Support Care Cancer
DOI 10.1007/s00520-015-2718-5
ORIGINAL ARTICLE
The association between glucocorticoid therapy and BMI z-score
changes in children with acute lymphoblastic leukemia
Marie-Louise Hyre Arpe 1,2 & Sascha Rrvig 1,2 & Karin Kok 1 & Christian Mlgaard 1,2 &
Thomas Leth Frandsen 3
Received: 4 December 2014 / Accepted: 23 March 2015
# Springer-Verlag Berlin Heidelberg 2015
Abstract
Purpose Few studies have addressed the common issue of
weight gain in children with acute lymphoblastic leukemia
(ALL) during early phases of treatment, and even fewer have
used the appropriate measure for weight fluctuation in chil-
dren, BMI-for-age z-scores (BAZs). The purpose of this study
is thus to measure the extent of the weight gain in BAZ during
the 150 first days of treatment and to identify factors associ-
ated with the weight gain. Furthermore, we wish to raise the
question of whether changes in treatment protocols automati-
cally should be followed by an evaluation of the nutritional
guidelines.
Method In this retrospective study, the medical records of 51
children with ALL treated with the NOPHO ALL 2008 pro-
tocol at Copenhagen University Hospital were assessed. Pa-
tient characteristics were extracted, and height, weight, and
age during the first 150 days of treatment were converted to
BAZ.
Results During 150 days of treatment, the proportion of
overweight/obese patients increased significantly from 9.8 to
33.3 %. The mean change in BAZ (BAZ) was +1 standard
deviation (0.021.16 vs. 1.121.44; p<0.001) and BAZ in-
creased significantly during periods with glucocorticoid (GC)
treatment but not in periods without GC. BAZ was larger in
* Thomas Leth Frandsen
thomas.leth.frandsen@regionh.dk
1
Pediatric Nutrition Unit, Copenhagen University Hospital,
Rigshospitalet, Denmark
2
Department of Nutrition, Exercise and Sports, Faculty of Science,
University of Copenhagen, Copenhagen, Denmark
3
Department of Child and Youth, Copenhagen University Hospital,
Rigshospitalet, Denmark
boys compared to girls, and BAZ was higher in patients
who were under/normal weight at diagnosis, compared to pa-
tients who were overweight/obese (1.261.29 vs. 0.04
0.41; p=0.032).
Conclusion BAZ increased significantly in children with
ALL during the initial treatment with the NOPHO ALL
2008 protocol. This is likely associated with the GC adminis-
tration and influenced by gender and initial BAZ.
Keywords Children . Cancer . Leukemia . Glucocorticoids .
BMI
Introduction
Acute lymphoblastic leukemia (ALL) is the most prevalent
type of cancer in children younger than 15 years of age [1].
Since the 1970s, the 5-year survival rate in children 15 years
of age has increased from approximately 60 % to almost 90 %
[2, 3], and the improved survival rate has been associated with
the extended use of asparaginase (ASP) and glucocorticoids
(GCs) in treatment [46]. Despite the improved cure rates,
more than 50 % of children undergoing ALL treatment expe-
rience serious adverse events [7]. In addition, most ALL pa-
tients experience additional adverse events at some level.
Some of these include weight gain, hyperglycemia, hyperlip-
idemia, fluid retention, and an increase in abdominal fat de-
position, all of which have been associated with GC treatment
[4, 6, 810]. Overweight during childhood can have long-term
consequences and has been linked to an increased risk of
cardiovascular diseases, type 2 diabetes, metabolic syndrome,
morbidity, and mortality [11]. In children with ALL,
overweight/obesity has also been associated with a diminished
treatment effect as well as an increased risk of complications
such as pancreatic toxicity and increased hospitalization

