1

Correlation of Hot Spot to Breast Separation in Patients Treated with Post-lumpectomy

Tangent 3D-CRT Using Field-in-Field Technique and Mixed Photon Energies
Nancy R. Gustafson, DVM, MS, DACVR (Radiation Oncology); Teri Burrier, BA,
RT(R)(T); Brittany Butler, BS, RT(R)(CT); Ashley Hunzeker, MS, CMD; Nishele Lenards,
PhD, CMD, RT(R)(T), FAAMD
Department of Health Sciences, University of Wisconsin – LaCrosse, LaCrosse, Wisconsin

Abstract:
Radiotherapy to an intact breast was previously determined to have a positive correlation
between breast separation measurement and hot spot dose. The hot spot dose increased as the
breast separation measurement increased. This retrospective study aimed to determine if this
correlation persisted despite current techniques including field-in-field blocking and mixed
photon energies. Radiation treatment plans on unilateral intact breasts from 90 female, early
stage breast cancer patients treated with lumpectomy were analyzed. Plans were created using
3D-CRT non-divergent opposing tangent beams, field-in-field technique, and 6 MV with or
without higher energy photons. Data collected included breast separation measurement, hot spot
point dose and location, number of beams, photon energy, CTV coverage and breast volume
coverage. Correlations between breast separation measurement and each of these values were
determined using Pearson Correlation regression analysis, one-way ANOVA, Pearson Chi-
square, Likelihood Ratio, and Linear-by-Linear Association tests. The positive correlation
between breast separation measurement and hot spot dose persisted despite incorporating FIF
and mixed photon energies. Correlations were also found between breast separation and the
number of beams as well as breast separation and photon energy. As the breast separation
increased, the hot spot dose, number of beams and photon energy all increased. There were no
correlations found between breast separation and CTV or breast volume coverage. It could be
argued that FIF and/or mixed photon energies give the medical dosimetrist the ability to create
treatment plans with homogenous dose distributions, thereby negating the previously established
positive correlation between breast separation measurement and hot spot dose. However, this
study showed, despite these current techniques, the correlation persists. This study determined
the dosimetrist should expect hot spots above prescription dose of 106%, 107% and 108% for
 2

small, medium and large breast separation sizes respectively. Additionally, adding a high energy
photon beam may be indicated with medium and large breast separations.

Keywords: Breast separation measurement; Hot spot point dose; Field-in-field tangents;
Dosimetry

Introduction:
Breast cancer treated with lumpectomy followed by radiation therapy is a successful
therapy for early stage breast cancer. This multi-modality treatment provides a more
cosmetically pleasing outcome compared to mastectomy, with comparable survival outcomes.1
Patients with intact breasts, post-lumpectomy, are most commonly treated with 3-dimensional
conformal radiation therapy (3D-CRT) using medial and lateral non-divergent opposing tangent
beams to the whole breast followed by a boost to the tumor bed. The goal of breast tangent
radiotherapy is to administer the desired dose to the entire breast volume, focusing the highest
dose to the lumpectomy site, while minimizing the hot spot (point of maximum dose).
The shape of the intact breast and the goal of administering dose to the chest wall
typically cause the appearance of high dose superficially. These high superficial doses tend to be
greater in patients with larger breast sizes. High doses can cause unacceptable acute and late
toxicities. Painful erythema and desquamation may occur acutely, while late effects may include
pigmentation, fibrosis, dimpling, atrophy, and asymmetry.2
Protocols exist to guide beam design, fractionation, and maximum dose. There are a
variety of techniques available to reduce hot spots; however, some have unacceptable
consequences. For example, hard wedges, with the heel positioned anterior, will attenuate the
superficial dose reducing some hot spots. However, hard wedges have been shown to have a
higher risk of causing cancer in the contralateral breast due to scatter.3 Other techniques such as
intensity-modulated radiation therapy (IMRT) or volumetric-modulated arc therapy (VMAT), are
less likely used due to undesirable high integral dose.4 Low dose irradiation is concerning for
radiation-induced cancer. Therefore, field-in-field (FiF) is a commonly used technique to
decrease the hot spots. Additionally, higher energy beams may be used in conjunction with or
instead of lower energy beams in patients with larger breast separations to aid in reducing hot
spots and improving chest wall coverage.
 3

