE & Penckofer S. (2007) Letter by Wallis and Penckofer regarding article,
"Calcium/vitamin D supplementation and cardiovascular events". Circulation, 116, 86.
43. Giovannucci E, Liu Y, Rimm EB, Hollis BW, Fuchs CS, Stampfer MJ & Willett WC.
(2006) Prospective study of predictors of vitamin D status and cancer incidence and
mortality in men. Journal of the National Cancer Institute, 98,451-459.
44. Bischoff-Ferrari H, Giovannucci E, Willett WC, Dietrich T, Dawson-Hughes B. (2006)
Estimation of optin1al serum concentrations of 25-bydroxyvitamin D for multiple health
outcomes. American Journal of Clinical Nutrition, 84, 18-28.
45. Palmer SC, McGregor DO, Macaskill P, Craig JC, Elder GJ & Strippoli GF. (2007) Meta-
analysis: Vitamin D compounds in chronic kidney disease. Ann intern Med, 147, 840-853.
46. Tonelli M. (2007) Vitamin Din patients with chronic kidney disease: nothing new under
the sun. Ann intern Med, 147, 880-881.
47. Kovcsdy CP, Ahmadzadeh S, Anderson JE, Kalantar-Zadeh K. (2008) Association of
activated vitamin D treatment and mortality in chronic kidney disease. Arch Intern Med,
168, 397-403.
48. Shoben AB, Rudser KD, deBoer IH, Young B, Kestenbaum B. (2008) Association of oral
calcitriol with improved survival in nondialyzed CKD. JAm Soc Nephrol, 19, 1613-1619.
20
Table 1. Baseline characteristics according to sex-specific 25(0H)D quartiles
Variable 1st quartile 2"" quartile 3rd q1.1artile 4th quartile P-value
Numbers 152 155 151 !56
25(0H)D (nmoVL) 30.6 ;;!: 6.9 45.6 6.3 58.3 :!: 6.4 78.9:!: 11.5 <0.001
Age (years) 72.9 :!: 6.7 69.9 6.6 69.2 6.5 67.1:!: 5. 1 <0.001
Females (%) 50.7 51.0 51. 0 50.0 0.912
B!vfi (kg/m
1
) 27.5 :!: 4.6 27.4 4.0 27.5 :!: 3.9 26.6 3.4 0.073
Waist to hip ratio 0.93 0.09 0.93 0.09 0.93:!: 0.09 0.90 :::0.09 0.013
Percentage body fat(%) 36.2 10.0 34.9 9.7 35.0 8.8 33.4 :!: 9.3 0.01 6
SBP (mmHG) 148 22 141:!: 20 140 21 139 19 <0.001
DBP (mmHG) 85 12 82 II 82 11 81 9 0.017
Arterial hypertension (%) 76.2 69.7 67.3 60.6 0.004
Fasting glucose (mmoVL) 6.24 1.34 6.03 :!: 0.95 6.26 1.81 5.97 1.10 0.182
HbAic(%) 6.0 0.7 6.0 0.5 6.0 0.8 5.90.6 0.296
Diabetes mellitus (%) 28.7 19.9 20.0 17.4 0.024
Impaired glucose tolerance (%) 28.7 34.4 33.3 26.5 0.629
Triglycerides (mmol/L) 1.5 0.9 1.5 0.7 1.5:1: 0. 8 1.4:1: 0.6 0.095
LDL cholesterol (mmol/L) 3.5 :1: 1.0 3.6 :1:0.9 3.7 0.9 3.7 ::: 0.8 0.117
HDL cholesterol (mmoi!L) 1.38:!: 0.43 1.37:1: 0.37 1.45 :1:0.40 1.53 :1: 0.45 0.001
Serum al bumin (giL) 40.7:!: 3.4 41.3 :1: 2.6 41.6:1:2.5 41.6 2.7 0.003
Smokers (ex and active)(%) 55.0 62.6 62.7 61.5 0.278
PTH (pmoi/L) 6.93 :1:2.91 6.33 :!: 2.29 6.18:1:3.1 1 5.24 1.60 <0.001
GFR (ml/minll. 73m
2
) 64.8:!: 11.0 63.8 10.4 62.3 :1:9.5 62.5 9.5 0.029
Prior CVD (%) 51.7 48.7 48.3 41.6 0.089
Physical activity (hours/day) 2.8 2.2 3.2 2.6 3.2 2.3 3.5 :!: 2. 7 0.008
25(0H)D, 25-hydroxyvitnrnin D; BMI, body mass index; SBP, systol ic blood pressure; DBP, diastolic
blood pressure; HbA 1 c, hemoglobin A 1 c; LDL, low density lipoprotein; HDL, high density lipoprotein;
PTH, parathyroid hormone; GFR, glomerular filtrati on rate; CVD, cardiovascular disease
Data are presented as percentages and as means standard deviations
Differences between groups were assessed by ;(-test and Analysis of Variance (A NOVA) with P for
trend
Materi al may be protected by copyright law (Title 17, U.S. Code)
Table 2. Cox proportional hazard ratios (with 95% CI) for all-cause and cardiovascular mortality
in the first when compared to the upper three sex-specific 25(0H)D quartiles
AU-cause mortality Cardiovascular mortality
Unadjusted
Upper three quartilcs 1.00 reference 1.00 reference
I st quartile 2.24 (1.28-3.92; p=0.005) 4.78 (1.95-11.69; p=O.OOI)
Modell
Upper three quartiles 1.00 reference 1.00 reference
1st quartile 1.70 (0.96-3.03; p=O.OQ9) 3.25 ( 1.30-8.15; p=O.O 12)
Model2
1st quartile 1.97 ( 1.08-3.58; p=0.027) 5.38 {2.02- 14.34; p=0.001)
Model 3
1st quartile 1.90 ( 1. 04-3.48; p=0.037) 4.70 (1.75- 12.62; p=0.002)
Model4
Upper three quarti les 1.00 reference 1.00 reference
1st quartile 1.93 (1.06-3.5 1; p=0.032) 5.02 (1.88-13.42; p=O. OOJ)
Model S
Upper three quartiles 1. 00 reference 1.00 reference
lst quarti le 1.88 (1.02-3.44; p=0.042) 5.33 ( 1. 97-14.45; p=0.001)
Model 1: Adjusted for age and sex
Model 2: Additionally adjusted for diabetes mellitus, ex- and active smokers, arterial hypertension,
high density lipoprotein cholesterol, glomerular filtration rate and waist to hip rat io
Model 3: Adj\ISted for all covariates from model 2 plus parathyroid hormone
Model 4: Adjusted for all covariates from model 2 plus physical activity
Model 5: Adjusted for all covariates from model 2 plus albumin
21
Figure Legend
Figure la: Kaplan Meier curve for all-cause mortality in the flrst and the upper three
25(0H)D quartiles.
Figure l b: Kaplan Meier curve for cardiovascular mortality in the first and the upper three
25(0H)D quartiles.
22
0,99
0,96
ns
~ 0,93
~
:::l
(/)
0,90
0,87
23
~ ~ t thr:: 25(0 H)O quartiles.-
0 2 4 6 8
Years of follow-up
1,00
0,98
_ 0,96
C'O
.:::
2:
:::::1
en 0,94
0,92
0,90
24
Highest three 25(0H)D quartiles
0 2 4 6 8
Years of follow-up
'

..

I
Addressing the health benefits and risks, involving
vitamin D or skin cancer, of increased sun exposure
Johan Moan*t*, Alina Carmen Porojnicu*, Arne Dahlbackt, and Richard B. Setlow*
*Department of Radiation Biology, Institute for Cancer Research, Montebello, 0310 Oslo, Norway; tDepartment of Physics, University of Oslo,
0316 Oslo, Norway; and Biology Department, Brookhaven National Laboratory, Upton, NY 11973-5000
Contributed by Richard B. Setlow, November 13, 2007 (sent for review September 5, 2007)
Solar radiation is the main cause of skin cancers. However, it also
is a main source of vitamin D for humans. Because the optimal
status of vitamin D protects against internal cancers and a number
of other diseases, a controversy exists: Will increased sun exposure
lead to net health benefits or risks? We calculated the relative yield
of vitamin D photosynthesis as a function of latitude with a
radiative transfer model and cylinder geometry for the human skin
surface. The annual yield of vitamin D is 3.4 and 4.8 times larger
below the equator than in the U.K. and Scandinavia, respectively.
In populations with similar skin types, there are dear latitude
gradients of all major forms of skin cancer, indicating a north-
south gradient in real sun exposure. Surprisingly, the incidence
rates of major internal cancers also increase from north to south.
However, the survival prognosis also improves significantly from
north to south. Reasons for these findings are discussed in view of
the role of vitamin D. In Norway, melanoma rates increased by a
factor of 6 from 1960 to 1990, while the prognosis improved in the
same period. After 1990, melanoma rates have remained constant
or even decreased in age groups <SO years, whereas the prognosis
has not improved further. These data, together with those for
internal cancers and the beneficial effects of an optimal vitamin 0
status, indicate that increased sun exposure may lead to improved
cancer prognosis and, possibly, give more positive than adverse
health effects.
body mass index I cutaneous malignant melanoma I squamous cell
carcinoma I ultraviolet radiation
T
here is a controversy as to whether increased sun exposure
to Western populations would prolong or shorten lifetime
expectancy, result in fewer or more cancer deaths, and, in
general, lead to health benefits or risks (1, 2). For years,
emphasis has been placed on the increasing time trends of
incidence and mortality rates of cutaneous malignant melanoma
(CMM) (3, 4) and, in contrast, on the protective role of vitamin
D regarding many types of internal cancer and other diseases
(5- 7). Too much sun exposure has been blamed for the high and
increasing incidence rates of CMM. However, solar radiation is
a major, if not the main, source of vitamin D in humans.
Therefore, a population' s increased sun exposure leads to im-
proved vit amin D status. The observation that the incidence and
mortality of several types of internal cancers decreases with
decreasing latitude in the United States and other countries
initiated the research on vitamin D- cancer relationships in the
1980s and 1990s. However, in some cases, there is an inverse
gradient of the rates of internal cancer with latitude (1), with the
rates being higher in regions with high annual UV fluences (New
Zealand and Australia) than in countries with low annual UV
fluences (Northern Europe, Scandinavia, and the U. K.), despite
the fact that the populations of these regions are closely related
genetically or, at least, have similar skin types, which is important
for the photosynthesis of vitamin D.
These issues have health consequences far beyond those of
cancer because a number of diseases are associated with inad-
equate vitamin D levels or low sun exposure: neurological,
cardiovascular, metabolic, immune, and bone diseases (2, 7).
668- 673 I PNAS I January 15, 2008 I vol. 105 I no. 2
Evolutionary arguments involving skin color also should be
taken into account. A dark skin color is found among Africans
and, possibly, early hominids who lived close to the equator (8).
This pigment may protect against skin cancer and folate photo-
degradation (8, 9). A white skin color developed later in our
history, as humans left Africa and went north. Because dark skin
needs about 6 times higher solar exposure for vitamin D
photosynthesis than white skin (10, 11) and because the fluence
rate of vitamin D-generating solar radiation decreases with
increasing latitude (Fig. lA), one can argue that skin whitening
may be related to the need for vitamin D and the lack of sunshine
at high latitudes.
Results and Discussion
Is CMM Caused by Solar Radiation? Because the mortality rates of
CMM are much higher than those of nonmelanoma skin cancer
(in some populations, more than a factor of 10 higher), this
problem is the most important one to solve regarding the
negative consequences of sun exposure. The solution is by no
means certain yet. A number of investigators disagree, as we
reviewed earlier (12, 13). The main arguments against the
concept that sun exposure causes CMM are that: (i) CMM is
more common among persons with indoor work than among
those people with outdoor work ('14, 15); (ii) in younger gener-
ations, more CMMs arise per unit skin area on partly shielded
areas (trunk and legs) than on face and neck (16); and (iii)
CMMs sometimes arise on totally shielded areas (acral CMM
and uveal melanomas). Although the connection between these
melanoma types and sun exposure is controversial (17-19), their
inclusion in the present discussion is justified because of the
possible involvement of vitamin D.
However, in our opinion, a significant fraction of CMMs is
related to sun exposure (16, 20). The main arguments for this
relationship are: (i) the north- south gradients in CMM incidence
between Scandinavia and Australia (16), (ii) before the advent
of the " top-less" fashion, few women developed CMM on the
breast area (13, 16), and (iii) in some animals (Sinclair swine,
Monodelphis domestica, the fish Xiphophorus, white horses,
angora goats, transgenic mice, etc.) UV exposure leads to CMM
(16). The reason that CMM incidence rates decrease with
decreasing latitude in Europe is likely because of differences in
skin color from region to region.
Seasonal Variations of the Vitamin D Status. As shown in Fig. 1B
(21- 30), a pronounced seasonal variation is evident in most of
the published investigations on 25(0H)D (the serum marker
of vitamin D status). Summer values can be > 100% larger than
winter values. In Troms0, Norway, at 700N, people have a
higher intake of vitamin D (mainly from cod liver) in the cod
Author contributions: J.M. designed research; A.C.P. and A.D. performed research; J.M .
A.C.P., and A.D. analyzed data; and J.M. and R.B.S. wrote the paper.
The authors declare no conflict of interest.
' To whom correspondence may be addressed. E-mail: johan.moan@fys.uio.no or
setlow@bnl.gov.
C 2008 by The National Academy of Sciences of the USA
www.pnas.org/ cgi/doi/ 1 0.1 073/ pnas.071 06151 OS
'

..

I
,..
A

c:
1.2
::>
1.0
!
"' u
0.8 c:
"'
::>
0:
c
0.6
g
!!:
"' c
"'
0.4
0>
0
0.2
(ij
::>
c:
0.0
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70
Latitude
8
100

summer
29
E
0 winter
.s
80
I
l
I!" I
l
c
0

'E
60
"' u
c
0
u
23 I 27
30
0
40
c
22
E 25
"'
21 28
"' >
20
I'
Q.
"'
N
0
20 25 30 35 40 45 50 55 60 65 70 75
Latitude
Fig. 1. Vitamin 0 as a function of latitude. (A) The dependency of annual
vitamin 0 photosynthesis on latitude, calculated by using the in vivo action
spectrum of pre-vitamin 0 synthesis (43) and known f luence rates of solar
radiation as earlier described (37). (8) Summer (fil led circle) and winter (empty
circle) values for 25(0H)O levels in di fferent populations living at different
latitudes. The numbers in the graph indicate the citation number in the
reference list.
season (January-March) than in the rest of the year (31 , 32).
The annual vitamin D photosynthesis is modest, compared
with further south (Fig. 1A). In Bergen, Norway (61 N), there
is no vitamin D photosynthesis from October to March (7).
Despite this fact, the serum level of 25(0H)D in a population
living in Troms0 is 30% higher in late summer than in late
winter (30). Thus, we can conclude that, even at such high
latitudes, the sun is an important source of vitamin D. This
finding is supported by controlled sun bed experiments, which
show that exposure to suberythemal doses gives 25(0H)D
contributions of 10- 50 nmoVliter in serum (33). Our recent
investigations (A.C.P., 0. S. Bruland, L. Aksnes, W. Grant, and
J.M., unpublished work) support this notion and even show
that a high sun bed-induced 25(0H)D level cannot be main-
tained by daily intakes of the recommended amount of vitamin
D (200 units in the form of cod-liver oil pills).
Seasonal Variations of Cancer Prognosis. Because our demonstra-
tion of t he prognostic advantage of diagnosis in late summer and
autumn [ ,.,20% difference in relative risk of death in these
seasons when the 25(0H)D status is optimal] (35), we conducted
several more detailed studies showing similar trends. Many
cancer forms are now on our Jist: prostate, breast, colon, and
lung cancers, as well as lymphomas and even melanomas (36-
40). Other investigators have found comparable results ( 41, 42).
These data argue for a positive role of sun induced-vitamin D in
cancer prognosis or that a good vitamin D status is advantageous
when in combination with standard cancer therapies.
North-South Gradi ents of Vitamin D. Our calculations, which are
based on known ozone levels, cloud covers, and the in vivo
Moan eta/.
action spectrum for photosynthesis of pre-vitamin D from
7-dehydrocholesterol ( 43), show that there is a pronounced
north- south gradient in vitamin D-generating solar radiation
(Fig. I A). It should be emphasized that, in contrast to earlier
investigations (16), we calculated the doses for a vertical
cylinder, expecting such a geometry to represent the human
body better than a horizontal, flat surface. With our approach,
the annual, equatorial fluence of vitamin D-inducing radiation
is = 3.4 times larger than that in the U.K. and = 4.8 times larger
than that in Scandinavia (Fig. 1 A). A crucial and as-yet-
unanswered question is: Are there north- south gradients in
sun exposure habits and in vitamin D intake? In Norway, we
know that the vitamin D intake is 10- 20% larger in the north
than in the south. This finding is mainly related to the
consumption of cod liver (32). However, the population's sun
exposure is definitely larger in the south than in t he north, as
shown by calculations as well as skin cancer epidemiological
investigations (see Fig. 1, ref. 13, and www.kreftregisteret.no).
Overall, therefore, there is probably no north- south gradient
in vitamin D status in Norway. This finding seems consistent
with the lack of north- south gradient in both cancer incidence
and prognosis, which is discussed later (36, 39, 40).
Different clinical searches for a latitudinal gradient in vitamin
D status do not agree. Zittermann et a!. ( 44) found a negative
25(0H)D gradient with increasing latitude, as expected, whereas
others found the opposite ( 45). Our review of international data
(Fig. lB) shows no significant gradient. It is surpris.ing that mean
population levels of vitamin D are similar in sunny regions like
Florida (46), Australia (47), and Northern Europe (48). We
found earlier that the incidence rates of the three major forms
of skin cancer increase from Norway to Australia, which is in
agreement with a large increase in annual UV fluence (16).
Thus, because the action spectrum of pre-vitamin D photosyn-
thesis and that of squamous cell carcinoma are similar ( 43, 49),
one should expect to find a vitamin D gradient. The answer to
this puzzle may be found either in the pattern of sun exposure
or in differences in vitamin D intake. The most likely explanation
of the discrepancy, however, is probably that 25(0H)D deter-
minations are not standardized well enough for international or
interlaboratorial comparisons (50, 51).
Pre-vitamin D and vitamin D are photolabile (52). These
compounds and some of their metabolites can be photode-
graded or photochemically changed while they are in the skin,
where solar radiation can reach them. Photolability may be the
reason that sun-induced vitamin D intoxication has rarely or
never been reported. Such intoxication was wrongly proposed
to be the evolutionary reason for dark skin colors of humans
living close to the equator (53). However, the photolability of
pre-vitamin D and vitamin D is not likely to explain the lack
of latitude gradients in 25(0H)D levels because the vitamin D
generation is almost linear up to UV exposures as high as three
or four minimum eryt hema doses (54). However, it should be
noted that this reference concerns a narrow wavelength band
of around 295 nm. In human skin radiation, around 295 nm can
convert = 65% of the 7-dehydrocholestrol to pre-vitamin D,
whereas solar radiation can convert only = 20% (43). In future
investigations, one should take into account the increase in
skin darkness of populations from north to south. Moreover,
in assessing latitude variations of vitamin D levels, one should
focus mainly on summer values or winter-summer differences.
Doing so would minimize the role of different vitamin D
intakes.
North-South Gradients of Cancer Incidence, Mortality, and Prognosis.
A number of investigations (6, 55, 56) indicate that, in some
populations, the incidence and/or mortality of a number of
cancers (prostate, breast, colon, etc.) increase with increasing
latitude. However, in contrast to this theory, the similarity of
PNAS I January 15, 2008 I vol. 1 OS I no. 2 I 669
"'
w
v
z
w
;:;
"'
5
Q
w
::E
'