during treatment [12, 13]. A recent study has even shown that
the cumulative time that an ALL patient is overweight/obese
between induction and maintenance phase can influence the
event-free survival and that the normalization of weight dur-
ing this period can improve outcomes [14].
Previously, nutritional guidelines have recommended an
energy-dense diet to patients with ALL, because their food
intake was limited as a result of chemotherapy-induced nau-
sea. Despite the advancements in anti-emetics and the en-
hanced use of GC, current nutritional recommendations at
many institutions are still based on this principle of an
energy-dense diet [15].
The different GC types, doses, and timing of GCs during
treatment [1618] make comparisons between countries and
protocols difficult. However, in the Nordic and Baltic coun-
tries (Denmark, Norway, Sweden, Finland, Iceland, Estonia,
Latvia, and Lithuania), the treatment is based on a shared
protocol created in 2008 by The Nordic Society of Pediatric
Hematology and Oncology (NOPHO ALL 2008) [7, 19]. The
aim of this study was thus to estimate the extent of weight gain
and to identify factors associated with weight gain during
ALL treatment, based on the NOPHO 2008 protocol. The
initial 150 days of treatment were the focus of this study be-
cause this period constitutes the most GC intensive period of
the NOPHO ALL 2008 protocol [7].
Method
Inclusion and exclusion of patients
This retrospective study is based on medical records of chil-
dren (age 118 years) with ALL, treated according to NOPHO
ALL 2008 at the Copenhagen University Hospital,
Rigshospitalet, cf. Fig 1. All medical records of children,
who were treated from July 2008 until February 15 2013, were
assessed. Sixty-seven patients met the inclusion criteria: 16
patients were excluded according to the exclusion criteria,
thus 51 patients fulfilled the entry criteria, cf. Fig. 2. Patients,
who were allocated to other cancer treatment regimes, and
patients who were categorized as high risk ALL after day 29
in treatment period were all excluded. The diagnosis of
Downs syndrome was also determined as an exclusion
criteria as patients with Downs syndrome have a genetic pre-
disposition of obesity and growth abnormalities [20].
Method of data collection
Factors such as age, the first date of treatment, ALL type, risk
group, gender, ethnicity, changes in treatment, as well as all
weight measurements were extracted from the patients med-
ical records. Due to the retrospective approach of this study,
the available data came from routine measurements of weight
Support Care Cancer
during treatment and were thus not performed systematically,
but an average 21 measurements per patient between day 1
and day 160 were obtained. Heights at day 1 and at day 150 or
the nearest measurement were noted. With the intention of
comparing the change in weight in children across different
ages, BMI-for-age z-score (BAZ) was chosen as primary end-
point. BAZ represents the number of standard deviation an
observation differs from the mean BMI in a given population,
at a certain age. In order to calculate BAZ, age and a height for
each weight were needed. The assumption that the height
increased linearly was adopted, and the height at each day
could thus be calculated:
End heightinitial height days of treatment
150 days
initial height
In order to investigate whether growth was influenced dur-
ing treatment, height-for-age z-scores (HAZ) was also calcu-
lated. The calculation of both HAZ and BAZ was based on
WHOs European reference material from 2007 [21].
The study population
Baseline characteristics of the study population are presented
in Table 1. The characteristics of the study population seem
comparable to other ALL populations in regard to age, gender,
ALL type, and risk group stratification [22, 23], cf. Table 1.
The mean age was higher than the median age, due to one
patients age (16.7 year) differing from the rest of the popula-
tion. This patient also raised the mean weight and height, but
the patient did not differ from the population in regard to BAZ
and HAZ. The average BAZ and HAZ score was close to zero,
and the prevalence of overweight children was similar to re-
sults from previous studies in Danish preschool children [24].
The presence of subjects with non-Danish ethnicity was also
similar to the rest of the Danish population [25], cf. Table 1.
Thus, the study population was a representable study group
for Danish ALL patients.
Method of data analysis
All data was processed in STATA or Excel. End values repre-
sent the value closest to day 150. Of the 51 patients, 13 pa-
tients had end values that differed 10 days or more from day
150, and all of these end values were measured before day
150. Analyses were based on these values or restricted inter-
vals for the end values (days 145155). The definition of
overweight/obesity was based on WHOs definition, in which
children <5 years are defined as overweight if BAZ2 and
obese if BAZ3, while children 5 years are categorized as
overweight if BAZ1 and as obese if BAZ2 [26]. The def-
inition of underweight in children was BAZ 2 [26].
Support Care Cancer

Non HR - inducon

PNS 60 mg/m2/d
PEG-ASP 1000UI/m2/d

*
IR & SR during consolidaon and reinducon GC and ASP does not dier
0 22 30 38

GC on DXM 10 mg/m2/d
PEG-ASP 1000UI/m2/d

HR inducon
* Stracaon IR/SR
I

39 43 57 71 85 92 99 106 113 121 127 134 141 150

DXM 10 mg/m2/d
PEG-ASP 1000UI/m2/d
HR GC o GC + ASP on
* (Consolidaon) (Reinducon)