Breast separation has not yet been correlated to hot spot when using the FiF technique
and mixed photon energies. Inhomogeneity of dose within the breast has been examined in prior
literature. Buchholz et al5 examined off-axis dose inhomogeneity and found a significant volume
of breast received greater than 110% of prescribed dose, which was more common in larger
breast sizes and lower quadrant tumor locations. Delaney et al6 evaluated inhomogeneity in
phantoms with small and large breasts. These studies by Buchholz et al5 and Delaney et al6
focused on inhomogeneity and tumor location in the breast tissue. Other studies assessed
collimator angle, irregular surface compensation, multiple-segment tangent fields to decrease hot
spots, planning target volume (PTV) margin reduction to reduce dose to organs at risk (OAR),
lack of dose specification, and when hypo-fractionation is employed.7-12 However, only two
studies, one by Das et al13 and one by Neal et al14, compared breast separation with resulting hot
spot. The study by Das et al13 found a variation of 2-27% in maximum dose depending on chest
wall separation and the study by Neal et al14 found 0.2 – 23.8% of the breast received an
absorbed dose outside the desired 95-105% of the prescribed dose. In the study by Neal14,
patients were treated with plans using compensators, wedges, and single energy (6 megavoltage
(MV)) photons. Prescribed doses ranged from 50 to 55 Gy in the Das et al13 study. Techniques
have changed since these studies were performed 21 and 23 years ago respectively.
This retrospective study aimed to identify a correlation between physical breast
separation and hot spot dose in the treatment of intact breast cancer using tangent FiF, single or
mixed energy photons, and a single standard prescription. Photon energy, number of beams,
clinical target volume (CTV) coverage, breast coverage and normalization were analyzed for a
correlation to breast separation. Medical dosimetrists spend much of their planning time trying to
design an optimal plan. Findings from this study may be useful to guide dosimetrists when
determining an appropriate compromise between tumor volume coverage and hot spot
occurrence and reduce treatment planning time.
Materials and Methods:
Patient
Treatment plans from 90 female patients treated between 2015 and 2018 were reviewed.
Three clinic sites participated in this study. Data from 30, 29 and 31 patients were obtained from
clinic sites 1, 2 and 3 respectively. Patients with unilateral early stage breast cancer treated with
lumpectomy followed by radiation therapy were included in the cohort. No patients were treated
 4

previously with radiation therapy. Plans with other beam arrangements such as an en face photon
beam and boost plans were excluded. The radiation treatment plans at all 3 sites were created
from non-diagnostic, non-contrast simulation CT scans. Each patient was positioned in the
supine position with their thorax on a Med-Tec slanted breast board, arms raised above their
head in arm cups or on a wing board, and a cushion under their knees. Radiopaque wires were
placed on medial, lateral, superior, and inferior borders of the breast and on the scar for breast
delineation purposes. Radiopaque markers were placed anteriorly and bilaterally at initial
leveling. The isocenter was set at time of simulation or during treatment planning.
Contouring
Contours drawn included CTV, heart, spinal cord, lungs, breast, scar, and scar wire. The
CTV, drawn by the physician, was the lumpectomy site, and was the primary target volume. A
PTV was not analyzed because it was not consistently contoured or not contoured at all. A GTV
was not contoured since the gross tumor was resected. Other volumes were contoured by the
medical dosimetrist or physician. The breast volume was the secondary target volume, and was
contoured by the medical dosimetrist according to the guidelines in the Radiation Therapy
Oncology Group (RTOG) Breast Cancer Atlas.15 These guidelines define the medial border as
the sternal rib junction, the lateral border as the mid axillary line excluding the latissimus dorsi
muscle, the anterior border as the skin, the posterior border as the superficial margins of the
pectoralis muscle and chest wall, the cranial margin as the second rib insertion, and the caudal
margin as the visual loss of breast tissue noted on the CT images. The breast volume was
cropped 5 mm from the skin.
Treatment Planning
The prescription was 180 cGy per fraction in 28 fractions for a total dose of 5040 cGy.
Treatments were administered once daily, Monday through Friday, using Varian Trilogy, Clinac
21EX or TrueBeam linear accelerators. Treatment plans of the intact breast were designed using
opposing medial and lateral 3D-CRT tangent beams with Pinnacle or Eclipse treatment planning
systems (TPS). The tangential edge of each opposing beam was positioned medially along the
chest wall. The number of beams varied from 2-4, with energies including 6 MV, a combination
of 6 MV with 10 MV, or a combination of 6 MV with 18 MV. The plans were normalized as
needed to provide adequate breast and CTV coverage. Coverage was defined as the percent of
organ (breast or CTV) receiving prescription dose. Beam and collimator angles were selected
 5