..

I
,.
A 7o
Colon cancer lung cancer
70
UK
60

60
Denmark
50 50
40 40
!:f Austra)la
New Zealand
30

Notway 30
20
0
20
enmartl
0
00
10

males

mates 10
0 females 0 females
0 0
Prostate and breast cancer
CMM
80
Ausb'alia 0 Oenmark
30
70

25
!:f 60

20
50
15
40
10
5
30

Prostate

males
0 Breast 0 females
0 0
20 30 40 50 60 20 30 40 50 60
Latitude Latitude
B
1,2
1,0
Cl: 0,8

0.4
0.2
0.0
1.2
1,0
0,8
lr

0,4
0,2
0,0
20
Colon cancer Lung cancer
Oenrnatk

Norway

New Zeal
Norway
0
0
AustraJia
UK

males

malos
0 females
0 f&makls
Prostate and CMM

mates
breast cancer
g:ostate 0 females
0 east

UK

Denmark Norway
Oenmar1<

Aus Sweden
Austr(p

0 0
Sweden
30 40 50 60 20 30 40 50 60
Latitude
Fig. 2. Cancer incidence and death rates as a function of latitude. (A) Incidence rates offour cancerforms in different countries as functions oft he mean latitude
of the country. Only countries populated predominantly by individua ls with skin types I and II are incl uded: Australia, New Zealand, Sweden, Norway, Denmark,
and U.K. cancer data represent averages for the period 1987- 1997. The data are from Cancer Incidence in Five Continents (ref. 34; see www-dep.iarc.fr). (8) The
ratios of death rates to incidence rates for the same countri es as considered in A Death rates are collected 2 years after incidence rates and represent averages
for the period 1989-1999. Cancer mortality data are obt ained from a WHO database (see www-dep.iarc.fr).
cancer incidence and mortality rates in Australia/New Zealand
and U.K./Scandinavia should be noticed (Fig. 2). For some
cancers, there may even be a significant inverse latitude
gradient (Fig. 2). Differences in sun exposure habits and skin
types can probably not explain this observation because the
inverse relationship remains when we use CMM incidence
rates as a crude measure for real UV exposures (Fig. 3).
However, taking more countries into consideration, we see
that no reliable north- south gradient can be extracted (Fig. 4).
There is a large variation of the incidence rates by factors of
= 50 and 5 for prostate cancer and breast cancer, respectively.
Even for countries at the same latitude, large differences are
found. From this epidemiological variation, we can conclude
that genetical, dietary, and environmental factors, other than
sun exposure, play major roles and may completely mask the
effects of vitamin D.
There might be a method to approach the problem from a
different point of view, namely by looking at prognosis. Accord-
ing to our experience with the Norwegian epidemiological data,
this method may be sensitive enough to study the connection
between vitamin D and cancer. The ratio of death rate to
incidence rate is a crude estimate of prognosis. In populations
with white skin tone included in Fig. 2B, there is an increasing
ratio of death rates to incidence rates with increasing latitude.
This finding indicates improved prognosis with decreasing lati-
tude (i.e., with increasing U V exposure). We find it unlikely that
cancer treatment is better in Australia than in the U.K. Further,
we conclude that the observations in Fig. 2B indicate, although
weakly, the beneficial role of sun-induced vitamin D for cancer
prognosis, in agreement with epidemiological findings (35- 40).
The Rise of Incidence Rates of Skin Cancer and Internal Cancers. It is
well known that the incidence rates of most cancers have been
increasing with time over many decades. Until 1990, this fact
also was universally true for all three major forms of skin
cancer (see ref. 16 and www.kreftregisteret.no). This finding
clearly indicates that, before 1990, the sun exposure of people
was constantly increasing with time. One would then expect
increasing vitamin D levels and decreasing rates of internal
670 I www.pnas.org/cgijdoi/ 10.1073/ pnas.071 0615105
cancers, the opposite of what is found. However, in addition to
vitamin D, a number of dietary and environmental factors
need to be considered. In many countries, people certainly
80 .-----------------------,-----------------------,
70
60
so
!:f 40
30
20
10
UK

Denmark
e uK

"''""""
Breast cancer
Aus1rB1a
J
--------....,Zealand
NOtWay

70 Lung cancer, males
60
!:!; so
40
30
20
10

UK
Oenmatk
Australia

NewZtWidiiU
SwedeJ NO<Way
UK

Lung cancer, females
Denmark
Australia

Norway -.
Swede' New Zealand
0
50 Colon cancer, males Colon cancer, females
40
lr
- 30
20
10
0
0 10 15 20 25 30 0 5 10 15 20 25 30 35
IRCMM !RCMM
Fig. 3. The incidence rates of prostate, breast, lung, and colon cancers as
functions of the incidence rates of CMM. Data are from the same sources as
those in Fig. 2.
Moan eta/.
'

..

I
isteret.no) and are presented as 2-year averages for the period 1960-2003.
Mortalit y dat a are retrieved f rom t he WHO mort alit y dat abase (see
www-dep.iarc.fr) and are presented as 2-year averages for the period 1960- 2003.
Data on seasonal variation of 25(0H)D were coll ected from a number of
invest igations done in healthy individuals ages 30- 50 years.
Cancer data were plotted against latitude or the age-adjusted incidence
rates of CMM as a measure of the UV exposure achieved. Simple linear
regression Sigma Plot 10 (Systat) was used to invest igat e the relationshi p.
Vitamin D Photosynthesis. We calculated t he annual fluence of vitamin D-
generating solar radiation as a function of latitude by using the action
spectrum for generat ion of pre-vitamin D in human skin (43) by applying a
1. Diffey 8 (2006) Do we need a revised publ ic health policy on sun exposure? Br 1
Dermato/ 154:1046- 1051.
2. Gillie 0 (2006) A new government policy is needed for sunli ght and vitami n D. Br 1
Dermato/ 154:1052- 1061.
3. Garbe C, Eigentler TK (2007) Diagnosi s and treatment of cutaneous mel anoma: State
of the art 2006. Melanoma Res 17:117-127.
4. Cummins Dl, et at. (2006) Cutaneous mal ignant melanoma. Mayo Clin Proc81:500-
507.
5. Bouillon R, eta/. (2006) Vitamin D, cancer. 1 Steroid Biochem Mol Bio/102:156-162.
6. Giovannucci E (2005) The epidemiology of vitamin D, cancer incidence and mortality:
A review (United States). Cancer Causes Contro/16:83- 95.
7. Holick MF (2004) Vitamin D: Importance in the prevention of cancers, type 1 diabetes,
heart disease, and osteoporosis. Am 1 Clin Nutr 79:362-371.
8. Jablonski NG, Chaplin G (2000) The evol uti on of human skin coloration. 1 Hum Evol
39:57-106.
9. Jablonski NG (1999) A possible link between neural t ube defects and ultraviolet light
exposure. Med Hypotheses 52:581-582.
10. Clemens Tl, Adams JS, Henderson SL, Holick MF (1982) Increased skin pigment reduces
the capacity of skin to synthesise vitamin D3. Lancet 1:74- 76.
11 . Chen TC, eta/. (2007) Factors that influence the cutaneous synthesis and dietary sources
of vitamin D. Arch Biochem Biophys 460:213-217.
12. Moan J, Porojnicu AC, Dahl back A (2006) in Skin Cancer Prevention, eds Ri ngborg U,
Brandberg Y, Breitbart EW, Grei nert R (lnforma Healthcare, New York), pp 179- 201.
13. Moan J, Porojnicu A C. Dahlback A (2007) Sunlight, Vitamin D, Skin Cancer, ed Reichrat h
J (Landes Bioscience, Austin, TX).
14. Beral v, Robi nson N (1981) The relationship of mal ignant melanoma, basal and
squamous skin cancers to indoor and outdoor work. Br 1 Cancer 44:886- 891.
15. Elwood JM, Jopson J (1997) Melanoma and sun exposure: An overview of published
studies. tnt 1 Cancer 73:198-203.
16. Moan J, Dahl back A (1993) in Environmental UV Photobiology. eds Bj0rn LO, Moan J,
Nultsch W, Young AR (Plenum, New York). pp 255-192.
17. Green A, eta/. (1999) A casecontrol study of mel anomas of the soles and palms
(Australia and Scotland). Cancer Causes Contro/ 10:21- 25.
18. Yu GP, Hu ON, McCormick SA (2006) Latitude and incidence of ocular melanoma.
Photochem Photobio/82:1621-1626.
19. Shah CP, et at. (2005) Intermi ttent and chronic ul traviolet l ight exposure and uveal
mel anoma: A meta-analysi s. Ophthalmology 112:1599- 1607.
20. Lea CS, eta/. (2007) Ambient UVB, melanoma risk in the United States: A case-control
analysis. Ann Epidemio/17:447- 453.
21. Ono Y, eta/. (2005) Seasonal changes of serum 25-hydroxyvitami n D, intact parathyroid
hormone levels in a normal Japanese population. 1 Bone Miner Metab 23:147- 151.
22. Carneval e V, eta/. (2001) longitudinal eval uation of vitamin D status i n healthy
subjects from southern Italy: Seasonal and gender di fferences. Osteoporos lnt
12:1026- 1030.
23. Bischof MG, Heinze G, Vierhapper H (2006) Vitamin D status and its relation to age and
body mass i ndex. Horm Res 66:211- 215.
24. Bouillon RA, Auwerx JH, Lissens WD, Pelemans WK (1987) Vitami n D status i n the
elderly: Seasonal substrate deficiency causes 1,25di hydroxycholecalciferol deficiency.
Am 1 C/in Nutr 45:755-763.
25. Fi nch PJ. et a/. ( 1992) Blunted seasonal variation in serum 25-hydroxy vitamin D.
increased risk of osteomalacia i n vegetarian london Asians. Eur 1Ciin Nutr46:509-51S.
26. Rejnmark l, eta/. (2004) Vitamin D insufficiency i n Greenlanders on a westerni zed fare:
Ethnic differences in cal citropic hormones between Greenl anders and Danes. Calcif
Tissue Jnt 74:255- 263.
27. Hine TJ, Roberts NB (1994) Seasonal variation in serum 25-hydroxyvitamin D3 does not
affect 1,25-dihydroxy vitami n D. Ann Clin Biochem 31:31-34.
28. Savolainen K, Maenpaa PH, Al hava EM, Kettunen K (1980) A seasonal difference in
serum 25-hydroxyvi tami n D3 in a Finnish population. Med Bio/58:49-52.
29. Lamberg-AIIardt C (1984) Vi tami n D intake, sunl ight exposure and 25hydroxyvitamin
D levels i n the elderly duri ng one year. Ann Nutr Metab 28:144- 150.
30. Vik T, Try K, Stromme JH (1980) The vitami n D status of man at 70 degrees north. Scand
1 C/in Lab Invest 40:227- 232.
31. Brustad M, eta/. (2004) Vitami n D status of middle-aged women at 65-71 degrees N in
relation to dietary intake and exposure to ultraviolet radiation. Public Health Nutr
7:327-335.
32. Johansson L, Solvoll K (1999) National survey for nutrition and physical activity, p. 45.
33. Moan J, lagunovaz. Porojnicu AC (2006) Proceedings of the meeting "Sunlight,
Vitamin D, Health" (House of Commons, london), pp 33- 40.
672 I www.pnas.org/ cgijdoi/ 10.1073/ pnas.071 0615105
radiation transfer model (66, 67). Global solar exposure (direct plus diffuse
exposure) was det ermined, approximat ing the human body by a horizontal
cylinder, excluding top and bottom. Total ozone columns measured by t he
TOMS satellite instruments were used in the calculations. The daily average
cloud cover f or each sit e was deri ved f rom measured reflectivities at an
ozone-insensitive channel of the same instrument. Further details of the
calculat ions can be found elsewhere (68, 69).
ACKNOWLEDGMENTS. This work was supported by Sigval Sergesen D.Y. og
hustru Nankis Foundation, The Research Foundation of the Norwegian Radi-
umhospit al, and Helse-S0r Norway. Brookhaven National Laboratory is oper-
ated by Brookhaven Associates, under contract with the U.S. Department of
Energy.
34. Curado MP, et a/., eds (2007) Cancer Incidence in Five Continents (lnt Agency Res
Cancer, lyon, France), Vol 9.
35. Robsahm TE, Tretli S, Dahl back A, Moan J (2004) Vitamin D3 from sunlight may improve
the prognosis of breast-, colon- and prost ate cancer (Norway). Cancer Causes Control
15:149-158.
36. Lagunova Z, eta/. (2007) Prostate cancer survival is dependent on season of diagnosis.
Prostate 67: 1362-1370.
37. Moan J, eta/. (2005) Solarradiation, vitami n D, survival rat e of colon cancer in Norway.
1 Photochem Photobiol 8 78:189-193.
38. Porojnicu AC. Robsahm TE, Hansen Ree A, Moan J (2005) Season of diagnosis i s a
prognosticfactor i n Hodgkin lymphoma. A possible rol e of sun induced vitamin D. Br 1
Cancer 93:571-574.
39. Porojnicu AC, et a/. (2007) Seasonal and geographical variati ons in lung cancer
prognosis in Norway. Does vitami n D f rom the sun play a role? Lung Cancer
55:263-270.
40. Porojnicu AC, eta/. (2007) Changes in ri sk of death from breast cancer with season and
latitude: Sun exposure and breast cancer survival in Norway. Breast Cancer Res Treat
102:323-328.
41. Lim HS, eta/. (2006) Cancer survival i s dependent on season of diagnosis and sunlight
exposure. Inti Cancer 119:1530-1536.
42. Zhou W, eta/. (2005) Vi tami n Dis associated wi th improved survival i n early-stage
non-small cell l ung cancer pati ents. Cancer Epidemiol Bi omarkers Prev 14:2303-
2309.
43. Maclaughlin JA, Anderson RR, Holick MF (1982) Spectral character of sunlight modu-
lates phot osynthesis of previtamin D3 and its photoisomers i n human ski n. Science
216:1001-1003.
44. Zittermann A, Schl eithoff SS, Koerfer R (2005) Putting cardiovascular disease and
vitamin D insuffi ciency into perspective. Br 1 Nutr 94:483-492.
45. Lips P, eta/. (2001) A global study of vitamin D status and parathyroid function in
postmenopausal women with osteoporosis: Basel ine data from the multiple outcomes
of raloxifene evaluation cl inical trial . 1 Clin Endocrinol Me tab 86: 1212-1221.
46. Levis S, et a/. (2005) Vitamin d def iciency and seasonal vari ation in an adult South
Florida population. 1 Clin Endocrinol Me tab 90: 1557-1562.
47. Pasco JA. eta/. (2004) Seasonal periodicity of serum vitamin D, parathyroid hormone,
bone resorption, and f ractures: The Gee long Osteoporosis Study. 1 Bone Miner Res
19:752-758.
48. Beadle PC, Burton Jl , leach JF (1980) Correlati on of seasonal variati on of 25-
hydroxycalciferol with UV radiation dose. Br 1 Dermato/ 1 03:289- 293.
49. de Gruijl FR, eta/. (1993) Wavelength dependence of skin cancer induction by ult ra-
violet irradiation of albino hairless mice. Cancer Res 53:53- 60.
50. Binkley N, eta/. (2004) Assay variation confounds the di agnosi s of hypovitami nosis D:
A call for standardization. 1 Clin Endocrinol Metab 89:3152-3157.
51. Hollis BW (2004) Editori al : The determinati on of circulating 25-hydroxyvitamin D: No
easy task. 1 Clin Endocrinol Metab 89:3149-3151.
52. Holick MF (1994) Vitamin D: Photobiology, metabolism, and cl inical applicati on. The
Liver: Biology and Pathobiology, eds Ari as I M, Boyer JL, Fausto N, Jakoby WB, Schachter
D. Shafritz DA (Raven, New York), 3rd Ed, pp 543-562.
53. Loomis WF (1967) Skin-pigment regulati on of vitami n-0 biosynthesis i n man. Science
157:501- 506.
54. Adams JS, Clemens Tl, Parrish JA, Holick MF (1982) Vi ta min-D synthesis and metaboli sm
after ultraviolet irradi ation of normal and vitami nD-deficient subjects. N Eng/ 1 Med
306:722-725.
55. Garland CF, eta/. (2006) The role of vi tamin D in cancer prevention. Am 1 Public Health
96:252- 261.
56. Schwartz GG, Ski nner HG (2007) Vitamin D status and cancer: New i nsights. CurrOpin
Clin Nutr Metab Care 10:6- 11.
57. Wortsman J, et al. (2000) Decreased bioavailability of vitamin Din obesity. Am 1 Clin
Nutr 72:690-693.
58. Ybarra J, Sanchez-Hernandez J, Perez A (2007) Hypovitaminosi s D. morbid obesity.
Nurs Clin North Am 42:19- 27.
59. Coory M, et a/. (2006) Trends for in situ and invasive melanoma in Queensland,
Austral ia, 1982- 2002. Cancer Causes Contro/17:21- 27.
60. Marks R (2002) The changing i ncidence and mortali ty of melanoma in Australia. Recent
Results Cancer Res 160:113- 121.
61. Marrett LD, Nguyen Hl, Armstrong BK (2001) Trends in the incidence of cutaneous
malignant melanoma in New South Wales, 1983-1996. tnt 1 Cancer 92:457-462.
Moan eta/.
'