0 22 30 38
HR treatment - paents are not included in this study
GC on

Fig. 1 ALL treatment at Copenhagen University Hospital,
Rigshospitalet. Children with ALL are initially stratified into HR and
non-HR groups depending on the leukocyte number and type of ALL.
After 29 days, a bone marrow sample is performed and patients are
restratified into SR, IR, or HR groups depending on response to the
induction treatment and cytogenetics [7, 21]. The figure is based on
data from a personal correspondence with Thomas Leth Frandsen about
unpublished data concerning NOPHO ALL-2008 3atreatment protocol.
DXM dexamethasone, GC glucocorticoids, HR high risk, IR intermediary
risk, PEG-ASP pegylated asparaginase, PNS prednisolone, SR standard
risk, * stratification

At binary issues, tests of proportion were performed. Chi2
tests were used when possible; Fishers test was performed
when n<40 or values <5, and in analysis of paired data,
McNemars test was executed. QQ plots were preformed in
order to test for normality of BAZ and HAZ.
Because BAZ was normally distributed, the parametric
paired t test was used to detect changes. When comparing
BAZ between subgroups, the parametric unpaired t tests
were performed. Two patients had HAZ that deviated from
a normal distribution. Thus, a non-parametric Wilcoxon rang
sum test including all subjects and as well as a paired t test
excluding the two subjects were performed. In all of the tests,
p values <0.05 were considered statistically significant.
In order to examine whether there was a significant differ-
ence between on and off GC periods, with and without ASP
treatment, data was divided into four periods based on the
timing of GC and ASP (Table 2). GC on periods were defined
as the coherent days during which the patients received GC, and
these periods include the days in which GC are phased out.
Results
BAZ increased significantly during the selected period of the
150 first days of treatment. When including the end values of
all of the patients (n=51, 0.021.16 vs. 1.121.44; p<0.001)
and also when constricting the end values to only include day
1505 days, a significant change in BAZ was observed (n=
31, 0.080.18 vs. 1.160.29; p<0.001) The change in BAZ
was heterogenic, and BAZ increased in 84.3 % of patients,

Fig. 2 Inclusion and exclusion 1 pt. was philadelphia 1 pt. died during 1 pt. with blasc plasmacytoid
flow diagram and criteria posive study periode dendric cell leukemia

67 pt. met the

51 paents
inclusion criterias

4 pt. (lack of 8 pt. (HR protocol 1 pt. with Downs

reported weights) aer day 29) syndrom

Inclusion:
Children with ALL treated according to NOPHO ALL 2008 (age 1-17.9)
ALL treated at the Copenhagen University Hospital, Rigshospitalet

Exclusion
High risk treatment after day 29
Discontinuation of treatment or transfer to another treatment regime
A present diagnose of Downs syndrome
Lack of weight- and height measurement
Ph positive ALL (different treatment regime)
Atypical ALL incl. dendritic cell leukemia (different treatment regime)
 Support Care Cancer

Table 1 Patient characteristics at time of diagnosis

Characteristics at diagnosis Values (prevalence/meanSD) Median and interval Standard population

n (Total) 51
n (Girls) 22 (43.1 %) A slightly higher ALL prevalence in boys [22]
n (Boys) 29 (56.9 %)
Age (years) 4.8 (1.88, 16.7) 25 [23]
BAZ 0.021.16
Adipositas
Overweight 5 (9.8 %) 8.95 % in Danish preschool children [24]
Obese 1 (1.96 %) 2.3 % in Danish preschool children [24]
HAZ 0.71 (1.19, 2.95)
Non-Danish ethnicity 10 (19.6 %) 20 % in Danish population (Statistics Denmark)
ALL phenotype
Pre B-ALL 46 (90.2 %) Pre B: 85 % in ALL population [22]
T-ALL 5 (9.8 %) T-ALL: 15 % in ALL population [22]
Risk group
Non-HR 42 (82.4 %) 85 % in ALL populationa [19]
HR 8 (15.7 %) 15 % in ALL populationa [19]
# 1 (1.9 %)
Reassessed risk (day 29)
IR 22 (43.1 %) 39 %a [19]
SR 29 (56.9 %) 61 %a [19]