with the goals of providing breast and CTV coverage, while minimizing inclusion of the heart
and lungs. Field-in-field blocking was used to decrease the hot spot dose. Each field had a
minimum of 3 monitor units. All plans were reviewed and approved by the attending physician.
Plan Comparison
Data recorded from the plans included measurement of separation, photon energy,
number of beams, beam arrangements, normalization, breast volume coverage, CTV coverage,
hot spot measurement in absolute dose and location of hot spot. The breast separation was
measured on the central axis slice, from the antero-medial skin margin to the postero-lateral skin
margin. Figure 1 shows the breast separation measurement along the tangent beam edge. Breast
separation sizes were divided into 3 categories: ≤ 18 cm (small), 18.1-22 cm (medium), and ≥ 22
cm (large). Measurements of hot spot, the maximum point dose within the treatment field, were
also recorded for each plan and categorized by location. Locations of the hot spot were defined
as upper inner quadrant (UIQ), lower inner quadrant (LIQ), upper outer quadrant (UOQ), and
lower outer quadrant (LOQ).
Statistical analyses included Pearson correlation regression analysis to determine
relationships between breast separation and each of the other variables. A P < 0.05 indicated
statistical significance. A one-way ANOVA test was performed to compare the mean breast
separation to photon energy as well as separation to hot spot location. Cross tabulation analyses
including Pearson Chi-Square, Likelihood Ratio and Linear-by-Linear Association were
performed to determine correlations between separation and hot spot location, and photon
energy. Cross tabulation analyses were also performed within each clinic site to determine
correlations between separation and photon energy, and breast coverage. Basic statistics such as
mean, median, mode, range and standard deviation were also determined for each nominal
variable.
Results:
Statistically significant correlations were found between breast separation measurement
and hot spot dose, photon energy, and number of beams. Correlations were not found between
separation and breast coverage or CTV coverage.
Separation and hot spot
Separation values ranged from 14.3-31.1 cm. Breast separation was divided into 3 groups
as was done in the study by Das et al13. The intention was to record data from 30 patients in each
 6