..

I
62. Pearce J, Barnett R, Kingham S (2006) Slip! Slap! Sl op! Cutaneous mali gnant
melanoma incidence and social status in New Zealand, 1995-2000. Health Place
12:239-252.
63. Staples MP, eta/. (2006) Non-melanoma ski n cancer in Australia: The 2002 national
survey and trends since 1985. Med 1 Aust 184:6- 10.
64. Berwick M, et at. (2005) Sun exposure and mortality from melanoma. 1 Nat/ Cancerlnst
97: 19>-199.
65. Giovannucci E, eta/. (2006) Prospective study of predictors of vitamin D status and
cancer i ncidence and mortality in men. 1 Nat/ Cancer lnst 98:451-459 .
Moan eta/.
66. Dahlback A. Stamnes K (1991) A new spherical model for computing the radiation field
available for photolysis and heating rate at twilight. Planet Space Sci 39:671- 683.
67. Stamnes K, TsaySC, WiscombeW, Jayaweera K (1988) Numerically stable al gori thm for
discrete-ordinate-method for radiative t ransfer in multi pl e scattering and emitting
layered media. Appl Opt 2502- 2509.
68. Moan J, Dahlback A. Henriksen T, Magnus K (1989) Biological amplif ication factor for
sunlight-i nduced nonmelanoma skin cancer at high latitudes. Cancer Res
69. Moan J, Dahlback A (1992) The relationship between skin cancers, solar radiation and
ozone depleti on. Br 1 Cancer 65:916-921.
PNAS I January 15, 2008 I vol. 105 I no. 2 I 673
"'
...
v
z
...
v
"'
5
Q
...
:::E
HHS Publication FDA 88-8234
Quality Control Guide
for Sunlamp Products
WHO Collaborating Centers for: J(;.
Standardization of Protection ' ~ \
Against Nonionizing Radiations IR
Training and General Tasks in
Radiation Medicine ~
Nuclear Medicine ~
~ ~
March 1988
Supersedes FDA 84-8234
U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES
Center for Devices and Radiological Health
Rockville, Maryland 20857
FOREWORD
In October 1982, the Food and Drug Administration established the Center for
Devices and Radiological Health (CDRH) by merging the Bureau of Medical Devices and
the Bureau of Radiological Health.
The Center develops and implements national programs to protect the public health
in the fields of medical devices and radiological health. These programs are intended to
assure the safety, effectiveness and proper labeling of medical devices, to control
unnecessary human exposure to potentially hazardous ionizing and nonionizing radiation,
and to ensure the safe, efficacious use of such radiation.
The Center publishes the results of its work in scientific journals and in its own
technical reports. These reports provide a mechanism for disseminating results of
CDRH and contractor projects. They are sold by the Government Printing Office and/or
the National Technical Information Service.
Also, CDRH technical reports in radiological health are made available to the World
Health Organization (WHO) under a memorandum of agreement between WHO and the
Department of Health and Human Services. Three WHO Collaborating Centers,
established under the Bureau of Radiological Health, continue to function under CDRH:
WHO Collaborating Center for Standardization of Protection Against Nonionizing
Radiations;
WHO Collaborating Center for Training and General Tasks in Radiation Medicine;
and
WHO Collaborating Center for Nuclear Medicine.
We welcome your comments and requests for further information.
ohn C. Villforth
Director
Radiological Health
PREFACE
The Performance Standard for Sunlamp Products (21 CFR 1040.20) became effective
May 7, 1980, and was amended effective September 8, 1986. Sunlamp products and
ultraviolet lamps manufactured on or after that date must conform to the applicable
provisions of the standard. In addition, manufacturers are required to certify that their
products comply with the standard and to furnish reports, to the Center for Devices and
Radiological Health, that clearly substantiate the product's compliance. Because sun-
lamp products are also medical devices, they are subject to the Medical Device Amend-
ments to the Federal Food, Drug, and Cosmetic Act, including the Good Manufacturing
Practices (GMP) regulations developed under the authority of that Act. All manu-
facturers of sunlamps and sunlamp products and ultraviolet lamps must establish and
implement GMPs to ensure that the devices conform to their specifications, and imple-
ment quality control procedures to ensure compliance with the Federal performance
standard.
This guide has been prepared to assist manufacturers in satisfying the G MP require-
ments and in developing and implementing quality assurance programs, including
appropriate testing. This guide also provides what we believe to be fair and standard
criteria by which manufacturers' programs will be evaluated. Because of variation in
product designs and manufacturing processes, a detailed step-by-step protocol would be
of limited application. The purpose here, therefore, is to set forth a general description
of the procedures and philosophy of quality assurance as appropriate for sunlamp
products and ultraviolet lamps. This document is intended for use in conjunction with
the "Reporting Guide for Initial Reports and Model Change Reports on Sunlamps and
Sunlamp Products."
We strongly emphasize that the manufacturer may adopt other alternative pro-
cedures that are equivalent to, or are as effective as, those methods described in this
guide. Each manufacturer's test procedures and testing program will be evaluated on an
individual case-by-case basis. If the alternative procedures constitute good manufac-
turing practices and provide adequate safeguards to ensure compliance with the
standard, the Center for Devices and Radiological Health (CDRH) would have no reason
to disapprove such a program.
However, neither the fact that CDRH did not disapprove a testing program, nor the
fact that such a program was established pursuant to this guide, relieves a manufacturer
from his obligations to comply with the Radiation Control for Health and Safety Act of
1968 and the Federal Food, Drug, and Cosmetic Act as amended.
~ ~
Walter E. Gundaker
Director
iv
CONTENTS
Foreword .... ... ... ...... .. ..... ... ...... ...... ........... .. ..
i i
Preface
iii
Abstract . ..... ............ ........ .... ..... .. .. ...... ... ...... vi
1. Quality Assurance Program . . . . . . . . . . . . . . . . . . . . . . . . . 1
Elements of a Qualit y Assurance Program . . . . . . . . . 1
Objectivity of Program. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Documentation and Records . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Validity of Measurement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Audit Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Personnel Training. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2. Preproduction Evaluation. . . . . . . . . 4
Design Review. . . . . . . . . . . . . . . . . . . . . 4
Engineering and Prototype Testing and Evaluation . . . 4
3. Components and Production Testing. . . . . . . 5
Component Evaluation and Testing . . . . . . . . . . 5
Production Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Qualitative Test ing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Quantitative Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Test Condit ions for Determination of Compliance . . . . . . . 7
Sampling, Audit, and Recycling. . . . . . . . . . . . . 8
References 9
Appendixes
A. Design Considerations for Compliance with the Sunlamp Standard . 11
B. Procedures for Laboratory Compliance Testing . . . . . . . 15
C. FDA Policy on Maximum Timer Interval and Exposure Schedule. . . . . 25
II
ABSTRACT
Office of Compliance. Quality Control Guide for Sunlamp Products. HHS Publication
88- 8234 (March 1988) pp. 30.
Sunlamp products and ultraviolet lamps manufactured on or after May 7,
1980 must conform to applicable portions of a performance standard (21 CFR
1040. 20, amended effective September 8, 1986), and their manufacture must be
in accord with the medical devices Good Manufacturing Practices (GMP) regu-
lations promulgated under the Food, Drug, and Cosmetic Act (FD&C). This
guide is intended to assist manufacturers of these products in satisfying both the
GMP requirements and in developing and implementing quality assurance (QA}
programs. This guide provides the criteria and the underlying philosophy used by
FDA to evaluate manufacturers' QA programs. This document is intended to be
used in conjunction with "Reporting Guide for Initial Reports and Model Change
Reports on Sunlamps and Sunlamp Products."
The mention of commercial products, their sources, or their use in
connection with material reported herein is not to be construed as either an
actual or implied endorsement of such products by the Depart ment of Health
and Human Services (DHHS) or the World Health Organization (WHO).
ui
QUALITY CONTROL GUIDE FOR SUNLAMP PRODUCTS
1. QUALITY ASSURANCE PROGRAM
1.1 ELEMENTS OF A QUALITY ASSURANCE PROGRAM
A viable quality assurance program ensures the performance of a product at a
designated level at the time of manufacture and throughout its useful life. It is
essential for a manufacturer to establish such a program to ensure compliance of a
product with the Federal performance standard and its conformance with design
specifications. The adequacy of a specific quality assurance and testing program must
be judged solely on its own merits and applicability to the product. The essential
elements of a quality assurance program include the following:
1. proper organization, personnel, facilities, and administrative procedures to en-
sure an objective and defensible program;
2. preproduction evaluation and testing of the product and testing of components
and material obtained from other manufacturers;
3. valid measurement techniques, calibrations, and treatment of measurement
uncertainties;
4. testing and evaluation of the product during and after production and statistical
analyses of data (the establishment of confidence limits and rejection criteria are
important components of such an evaluation};
5. procedures to ensure that only accepted units are distributed and that any
rejected units, if modified and recycled back into the production sequence,
undergo the same level of scrutiny as other units that have been found accep-
table;
6. an audit procedure to measure the degree of conformance and the effectiveness
of the quality assurance program, including product audits;
7. lifetime and reliability testing procedures or other suitable means to determine
whether the product will continue to meet its design specifications during its
useful life;
8. a continuous review of the product's performance throughout its useful life and a
procedure to ensure that any problems discovered are corrected, and that appro-
priate design changes are made to eliminate them from subsequent units
produced;
9. documentation control to assure that all manufacturing, component, and finished
product specifications adequately reflect the product design and provide an
effective and consistent mechanism for changes of specifications or quality
control procedures; and
10. quality control records, including a device master record and a device history
record.
1.2 OBJECTIVITY OP PROGRAM
The quality assurance program should be carefully designed and administered to
ensure the objectivity of the program. Certain safeguards against personal prejudice
and conflicting interests should be built into the system to avoid biased decisions. This
may present a problem in small manufacturing firms that employ only a few persons.
However, the division of responsibility between production and quality control personnel
should be maintained. This may be accomplished, for example, if one person, who is
responsible for the receiving, inspection, and accountability of components, is also
designed responsible for in-process checks and finished product inspection. Another
person should be held responsible for all the production processes. This arrangement (or
similar ones) may sustain the objectivity of the quality control program by dividing the
responsibilities.
1.3 DOCUMENTATION AND RECORDS
Complete documentation of the quality assurance program must be maintained as
required by 21 CFR 1002.30. This should include such items as description of tests
performed and their sequence, description of the measurement instrumentation and
techniques, rejection criteria or confidence limits and justification for choosing such
limits, methods of data analysis and sampling plans. If all the units produced are not
tested, a sampling technique must be used and must be scrupulously documented and
followed. Standardized forms for data recording should be used. The forms should
provide space for logging all information relevant .to the product and the test, including
the product model and serial numbers, date of test, date of manufacture, samples per
lot, results of test, identification of personnel performing and reviewing the test, and
model and serial numbers of test instruments. A device master record as required by 21
CFR 820. 181 must be established for each type of device and include, or refer to the
location of, the information concerning:
1. the device specification, including drawings and component specifications;
2. production process specifications, including the appropriate equipment specifi-
cations, production methods, and production procedures;
3. packaging and labeling specifications; and
4. quality assurance procedures and specifications, including quality assurance
checks used and the quality assurance apparatus used. The device master record
shall be prepared, dated, and signed by a designated individual(s).
A device history record as required by 21 CFR 820.184 must also be maintained and
it shall include, or refer to the location of, the information concerning the dates of
manufacture, the quantity manufactured, the quantity released for distribution, and any
control number used.
Among small manufacturers of sunlamp products, the manufacturing process may be
limited to assembling components into the final product, and there should be no problem
with communication and managing operations. In this situation, written manufacturing
procedures may be reduced to a practical extent. In some cases, training and work
experience may be considered valid substitutes for written manufacturing procedures.
However, manufacturers must still ensure uniformity of procedures and keep records of
test results necessary to ensure the quality of the product.
2
Records required by the Radiation Control for Health and Safety Act are to be kept
for 5 years (21 CFR 1002.31(a)), and for no less than 2 years if required by the Food,
Drug, and Cosmetic Act (21 CPR 820.180(b)). Test records, complaints, trends analyses,
failure analyses, and audits should be used to identify and provide solutions to quality
assurance problems.
1.4 VALIDITY OF MEASUREMENT
A valid program of measurements made as part of quality assurance testing involves
periodic evaluation of the test methods and calibration of the instruments. A schedule
to recalibrate measuring instruments against an acceptable standard is important to
ensure a consistent measurement capability. Calibration services from the instrument
manufacturer or other qualified outside sources, with traceability to National Bureau of
Standards primary measurement standards, should be used when adequate calibration
facilities are not available within a company.
The instrumentation should be carefully specified and selected to ensure that it is
capable of performing the tests required by the standard. For example, instruments for
ultraviolet (UV) radiation measurements should have sufficient spectral sensitivity in
the UV region to measure the anticipated radiation levels from the product. After the
measurement instruments have been obtained, they need to be evaluated for proper
performance and operation. Control charts and logs of instrument performance and
calibrations are essential, as well as service and maintenance schedules. Any changes of
the operator, or in the instrument, its components, its techniques of use, calibration
standards, or other relevant components of the measurements system should be docu-
mented.
1.5 AUDIT PROCEDURES
The purpose of the quality audit is to check the degree of conformance of the quality
control procedures and the effectiveness of the quality assurance program, and to
ensure that all procedures established are adequate and consistently followed. The
quality audits are in addition to the routine quality assurance program and should
include, but not be limited to, product audits. A product audit is an independent
evaluation of product quality to determine if the product complies with the standards
and if the product conforms to design specifications. Random sampling is the only
acceptable method in selecting units for product audit. The person responsible for the
audit should not have responsibility for the matter being audited but must have
sufficient training and experience in the manufacturing process and testing procedures.
For small manufacturers, a simple checklist with appropriate instructions may suffice.
Documentation of audits and their results must be kept; results of findings should be
reviewed by management. Audit frequency should be consistent with findings, but the
interval between audits should not exceed 1 year.
1.6 PERSONNEL TRAINING
All quality assurance and production personnel shall have the necessary training to
perform their assigned responsibilities adequately. They should have a thorough
understanding of their jobs and be made aware of defects and errors likely to be encoun-
tered in performing their assigned functions.
3
2. PREPRODUCTION EVALUATION
Preproduction evaluation and testing should include a review of the design,
evaluation of components and material obtained from other manufacturers, and
engineering and prototype testing to confirm that the product as designed can be
manufactured in compliance with the standard and in conformance with the design
specifications.
2.1 DESIGN REVIEW
Before the manufacture of the product begins, the product design must be reviewed
to determine if the product will comply with the Federal performance standard and