ALL acute lymphoblastic leukemia, BAZ BMI-for-age z-score, HAZ height-for-age z-score, HR high risk, IR intermediary risk, Pre B-ALL pre B cell acute
lymphoblastic leukemia, SR standard risk, SD standard deviation, T-ALL T cell acute lymphoblastic leukemia, # unknown
a
Personal correspondence with Thomas Leth Frandsen about unpublished data concerning NOPHO ALL-2008 3atreatment protocol

while 15.7 % of patients experienced a decrease in BAZ. An
average of a 12 % increase in bodyweight was observed, and
the proportion of overweight/obese patients had increased sig-
nificantly (9.8 vs. 33.3 %; p<0.001). None of the subjects
characterized as overweight/obese at day 1 became normal/
underweight within the 150 days, and the prevalence of obe-
sity increased significantly from 1.96 to 15.7 %; p=0.016. The
decrease in HAZ from diagnosis to day 150 was significant
when all subjects were included in a Wilcoxons signed rank
sum test (WSRS: 0.71 (1.19; 2.95) vs. 0.04 (2.04; 2.46)
p<0.001) and also when two subjects with deviating HAZ
were excluded in the paired t test (paired t test: 0.71 (1.19;
2.95) vs. 0.06 (2.04; 2.46); p<0.001).
When analyzing the four different treatment intervals,
Table 2, a significant BAZ increase was apparent after 15
29 days of GC therapy (period 1) (0.141.05 vs. 0.30
1.35, p=0.001). During treatment with GC and ASP (period
3) BAZ also increased significantly (0.991.33 vs. 1.59
1.53; p=0.023), but during ASP, monotherapy BAZ did not
increase (period 2) (0.551.52 vs. 0.511.25, p=0.815).
In order to examine whether the BAZ change differed be-
tween GC on periods and GC + ASP on periods, periods 2 and
3 were compared (n=7, 0.561.62 vs. 0.780.82; p=0.711).
Due to the limited number of subjects with measurements
during both periods, an unpaired test was also performed, in
which non-paired values were included (n=22/16, 0.280.84

Table 2 Periods and argumentation

Periods Days n Explanation of the interval Initial BAZ End of period BAZ p value

1 129 (1529) 43 GC on period: GC for at least 14 days and 0.141.05 0.301.35 0.007
before PEG-ASP at day 30
2 42 (3742)86 (8691) 16 GC off period: PEG-ASP without GC (PEG-ASP at 0.551.52 0.511.25 0.815
days 43, 57, 71, and 85)
3 86 (8691)116 (116121) 12 GC + ASP on period: PEG-ASP at day 85, GC from 0.991.33 1.581.53 0.023
day 92. GC is decreased at days 113121
4 1end point 51 First and last day before the maintenance phase 0.021.16 1.121.44 <0.001

GC glucocorticoid, PEG-ASP pegylated asparaginase

Support Care Cancer

vs. 0.590.94; p=0.296). Both tests showed no significant
difference between GC on and GC + ASP on periods.
Subgroup analysis
prevalence of T cell ALL (T-ALL) in boys compared to girls
(Table 4). The BAZ for the two boys with T-ALL was 1.08
and 1.42, respectively. The mean BAZ for boys with pre B
cell ALL (B-ALL) was 1.81.