group. The small separation group was less abundant. Therefore, the resulting numbers were 19
patients (21%) in the small separation group, 37 patients (41%) in the medium group and 34
patients (38%) in the large group. Hot spot doses ranged from 5204.6 to 5669 cGy, which is
103.3 to 112.5% above prescription dose (5040 cGy) respectively. The mean was 5404.9 cGy
(107.2%), median was 5393 cGy (107%) and mode was 5384 cGy (106.8%) for all groups
combined (Table 1). There was a statistically significant positive correlation between separation
and hot spot dose with an r-value of 0.365 and a P = 0.0. This indicated that as the separation
increased; the hot spot dose increased (Figure 2).
Separation and photon energy
There was a correlation between separation and photon energy. The r-value was 0.26
with P = 0.01, which indicated the larger breast separation required a higher photon energy.
Table 2 shows the photon energy combinations with associated number of patients and mean
breast separation for each group. A one-way ANOVA compared the mean separation values to
the photon energy groups, which revealed a significant correlation. The F-value was 7.2 with P =
0.0. A cross tabulation analysis also confirmed this correlation. The Pearson Chi-Square value
was 22.81 with P = 0.01, Likelihood Ratio value of 26.22 and P = 0.003, and Linear-by-Linear
Association value of 4.5 and P = 0.03. Additionally, a comparison of separation with photon
energy was performed within each clinic site. The correlation was significant within 2 of 3 clinic
sites. Table 3 shows the cross tabulation tests and values for each site. Site 2 did not have a
statistically significant correlation between separation and photon energy.
Separation and number of beams
The number of beams ranged from 2-4. There was a correlation between separation and
number of beams with an r-value of 0.53 and P = 0.0. This indicated that as separation
measurements increased; the number of beams increased. Figure 3 shows the correlation between
separation (Y axis) and number of beams (X axis).
Separation and breast coverage; separation and CTV coverage
Pearson correlation tests revealed there were no correlations found between separation
and breast coverage (r-value = 0.095 with P = 0.37) or between separation and CTV coverage (r-
value = -0.07 with P = 0.49). However, Pearson correlation tests between separation and CTV
coverage within two of the three sites did reveal correlations. Site 1 had a negative correlation
suggesting as separation increased; CTV coverage decreased. Site 2 had a positive correlation in
 7

which CTV coverage increased with an increase in separation and the negative correlation in site
3 was not statistically significant. The correlation between separation and breast coverage within
each site was compared. Site 3 had a statistically significant negative correlation, while the
correlations in sites 1 and 2 were not statistically significant. Table 4 shows the r-value and p-
value for each site correlating separation with breast coverage and CTV coverage. The range of
breast volume coverage in this study was 44.4-96.6% with a mean of 79.28%, median of 82.6%
and mode of 86.4%. The CTV coverage ranged from 22.2-100% with a mean of 95.2%, median
of 99.9% and mode of 100%.
Separation and hot spot location
A one-way ANOVA test revealed no correlation between separation and hot spot location
(F = 1.12 with P = 0.35). This was confirmed with a cross tabulation analysis, which included
Pearson Chi-Square with a value of 4.39 and P = 0.62, Likelihood Ratio value of 4.51 and P =
0.61 and Linear-by-Linear Association value of 0.08 and P = 0.77. Table 5 shows the number of
patients in each hot spot location for each breast separation size used for the cross tabulation
analysis.
Discussion:
This study was performed to determine if the FiF technique and/or the use of mixed
energies negated the previously established correlation of increasing hot spot point dose with
increasing breast separation measurement. The results verified, despite FiF techniques and mixed
energies, the correlation persists. The correlation between separation and hot spot was previously
established. It was determined in the studies by Das et al13 and Neal et al14 that hot spot dose
increased with an increase in separation and hot spots ranged from 0.2 – 27%. The hot spots in
this study ranged from 3.3-12.5%. The smaller range may be due to FiF blocking and mixed
energies, which were not used in previous comparisons. The FiF technique provides blocking of
defined areas of high dose while maintaining dose to low or adequately dosed areas per beam of
radiation using multi-leaf collimation. The overall desired effect is to achieve a homogenous
dose distribution, which reduces the hot spot dose, while maintaining the prescription dose. A
possible reason for this persistent correlation could be the limitations of the FiF technique and/or
different planning styles amongst medical dosimetrists. Blocking of high dose areas can decrease
dose along the chest wall and superficial margins of the breast. The medical dosimetrist,
physicist and/or physician often choose the higher dose to maintain coverage.
 8

Another method to achieve homogenous dose distribution is to combine photon energies.