other applicable criteria for product safety, including electrical and mechanical safety.
This review allows the manufacturer to perform a step-by-step evaluation of the
requirements for the product and the means used to fulfill these requirements. In
addition to the performance requirements, the conditions and environments under which
the product will be used should be considered carefully. For example, if a product is
intended to be used at a commercial facility and an alternate timer such as a coin box is
permit ted, then a mechanism must be incorporated into the coin box to allow the user to
terminate exposure before completion of the timer cycle and to prevent the timer from
operating continuously in excess of the recommended maximum exposure time. The
coin box timer must also provide intermediate exposure times compatible with the
recommended schedule. The goggles should be evaluated to ensure adequate eye
protection for the user under all possible exposure positions and uses. The location of
controls, particularly the radiation emission control, should be convenient for the user
to reach in case of emergency. The contents and locations of the required labels should
be carefully reviewed to assure they comply with the standard and policies. The
development of a recommended exposure schedule for the particular product design is of
utmost importance and should be checked against the agency guidelines. In general, an
adequate safety factor regarding the use of the product should be built into the product
design, because the product is not only expected to be in compliance at the time of
manufacture but also during its entire useful life (see Appendix B for details of
compliance requirements).
Further, the manufacturing process must be reviewed to assure that the design basis
for the product, components, and packaging is correctly translated into approved
specifics t ions.
The approved design specifications must be adequately documented. The
manufact uring process must be described in writing and implemented according to the
written description in order to ensure the finished product is manufactured acording to
the approved design. Any changes in the design and/or manufacturing process must be
reviewed, approved, and dated before implementation.
2.2 ENGINEERING AND PROTOTYPE TESTING AND EVALUATION
Engineering models and prototypes should be built and tested thoroughly to ensure
that the product can be manufactured as designed and be in compliance with the
performance standard during the product's useful life.
4
This testing must include adequate transportation tests, tests of performance under
expected environmental condidtions, use and abuse testing, and accelerated life tests.
A consistent effort is required to keep drawings and prototypes up-to-date and accurate.
Any design changes must be properly documented, tested, and evaluated in relation to
product safety.
3. COMPONENTS AND PRODUCTION TESTING
3.1 COMPONENT EVALUATION AND TESTING
Components should be received, stored, and handled in a manner to prevent damage,
mixup, contamination, and other adverse consequences. Components should be in-
spected, sampled, and tested for conformance to specifications because certain compo-
nents can affect the compliance of ultraviolet lamps and/or sunlamp products. For
example, glass structure or an electrical or mechanical component, such as timer,
eyewear, or emission control, will definitely affect the product's compliance. (Tags and
labels are also important for compliance.)
Whether such components are fabricated by the manufacturer of the ultraviolet
sunlamp or sunlamp product, or purchased from a vendor, proper evaluation of the
components is essential in determining if the ultraviolet or sunlamp product will comply
with the applicable standard after manufacture. Manufacturers who obtain components
from different vendors may choose to rely on a supplier's certificate of analysis and
approval in lieu of component testing. These certificates should be made part of the
quality assurance documentation.
However, the responsibility for product compliance rests with the manufacturer of
the final product rather than with the supplier of components or material. Therefore,
any decision to totally rely on a supplier's evaluation must be carefully considered; some
type of incoming acceptance examination or testing is recommended. An exception
applies to the lamp manufacturers' certification of compliance for their UV lamps. This
certification is acceptable because the lainp manufacturers are legally liable for the
validity of that certification.
In general, three points must be considered in evaluating components prior to their
use in the ultraviolet lamp or sunlamp product.
1. Performance specification for components should be developed, taking into
consideration both the requirements of the performance standard and a tolerance
analysis of the product. A tolerance analysis is a calculation of a product's
performance range that can result from component variations. Such an analysis
would, whenever possible, be experimentally validated.
2. All components should be tested to verify that their performance is within the
range of specifications required of them by the manufacturer of the final
product. Switches and timers may need to undergo accelerated life testing to
ensure reliability. Durability and ruggedness of. such components may be appro-
priately tested under expected environmental conditons. Components also may
be appropriately tested after assembly of a subsystem or entire product.
3. If at any time the design specifications or materials of a component are changed,
the design of the entire product should be reviewed to ensure that the change
does not in any way affect the compliance of the product with the performance
standard.
5
3.2 PRODUCTION TESTING
In designing the production quality assurance and testing program, it is first
necessary to identify the parameters that need to be examined or tested in order to
determine compliance with the performance standard. The identification and critical
nature of these parameters depend to a large extent upon the design, function, and
anticipated use of a specific product. Some examples of such parameters are the
irradiance ratio, reliability of timers and switches, the inclusion and durability of labels,
mechanical construction, electrical safety, and so forth. Some examinations or tests
may consist of simple inspections. For example, the product should be inspected to
determine that all labels are in the correct positions and properly attached to the
product. However, other tests, such as those involving the irradiance ratio of the
ultraviolet lamp or sunlamp product and the spectral transmittance of the protective
eyewear, are more complicated and require design of test methods and the use of
defensible measurement instrumentation and techniques.
Testing will be of two types: qualitative and quantitative. Qualitative testing refers
to a functional test of performance features and inspections, such as tests that confirm
the inclusion of labels or instructional material. Quantitative testing refers to the
measurement of variable parameters such as the spectral irradiance of the ultraviolet
lamp. Individual tests may be designed for each parameter.
Rejection or acceptance criteria need to be determined for different tests. In
qualitative tests, it is easy to establish the criteria because one is faced with a "has" or
"has not" or "functions" or "does not function" situation. The establishment of rejection
or acceptance criteria for the quantitative test is more involved. An example is the
measurement of ultraviolet radiation from an ultraviolet lamp. The rejection or
acceptance criteria must assure the manufacturer that ultraviolet radiation is within
the irradiance ratio limits for sunlamp products as specified in the sunlamp products
performance standard, even when the total measurement uncertainty is taken into
account. If the testing is not done on every unit, it may be necessary to have more
stringent criteria than if every unit were tested.
3.3 QUALITATIVE TESTING
Qualitative testing should be conducted on all of the units produced and should
include checks or testing of the following:
1. the presence, secure attachment, and proper content of a label certifying com-
pliance of the product;
2. the presence, secure attachment, and proper content of the label that identifies
the manufacturer and place and date of manufacture;
3. in the case of ultraviolet lamps, the presence, secure attachment, and proper
content of the warning label;
4. in the case of sunlamp products, the presence, secure attachment, and proper
content of aU hazard warning labels;
5. the presence and proper content of instructions to the users;
6. the presence and proper function of the timer;
6
7. the presence and proper functioning of a control that permits users to manually
terminate radiation emisssion from the sunlamp product at any time without
disconnecting the electrical plug or removing the ultraviolet lamp;
8. the presence and required number of protective eyewear devices accompanying
the sunlamp product;
9. the proper functioning of the ultraviolet lamps; and
10. the presence and proper contents of user instructions for both ultraviolet lamps
and sunlamp products.
3.4 QUANTITATIVE TESTING
Quantitative testing includes the following types of tests:
1. measurement of ultraviolet radiation output to determine if the spectral irra-
diance is within the expected range for the product and its recommended
exposure schedule;
2. measurement of the irradiance output within the specified wavelength ranges to
determine the irradiance ratio;
3. measurements of protective eyewear spectral transmittance, to determine that it
does not exceed the values required by the performance standard; and
4. measurement of timer acuracy to determine that the error is not greater than
10 percent of the maximum timer interval.
There are many ways to conduct these measurements. CDRH does not object to any
method that would, when correctly executed, yield the correct physical values. Pro-
cedures for compliance testing at FDA's Winchester laboratory are provided in Appendix
B for manufacturers' reference only. Procedures for developing the recommended
exposure schedule are provided in Appendix C.
3.5 TEST CONDITIONS FOR DETERMINATION OF COMPLIANCE
In order for the product to have a valid certification (21 CFR 1010.2), the test on
which the certificaton is based must meet the following conditions:
1. The tests for the irradiance ratio and to develop and verify the recommended
exposure schedule (and/or to establish lamp compatability) and protective
eyewear transmittance must account for all measurement errors and statistical
uncertainties in the measurement process and, wherever applicable, for changes
in radiation emission or degradation in radiation safety with age of the product.
2. The measurements to check irradiance ratio, recommended exposure (and/or
lamp compatability) and protective eyewear transmittance must be made under
the operational conditions of lamp voltage, current, and user position as recom-
mended by the manufacturers. The measuring instrument must be positioned at
the recommended exposure position orientation and so oriented as to detect the
maximum amount of radiation detectable by the instrument.
7
3 The timepiece used to determine timer accuracy is accurate and reliable. It is
regularly calibrated.
4. It must provide procedures to check for all items mentioned in 3.3 and must also
provide for checksheets to include date, initials of tester, and product identifying
marks.
3.6 SAMPLING, AUDIT AND RECYCLING
3.6.1 Sampling
CDRH recommends testing 100 percent of the products to determine compliance.
However, in the case of products built in large quantity on a production line, a sampling
plan that randomly allows selection of some units for some tests and inspections may be
appropriate. When this is done, it is the responsibility of the manufacturer to
demonstrate that the sampling plan will ensure compliance of the product with the
standard and with the design specifications. A random sampling scheme must be truly
random, i.e., the probability of a unit being selected is the same for all produced units.
The so-called "random" selection of units from a production line based on a judgment of
randomness by an individual does not constitute true random sampling. A scheme that is
based on generating random numbers or other specified statistical schemes is a more
objective basis for random sampling. Examples of sampling plans are contained in Mil-
Std-1050 and Mil-Std-414 (See References 1 and 2).
In addition to random sampling to test for compliance, any systematic problem areas
may warrant testing; e.g., at the beginning of a production run, or at the beginning of
each day, or when a new assembler begins work on a product, and so forth.
3.6.2 Product Audit
A product audit is a GMP requirement. It is desirable to establish such an audit to
test the validity of the quality control and testing program. A product audit includes
testing previously tested and accepted products to provide a check on the measuremnt
process and the acceptance procedures. A specific sampling procedure should be used to
select accepted units that should be subjected to all the tests and inspections of the
product quality assurance program. Repeated failures of audited samples indicate
problems in the product design or the quality assurance program, which must be traced
and corrected.
3.6.3 Recycling
Units or lots that fail to conform to specifications and which are refurbished,
repaired, reworked, and so forth must be subjected to the testing and scrutiny and the
same acceptance or rejection criteria to which the original was subjected for the
characteristics that may be adversely affected by the reprocessing.
8
REFERENCES
1. United States Department of Defense. Sampling Procedures and Tables for
Inspection by Attributes. Military Standard 105D (April 27, 1963).
2. United States Department of Defense. Sampling Procedures and Tables for
Inspection by Variables for Percent Defective. Military Standard 414 (June 11,
1957).
Copies of these documents (Mil-Std 105D and Mil-Std 414) are for sale by
Superintendent of Documents
U.S. Government Printing Office
Washington, D.C. 20402
9
i
I
l
I
I
1.
l
APPENDIX A
DESIGN CONSIDERATIONS FOR COMPLIANCE WITH THE SUNLAMP STANDARD
A. FOR ULTRAVIOLET LAMPS
1. Each ultraviolet lamp, unless exempted, must have on t he product a permanent
label which contains (21 CFR 1040.20(d)(2}) (i) the words "Sunlamp-DANGER-
Ultraviolet radiation. Follow instructions," (ii) the model identification and (iii)
the words "Use ONLY in fixture equipped with a timer."
2. Each ultraviolet lamp must be permanently labeled or marked in such a manner
that the name of the manufacturer and month and year of manufacture (all of
which may be expressed in code) can be determined (21 CFR 1040.20(d)(3)).
Labels or marks shall be legible throughout the useful life of the lamp.
3. When a lamp is sold separately from a sunlamp product, the lamp packaging
uniquely associated with an ultraviolet lamp (i.e., innermost lamp packaging)
must bear a label or tag that contains the full name and address of the
manufacturer, place of manufacture and month and year of manufacture if such a
label or tag is not affixed or inscribed on the lamp itself (21 CFR 1040.20(d)(3)
and 21 CFR 1010.3(a)). If the lamp is sold under a name other than the
manufacturer, the full name and address of the company under whose name the
lamp is sold may be set forth, provided sufficient information is furnished to
CDRH to allow the Agency to identify the manufacturer. The place of manu-
facture may be expressed in code. However, no specific place of manufacture
identification will be required if the place of manufacture is the same as the
address of the manufacturer stated on the label, or if the manufacturer has only
one place of manufacture for such lamps and has identified that place of manu-
facture to CDRH. Themonth and year of manufacture must appear in full, with-
out abbreviation, as in the following example - Manufactured: December 1998.
However, when the date of manufacture is provided in code on the lamp, CDRH
will not object to the omission of the date of manufacture on the packaging if the
key to the ~ t e code and the location of the coded information are provided on
the packaging in a manner that will allow the date to be readily decoded. The
month and year of manufacture on the packaging must be the same as that which
appears on toe lamp.
4. The lamp packaging uniquely associated with the ulraviolet lamp not accom-
panying a sunlamp product must also bear a label or tag which contains a
statement by the manufacturer certifying that the lamp conforms to the require-
ments of the Federal performance standard for such lamps if the certificat ion
label or tag ia not affixed or inscribed on the lamp itself (21 CFR 1040. 20(d)(3)
and 21 CFR 1010.2(b)). The certification statement can be in any of the
following acceptable formst
"Product complies with Dl:lHS radiation performance standard, 21 CFR
1040.20"
"Product complies with DHHS radiation performance standards, 21 CFR
Chapter 1, Subchapter J"
11
"Product complies with applicable DHHS standards under the Radiation
Control for Health and Safety Act of 1968."
Other certification statements may be used as long as they are clear and
unambiguous.
5. The ratio of the irradiance within the wavelength range of greater than 200
nanometers through 260 nanometers to the irradiance within the wavelength
range of greater than 260 nanometers through 320 nanometers must not exceed
0.003 at any distance and direction from the lamp {21 CFR 1040.20{c)(1)).
6. The lamp must not be capable of insertion and operation in the "single-contact
medium screw" and/or "double-contact medium screw" lampholders (21 CFR
1 040.20{c)6).
7. Instructions containing information as required by 21 CFR 1040.20(f) must be
provided to the purchasers of the lamp not accompanying a sunlamp product.
8. A clear identification by brand and model designation of all lamp models for