Subgroup analyses were performed in order to test whether the

Discussion
change in BAZ was affected by subjects initial BAZ, gender,
age, risk group, and reassessed risk group (Table 3). Patients
A significant increase in BAZ or BMI during ALL treatment
who were under or of normal weight at diagnosis (n=46)
has been observed in other studies [6, 2729]; however, due to
experienced a more extensive increase in BAZ compared to
different treatment periods and protocols, it was not possible
patients who were initially categorized as being overweight or
to compare the extent of the BAZ change between studies. In
obese (1.261.29 vs. 0.040.41; p=0.027). There was a
this study, BAZ increased in 84.3 % of the patients and de-
tendency toward a larger change in BAZ for boys (n=29)
creased in 15.7 % during the first 150 days of treatment. The
compared to girls (n=22) (1.401.08 vs. 0.781.47; p=
average increase was 1 BAZ, but the results were very diverse
0.088), and this result became significant when the end values
and seemed to be influenced by other factor related to the
were restricted to days 145155 (boys (n=18): 1.800.98 vs.
individual patients. The diversity, which has also been ob-
girls (n=13): 1.291.33; p=0.009). In the subgroup analysis
served in a study by Ethier et al. (2012) [12], could be ex-
of age, the median was used to divide the subjects into two age
plained by variations in GC receptors, mineral corticoid recep-
groups, and a tendency toward a greater increase in BAZ in
tors, variations in the 11-hydroxysteroid dehydrogenase
children 4.8 years of age compared to children >4.8 years
1/11-hydroxysteroid dehydrogenase 2 system, and cross reg-
was observed; however, this was non-significant (1421.57
ulation by other hormones such as aldosteron [8, 3032], all of
vs. 0.850.89; p=0.126). In order to test whether the ob-
which could affect the individual GC sensitivity and thus the
served difference between boys and girls was caused by a
weight gain [9]. Factors related to different aspects of treat-
difference in distribution of other factors, the distribution of
ment, such as nausea and ability to be physically active, could
age, initial BAZ, HAZ, ALL type, and initial and reassessed
also influence energy balance [3335]. In addition, some of
risk was analyzed. The only significant result was a higher
the end values were more than 10 days prior to day 150, and
these patients were thus less exposed to GC treatment.
Table 3 Subgroup analysis Furthermore, individual factors such as initial weight, age,
gender, and risk group have been tested for in this study. The
Parameter n BAZ meanSD p value initial weight seemed to influence BAZ, as patients who
Initial BAZ
were initially categorized as being under/normal weight had
Normal/underweight 46 1.261.29 0.032*
a significant higher increase in BAZ than patients who were
Overweight/obese 5 0.040.41 initially categorized as being overweight/obese. The high in-
Initial age crease in BAZ in under/normal weight patients could reflect a
4.8 years 25 1.421.57 0.126 healthy weight catch-up in underweight patients, who experi-
>4.8 years 26 0.850.89 enced a weight loss prior to the initiation of treatment. How-
Sex
ever, only two of the 46 under/normal weight patients were
Boys 29 1.401.08 0.088a
underweight at day 1, and the mean change in BAZ for these
Girls 22 0.781.47
Initial risk group two patients did not exceed the average of the under/normal
HR 8 1.231.02 0.827 weight group. The observed increase in the prevalence of chil-
Non-HR 42 1.121.35 dren categorized as overweight/obese during treatment has
Reassessed risk group also been shown in other studies [2729]. However, only
IR 22 0.871.56 0.206 few of these studies tested for significance and the initial prev-
SR 29 1.331.03 alence of obesity differed extensively between the studies. The
Phenotype 46 1.151.33 0.946 variation in prevalence may reflect cross country variation,
Pre B-ALL 5 1.200.89 differences in reference materials, or cutoff values to define
T-ALL overweight/obesity.
BAZ BMI-for-age z-score, HR high risk, IR intermediary risk, Pre B-ALL In our study, a difference in BAZ was also observed be-
pre B cell acute lymphoblastic leukemia, SR standard risk, SD standard tween genders. This could be a reflection of other unequally
deviation, T-ALL T cell acute lymphoblastic leukemia distributed factors between the genders; however, neither age,
a
Non-significant trend initial BAZ, risk group, reassessed risk stratification, nor
*Significant result change in HAZ were significantly different between boys

