Clinical and Translational Radiation Oncology 28 (2021) 118–123

Contents lists available at ScienceDirect

Clinical and Translational Radiation Oncology

journal homepage: www.elsevier.com/locate/ctro

Original Research Article

Dose constraints for whole breast radiation therapy based on the quality
assessment of treatment plans in the randomised Danish breast cancer
group (DBCG) HYPO trial
M.S. Thomsen a,⇑, M. Berg b, S. Zimmermann c, C.M. Lutz a, S. Makocki d, I. Jensen e, M.H.B. Hjelstuen f,
S. Pensold g, M.P. Hasler h, M.-B. Jensen i, B.V. Offersen j
a
Department of Medical Physics, Aarhus University Hospital, Aarhus, Denmark
b
Department of Medical Physics, Lillebaelt Hospital, Vejle, Denmark
c
Department of Oncology, Odense University Hospital, Odense, Denmark
d
Dept of Radiation Oncology and Faculty of Medicine Carl Gustav Carus, Technische Universität Dresden, Dresden, Germany
e
Department of Medical Physics, Aalborg University Hospital, Aalborg, Denmark
f
Department of Radiotherapy, Stavanger University Hospital, Stavanger, Norway
g
Praxis for Radiotherapy, Academic Teaching Hospital Dresden-Friedrichstadt, Dresden, Germany
h
Department of Oncology, Sørlandet Hospital, Kristiansand, Norway
i
Danish Breast Cancer Cooperative Group, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark
j
Department of Experimental Clinical Oncology and Department of Oncology, Aarhus University Hospital, Denmark

a r t i c l e i n f o a b s t r a c t

Article history: Purpose: Quality assessment of the treatment plans in the Danish Breast Cancer Group (DBCG) HYPO trial
Received 21 February 2021 was carried out based on prospectively reported dosimetric parameters and evidence-based dose con-
Revised 27 March 2021 straints for whole breast radiation therapy were derived.
Accepted 28 March 2021
Materials and methods: From 2009 to 2014, 1882 patients (pts) were randomised between
Available online 6 April 2021
50 Gy/25fractions (fr) versus 40 Gy/15fr. Doses to CTVp_breast (V95%, V107%-V110%, Dmax, and in addition
for 40 Gy plans V105%-V107%), ipsilateral lung (V20Gy/V17Gy), heart (V20Gy/V17Gy, V40Gy/V35Gy), and left ante-
Keywords:
rior descending coronary artery (LADCA) (Dmax) and use of respiratory gated technique were prospec-
Breast cancer
Radiation therapy
tively reported to the DBCG database. After end of accrual, these dosimetric parameters from all plans
Fractionation in the trial were compared to the pre-specified treatment constraints.
OAR dose constraints Results: In total, 1854 pts from eight radiation therapy (RT) centres in three countries were treated. No
statistically significant differences were found between the results for 40 Gy and 50 Gy plans, except
for CTVp_breast hot-spot volume (V107%-V110%). Of the 40 Gy pts, 90% with CTVp_breast > 600 mL and
95% with CTVp_breast  600 mL had a CTVp_breast hot-spot volume (V105%-V107%) <2%. In 95% of the
50 Gy plans, the CTVp_breast absolute hot-spot volume (V107%-V110%) was <0.5 mL and 1.7 mL for
CTVp_breast  600 mL and > 600 mL, respectively. Compliance was >99% for both heart and lung con-
straints. Largest deviation from protocol constraints was found for the volume of CTVp_breast covered
with 95% of the prescription dose or more (V95%). The CTV dose coverage (V95%) was >94.3% in 95% of
the right-sided pts, whereas the figures for 95% of the left-sided pts treated with and without respiratory
gating were 93.2% and 88.8%, respectively.
Conclusion: A high degree of compliance with protocol dose constraints was found for treatment plans in
the DBCG HYPO trial. New constraints for dose to organs at risk and high-dose volumes in the breast are
suggested for breast-only RT planning.
Ó 2021 The Authors. Published by Elsevier B.V. on behalf of European Society for Radiotherapy and
Oncology. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/
licenses/by-nc-nd/4.0/).

1. Introduction
⇑ Corresponding author at: Medical Physics, Department of Oncology, Aarhus
University Hospital, Palle Juul-Jensens Boulevard 99, Level D103, DK-8200 Aarhus
N, Denmark. For decades, the standard fractionation for radiation therapy
E-mail address: [email protected] (M.S. Thomsen). (RT) after breast conserving surgery was 50 Gy/25fractions (fr),

https://doi.org/10.1016/j.ctro.2021.03.009
2405-6308/Ó 2021 The Authors. Published by Elsevier B.V. on behalf of European Society for Radiotherapy and Oncology.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
M.S. Thomsen, M. Berg, S. Zimmermann et al. Clinical and Translational Radiation Oncology 28 (2021) 118–123