which the lamp manufacturer claims equivalency (see CDRH policy on lamp
compatability dated September 2, 1986).
B. FOR SUNLAMP PRODUCTS
1. Each sunlamp product must have a label that contains the information as required
by 21 CFR 1040.20{d){l)(i) through (vi), including the warning statement,
designation of the ultraviolet lamp type which is to be used in the product,
instructions concerning minimum use distance, recommended exposure schedule,
and a statement of the time it may take before the expected results appear. The
design must clearly define the contents and locations of the labels required in
Section 1040.20{d), and their permanence and visibility must be assured. See
CDRH policy statement on warning label dated June 25, 1985, and policy
statement on "Maximum timer interval and exposure schedule" {Appendix C).
2. Each sunlamp product must bear a label or tag that contains the full name and
address of the manufacturer, place of manufacture and month and year of
manufacture {21 CFR 1040.20(d) and 21 CFR 1010.30(a)). If the sunlamp product
is sold under a name other than the manufacturer, the full name and address of
the company under whose name the product is sold may be set forth, provided
sufficient information is furnished to CDRH to allow the Agency to identify the
manufacturer. The place of manufacture may be expressed in code.
However, no specific place of manufacture identification will be required if the
place of manufacture is the same as the address of the manufacturer stated on
the label or if the manufacturer has only one place of manufacture for such
products and has identified that place of manufacture to CDRH. The month and
year of manufacture must appear in full, without abbreviation, as in the following
example - Manufactured December 1998.
3. Each sunlamp product must also bear a label or tag which contains a statement
by the manufacturer certifying that the product conforms to the requirements of
the Federal performance standard for such products {21 CFR 1040.20{d) and 21
12
CPR 1010.2(b)). The certification statement can be in any of the following
acceptable forms:
"Product complies with DHHS radiation performance standard, 21 CPR
1040.20"
"Product complies with DHHS radiation performance standards, 21 CPR
Chapter 1, Subchapter J"
"Product complies with applicable DHHS standards under the Radiation
Control for Health and Safety Act of 1968."
Other certification statements may be used as long as they are clear and
unambiguous.
4. The ratio of the irradiance within the wavelength range of greater than 200
nanometers through 260 nanometers to the irradiance within the wavelength
range of greater than 260 nanometers through 320 nanometers shall not exceed
0. 003 at any distance and direction from the product (21 CPR 1040.20(c)(1)). In
meeting this requirement, a manufacturer of sunlamp products may rely on a
certification of compliance made by the manufacturer of the ultraviolet lamp.
5. A sunlamp product must incorporate a timer with multiple timer settings
adequate for the recommended exposure time intervals for different exposure
distances and expected results of the product as specified on the product label.
The maximum timer interval shall not exceed the recommended maximum
exposure time. The timer must have an accuracy of 10 percent of the maximum
timer interval.
6. A sunlamp product must incorporate a control to enable the user to manually
terminate radiation emission from the sunlamp product at any time, without
disconnecting the electrical plug or removing the uv lamps.
7. The timer may not automatically reset and restart the sunlamp p r o u ~ t for a
period greater than the unused portion of the timer cycle, when emission was
terminated.
8. The spectral transmittance of the protective eyewear must not exceed a value of
0.001 over the wavelength range of greater than 200 nanometers through 320
nanometers and a value of 0.01 over the wavelength range of greater than 320
nanometers through 400 nanometers, and must be sufficient over the wavelengths
greater than 400 nanometers to enable the user to see clearly enough to read the
labels and reset the timer. There is protective eyewear on the market that
appears to comply with this requirement. The sunlamp product manufacturer
could require a certification by the protective eyewear manufacturer that the
product complies with the applicable portion of the standard.
9. Instructions containing all the information required by 21 CPR 1040.20(!)(1) must
be provided to the purchasers of the product.
13
14
APPENDIX B
FOOD AND DRUG ADMINISTRATION
WINCHESTER ENGINEERING AND ANALYTICAL CENTER
PROCEDURES FOR LABORATORY COMPLIANCE TESTING
OF SUNLAMP PRODUCTS AND ULTRA VIOLET LAMPS
INTENDED TO BE USED IN SUNLAMP PRODUCTS
FEBRUARY, 1981
INTRODUCTION
The purpose of this document is to establish procedures for laboratory compliance
testing of Sunlamp Products and Ultraviolet Lamps intended to be used in Sunlamp
Products that are certified by the manufacturer as in compliance with the "Sunlamp
Products: Performance Standard" published in 21 CFR Parts 1010 and 1040. The
procedures described in this document are applicable to any ultraviolet la mp and
product containing such lamps intended for irradiation of any part of the human body by
light (of wavelength in air less than 320 nanometers) to induce skin tanning. Definitions
of terms used in this document are identical to definitions published in 21 CFR Parts
1000 through 1040.
*SPECIAL INSTRUCTIONS*
PERFORM ALL OPERATIONS WITH EXTREME CARE. ULTRAVIOLET RADIATION,
OZONE AND ELECTRICAL SHOCK ARE ALL HAZARDS
INHERENT WITH THESE TESTS.
GENERAL INSTRUCTIONS
1. Laboratory control of sunlamp products and ultraviolet lamps intended to be used in
sunlamp products shall be maintained as specified in the FDA Regulatory Procedures
Manual, the Analyst Guidance Manual, and Compliance Program 7390.804E
"Compliance Testing of Sunlamp Products at WEAC."
2. Upon notification of assignment of a Sunlamp/Ultraviolet Lamp sample for analysis,
the analyst shall arrange with the Sample Custodian to transfer the sample to the
testing laboratory.
3. All test data shall be recorded on Analyst Worksheets (FD-431 and FD-431a) and
appropriate special data sheets.
4. The sample number shall be acquired from the "Official Seal" (FD-414a) on the
sample and verified with the "Collection Report" (FD-464) and be clearly printed on
all worksheets. If no "Official Seal" is present on the sample, consult your
supervisor.
5. Before breaking the seal and removing the sample from its shipping carton, the
analyst shall initiate an Analyst Worksheet (FD-431) in accordance with Chapter 9 of
the Analyst Operations Manual.
15
6. All accompanying materials shall be read and understood before any testing
procedures are initiated. Consult your supervisor if any problems are encountered
with the materials.
COMPLIANCE TESTING
I. Introduction
The compliance tests delineated in this part are performend on all Sunlamps and
Sunlamp Products collected for analysis at WEAC under compliance program
7390.804E.
II. Sunlamp Products
A. Performance
1 Measurement of Irradiance Ratio Limit, Timer Accuracy, and Transmission of
Protective Eyewear shall be carried out according to the procedures outlined
in Appendices A through C [of this Appendix] respectively.
2. Radiation Emission Controls
a. Verify that there is a user control on the sunlamp product that will
terminate radiation emission at any time without disconnecting the elec-
trical plug or removing the ultraviolet lamp.
b. Verify that when radiation emission from a sunlamp product has been
terminated for any reason, including termination by a timer, the product
will not automatically recycle and restart.
B. Labels
1. Identification Labeling
a. Visually examine the identification label for (a) presence, (b) permanence,
(c) legibility, (d) accessibility, and (e) correctness.
b. Record the name of the manufacturer and the month and year of
manufacture.
2. Certificaton Labeling
a. Visually examine the certification label for (a) presence, (b) permanence,
3. Verify that the following statement is present on the product warning label:
a. "DANGER--Ultraviolet radiation. Follow instructions. Avoid over-
exposure. As with natural sunlight, overexposure can cause eye and skin
injury and allergic reactions. Repeated exposure may cause premature
aging of the skin and skin cancer. WEAR PROTECTIVE EYEWEAR;
16
1
I
l
\
(
FAILURE TO MAY RESULT IN SEVERE BURNS OR LONG-TERM INJURY
TO THE EYES. Medications or cosmetics may increase your sensitivity to
the ultraviolet radiation. Consult physician before using sunlamp if you
are using medications or have a hist ory of skin problems or believe
yourself especially sensitive to sunlight. If you do not tan in the sun, you
are unlikely to tan from the use of this product."
Verify that the following recommendations, directions, warnings, and state-
ments are present on the product warning label:
b. A recommended exposure position (may be expressed in distance or though
the use of markings or other means, i.e., product design).
c. A warning that exposure at distances less than the minimum use distance
is not recommended.
d. A recommended exposure schedule that complies with CDRH policy state-
ment dated August 21, 1986.
e. A statement of the time it may take before the expected results appear.
f. Designation of the ultraviolet lamp type which is to be used in the product.
C. Instructions
Verify that the users' instructions contain the following:
1 . A reproduction (color optional) of the label described in 3 above, prominently
displayed at the beginning of the instructions.
2. A statement of the maximum number of people who may be exposed to the
product at the same time and a warning that only that number of protective
eyewear has been provided.
3. Instructions for obtaining repairs and recommended replacement components
and accessories which are compatible with the product, including compatible
eyewear, ultraviolet lamps, timers, reflectors, and filters, and which will, if
followed, result in continued compliance with the Standard.
III. Ultraviolet Lamps
A. Performance
1. Measurement of Irradiance Ratio Limit shall be carried out according to the
procedures outlined in Appendix A [of this appendix].
2. Compatability of Lamps
a. Determine visually, by physical measurement, or trial that the lamp
cannot be inserted into either a "Single-contact medium screw" or
"Double-contact medium screw" lampholder.
b. For lamps claiming "equivalency" to other lamps, conduct the testing as
specified in our policy statements of August 21 and September 2, 1986.
17
B. Labels
1. Identification Labeling
a. Visually examine the identification label for (a} presence, (b) permanence,
b. Record the name of the manufacturer and the month and year, or codes, of
manufacture as marked on the inscribing tags, or labels, or on the
packaging.
2. Certification Labeling
a. Visually examine the certification label as required under 1010.2 for: (a}
presence, (b) permanence, (c) legibility, (d) accessibility, and (e)
correctness.
3. Other Labeling
Determine the presence of labels on the lamp stating:
a. radiation. Follow instructions."
b. The model identification
c. "Use ONLY in fixture equipped with a timer."
4. If the labeling in 1 and 2 above is not permanently affixed or inscribed on the
lamp but is on the lamp packaging, determine that the name of the
manufacturer and month and year are permanently affixed or inscribed on the
exterior surface of the ultraviolet lamp so as to be legible and readily
accessible to view. The name of the manufacturer and month and year of
manufacture required to be permanently affixed or inscribed on the exterior
surface of the lamp may be expressed in code or symbols.
C. Instructions
Verify that the users' instructions contain the following:
1. A reproduction (color optional) of the label required in 3 above, prominently
displayed at the beginning of the instructions.
2. A warning statement with the words "DANGER--Ultraviolet radiation. Follow
instructions. Avoid overexposure. As with natural sunlight, overexposure can
cause eye and skin injury and allergic reactions. Repeated exposure may
cause premature aging of the skin and skin cancer. WEAR PROTECTIVE
EYEWEAR; FAILURE TO MAY RESULT IN SEVERE BURNS OR LONG-TERM
INJURY TO THE EYES. Medications or cosmetics may increase your
sensitivity to the ultraviolet radiation. Consult physician before using
sunlamp if you are using medications or have a history of skin problems or
believe yourself especially sensitive to sunlight. If you do not tan in the sun,
you are unlikely to tan from the use of this product."
3. A warning that the instructions accompanying the sunlamp product should
always be followed to avoid or to minimize personal injury.
18
l
t
I
4. A clear identification by brand and complete model designation of all lamp
models for which the lamp manufacturer claims equivalency.
APPENDIX A
TEST PROCEDURE FOR MEASUREMENT OF IRRADIANCE RATIO LIMIT
USING THE
OPTRONICS 747 SPECTRORADIOMETER
AND
HP 9825 SYSTEM CONTROLLER
I. Initial Start-Up
A. Spectroradiometer Switch On
1. Switch on the Power console (the unmarked chassis below the calculator).
2. Switch on the Control console. Verify that the pre-amp light below the PM
housing on monochronometer is on. Shutter the PM tube.
3. Switch on the Pacific Photometric Instruments, push the "reset" and set the
control to 2.42, verify that the PM housing fan is operating.
4. Set the following "switches" to the noted position: "Filter wheel", Auto; "Pbs
cooler", off; "Response time", 0.3; "Jog'', normal; "Sphere position", off.
5. "Ranging" manual, use pushbuttons to set to the Xl0-0 range.
6. Set switches for "DC gain", "PMT," "Chopper," off.
7. "Wavelength," stop.
8. Continue with B while the system warms up for 90 minutes.
B. System Controller Switch On
1. Remove any dust covers, especially from t.he HP 9872A Plotter.
2. Place the 69408 Multiprogrammer in Local mode then switch on.
3. 4880 Coupler Qlli the "SRQ" Inhibit switch should be out.
4. Turn on the 3455 DVM, 3495A Scanner, 9872A Plotter, 98668 Printer and the
9825 Calculator.
5. Turn on the two Kepco JQE power supplies, Kepco BHK (AC only, DC off) and
the Sorenson ACR 3000.
6. Switch the Oriel C73-14 on.
19
II. Set-Up
A. Set 747 signal to -0.001 to 0.001 (X10-0 range) using Zero Offset and manual
range change.
B. Set the high voltage to 8.000 (800 volts).
C. Adjust the Zero potentiometer on the pre-amp housing to null out dark current on
the 10-4 range.
D. Set to Auto range.
E. Verify that the 3-in. integrating sphere is in place with the aperture towards the
reference lamp.
F. Verify that the proper slits are in place, normally they are 1 nm entrance and
exit and 2 nm middle.
1. Turn HeNe laser on.
2 . Check laser alignment on optical bench.
3. Check that laser is centered on aperture plug and that aperture is normal to
laser beam.
4. Remove aperture plug.
III. Wavelength Check
A. Enter erase a on the 9825 and press "Execute". Insert tape #14 and press "Load"
and "Execute". Enter ldk 1 and press "Execute". Place the sunlamp overlay over
the special function keys. Press "clr 704".
B. Switch the 6940 Multiprogrammer to Remote, press "shut down" on the 9825.
C. Set the wavelength to 631.9, select jQg mode, .1 nm jog interval and 20 nm/sec
scan speed.
D. Press "Run" on the calculator, respond to displayed prompts. If no answer is
appropriate or for no, press "Continue" without any entry. Start wavelength 632,
stop 634.
E. Follow program prompts.
F. Turn laser off.
G. Place the Hg lamp attached to the Oriel C73-14 at the aperture. CAUTION: The
lamp housing is hot and the lamp is producing UV.
H. Repeat A through C using:
1. Set 251.9, start 252, stop 254.
2. Set 311.9, start 312, stop 314.
20
I
i
I
}
3. Set 363. 9, start 364, stop 366.
4. Set 434.9, start 435, stop 437.
5. Set 544. 9, start 545, stop 54 7.
I. Turn off C73-14 and remove lamp from aperture.
J. Complete the wavelength calibration chart. If the mean deviation is greater than
.5 nm, make wavelength drive adjustments as described in the Model 747
instruction manual.
IV. Calibration Check
A. Place the alignment jig in the standard lamp bi-pin socket. Turn laser on and
verify or establish socket alignment so that the beam is centered and normal to
the plane of the bi-pin. Turn the laser off.
B. Establish 50 em. distance from sphere aperture to the alignment jig using the
50-cm reference gauge.
C. Set the 10-turn potentiometer located above the plotter to between 4 and 6.
D. Set the radiometer to 199.9 nm, 1 nm jog interval, and 50 nm/sec scan speed.
E. Place working standard in the socket, place baffles and shields to reduce stray
light on the radiometer.
F. Press "Run" on the calculator. Respond to displayed prompts. Start wavelength
200, end 400, current 7.9
G. After 15 minutes adjust the potentiometer so that the DVM reading is between
0. 7899219 and 0. 7899999 and press "Continue".
H. Compare luminous intensity values at 250, 270, 300, 350, and 400 nm with the
working standard values. If they differ by more than 4 percent, a new calibration
factor will have to be determined using a standard lamp.
V. Sunlamp Test
A. Perform erase a "Execute" on the 9825, insert tape # 13 and press "Load" and
"Execute", enter ldk 1 and press "Execute".
B. Set up the sample at the specified minimum use distance from the detector
aperture, measuring the distance as described in the sunlamp manufacturer's
instructions. Gauge rods are available for the most common distances. Another
distance may be used; however, it must be no less than the manufacturer's
minimum use distance.
C. Lock the timer on. Arrange baffles to reduce stray light. Check that the lamp
switch is off and plug the 120v sunlamp into the Sorenson supply.
D. Press "Run" on the calculator, answer the display prompts. Turn the lamp switch
on and adjust the variac to 130v using the DVM.
21
E. After 5 minutes, or the manufacturer's recommended warm-up time, check and
adjust the variac as necessary and press "Continue".
F. At the end of the data run, respond to the prompts. If plotting is desired make
sure there is paper in place on the plotter.