Table 4 Test of difference in age,
initial BAZ, HAZ, ALL type, Parameter Boys (n=29) meanSD
initial risk group, and reassessed
risk group between sex Age
Initial BAZ
HAZ 0.640.49
Parameter
Phenotype
T-ALL
Pre B-ALL
Initial risk
HR
Non-HR
Reassessed risk
IR
SR
BAZ BMI-for-age z score, HAZ height-for-age z-score, HR high risk, IR intermediary risk, Pre B-ALL pre B cell
acute lymphoblastic leukemia, SR standard risk, SD standard deviation, T-ALL T cell acute lymphoblastic
leukemia
*Significant result
and girls. The only significant difference was the higher prev-
alence of T-ALL in boys, but since BAZ for the two boys
with T-ALL was lower than the average BAZ for boys with
B-ALL, the more extensive change BAZ in boys cannot be
explained by the higher prevalence of T-ALL among boys,
and the findings suggests that the change in BAZ is due to
differences between genders. In contrast, other studies have
indicated that girls are in greater risk of becoming overweight
or obese during ALL treatment, even though boys received a
larger cumulative corticosteroid dose throughout a prolonged
protocol in these studies [28, 29].
Another tendency observed in our study was a more exten-
sive increase in BAZ for children younger than 4.8 years. This
tendency could simply reflect that an absolute weight gain will
result in a larger relative weight gain in younger children due
to their lower weight, thus affecting BAZ to a higher degree,
but may also be affected by the developmental stage or the
onset of the BMI rebound. Even though the frequency of tube
feeding was not recorded, tube feeding could have influenced
the age-specific results, as this nutritional approach in general
is more frequently used in younger children. It should be not-
ed that only a tendency was observed and that a selection of
other age intervals could have led to other results.
In addition to the individual factor associated with weight
gain, it was also investigated whether the weight gain was
associated with the GC treatment. In our study, BAZ only
increased significantly in the GC on periods, and there were
no significant difference between GC on periods compared to
GC + ASP on periods. These results support the likeliness of
an association between GC treatment and weight gain. The
change in BAZ was observed already during induction treat-
ment days 129 (1529), and similar results have been ob-
served in other studies after 4 weeks of GC treatment [6, 27,
Support Care Cancer
Girls (n=22) meanSD p value
6.213.22 6.323.22 0.919
0.031.28 0.001.01 0.927
0.590.56 0.819
n (Boys) n (Girls) p value
5 0 0.040*
24 22
5 23 1.000
3 19
13 9 0.780
16 13
28], but in these studies, no significant correlation between
GC and BAZ was found. Chow et al. (2007) did however
find a significantly higher increase in BAZ in patients receiv-
ing a high GC dose compared to a low GC dose [29].
The mechanism behind GCs influence on weight gain in
still elusive. Studies have shown an increase in energy intake
in children with ALL during treatment [36, 37], and cell stud-
ies have shown an interaction between GC and appetite regu-
lating mechanisms [3840]. It is thus plausible that the ob-
served weight gain is related to an increased energy intake,
due to an enhanced appetite caused by a GC interaction. Re-
sults from in vitro studies have shown that GC increased li-
polysis in cells from peripheral fat tissue and increased lipo-
genesis in cells from abdominal fat tissue [41, 42]. However, it
was not possible to find studies that examined these effects,
in vivo, nor did the examined studies address whether the
increased fat deposition exceeded the degradation of fat tissue.
Other in vitro studies have shown that GC influences mecha-
nism associated with muscle degradation, including proteolyt-
ic effects and inhibitory effects on protein synthesis and on
growth and transcription factors [4345]. The development of
muscle atrophy was however not examined in any of the
assessed in vivo studies performed on ALL children. In order
to evaluate the significance of the weight gain, it would be
relevant to examine the change in body composition. Due to
the unlikeliness of muscle development during stress metab-
olism, the observed BAZ increase would reflect increased fat
mass and/or increased water retention. The impact of the
weight gain may be more extensive than observed, if it is
accompanied by a reduction in muscle mass, which would
also lower the resting energy expenditure. A decrease in en-
ergy expenditure is common, as the level of physical activity
often decreases in patients during ALL treatment, thus
Support Care Cancer
contributing to weight gain [36]. In addition, GC has also been
associated with inhibited growth, and the decrease in HAZ
observed in our study would also make weight gain a more
significant contributor to BAZ.
Even though the composition of and the mechanism behind
the weight gain cannot be completely established, the results
indicate that GC treatment can induce weight gain. Since
weight gain and an increased body weight during treatment
has been associated with diminished treatment effect, as well
as long-term consequences, and since the nutritional status is
one of the few modifiable prognostic risk factors [14], we
suggest that future nutritional guidelines for patients with
ALL not only address weight loss but also consider weight
gain and encourage physical activity.
Conclusion
Our data suggest that treatment with GC results in a significant