because early experience with moderately hypofractionated breast 2.2. Delineation of target volumes and organs at risk
RT resulted in excess radiation-associated loco-regional morbidity
[1,2]. Since then, 3D CT-based treatment planning has emerged. The CTVp_breast was defined according to DBCG guidelines,
Furthermore, the a/b ratio for late normal tissue changes in the which was identical to the ESTRO guidelines [8,9]. CTVp_boost
breast has been reported to be around 3 Gy, supporting modest was defined as a 5 mm expansion from the volume including sur-
increase to fraction size above 2 Gy [3]. Such schedules were tested gical clips and a relevant part of the surgical cavity guided by pre-
in randomised trials in Canada and the UK. The Canadian trial surgical imaging and the surgical report, and cropped inside the
tested 42.5 Gy/16fr versus 50 Gy/25fr [4], whereas the UK START CTVp_breast. PTVs were constructed from the CTVs following the
Trial B tested 40 Gy/15fr/3weeks versus 50 Gy/25fr/5weeks [5,6]. institutional standard CTV-PTV margins for whole breast RT, and
The favourable results from these trials paved the way for more cropped to 5 mm below the skin. Heart, ipsilateral lung and con-
widespread use of moderate hypofractionation for breast RT. tralateral breast were defined as organs at risk (OAR) [10].
After the initial encouraging results from the START Trial B [5],
the Danish Breast Cancer Group (DBCG) RT Committee decided to 2.3. Treatment planning
reintroduce moderately hypofractionated RT through a clinically
controlled randomised trial, including patients with an indication The recommended treatment technique was a one-isocenter
for breast-only RT. The fractionation design of the DBCG HYPO trial technique with two tangential fields with parallel posterior field
was similar to the START Trial B, however with an additional con- borders. Wedges or field-in-field segments were applied to obtain
straint for breast doses in the range of 105% to 107% of the pre- a homogeneous dose distribution. However, inverse optimized
scription dose in the 40 Gy arm compared to the 50 Gy arm. The IMRT planning was also an option. The institutional dose calcula-
primary endpoint in the DBCG HYPO trial was 3-year breast tion algorithm was used. There was no restriction on the photon
induration grade 2–3. energy used except that the 95% isodose curve should cover the
A quality assessment (QA) program was set up to ensure that breast 5 mm below the surface. The CTVp_breast was to be covered
the technical guidelines were followed in the participating centres with doses in the range 95%-105% (40 Gy plans) and 95%-107%
(4 in Denmark, 2 in Norway and 2 in Germany). For each individual (50 Gy plans) of the prescription dose and Dmax  110%. For the
treatment plan, a selection of dosimetric parameters were prospec- 40 Gy arm, a volume of up to 2% of the CTVp_breast could receive
tively reported to the DBCG database. After end of accrual, analysis a dose between 105% and 107% of the prescription dose
of these parameters evaluated the compliance of the participating (V105%-V107%2%). In both treatment arms, an absolute volume of
centers with the trial protocol. Recently, the first results from the the CTVp_breast of up to 2 mL could receive a dose in the range
trial were reported with median 7.3-year follow-up. The nine- 107%-110% of the prescription dose (V107%-V110%<2mL). In this
year loco-regional recurrence risks in the DBCG HYPO study were way, hot spots in the breast were defined for doses >105% for
3.3% (95% Cl, 2.0% to 5.0%) and 3.0% (95% Cl, 1.9% to 4.5%) in the 40 Gy plans and >107% for 50 Gy plans, respectively. No part of
50 Gy and 40 Gy arm, respectively, and radiation-associated car- the CTVp_breast except for build-up regions should be covered
diac and lung disease was seldom [7]. This supported that the dose with doses <95% of the prescription dose (V95%95%), correspond-
constraints applied for the treatment planning in this study were ing to under-dosed volumes.
appropriate, and the dosimetric results from the study may be used The maximum dose to <25% of the ipsilateral lung was 20 Gy
to improve the constraints for future planning. (50 Gy plans) and 17 Gy (40 Gy plans) (V20Gy/V17Gy  25%). <10%
and 5% of the heart should receive 20 Gy and 40 Gy, respectively,
in 50 Gy plans and 17 Gy and 35 Gy, respectively, in 40 Gy plans
2. Materials and methods
(V20Gy/V17Gy  10% and V40Gy/V35Gy  5%). Delineation of the left
anterior descending coronary artery (LADCA) and contralateral
Patients older than 40 years with a T1-2 N0-N1(mic) invasive
breast was optional. If LADCA was delineated, DLADCA,max was
adenocarcinoma or ductal carcinoma in situ (DCIS) referred to
17 Gy and 20 Gy in 40 Gy and 50 Gy plans, respectively. The dose
postoperative whole breast RT were eligible for inclusion in the
to the contralateral breast should be as low as possible. It is note-
DBCG HYPO trial. No RT of regional nodes was allowed. Stratifica-
worthy that the constraints for 50 Gy/25fr were all based on con-
tion factors were breast size (smaller/larger than 600 mL), boost
sensus (in lack of evidence) in the DBCG RT Committee dating
(yes/no), systemic treatment and institution. The patients were
back to the days where CT-based RT planning was introduced.
randomized to whole breast irradiation 50 Gy/25fr/5weeks versus
The committee was worried about excess morbidity after
40 Gy/15fr/3weeks. Boost was prescribed as sequential 5–8 fr of
hypofractionated RT, so the constraints were deliberately stricter
2 Gy according to national or institutional guidelines (supplemen-
for the 40 Gy plans. Highest priority was given to CTVp_boost or
tary material in [7]). A written informed consent according to Good
tumor bed if no boost, thereafter the priority was
Clinical Practice guidelines and local and national rules of partici-
heart > ipsilateral lung > CTVp_breast > PTVp_breast > contralatera
pating institutions was obtained. The trial was approved by the
l breast, however, every treatment plan was to be balanced and
Ethics Committee in Region Midt, Denmark, on behalf of all partic-
approved after considering information about the patient, comor-
ipating Danish centres, and by local Ethics Committees for the non-
bitidy, tumour characteristics and other treatment factors.
Danish centres (clinicaltrials.gov number NCT00909818).
2.4. Data available for analysis
2.1. Planning CT scanning
The dosimetric parameters for CTVp_breast (V95%, V107%-V110%,
Treatment preparation was according to DBCG RT guidelines. Dmax, and in addition for 40 Gy plans V105%-V107%), ipsilateral lung
The neck and breast region including both lungs was CT scanned (V20Gy/V17Gy), heart (V20Gy/V17Gy, V40Gy/V35Gy) and LADCA (Dmax)
with the patient positioned in the institutional standard fixation were prospectively reported to the DBCG database. Prior to analy-
for this patient group. The only requirement was that a daily repro- sis, the data quality was investigated, and if values deviated much
ducibility of approx. 5 mm should be achieved during the course of from expected values, the centre was asked to confirm the value.
treatment. The institutional guideline for respiratory gating and Use of respiratory gating was added to the database in 2012 [7],
image verification were followed and had to be identical for the thus, data on gating prior to 2012 was collected retrospectively
two treatment arms.. from the centres.
119
M.S. Thomsen, M. Berg, S. Zimmermann et al. Clinical and Translational Radiation Oncology 28 (2021) 118–123