APPENDIX 8
TEST PROCEDURE FOR
SUNLAMP TIMER ACCURACY
1. Determine maximum timer interval
2. Compute maximum allowable timer error (10 percent of maximum interval)
3. Test accuracy of timer at maximum interval
a. Set timer to its maximum
b. Simultaneously start timer and suitable timing device (stopwatch, clock, etc).
c. When interval is complete, compute difference between sunlamp timer setting
and measured length of time.
d. Compare timer error with maximum allowed error to determine compliance
status.
e. Repeat steps a-d.
4. Test accuracy of timer at the 1 minute setting.
a. Perform steps 3a-3e above using a setting of 1 minute instead of the maximum
interval setting. If there is no setting for 1 minute, use the available setting
closest to 1 minute.
5. Test accuracy of the timer at the minimum marked setting, provided it is different
from .the setting used in step 4 above.
a. Perform steps 3a-3e above, using the minimum marked timer setting instead of
the maximum interval setting.
APPENDIX C
TEST PROCEDURE FOR
EVALUATION OF PROTECTIVE EYEW.EAR
This test is to be performed using a Carey Model 118 Spectrophotometer.
22
1. Initial set-up control settings:
"Power" switch ON
"Function" switch 1096T
"Mode" switch SB
"Source Select" switch VIS
"UV Source" switch OFF
"VIS Source" switch HIGH
2. Open the sample compartment and remove both cell holders. Place the universal
positioning jig in the sample chamber. Clamp the protective eyewear to be tested
in the clip and adjust the base and arm so that one lens intercepts the sample beam
about halfway across the chamber. The sample beam is a thin vertical green light
in the front of the chamber. The lens should face left so the beam is incident on
the front. Orient the lens so that it is perpendicular to the beam and intercepts the
entire beam. A piece of white paper will help to locate the beam and to ensure
that the entire beam is intercepted. Make sure no part of the eyewear or
positioning jig intercepts any of the reference beam that also runs across the
chamber. Close the chamber and turn "Source Select" switch OFF.
3. Set the "Source Select" switch to UV and the "UV Source" switch to ON. Note the
time to allow a 30-minute warmup of the UV lamp.
4. Control settings for data run:
"Beam Interchange" switch NORMAL
"Mode" switch AUTO SLIT
"Period" switch !
"Function" switch 1096T
"Wavelength" switch OFF
"Chart" switch OFF
"Scan Speed" switch 2 nm/sec
,;Chart Selector" switch 50 mm/in
"Baseline" switch IN
"Baseline UV-VIS" switch UV
5. Using the knob on the left side of the cabinet, set the wavelength to 375 nm.
6. Set the chart speed so that the pen will start on a line. Turn on wavelength and
chart drive toggle switches.
7. Turn "Scan Switch" to"-". When wavelength reaches 200 nm, turn off.
8. Turn off pen and chart switches and turn "Mode" switch to SB.
9. Open the sample chamber, remove the universal positioning jig and eyewear, and
replace the cell holders.
10. Put eyewear on. Determine if labels and timer graduations can be read.
determine if there are light leaks around the eyewear.
23
Also
24
APPENDIX C
POLICY ON MAXIMUM TIMER INTERVAL AND EXPOSURE SCHEDULE
FOR SUNLAMP PRODUCTS
25
DEPARTMENT OF HEALTH & HUMAN SERVICES
Aug. 21, 1986
Public Heal th Service
Rockville MD 20857
TO: ALL MANUFACTURERS, IMPORTERS AND POTENTIAL MANUFACTURERS OF
SUNLAMP PRODUCTS
SUBJECT: POLICY ON MAXIMUM TIMER INTERVAL AND EXPOSURE SCHEDULE
FOR SUNLAMP PRODUCTS
BACKGROUND:
The amended performance standard for sunlamp products (21 CFR 1040.20) was
published in the September 6, 1985 issue of the Federal Register and will become
effective September 8, 1986. Any sunlamp product manufactured on or after that date
must comply with the amended standard.
The ten (10) minute maximum timer interval requirement was removed from the original
performance standard since there are newer sunlamp products on the market for which
ten (10) minutes is not appropriate. The maximum timer interval now depends on the
intensity and spectral distribution of ultraviolet (UV) radiation emission of each
individual model of sunlamp product and must not exceed the maximum recommended
exposure time provided on the required product warning label. Therefore, sunlamp
product manufacturers must develop an exposure schedule and establish the maximum
recommended exposure time (and therefore the maximum timer interval) based on the
characteristics of their particular products.
The intended purposes of a sunlamp product t imer are to provide for r.eliable control of
exposures and to limit acute (and delayed) damage from unintentionally long exposures.
However, the maximum timer setting should also allow for selection of exposure times
needed to build up and maintain a tan. The maximum timer interval is in no way to be
considered as a safe limit; all ultraviolet radiation is potentially hazardous.
The standard requires the manufacturer to provide an exposure schedule in the product
warning label. The purpose of the exposure schedule is to allow a person to gradually
build-up skin pigmentation and to maintain a tan while controlling the risk of acute
injury and delayed adverse effects. Since the UV radiation dose that causes a barely
discernible pink coloration (minimal erythemal dose or MED) is not the same for
different skin types, the exposure schedule for first time users will depend on the skin
type of the user. Furthermore, suberythemogenic doses of UV radiation received at 24
hours intervals initially lead to lowering of the erythema and tanning thresholds.
Therefore, the exposure schedule and maximum recommended exposure time should be
constrained by the potential for erythema as well as the quantity of radiation necessary
to achieve and maintain a tan.
POLICY:
The Center for Devices and Radiological Health (CDRH) will use the following criteria
to evaluate the adequacy of the exposure schedule and the recommended maximum
exposure time (and therefore the maximum timer interval):
26
1) The maximum recommended exposure time (and maximum timer interval) must
not exceed a value which will result in an exposure of four {4) times the minimal
erythema dose (MED) for untanned Type II skin (always burns, then tans slightly).
This is based on the CDRH Erythema Action Spectrum [proposed action spectrum
of Commission lnternationale de L'Eclairage (CIE) modified by CDRH]. See
Appendix A [of this appendix] for the action spectrum and weighting factors and
equations needed to derive it.
The formula for determining the recommended maximum exposure time, "Te" in
seconds is:
624JIM
2
T =---
e "'V.R .
,t_ l I
where Standard MED = 156J/M
2
at 296 nm
Vi = weighting factor
Ri = irradiance in W/M
2
2) The recommended maximum exposure time must not exceed a value which will
result in an exposure of four (4) times the minimal melanogenic dose (MMD) for
untanned Type II skin. This is based on the melanogenic action spect rum
developed by Parrish et al. (1982). See Appendix B [of this appendix] for this
action spectrum.
The formula for determining the recommended maximum exposure time, "Tm" in
seconds is:
T =
m
1836J/M
2
"'J.R .
,t_ I I
where standard MMD = 459J/M
2
at 296 nm
Ji = weighting factor
Ri = irradiance in W/M
2
3) The recommended exposure schedule should provide for exposures of no more
than 0. 75 MED three times the first week, gradually increasing the exposure the
following weeks until maximum tanning has occurred (approximately four weeks
total) and then provide for maintenance of a tan by biweekly or weekly exposures
of up to four (4) MEDs or four (4) MMDs, whichever is less.
CDRH believes that the above criteria balance the need to limit acute (and delayed)
damages from unintentionally long exposure and the need to provide for single exposure
durations adequate to achieve and maintain a tan.
Appendices
Walter E. Gundaker, Director
Radiological Health
27
.;':
. ..1
l
0
9
0
-4
f -3
w
e
-2
t
n
9
F
a
(
t
0
-1
r 0
s
Weighting factors for erythema, Vi
250 275 300 325 350
Wavelength (nm)
The equations describing the curve are:
V. (A)= 1.0(250 < A < 302 nm)
I
V. (A)= 10
0114
<302-A> (302 <A < 325 nm)
'
V. (A)= 10
0161 059
-A> (325 <A< 405 nm)
'
28
375
Appendix A
(page 1)
400
1.0E +00
w
e
i
l .OE-{)1
t
i
n
g
F
1.0E02
a
(
t
0
r
l.OE-03
Ji
l.OE-{)4
Weighting factors for melanogenesis, Ji
250 300 350
Wavelength (nm)
400
Appendix B
(page 1)
450
The MMD as function of wavelength has been interpolated (using log MMD) from the
action spect rum for melanogenesis of Type II skin (Parrish et al. 1982)
29
Appendix B
(page 1)
Parrish Melanogenesis Type. II Skin 1982 Normalized to 296 nm
Wavelength Ji Wavelength J Wavelength J
250
251 .374828 303 .. 750391 355 .972644E-03
252 .371428 304 .690261 356 .944186E-03
253 .366714 305 .502296 357 .916645E-03
254 .364225 306 .36551 358 .890022E-03
255 .360783 307 .265997 359 .863859E-03
256 .35734 308 .193565 360 .838613E-03
257 .353943 309 . 14087 361 .813826E-03
258 .350547 310 .102497 362 .789958E-03
259 .347196 311 .745893E-01 363 .767007E-03
260 .343891 312 .054301 364 .747729E-03
261 .340632 313 .395016E-01 365 .722942E-03
262 .337419 314 .341137E-01 366 .666943E-03
263 .334206 315 .294593E-01 367 .615075E-03
264 .331039 316 .254384E-01 368 .567338E-03
265 .327672 317 .219683E-01 369 .523272E-03
266 .327413 318 .189709E-01 370 .48288E-03
267 .326954 319 .163821E-01 371 .44547E-03
268 .326449 320 .141467E-01 372 .410953E-03
269 .32599 321 .122143E-01 373 .379097E-03
270 .325531 322 .105481E-01 374 .34972E-03
271 .325072 323 .. 91137E-02 375 .322593E-03
272 .324613 324 .786745E-02 376 .297577E-03
273 .324154 325 .679336E-02 377 .274534E-03
274 .323695 326 .586616E-02 378 .253236E-03
275 .323236 327 .506748E-02 379 .233591E-03
276 .321445 328 .437483E-02 380 .215506E-03
277 .319609 329 .377812E-02 381 .213532E-03
278 .317865 330 .326265E-02 382 .211558E-03
279 .316075 331 .281741E-02 383 .209963E-03
280 .314285 332 .243276E-02 384 .207702E-03
281 .312541 333 .210089E-02 385 .205821E-03
282 .31075 334 .181447E-02 386 .203939E-03
283 .351694 335 .176123E-02 387 .202057E-03
284 .398008 336 .170982E-02 388 .200221E-03
285 .450427 337 .165978E-02 389 .189296E-03
286 .509732 338 .161113E-02 390 .196594E-03
287 .576885 339 .156385E-02 391 .194804E-03
288 .652851 340 .151841E-02 392 .193014E-03
289 .738778 341 .147388E-02 393 .191224E-03
290 .836088 342 .143074E-02 394 .18948E-03
291 .861518 343 .138897E-02 395 .187735E-03
292 .887498 344 .134812E-02 396 .186037E-03
293 .91435 345 .130864E-02 397 .184339E-03
294 .94212 346 .127054E-02 398 .18264E-03
295 .970625 347 .123336E-02 399 .180988E-03
296 1.00000 348 .11971E-02 400 .179336E-03
297 .990959 349 .116222E-02 401 .177683E-03
298 .982054 350 .112825E-02 402 .176077E-03
299 .973287 351 .10952E-02 403 .17447E-03
300 .96429 352 .106307E-02 404 .172864E-03
301 .886993 353 .103232E-02 405 .171257E-03
30 *U. s.COII(RNIIENT PRINTING OH 1Ctt19B8-201-850: 60453
Report to Congress
Labeling Information on the Relationship Between the Use of Indoor
Tanning Devices and Development of Skin Cancer or
Other Skin Damage
Submitted Pursuant to Section 230 of the Food and Drug Administration
Amendments Act of 2007
____________ Date _____ _
Andrew C. von Eschenbach, M.D.
Commissioner of Food and Drugs
Executive Summary
Section 230 of the Food and Drug Administration Amendments Act of 2007 (FDAAA)
requires FDA to make certain determinations regarding the effectiveness of warning labels for
indoor tanning devices in conveying information to consumers regarding the risks that such
devices pose for the development of irreversible damage to the eyes and skin, including skin
cancer. FDA is required to conduct consumer testing in making these determinations, and to
submit a report to Congress providing these determinations and identifying measures being
implemented to significantly reduce the risks posed by indoor tanning devices. FDA is
submitting this report to Congress in response to that mandate.
Based on its analysis of the results of the consumer study required by section 230 of FDAAA,
FDA has determined that there are warnings that are capable of adequately communicating
the risks of indoor tanning, and that a modified warning statement label may more effectively
convey these risks than the current labeling requirements. FDA has also determined that
changes to the positioning requirements for the warning statement label may communicate
such risks more effectively.
As a result of these determinations, FDA is considering amending the warning label
requirements for sunlamp products to include specific formatting requirements to more clearly
and effectively convey the risks that these devices pose for the development of irreversible
damage to the eyes and skin, including skin cancer. To significantly reduce the risks
associated with the use of these devices, FDA is also considering amending the performance
requirements in the sunlamp products performance standard, and has begun educational
outreach efforts to better inform consumers about the risks of indoor tanning.
Table of Contents
I. Introduction ................ ................................ ................................ ................................ ..... 1
II. Background ..... ........ .... .................... ........ .... ............................ .... ............................ .... .... 2
A. Current Label Requirements and FDA Policy .. ... .. .. .. .. .. .. .. .. .. .. .. .. .. ... ... .. .. .. .. .. .. .. .. . 2
B. Reviewing the FDA Performance Standard for Sunlamp Products .. .... .. ... ... ... .. .. . 3
III. FDA Actions in Response to Section 230 ................................................................... .4
A. Consumer Testing of Warning Label ... .................... .... .. .. .... .................... .... .. .. ... .4
L. Research Objectives .. .... ... .. .. .. .. ... ... ... .. ... ... .... ... .. .. .. .. ... ... ... .. ... ... .... ... .. . 4
2. Methodology ........................................................................................ 4
3. Warning Statements Tested in Consumer Focus Groups .................... 5
4. Key Findings ... .... ... .. ....... ... .. ....... .... .... ... .. ....... ... .. ....... .... .... ... .. ....... ... . 6
5. Location of Warning Statement.. .... .. .. .. .. .. .. .. .. .. .. ... .. .... ... .. .. .. .. .. .. .. .. .. .. . 7
B. Measures to Reduce Risks Associated with Sunlamp Products .. .. ... ... .. .. ... .... ... .. . 7
1. Labeling and Performance Standard ................................................... 7
2. Educational Outreach .. .... .... .... .................... .... .... .... .................... .... .... 8
IV. Conclusion ... .. .. ... .... .. ... .. .. .. .. .. ... ... .. .. ... .... .. ... .. .. .. .. .. ... ... .. .. ... .... .. ... .. .. .. .. .. ... ... .. .. ... .... .. ... . 8
I. Introduction
On September 27, 2007, Congress passed the Food and Drug Administration
Amendments Act of 2007, Public Law 110-85 (FDAAA). Section 230 of FDAAA
provides:
SECTION 230. REPORT BY THE FOOD AND DRUG
ADMINISTRATION REGARDING LABELING
INFORMATION ON THE RELATIONSHIP BETWEEN THE
USE OF INDOOR TANNING DEVICES AND DEVELOPMENT
OF SKIN CANCER OR OTHER SKIN DAMAGE.
(a) IN GENERAL- The Secretary of Health and Human
Services (referred to in this section as the "Secretary"), acting
through the Commissioner of Food and Drugs, shall determine-
( 1) whether the labeling requirements for indoor
tanning devices, including the positioning requirements,
provide sufficient information to consumers regarding the risks
that the use of such devices pose for the development of
irreversible damage to the eyes and skin, including skin cancer;
and
(2)(A) whether modifying the warning label required on
tanning beds to read, ''Ultraviolet radiation can cause skin
cancer," or any other additional warning, would communicate
the risks of indoor tanning more effectively; or
(B) whether there is no waming that would be
capable of adequately communicating such risks.
(b) CONSUMER TESTING.- In making the determinations
under subsection (a), the Secretary shall conduct appropriate consumer
testing to determine consumer understanding of label warnings.
(c) REPORT.- Not later than 1 year after the date of the
enactment of this Act, the Secretary shall submit to the Congress a
report that provides the determinations under subsection (a). In
addition, the Secretary shall include in the report the measures being
implemented by the Secretary to significantly reduce the risks
associated with indoor tanning devices.
This provision requires the Food and Drug Administration (FDA) to determine
whether existing labeling requirements for indoor tanning devices provide sufficient
information to consumers regarding the risks these devices pose for the development
of ineversible damage to the eyes and skin, including skin cancer; whether modifying
the required warning label would better communicate such risks; or whether there is
no warning capable of adequately communicating these risks. In making these
determinations, FDA must conduct appropriate consumer testing to determine
consumer understanding of label warnings. FDA must also submit a report to
1
Congress by September 27, 2008, providing these detenninations and identifying the
measures being implemented to significantly reduce the risks associated with such
devices. FDA is submitting this report to Congress as required by section 230.
II. Background
A. Current Label Requirements and FDA Policy
FDA promulgated the sunlamp products
1
performance standard, 21 Code of
Federal Regulations (C.P.R.) 1040.20, in 1979, 44 Fed. Reg. 65,352 (November
9, 1979), and most recently amended it in 1985, 50 Fed. Reg. 36,548 (September
6, 1985). This regulation requires each sunlamp product to have a label that
contains a warning statement with the words:
DANGER-Ultraviolet radiation. Fo1low instructions. A void
overexposure. As with natural sunlight, overexposure can cause eye
and skin injury and allergic reactions. Repeated exposure may cause
premature aging of the skin and skin cancer. WEAR PROTECTIVE
EYEWEAR; FAILURE TO MAY RESULT IN SEVERE BURNS OR
LONG-TERM INJURY TO THE EYES. Medications or cosmetics