increase in BAZ and the extent to which BAZ increased was
influenced by the gender and initial BAZ. The results of this
study raise the question of whether protocol writers should
consider to evaluate nutritional guidelines, when making sub-
stantial changes to the treatment protocol. However, due to the
retrospective design and the lack of evaluation of body com-
position in this study, our results need to be confirmed in a
prospective study in which an assessment of changes in body
composition and BAZ is included for the entire treatment
period.
Conflict of interest Nothing to declare.
References
1. National Cancer Institute, US National Institutes of Health (2014)
General Information About Childhood Acute Lymphoblastic
Leukemia (ALL). http://www.cancer.gov/cancertopics/pdq/
treatment/childALL/HealthProfessional. Accessed 5 March 2014
2. Smith MA, Seibel NL, Altekruse SF et al (2010) Outcomes for
children and adolescents with cancer: challenges for the twenty-
first century. J Clin Oncol 28:26252634
3. Pui CH, Gajjar AJ, Kane JR et al (2011) Challenging issues in
pediatric oncology. Nat Rev Clin Oncol 8:540549
4. Kawedia JD, Kaste SC, Pei D et al (2011) Pharmacokinetic, phar-
macodynamic, and pharmacogenetic determinants of osteonecrosis
in children with acute lymphoblastic leukemia. Blood 117:2340
2347
5. Pession A, Valsecchi MG, Masera G et al (2005) Long-term results
of a randomized trial on extended use of high dose L-asparaginase
for standard risk childhood acute lymphoblastic leukemia. J Clin
Oncol 23:71617167
6. Chow EJ, Pihoker C, Friedman DL et al (2013) Glucocorticoids and
insulin resistance in children with acute lymphoblastic leukemia.
Pediatr Blood Cancer 60:621626
7. Frandsen TL, Heyman M, Abrahamsson J et al (2013) Complying
with the European Clinical Trials directive while surviving the ad-
ministrative pressurean alternative approach to toxicity registra-
tion in a cancer trial. Eur J Cancer 50:251259
8. Gross KL, Cidlowski JA (2008) Tissue-specific glucocorticoid ac-
tion: a family affair. Trends Endocrinol Metab 9:331339
9. Iughetti L, Bruzzi P, Predieri B, Paolucci P (2012) Obesity in pa-
tients with acute lymphoblastic leukemia in childhood. Ital J Pediatr
38:111
10. Bernbeck B, Christaras A, Krauth K et al (2004) Bone marrow
oedema and aseptic osteonecrosis in children and adolescents with
acute lymphoblastic leukaemia or non-Hodgkin-lymphoma treated
with hyperbaric-oxygen-therapy (HBO): an approach to cure?
BME/AON and hyperbaric oxygen therapy as a treatment modality.
Klin Padiatr 216:370378
11. Daniels SR, Arnett DK, Eckel RH et al (2010) Overweight in chil-
dren and adolescents pathophysiology, consequences, prevention,
and treatment. Circulation 111:19992012
12. Ethier MC, Alexander S, Abla O et al (2012) Association between
obesity at diagnosis and weight change during induction and sur-
vival in pediatric acute lymphoblastic leukemia. Leuk Lymphoma
53:16771681
13. Butturini AM, Dorey FJ, Lange BJ et al (2007) Obesity and out-
come in pediatric acute lymphoblastic leukemia. J Clin Oncol 25:
20632069
14. Orgel E, Sposto R, Malvar J et al (2014) Impact on survival and
toxicity by duration of weight extremes during treatment for pedi-
atric acute lymphoblastic leukemia: a report from the childrens
oncology group. J Clin Oncol 32:13311337
15. Royal College of Nursing. Nutrition in children and young people
with cancerRCN guidance (2010) http://www.rcn.org.uk/__data/
assets/pdf_file/0010/338689/003805.pdf. Accessed 20 Aug 2014
16. Mricke A, Zimmermann M, Reiter A et al (2010) Long-term re-
sults of five consecutive trials in childhood acute lymphoblastic
leukemia performed by the ALL-BFM study group from 1981 to
2000. Leukemia 24:265284
17. Vora A, Mitchell C, Rowntree C, et al. (2011) Guidelines for treat-
ment of children and young persons with acute lymphoblastic leu-
kaemia and lymphoblastic lymphoma. https://www.ctsu.ox.ac.uk/
research/mega-trials/leukaemia-trials/ukall-2003/interim-
guidelines. Accessed 17 July 2014
18. Silverman LB, Stevenson KE, O'Brien JE et al (2010) Long-term
results of Dana-Farber Cancer Institute ALL Consortium protocols
for children with newly diagnosed acute lymphoblastic leukemia
(1985-2000). Leukemia 24:320334
19. Toft N, Birgens H, Abrahamsson J et al (2013) Risk group assign-
ment differs for children and adults 1-45 yr with acute lymphoblas-
tic leukemia treated by the NOPHO ALL-2008 protocol. Eur J
Haematol 90:404412
20. Dalton VK, Rue M, Silverman LB et al (2003) Height and weight in
children treated for acute lymphoblastic leukemia: relationship to
CNS treatment. J Clin Oncol 21:29532960
21. World Health Organization. Child growth standards. http://www.
who.int/childgrowth/software/en/. Accessed 1 Mar 2013
22. Brunning RD, Borowitz M, Matutes E et al (2001) Precursor B-cell
and T-cell neoplasms. In: Jaffe ES, Harris NL, Stein H, Vardiman
JW (eds) World health organization classification of tumorstu-
mours of haematopoietic and lymphoid tissues. IARCPress Inc,
Lyon, pp 109117
23. Hjalgrim LL, Rostgaard K, Schmiegelov K et al (2003) Age- and
sex-specific incidence of childhood leukemia by immunophenotype
in the Nordic countries. J Natl Cancer Inst 95:15391544
24. Larsen LM, Hertel NT, Mlgaard C et al (2012) Prevalence of
overweight and obesity in Danish preschool children over a 10-
year period: a study of two birth cohorts in general practice. Acta
Paediatr 101:201207