One may argue using high energies mixed with low energies to achieve a homogenous dose
distribution in larger breast sizes may remove the correlation between separation and hot spot
dose. All patients in this study were treated with low energy photons (6 MV) alone or mixed with
high energy photons (10 or 18 MV). None of the patients were treated with high energy photons
alone. High energies alone may result in inadequate skin coverage of the breast tissue. This study
found a combination of high photon energies with low photon energies were used in larger breast
patients, whereas low energy alone was used in smaller breast sizes. The addition of high
energies did not remove the correlation between separation and hot spot dose. This may be due
to the FiF limitation mentioned above. Interestingly, it was found when assessing the sites
separately for this correlation, site 2 was not statistically significant. This is likely because site 2
combined 6 MV with 10 MV, whereas sites 1 and 3 combined 6 MV with 18 MV. Although 10
MV can be an ideal energy for treating intact breasts, it may not a large enough difference from 6
MV to result in statistical significance. The number of beams was also correlated with breast
separation. The larger breast size required more beams. This is likely due to the addition of the
high energy photons.
Correlations between separation with breast coverage and separation with CTV coverage
were not found. These absent correlations and the large range of breast volume coverage (44.4-
96.6%) and CTV coverage (22.2-100%) in this study may be due to variations in treatment
planning between clinic sites, physician preference in coverage, breast size, CTV location, and
effects from FiF technique. Each site was analyzed separately, and some significant correlations
were found, which may be explained by differing planning techniques and objectives. The breast
contours in all patients were drawn according to the RTOG guidelines, after treatment planning,
for use in this study. However, the RTOG study recognized some variability in defining breast
borders, and therefore inconsistencies between contours exist. It is common for physicians to
forego contouring breast tissue. Placement of the beams is often dictated by the location of the
tumor and the amount of heart and lung in the field. Some antero-medial and postero-lateral
areas of breast may be outside of the beams, thus decreasing the breast volume coverage. Figure
4 shows an example of beam placement for ideal breast coverage compared to reduced breast
coverage due to beam placement. Additionally, some physicians may prefer to focus the
coverage on the CTV while sparing heart and lung rather than focus on breast coverage. The
 9

higher CTV mean, median and mode, compared to breast coverage, are indicative of the
physicians’ priority on CTV coverage. There were 10 patients with CTV coverage less than 90%
of prescription dose. These CTV contours extended to the skin and therefore were not cropped 5
mm. The superficial location combined with the FiF technique to decrease hot spots may explain
the lack of coverage. The attending physicians may have approved this lack of coverage with the
intent for the boost treatment to compensate. There were 70 patients with less than 90% breast
volume coverage of prescription dose likely due to beam arrangement as well as effects from FiF
blocking. Figure 5 shows an example of decreased CTV coverage due to the tumor location,
shape of the breast and decreased dose along the breast margin secondary to FiF blocking.
Conclusion:
Radiation treatment of intact breast following lumpectomy has the goal of tumor control
with a desirable cosmetic outcome. Cosmetic outcome is a late effect of radiation, which is
dependent upon dose per fraction and total dose. High doses created by hot spots can cause
undesirable effects such as pigmentation, scarring, dimpling, atrophy, and asymmetry.2
Reduction of hot spots can be challenging, especially in larger breast sizes. The primary
advantage of FiF technique is the ability to lower the undesirable high dose in a step-like manner
without the use of wedges or resorting to arc therapy. One disadvantage of FiF is the potential for
decreased dose superficially and along the chest wall as areas of high dose are blocked to a level
preferred by the physician. The addition of high energy photon beams can compensate for this
lack of chest wall coverage. Dosimetrists devote much of their planning time attempting to find
the optimal blocking schemes and combinations of photon energies. This study showed despite
FiF technique and mixed photon energies, the hot spot dose increased with increasing breast
separation measurement. The hot spot ranged from 3.3-12.5% above prescription, which was
lower than previous studies. Additionally, the number of beams and photon energies increased
with increasing separation. Recommendations from this study for treatment planning include
adding high energy photons for breast separation measures greater than 18 cm and accepting hot
spots around 5350 cGy (106%) for small breasts, 5400 cGy (107%) for medium breasts and 5450
cGy (108%) for large breasts.
The limitations of this study include the retrospective nature of the study and the
variations between the 3 clinic sites. Correlation of separation with breast coverage and CTV
coverage within the sites separately were statistically significant; however, significance was lost
 10

when the sites were assessed together. Future studies with a larger patient cohort may help
confirm or more accurately identify these hot spot and photon energy recommendations.