may increase your sensitivity to the ultraviolet radiation. Consult
physician before using sunlamp if you are using medications or have a
history of skin problems or believe yourself especially sensitive to
sunlight. If you do not tan in the sun, you are unlikely to tan from the
use of this product.
21 C.P.R. 1040.20(d)(1)(i). The regulation does not specify requirements for the
format in which these words must appear, or the exact location on the product that
the warning label must appear, as long as it is "permanently affixed or inscribed
on an exterior surface of the product when fully assembled for use so as to be
legible and readily accessible to view by the person being exposed immediately
before the use of the product." 21 C.P.R. 1040.20(d)(3)(i).
FDA also issued a letter dated June 25, 1985, the warning label to
sunlamp product manufacturers outlining FDA policy. The policy letter states:
The intended purpose of the warning label required on sunlamp
products is to provide that information necessary for the consumer to
make an informed decision regarding the risks of using sunlamp
products and to provide adequate directions for skin tanning.
Therefore, the label must be legible and conspicuously placed on the
1
"Sunlamp products" and "indoor tanning devices" have the same meaning. This report uses the
former term because our performance standard refers to "sunlamp products."
2
"Policy on Warning Label Required on Sunlamp Products," available at
http://www. fda. gov/cdrh/radhlth/pdf/sunpolO l .pdf .
2
product so as to render it likely to be read by the user under normal
conditions of purchase and use. The Agency will consider sunlamp
products to be noncompliant with the performance standard under
Section 358(a)(l) of the RCHSA [Radiation Control for Health and
Safety Act]
3
and misbranded under Section 502(c) of the FD&C Act
if the required product label is not legible and accessible to view for
the following reasons:
1. The label required under 21 C.P.R. 1040.20(d)(l) does not appear
on a prominent part or panel which is presented or displayed under
normal conditions of purchase and/or use.
2. Adequate space is not provided for the required label or the label is
not prominently displayed on the device.
3. The normal individual can not read the label from a distance of one
meter because of inadequate lettering size and background contrast.
Lettering of ten ( 1 0) millimeters (height) for the word "DANGER"
and five (5) millimeters for the rest of the label information is
recommended to meet the visibility requirements.
B. Reviewing the FDA Performance Standard for Sunlamp Products
Research on the risks and benefits of ultraviolet radiation (UV) exposure has
advanced since the mid-1980s. FDA has been active in the development of
international standards in this area, particularly the International Electrotechnical
Commission (lEC) International Standard 60335-2-27 for sunlamp products. This
standard has been periodically updated; the most recent amendments were
published on April 27, 2007.
On February 9, 1999, FDA's Center for Devices and Radiological Health (CDRH)
pubbshed an Advance Notice of Proposed Rulemaking (ANPRM). 64 Fed. Reg.
6,288 (Docket No. 98N-1170). The purpose of the ANPRM was to seek input on
several proposed amendments to the performance standard for sunlamp products
that FDA was consideri ng at the time. One of the changes considered was to
update the warning statement to simplify the wording and highlight the risk of
skin cancer. Other changes considered included: (1) updating the recommended
exposure schedule and incorporating it into the standard itself (It currently
appears in an FDA policy letter dated August 21 , 1986.
4
); (2) requiring that the
warning statement appear in all catalogs, brochures, and specification sheets
pertaining to sunlamp products; and (3) developing a new system for rating the
3
The Safe Medical Devices Act of 1990 (Pub. L. No. 1 0 1-629) transferred the RCHSA' s provisions
from the Public Health Service Act to the Federal Food, Drug, and Cosmetic Act (FDCA). Section
358(a)(l) of the RCHSA can be found at section 534(a)(l) of the FDCA, 21 U.S.C. 360kk(a)( I).
4
"Policy on Maximum Timer Interval and Exposure Schedule for Sunlamp Products," available at
http://www. fda.gov/cdrh/radhJth/pclf!sunpoiO l.pclf.
3
biological efficacy of UV lamps to facilitate the choice of proper replacement
lamps.
Since 1999, CDRH has conducted clinical research on the effects of UV exposure.
FDA's Office of Science, Office of Women's Health, Center for Food Safety and
Applied Nutrition, and the National Cancer Institute (NCI) supported research on
UV risks in the general public and in tanners in particular. The results of these
studies have been published in numerous scientific publications (available on
request). The data generated in the FDA/NCI studies helped to define typical
erythema (sunburn) doses used to calculate exposure schedules and select the
maximum timer interval for tanning devices. They also showed that the UV
exposures typically provided by sunlamp products are excessive, and that
comparable cosmetic effects can be produced with cumulative exposures that are
only one-third or one-fourth the levels currently used.
FDA intends to use the results from the FDA/NCI research and the experience we
gained from working with international experts on the IEC standard when
considering any proposed changes to the FDA performance standard for sunlamp
products.
III. FDA Actions in Response to Section 230
A. Consumer Testing of Warning Label
1. Research Objectives
To conduct the consumer testing required by section 230(b) ofFDAAA, FDA
retained a contractor to evaluate the effectiveness of the warning labels on indoor
tanning devices in communicating the risks that use of the devices pose for the
development of irreversible damage to the eyes and skin, including skin cancer.
The objective of the focus group study was to explore consumers' perceptions and
understanding of the warning statement on the labels of sunlamp products (indoor
tanning equipment, tanning beds, and tanning booths) currently required by 21
C.P.R. 1040.20(d)(l)(i), as compared to their understanding of an alternative
warning statement. The groups explored the labels' content, language,
messaging, order, and format. In addition, the testing assessed the participants'
reaction to the location of the warning label on the device.
2. Methodology
The contractor used six focus groups, three in Rockville, Maryland, on
October 3, 2007, and three in Baltimore, Maryland, on October 18, 2007. A total
of 48 respondents participated in the focus groups. In each location, there was
one of each of the following groups:
4
Group 1: Teenagers in high school, ages 14-17, who either used or
considered using indoor tanning equipment.
Group 2: Adults with a college degree, ages 18 and older (majority 18-
30), who either used or considered using indoor tanning
equipment.
Group 3: Adults without a college degree, ages 18 and older (majority
18-30), who either used or considered using indoor tanning
equipment.
Each group had a combination of experienced indoor tanners and participants who
have never used indoor tanning equipment from both sexes. Prior to being shown
the labels, the participants were asked for their general opinions about indoor
tanning. In most of the focus groups, and without prompting, participants
mentioned some of the dangers associated with indoor tanning- specifically,
acute damage to the skin and skin cancer.
Participants were then given each warning statement (the one currently required
and an alternative, see below) one at a time and asked a series of questions
regarding the information contained in the warning, their understanding of it, and
how it might influence their behavior. Finally, participants were shown a life-size
picture of an indoor tanning bed and asked to evaluate where they would most
likely notice and read the warning statement label.
3. Warning Statements Tested in Consumer Focus Groups
Warning Statement Currently Required
5
DANGER - Ultraviolet radiation. Follow instructions. A void overexposure.
As with natural sunlight, overexposure can cause eye and skin injury and
allergic reactions. Repeated exposure may cause premature aging of the skin
and skin cancer.
WEAR PROTECTIVE EYEWEAR; F AlLURE TO MAY RESULT IN
SEVERE BURNS OR LONG-TERM INJURY TO THE EYES.
Medications or cosmetics may increase your sensitivity to the ultraviolet
radiation. Consult physician before using sunlamp if you are using
medications or have a history of skin problems or believe yourself especially
sensitive to sunlight. [f you do not tan in the sun, you are unlikely to tan from
the use of this product.
5
As explained above, the sunlamp products performance standard does not contai n formatting
requirements for the waming statement. FDA chose the formatting used in the study based on the use
of such formatting on certain tanning beds currently on the market.
5
Alternative Warning Statement Presented to Consumer Focus Groups
DANGER- Ultraviolet Radiation
A void overexposure- It may cause severe bums
Read instructions carefully
Ultraviolet Radiation causes:
Skin Cancer
Injury to the Eyes and Skin
Skin Aging
WEAR PROTECTIVE EYEWEAR TO PREVENT EYE INJURY
Certain medicines or cosmetics can increase your sensitivity to ultraviolet
radiation- Consult your physician before tanning
In this report, we refer to the two warning statements as "current" and
"alternative." This terminology was not used with the participants during the
focus group testing. Instead, the labels were referred to as "Label 1" and
"Label 2," and the order/numbering was alternated between the Baltimore and
Rockville sites to minimize order bias.
4. Key Findings
The majority of participants found the alternative warning statement easier to
understand than the current one. Most stated that they found the message to be
streamlined, and not as ambiguous as the current label.
The results of the testing indicate that the format of the current label was the
characteristic that made it least effective in communicating the dangers associated
with indoor tanning. Most participants found that the format of the current label
made it difficult to focus on and read due to the paragraph format and length of
the label. Responses indicated that seeing a paragraph-style warning statement
label in a real-world situation would cause them to disengage and ignore the
information.
Almost all participants ( 45 out of 48, the exceptions were three participants in one
of the teen groups) said they would be more likely to read the alternative label.
They stated that the shorter length and bulleted format made it easier to focus on
the risks and directives. Most participants also found the alternative label easier
to understand because of its clarity and simplicity. Comments explained that the
6
streamlined format and messaging made it more attention-grabbing and easier to
process. Participants stated that they were better able to pay attention to the range
of messages. Most said the alternative statement sent a stronger message about
the dangers associated with indoor tanning equipment.
Focus group participants made a number of recommendations for improving the
warning statement labels, including:
enlarge the font size and use the color red for the danger statement;
describe the information in order of importance;
move the need to wear protective eyewear higher up in the statement due
to its importance as indicated by its capitalization;
keep the term "long-term" to describe eye injury because that tenn is more
foreboding;
keep the descriptor "premature" to describe skin aging because it causes
more alarm; and
refer to the "exposure schedule," not "instructions," so that users will
understand what is meant by "overexposure."
Some participants expressed confusion with respect to the directive to "consult
your physician before tanning." Participants did not understand that this message
communicated that tanning and concurrent use of certain medicines and cosmetics
can be dangerous.
5. Location of Warning Statement
There was no single suggested location for the warning statement that a majority
picked as optimal. Participants suggested positioning the label in such places as
next to the control panel, centered on the canopy, or placed on the head side of the
canopy. Participants also recommended that the warning statement label should
be placed away from other labels on the tanning bed, so as not to detract from the
label's importance.
B. Measures to Reduce Risks Associated with Sunlamp Products
1. Labeling and Performance Standard
After reviewing findings from the consumer focus group study, FDA has
determined that an alternative warning may more effectively communicate the
risks of indoor tanning. The consumer focus group study serves as an important
step to inform FDA's deliberations regarding a more effective manner to
communicate the risks of using indoor tanning devices. FDA has also determined
7
that changes to the positioning requirements for the warning statement label may
communicate such risks more effectively.
In light of these determinations, FDA is considering amending the warning label
requirements for sunlamp products to include specific formatting requirements to
more clearly and effectively convey the risks that these devices pose for the
development of irreversible damage to the eyes and skin, including skin cancer.
FDA is also considering amending the performance requirements for sunlamp
products to ha1monize them with the IEC Intemational Standard. FDA is
continuing to evaluate labeling requirements for sunlamp products and may
identify further changes before undertaking any rulemaking to amend 21 C.F.R.
1040.20 with respect to the use of a warning label on sunlamp products.
2. Educational Outreach
FDA also believes that consumer education is an important component in
significantly reducing the risks associated with indoor tanning devices. To that
end, FDA has begun educational outreach efforts to better inform the target
audience before they go to a tanning salon. As part of this educational process,
FDA has expanded and revised its "Tanning" Web site, which may be found at
http://www.fda.gov/cdrh/tanning/. In a recent issue of FDA & You: News for
Health Educators and Students, a newsletter targeting teens and health educators,
FDA published an article entitled, "The Truth About Tanning: What You Need to
Know to Protect Your Skin," that focuses on how to avoid the risks posed by UV
radiation.
6
Accompanying this article is a lesson plan for health educators to help
students understand the risks associated with tanning and UV exposure. FDA
intends to publish reminders in future newsletters. FDA also provides links to
tanning information on its Consumer Information Web page at:
(http://www. fda. gov I cdrh/consumer/).
IV. Conclusion
As required by section 230 of FDAAA, FDA conducted consumer testing to
determine their understanding of label wamings for sunlamp products. As a result of
this testing, FDA has determined that certain modifications to the labeling
requirements for sunlamp products may communicate the risks of indoor tanning
more effectively and is considering rulemaking. FDA is also considering changes to
the performance requirements of the sunlamps performance standard, has undertaken
educational outreach including an expanded Website, and will continue its efforts in
this area as a means of significantly reducing the risks associated with sunlamp
products.
6
FDA & You, Issue No.7 (2005), available at http://www.fda.gov/cdrh/fdaandyou/issue07.html#l.
8
9
Published by Oxford Utliversity Press on behaU of tht Inttrnational Epidemiological Association l11ternational Journal of Epidemiology 2006;35: 1514-1521
Th<' Author 2006: all rigl11s reserved. Advance Access publication 30 August 2006 doi : I 0. 1 093/ ijeldyll 97
Risk factors for skin cancers: a nested
case-control study within the Nurses'
Health Study
Jiali Han, L
3
* Graham A Colditz LZ and David J HunterL
2
,
3
Accepted 2 August 2006
Background Constitutional factors and sun exposure a re associated with skin cancer risk.
However, these relations are complex and differ according tO skin cancer type.
Methods We examined the associations of constitutional risk factors and sun exposure
with the risks of three types of skin cancer simultaneously and evaluated the
interaction between constitutional susceptibility and sun exposure in a nested
case- comrol study within the Nurses' Health Study [200 melanoma, 275
squamous cel l carcinoma (SCC), and 283 basal cell carcinoma (BCC) cases, and
804 controls]. Information regarding skin cancer risk factors was obtained from
the retrospective supplementary questionnaire.
Results Constitutional susceptibility was an independent risk factor for all three types
of skin cancer. Sunlamp usage or tanning salon attendance was a risk fa ctor
for melanoma after adjusting for potential confounding variables (OR for ever vs
never usage, 2.06, 95% CI 1.30-3.26). Higher sun exposure whi le wearing a
bat hing suit was an independent risk [actor for all three types of skin cancer. We
observed a significant interaction between constitutional susceptibil ity and sun
exposure while wearing a bathing suit on melanoma risk (P, interaction, 0.03);
women with the highest susceptibility and highest exposure had an OR of
8.37 (95% Cl 3.07-22.84) . This interaction was weaker and non-significant for
sec and Bee.
Conclusions These data largely confirm past studies on risk [actors for skin cancer but provide
evidence of difference on the strength of these risk factors for melanoma
compared with SCC and BCC.
Keywords constitutional susceptibility, sun exposure, skin cancer
Skin cancer is the most common form of cancer in the US and
accounts for -1 million new cases per year, including -55 000
cases of cutaneous malignant melanoma (hereafter called
melanoma).l,
2
There are three major types o[ skin cancer.
Melanoma is the most fatal form. The most common type of
1
Channing Laborarory, Depamnenr of Medicine, Brigham and Women's
HospitaL and Harvard Medical School, 181 Longwood Avenue, Bosron, MA