25. Statistics Denmark. Immigrants and their descendents. http://www.
dst.dk/en/Statistik/emner/indvandrere-og-efterkommere/
indvandrere-og-efterkommere.aspx. Accessed 9 May 2013
26. WHO. A healthy lifestylebody mass indexBMI, World Health
OrganisationRegional office for Europe. http://www.euro.who.
int/en/what-we-do/health-topics/disease-prevention/nutrition/a-
healthy-lifestyle/body-mass-index-bmi. Accessed 22 Apr 2013
27. Esbenshade AJ, Simmons JH, Koyama T et al (2011) Body mass
index and blood pressure changes over the course of treatment of
pediatric acute lymphoblastic leukemia. Pediatr Blood Cancer 56:
372378
28. Withycombe JS, PostWhite JE, Meza JL et al (2009) Weight pat-
terns in children with higher risk ALL: a report from the Children's
Oncology Group (COG) for CCG 1961. Pediatr Blood Cancer 53:
12491254
29. Chow EJ, Pihoker C, Hunt K et al (2007) Obesity and hypertension
among children after treatment for acute lymphoblastic leukemia.
Cancer 110:23132320
30. Rang HP, Dale MM, Ritter JM et al (2012) Rang and Dales's phar-
macology. Elsevier, Spain
31. Tomlinson JW, Walker EA, Bujalska IJ et al (2004) 11-beta-
hydroxysteroid dehydrogenase type 1: a tissue-specific regulator
of glucocorticoid response. Endocr Rev 25:831866
32. Bagamasbad P, Denver RJ (2011) Mechanisms and significance of
nuclear receptor auto- and cross-regulation. Gen Comp Endocrinol
170:317
33. Pui CH, Robison LL, Look AT (2008) Acute lymphoblastic leukae-
mia. Lancet 371:10301043
34. Tyc VL, Mulhern RK, Fairclough D et al (1993) Chemotherapy
induced nausea and emesis in pediatric cancer patients: external
Support Care Cancer
validity of child and parent emesis ratings. J Dev Behav Pediatr
14:236241
35. Schacke H, Docke WD, Asadullah K (2002) Mechanisms involved
in the side effects of glucocorticoids. Pharmacol Ther 96:2343
36. Jansen H, Postma A, Stolk RP, Kamps WA (2009) Acute lympho-
blastic leukemia and obesity: increased energy intake or decreased
physical activity? Support Care Cancer 17:103106
37. Reilly JJ, Brougham M, Montgomery C et al (2001) Effect of glu-
cocorticoid therapy on energy intake in children treated for acute
lymphoblastic leukemia. J Clin Endocrinol Metab 86:37423745
38. Lim CT, Kola B, Korbonits M (2010) AMPK as a mediator of
hormonal signalling. J Mol Endocrinol 44:8797
39. Pritchard LE, Turnbull AV, White A (2002) Pro-opiomelanocortin
processing in the hypothalamus: impact on melanocortin signalling
and obesity. J Endocrinol 172:411421
40. Spencer SJ, Tilbrook A (2011) The glucocorticoid contribution to
obesity. Stress 14:233246
41. Peckett AJ, Wright DC, Ridell MC (2011) The effects of glucocor-
ticoids on adipose tissue lipid metabolism. Metabolism 60:1500
1510
42. Arnaldi G, Scandali VM, Trementino L et al (2010) Pathophysiology
of dyslipidemia in Cushings syndrome. Neuroendocrinology 92:86
90
43. Pereira RMR, Freire de Carvalho J (2011) Glucocorticoid-induced
myopathy. Jt Bone Spine 78:4144
44. Schakman O, Gilson H, Kalista S, Thissen J (2009) Mechanisms of
muscle atrophy induced by glucocorticoids. Horm Res Paediatr 72:
3641
45. Mauras N (2009) Can growth hormone counteract the catabolic
effects of steroids? Horm Res Paediatr 72:4854