Acknowledgements:
We would like to thank the Statistical Consulting Center at UW-La Crosse for its
assistance with statistical correlations between breast separation measurement and the recorded
variables. However, any errors of fact or interpretation remain the sole responsibility of the
authors.
 11

References
1. Onitilo AA, Engel JM, Stankowski RV, Doi SAR. Survival comparisons for breast
conserving surgery and mastectomy revisited: community experience and the role of radiation
therapy. Clin Med Res. 2015;13(2):65-73. https://dx.doi.org/10.3121/cmr.2014.1245.
2. Whelan TJ, Levine M, Math JJM, Kirkbride P, Skingley P. The effects of radiation therapy on
quality of life of women with breast carcinoma. Cancer. 2000;88(10):2260-2266.
https://dx.doi.org/10.1002/(SICI)1097-0142(20000515)88:10%3C2260::AID-
CNCR9%3E3.0.CO;2-M.
3. Njeh CF, Suh TS, Orton CG. Radiotherapy using hard wedges is no longer appropriate and
should be discontinued. Med Phys. 2016;43(3):1031-1034.
https://dx.doi.org/10.1118/1.4939262.
4. Abo-Madyan Y, Aziz MH, Aly MMOM, et al. Second cancer risk after 3D-CRT, IMRT and
VMAT for breast cancer. Radiother Oncol. 2014;110(3):471-476.
https://dx.doi.org/10.1016/j.radonc.2013.12.002.

5. Buchholz TA, Gurgoze E, Bice WS, Prestidge BR. Dosimetric analysis of intact breast
irradiation in off-axis planes. Int J Radiat Oncol Biol Phys. 1997;39(1):261-267.
https://dx.doi.org/10.1016/S0360-3016(97)00292-7.
6. Delaney G, Beckham W, Veness M, et al. Three-dimensional dose distribution of
tangential breast irradiation: results of a multicentre phantom dosimetry study. Radiother
Oncol. 2000;57(1):61-68. https://dx.doi.org/10.1016/S0167-8140(00)00262-0.
7. Buchholz TA, Bilton S, Gurgoze E, et al. Isoseparation curves: a mechanism for optimizing
off-axis dose homogeneity of intact breast irradiation. Radiat Oncol Investig. 1998;6(4):191-
198. https://dx.doi/10.1002/(SICI)1520-6823(1998)6:4%3C191::AID-ROI7%3E3.0.CO;2-T.
8. Emmens DJ, James HV. Irregular surface compensation for radiotherapy of the breast:
correlating depth of the compensation surface with breast size and resultant dose distribution.
Br J Radiol. 2010;83(986):159-165. https://dx.doi.org/10.1259/bjr/65264916.
9. Mayo C, Lo YC, Fitzgerald TJ, Urie M. Forward-planned, multiple-segment, tangential fields
with concomitant boost in the treatment of breast cancer. Med Dosim. 2004;29(4):265-270.
https://dx.doi.org/10.1016/j.meddos.2003.12.003.
10. Basaula D, Quinn A, Walker A, et al. Risks and benefits of reducing target volume margins
in breast tangent radiotherapy. Australas Phys Eng Sci Med. 2017;40(2):305-315.
 12