021 15, USA.
2
De(>artment of Epidemiology, Harvard School of Public Health, Boston,
MA, USA.
3
The Program in Molecular and Genetic Epidemiology, Harvard School of
Public Health. 665 Humingtou Avenue, Boston. MA 02115, USA.
Corresponding author. J iali Han, Channing Laborarory, Brigl1am and
Women's Hospiral. and Harvard Medical School. l S I Longwood Avenue,
Boston, MA 02115. USA. E-mail: jiali.han@channing.harvard.edu
non-melanoma skin cancer is basal cell carcinoma (BCC),
followed by squamous cell carcinoma (SCC). The carcinogenic
effects of sunlight exposure have been demonstrated in the
aetiology of both melaJ1oma and non-melanoma skin can-
cers.
3
-
7
Although certain host factors and sun exposure are
thought to be associated with the development of skin cancer,
the relations are complex and may differ according to the type
of skin cancer. Risk factors for melanoma have been evaluated
previously.B-II Few studies have directly compared risk factors
for melanoma and non-melanoma skin cancers.
12
13
Previous
epidemiological studies have suggested that melanoma and
BCC arise from intermittent sun exposure and childhood sun
exposure,
14
whereas sec has been associated with cumulative
sun exposure.
3
15
In addition, it remains unclear how con-
stitutional susceptibility and sun exposure interact to determine
1514
skin cancer risk. The purpose of the study was to examine the
associations of constitutional risk factors and sun exposure and
their interactions with the risks of the three types of skin cancer
simultaneously in a nested case- control study within the
Nu rses' Health Study (NHS) cohort.
Methods
St udy population
The NHS was established in 1976, when 12I 700 female
registered nurses between the ages of 30 and 55 completed a
self-administered questionnai re on their medical histories and
baseline health-related exposures. Updated information has
been obtained by questionnaires every 2 years. Between I989
and 1990, blood samples were collected from 32 826 of the
cohort members. Because the aims of our research included
the evaluation of DNA-based markers of susceptibility, the
baseline of the study was the blood collection in 1989- 90.
Eligible cases in this study consisted of Caucasian women wi th
incident skin cancer from the subcohon who gave a blood
specimen, including sec and BCC cases with a diagnosis
anytime after blood collection up to l Jw1e 1998 and melanoma
cases up to I June 2000. All the cases had no previously
diagnosed skin cancer. All available pathologically confirmed
melanoma and SCC cases and 300 self-reported BCC cases
randomly selected from about 2600 available self-reported BCC
cases were included. The validity of self-report of BCC is high
in this medically sophisticated population (90%).
16
A common
control series (case : control = J: 1} was randomly selected
from participants who gave a blood sample and were free of
diagnosed skin cancer up to and including the quest ionnaire
cycle in which the case was diagnosed. One control was
matched to each case by year of birth (:!: I year). At the time we
selected cases and controls, 47 cases ( 19 melanoma, 11 SCC,
and 17 BCC) and 69 controls were deceased. As we wished to
obtain additional information by supplementary questionnaire,
we randomly selected a second matched living control when
the firs t control was deceased. The nested case-control study
consisted of 200 melanoma cases, 275 sec cases, 283 BCC
cases, and 804 matched controls. We mailed to 758 living cases
and 804 li ving controls a supplementary questionnaire on
lifetime sun exposu re and other skin cancer risk factors. In
tOtal, 695 cases responded, 15 cases refused w participate, and
48 cases did not respond after three mailings (participation
rate = 92%). Among controls, 713 responded, 9 refused, and
82 did not respond (participation rate = 89%) . The study
protocol was approved by the Committee on Use of Human
Subject.s of the Brigham and Women's Hospital, Boston, MA.
Exposure data
Information regarding skin cancer risk factOrs was obtained
from the retrospective supplementary questionnaire. The retro-
spective supplementary questionnaire was collected in 2002
and consisted of questions in three major areas: (i) pigmenta-
tion, constitutional, and susceptibility factors, such as skin
colour, hair colour, chi ldhood tendency to burn or tan, and the
number of palpably raised moles on arms; (ii) h istory of
residence (states and towns), sun exposure habits, and severe
sunburns at different ages (during childhood and adolescence
RISK FACTORS FOR SKIN CANCERS 1515
and then by decade o[ adult life up to date of questionnaire
return); and (iv) family history of skin cancer (father, mother,
and sibli ngs). In addition, the J J states of residence of
cohort members at baseline were grouped into three regions:
Northeast (Connecticut, Massachusetts, Maryland, New Jersey,
New York, and Pennsylvania), Northcentral (Michigan and
Ohio), and West and South (California, Texas, and Florida).
In order to estimate sunlight exposure for each subject.
an ultraviolet (UV) database for 50 US states was developed.
The database used reports from the Climatic At las of the US,
which reported mean daily solar radiation (in Langleys) at the
earth's surface for weather stations around the country.
17
The
records of average annual solar radiation for January and July
were extracted to represent winter and summer radiation,
respectively. The mean solar radiation for each individual's past
(at different age categories) and current residences was derived
from the lJV values measured at the nearest weather station.
Both summer (Us) and winter (Uw) radiation indices were
developed for the residence of each age category. A cumulative
lifetime sun exposure was developed by combining the
residence-linked UV value and hours spent outdoors at
diffe rence age categories obtained from the supplementary
questionnaire. For example, we defined a cumulative lifetime
intermittent (recreational) sun exposure variable for this
behaviour as follows: in each age category (Y represents the
number of years in each age category), we asked questions
about average frequency and duration of sun exposu re whil e
wearing a bathing suit per year in summer (Fs and Ds represent
average frequency and duration in summer, respectively)
and in winter (Fw and Dw represent average frequency and
duration in winter, respectively). For each age category, an
individual's sun exposure for such behaviour was equal to
(UsFsDsY + Uw-FwDwY). We summed up this variable for
each age category as a cumulative lifetime sun exposure while
wearing a bathing suit.
Statistical methods
Unconditional logistic regression was employed to calculate
odds ratios (ORs) and 95% confidence intervals (Cis) to assess
the risk of each type of skin cancer compared with the common
control series. Tests for trend were calculated when appropriate
to assess the effects of multiple levels of exposure. To
summarize multiple variables, we const ructed a multivariate
confounder score t.o create a constitutional susceptibility score
for each type of skin cancer.
18
Briefly, we applied the logistic
regression coefficients from a multivariate model including age,
natural skin colour, natural hair colour, child or adolescent
tendency to burn, and the number of palpably raised moles on
arms, to each individual's values for the latter four of these
variables and summed the values to compute a susceptibility
risk score in the logit scale. We used this score to define women
with low, intermediate, and high constitutional susceptibility
based on tertiles among controls. We performed a statistical
comparison of risk factors for the three types of skin cancer
using polychotomous logistic regression modcls.
19
This pro
gramme provides formal tests of the differences in magnitude of
the beta estimate of each risk factor for the outcome categories.
In the interaction analyses, the constitutional susceptibility
score and cumulative sun exposure while wearing a bathing
1516 INTERNATIONAL JOURNAL OF EPIDEMIOLOGY
suit were categorized imo reniles with cutpoints based on Table 1 Constitutional risk factors for skin cancer in this
the distribution of controls. We modelled these two variables case-control study nested withi n the Nurses' Health Study
as categorical variables and used the li kelihood ratio test to
compare nested models that included terms for all com-
binations of these two risk factors with the models with
indicator variables for the main effects only. All P-values were
two-sided.
We examined potential recall bias by comparing the
individual responses to three questions with the same wording
asked in both the !982 prospective questionnaire and the 2002
retrospect ive supplementary questionnaire (natural hair colour,
childhood or adolescent tendency to tan after repeated sun
exposure, and childhood or adolescence tendency to burn after
two or more hours of sunlight exposure)
20
The reliability
between two responses was measured by the Kappa statistics.
The shift of the absolute level of the responses to these three
quest ions after the diagnosis was measured by mean change.
The age-adjusted ORs were shown to illustrate the impact of
recall bias for these variables.
Results
Constit utional risk factors of skin cancer
At the beginning of the follow-up of this nested case-control
study, the nurses were between 43 and 68 years of age with the
mean age of 58.7 years. The mean age at iJ1cident diagnosis of
melanoma cases was 63.4 years and that of SCC cases and BCC
cases was 64.7 and 64.0 years, respectively. The constitutional
risk factors for skin cancer are shown ill Table l. Women with
fair skin colour or red hair colour were more likely to be
diagnosed '-vitb skin cancer, particularly melanoma, compared
with those with darker pigmentation. Cases of each type of skin
cancer had greater childhood or adolescence tendency to burn
and less tendency to tan. Cases of each type of skin cancer,
especially melanoma, were more likely to have more moles on
arms than controls.
We created a constitutional susceptibility score to summarize
these constit utional risk factors (Table 2). The association of
tendency to tan with skin cancer risk was abolished after the
above constitutional risk factors were mutually adjusted for,
and, therefore, it was not integrated illlo the constitutional
susceptibility score. Higher score predicted higher risk of skin
cancer according to the combined effect of lighter natural skin
colour, lighter natural hair colour, greater childhood or
adolescent tendency to burn. and more moles. Each component
contributed to the score to a different extent. For example,
women with six or more moles on arms had an age-adjusted
OR of 3.53 (95% CI 2.01-6.19) for melanoma risk. Red hair
colour was strongly associated with melanoma risk in our study
(age-adj usted OR, 4.74; 95% CI 2.47-9.09). The other two
components of the score, i.e. natural skin colour and childhood
or adolescent tendency to burn had risk estimates of -2 for
melanoma. The risk for the highest terrile of the susceptibility
score was -3.5-fold for melanoma, and 3-fold for sec and
BCC, compared with the lowest tertile. The ORs for constitu
tiona! susceptibility score slightly changed but remained
significant in mult ivariate models (Table 2); and this variable
significantly increased the goodness-of-fi t of the model for
three cancer types.
Common
Melanoma sec BCC controls
(11 = 200) (n = 275) (11 = 283) (11 = 804)
fl (%) 11 (%) It (%) 11 (%)
Natural ski n colour
Fair t25 (68.3) 1 56 (61.7) 159 (6 1.9) 348 (49.0)
Medium 56 (3Q.6) 92 (36.4) 94 (36.6) 320 (45.1)
Olive 2 ( l.l ) 5 (2.0) 4 (1.6) 42 (5.9)
Natural hair colour
Black or dark brown 58 (31.9) 102 (40.3) 82 (3 1.8) 302 (42.5)
Light Brown 73 (40. 1) 93 (36.8) 114 (44.2) 300 (42.2)
Blonde 29 ( 15.9) 44 (17.4) 49 (19.0) 85 (12.0)
Red 22 (1 2. t ) 14 (5.5) t 3 (5.1) 24 (3.4)
Skin reaction to 2 or more hours of sunlight in childhood
or adolescence (tendency to burn)
Practically none
Some- redness only
Bum
Painful burn
12 (6.6) 12 (4.8) 17 (6.7) 92 (13.0)
62 (33.9) 94 (37.3) 82 (32.2) 327 (46.3)
72 (39.3) 81 (32. 1) 91 (35.7) 201 (28.4)
37 (20.2) 65 (25.8) 65 (25.5) 87 (12.3)
Skin tan after repeated sun exposure in childhood or
adolescence (tendency t o tan)
.Practically none 42 (23.3) 47 08.7) 42 (16.4) 101 (14.2)
Light ran 53 (29.4) 83 (32.9) 79 (30.9) 189 (26.7)
Average tan 65 (36.1) 98 (38.9) 107 (41.8) 323 (45.6)
Deep tan 20 ( ll.J) 24 (9.5) 28 (10.9) 96 (13.5)
Palpably raised moles on arms
None
1- 2
3- 5
;;:6
73 (47.4) 132 (59.5) )24 (53.9) 388 (6 J. 3)
35 (22.7) 51 (23.0) 59 (25.7) 148 (23.4)
20 (13.0) 21 (9. 5) 28 (12.2) 59 (9.3)
26 (16.9) t8 (8.1) 19 (8.3) 38 (6.0)
The percentages may nor add up to I 00 due to rounding. The numbers do
not add up to total due to missing values.
Sunlight exposure and oth er risk factors of
skin cancer
Sunlight exposure and other risk factors of skin cancer arc
presented in Table 2. In age-adjusted models, family history
of skin cancer, the number of lifetime severe sunburns, and
cumulative sun exposure while wearing a bathing suit were
significantly associated with all three types of skin cancer.
A family history of skin cancer remained significant for the
three types of skin cancer in multivariate models. Fu rther-
more, melanoma risk was associated with both famil y history
of melanoma (OR. 1.81; 95% Cl 0.99-3.29) and that of
non -melanoma skin cancer (OR, 1.49; 95% CI 0.99- 2.25). For
the risks of sec and BCC, neither of them was associated
with family history of melanoma (OR for SCC, 1.17; 95% CI
0.67- 2.02; OR for BCC, 1.04; 95% CI 0.60-1.81) . Family
history of non-melanoma skin cancer was associated with the
risks of SCC (OR. 1.86; 95% CI 1.29-2.68) and BCC (OR. 2.65;
95% CI 1.86- 3.76).
For the number of lifetime severe sunburns that blistered,
compared 'Arith the age-adjusted ORs, the multivariate ORs
Table 2 Risk factors for skin cancer in tbis case-control study nested within the Nurses' Health Study
Common controls
Melanoma t1 = 200 SCC t1 = 275 BCC , = 283 n = 804
-----------------
Age-adjusted Multivariate Age-adjusted Multivariate Age-adj usted Multivariate P, for
11 (%) ORa ORb 11 ("/<>) ORa _____ 11 (_"lo) ORa ORb n i"lo_) heterogeneit/
Constitutional s usceptibility score (tertile)d
Low 32/270c 1.00 1.00 55/278 1.00 1.00 52/267 1.00 1.00
Intermediate 62/299 1.74 (1.10-2. 75) 1.65 (0.99-2.76) 76/265 1.47 ( 1. 00-2.16) 1.67 ( 1.05-2.65) 75/270 1.43 (0.96-2.11) 1.40 (0.90-2.19)
High 106/235 3.67 (2.37- 5.66) 3.26 (2.05-5.21) 144/26 1 2.89 (2.02-4.13) 3.15 (2.0 1-4.94) 156/267 2.99 (2.09-4.28) 2.72 ( 1.85-4.00)
Trend <0.001 <0.001 < 0.001 <0.001 <0.001 <0.001
LRT
1
<0.001 <0.001 <0.001
Family history of skin cancer 0.20
No 92 (54.4) 100 100 132 (56.4) 1.00 1.00 114 (47.7) 100 100 468 (68.5)
Yes 77 (45.6) I. 72 ( 1.22-2.44) 1.53 (1 .05- 2.23) 102 (43.6) 1.72 ( 1.27- 2.35) 1.50 ( 1.08-2.08) 125 (52.3) 2.43 ( 1.79-3.29) 2.05 (J .49- 2.81) 215 (31.5)
LRT 0 01 0.009 <0.001
SunJamp use or tanning salon attendance 0.1 5
No 140 (76.9) 1.00 1.00 212 (83.8) 1.00 1.00 215 (83.0) 1.00 1.00 625 (87.8)
Yes 42 (23.1 ) 1.98 (1.30-3.02) 2.06 (1.30-3.26) 41 (16.2) 1.47 (0.98-2.22) 1.44 (0.93-2.24) 44 (17.0) 1.45 (0.97-2.16) 1.32 (0.87-2.03) 87 (12.2)
LRT 0.01 0.24 0.35
Lifetime sever e sunburns which blistered 0.43
None 29 (17. 1) 1.00 1.00 46 (20.0) 1.00 1.00 49 (20.1 ) 1.00 1.00 231 (34.5)
1-4 48 (28.2) 1.63 (0.99- 2.68) 1.43 (0.85- 2.42) 80 (34.8) 1.85 (1.23- 2.79) 1.65 (1.08- 2.53) 82 (33.6) 1.70 (1.14-2.54) 1.38 (0.90-2.11) 226 (33.7)
5-9 33 (19.4) 2.57 (1.48-4.48) 175 (0.97-3.16) 46 (20.0) 2.47 ( 1. 53-3.98) 1.88 (1.14-3.11) 49 (20. 1) 2.33 (147- 3.71) 1. 55 (0.95-2.55) 98 (14.6)
;;.Jo 60 (35. 3) 3.90 (2.36-6.47) 2.24 (1.30-3.84) s8 (25.2) 2.72 (1.73-4.29) 167 (1.02-2.72) 64 (26.2) 2.62 (1.69-4.06) 1.37 (0.84-2.21> 11s (17.2)
Trend <0.001 0.003 <0.001 0.04 <0.001 0.15
LRT 004 0.01 0.37
Cumulative sun expos ure while weari ng a bathing s uit (tertile) 0. 1 5
Low 37 (20.6) 1.00 1.00 58 (23.7) 1.00 1.00 55 (2 1. 5) 1.00 1.00 227 (33.2)
Intermediate 47 (26.1) 1.25 (0.78-2.00) 1.20 (0.73-1.97) 74 (30.2) 1.28 (0.87-1.90) 1.28 (0.85-1.93) 92 (35.9) 1.66 (1.13-2.43) 1.71 ( 1.14-2.56) 228 (33.4)
High 96 (53.3) 2.58 (1.69-3.94) 2.37 (1.51-3.73) 113 (46.1) 1.97 (1.37-2.85) 2.1 5 ( 1.45-3.19) 109 (42.6) 1.95 (1.34-2.83) 2.05 (1.38-3.06) 228 (33.4)
Trend <0.001 <0.001 <0.001 <0.001 <0.001 <0.001
LRT <0.001 < 0.001 <0.001
Geographic region at baseline 0. 18
Northeast 119 (59.5) 1.00 100 141 (51.3) 100 1.00 141 (49.8) 1.00 1.00 448 (55.7)
Northcentral 35 ( 17.5) 0.74 (0.49-1.12) 0.75 (0.48-1.1 7) 48 (17.5) 0.83 (0. 57-1.20) 082 (0.56-1.21) 58 (20.5) 1.00 (0.70- 1.43) 1.04 (0.71 - 1.51) 182 (22.6)
West a nd South 46 (23.0) 1.08 (0.74-1.60) 1.08 (0.72- 1.64) 86 (3 1. 3) 1. 54 ( 1.1 2-2.14) 1.50 ( 1.07- 2.1 2) 84 (29.7) 1.56 (1.13- 2.17) 1.65 (1. 17-2.34) 174 (2 1.6)
LRT 0.32 O.Ql 0.01
The percentages may not add up ro I 00 due to rounding. The numbers do not add up to rota I due to missing values.
Unconditional logistic regression adjusted for age.
b Unconditional logistic regression adjusted for age and risk factors listed in Table 2.
c P-value was generated using polychowrnous logistic tests in multivariate models.
d Wt constructed a confounder scon to crc;ne a constitutional suscept ibility score for each type of ski11 cancer. We summed natural skin colour, natural hair colour, child or adokscl.'nt tendency to
burn, and the number of palpabl y raised moles on anns to compme a susceptibility risk score in the logit scale.
" The number of cases and controls
r Likelihood ratio test of the goodness lit of the model for each variable.
::::
(/)
" 'T1
>
(}
-1
0
:::0
(/)
'T1
0

(/)
!A
z
(}
>
z
(}
tn

(/)
>-'
\II
>-'
'I

Responsive Documents - CREW: FTC: Regarding Indoor Tanning Association: 5/15/14 - Batch Two, Part Two

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Responsive Documents - CREW: FTC: Regarding Indoor Tanning Association: 5/15/14 - Batch Two, Part Two

Uploaded by

Copyright:

Available Formats

Dietary Supplement Fact Sheet: Vitamin D

In addition, observational data show that the

You might also like