https://dx.doi.org/10.1007/s13246-017-0529-3.
11. Kantorowitz DA. The impact of dose-specification policies upon nominal radiation dose
received by breast tissue in the conservation treatment of breast cancer. Int J Radiat Oncol
Biol Phys. 2000;47(3):841-848. https://dx.doi.org/10.1016/S0360-3016(00)00453-3.
12. Jagsi R, Griffith KA, Heimburger D, et al. Choosing wisely? Patterns and correlates of the
use of hypofractionated whole-breast radiation therapy in the state of Michigan.
Radiation Oncology Quality Consortium. Int J Radiat Oncol Biol Phys. 2014;90(5):1010-
1016. https://dx.doi.org/10.1016/j.ijrobp.2014.09.027.
13. Das IJ, Cheng CW, Fein DA, Fowble B. Patterns of dose variability in radiation prescription
of breast cancer. Radiother Oncol. 1997;44:83-89. https://dx.doi.org/10.1016/S0167-
8140(97)00054-6.
14. Neal AJ, Torr M, Helyer S, Yarnold JR. Correlation of breast dose heterogeneity with breast
size using 3D CT planning and dose-volume histograms. Radiother Oncol. 1995;34(3):210-
218. https://dx.doi.org/10.1016/0167-8140(95)01521-H.
15. White J, Tai A, Arthur D, Buchholz T, MacDonald S, et al. Breast Cancer Atlas for Radiation
Therapy Planning: Consensus Definitions. Radiation Therapy Oncology Group (RTOG) Web
site. https://www.rtog.org/LinkClick.aspx?fileticket=SQhssxHu7Jg%3d&tabid=227.
Accessed June 28, 2018.
 13

Figures

Figure 1. Measurement of breast separation along tangent beam edge.

 14

(cGy)

(cm)

Figure 2. Separation (S) (cm) is on the X axis and hot spot dose (MAX) (cGy) is on the Y axis.
The data points represent the hot spot (maximum point dose) values (cGy) and the solid line is
the regression line.
 15

(cm)

Figure 3. Correlation of separation (S) with number of beams (N_F). The data points are the
separation measurements and the solid line is the regression line.
 16

Figure 4. Comparison of beam arrangements for less than ideal breast coverage (above)
and ideal breast coverage (below).
 17

Figure 5. Coverage of the CTV is less than ideal due to the tumor location, shape of the breast
and decreased dose along the breast margin secondary to FiF blocking.
 18

Tables

Table 1. Hot spot dose range, mean and median for each separation size.
Separation Size (cm) Range (cGy) Mean (cGy) Median (cGy)
18 5204.6 - 5550.5 5349 5298.9
18.1 to 22 5255.5 - 5664.7 5400.3 5392.6
> 22 5277.7 - 5669 5441.1 5436

Table 2. Photon energy combinations with associated number of patients and mean breast
separation for each group.
Group Energy of Each Total Number of Mean Breast
Number Beam (MV) Patients in Group Separation (cm)
1 6, 6 41 19.50
2 6, 18, 6, 18 30 23.62
3 6, 6, 18 4 23.58
4 6, 10, 6, 10 8 23.33
5 6, 6, 10 4 21.18
6 6, 10 3 21.10

Table 3. Cross tabulation analyses for each site comparing separation to photon energy.
Value / p-value Value / p-value Value / p-value
Site 1 Site 2 Site 3
Pearson Chi-Square 11.02 / 0.026 5.253 / 0.512 8.636 / 0.071

Likelihood Ratio 12.59 / 0.013 6.598 / 0.36 9.83 / 0.043

Linear-by-Linear 7.643 / 0.006 0.934 / 0.334 4.803 / 0.028

Association
 19

Table 4. Correlation of separation with breast coverage and separation with CTV coverage
within each site.
r-value / p-value r-value / p-value r-value / p-value
Site 1 Site 2 Site 3
Breast coverage -0.097 / 0.61 0.129 / 0.504 -0.43 / 0.016
CTV coverage -0.386 / 0.035 0.435 / 0.018 -0.198 / 0.285

Table 5. Cross tabulation data for breast separation group and hot spot location. Actual number
of patients within each group and the associated percent number of patients within the breast
separation group are listed.
UOQ UIQ LOQ LIQ Total
Small 9 4 3 3 19
47.4% 21.1% 15.8% 15.8% 100%
Medium 15 4 6 12 37
40.5% 10.8% 16.2% 32.4% 100%
Large 13 9 5 7 34
38.2% 26.5% 14.7% 20.6% 100%
Total 37 17 14 22 90
41.1% 18.9% 15.6% 24.4% 100%