2.5. Data analysis respectively, corresponding to 90% of the large-breasted and 95% of
the small-breasted patients having treatment plans with a
Compliance with the protocol treatment constraints and 50th CTVp_breast hot-spot volume (V105% -V107%) up to 2% (Fig. 1b). Lar-
(median), 90th, and 95th percentiles for dosimetric parameters ger breast hot-spot volumes were observed in 50 Gy plans com-
were determined for both treatment arms, and for various sub- pared to 40 Gy plans (p = 0.001) as 2.1% (small breasts) and 4.7%
groups of patients according to laterality, use of respiratory gating, (large breasts) of the 40 Gy patients received breast doses above
and CTVp_breast volume. Student’s t-test in MS Excel version 2010 107% of the prescription dose, whereas the corresponding figures
was used to test for statistical significance (p < 0.05). for 50 Gy patients were 9.0% and 14.8% (Fig. 1c). However, in 95%
of the 50 Gy plans the CTVp_breast absolute hot-spot volume
(V107%-V110%) was<0.5 mL and 1.7 mL for small and large breast,
3. Results respectively.
For both lung and heart constraints, compliance was > 99%
Between May 2009 and March 2014, 1854 eligible patients (table 2). Delineation of the heart was not carried out in 57 (6%)
were treated [7]. RT characteristics from their treatment plans of the right-sided patients, whereas data was missing for only
are presented in table 1. Table 2 shows the compliance with the three left-sided patients. Since it was optional to delineate LADCA,
predefined dose constraints. Data on modified treatment plans a higher number of missing data was expected for LADCA. For ipsi-
was reported for 5 pts (0.3%) and of these, 3 pts had a new plan lateral lung, the highest lung V20Gy/17Gy values were seen for right-
during the treatment course, whereas the remaining two were sided patients, followed by FB and gated left-sided patients,
replanned prior to treatment start. respectively (Fig. 2a). No statistically significant difference was
The largest number of deviations from the constraint was found found between the lung values for 40 Gy plans and 50 Gy plans
for the dose coverage (V95%), however, no statistically significant with p = 0.50 (right), p = 0.06 (left, gated), and p = 0.53 (left, FB).
difference was found between the under-dosed volumes of The heart data are shown in Fig. 2b and c. Also, for the heart no sta-
CTVp_breast in the 50 Gy and 40 Gy plans (Fig. 1a). The best dose tistically significant difference was found when comparing 40 Gy
coverage of CTVp_breast was reported for right-sided patients with plans and 50 Gy plans with p-values for V20Gy/17Gy of 0.31 (right),
V95%>94.3% in 95% of the plans, whilst the coverage in left-sided 0.27 (left, gated) and 0.80 (left, FB) and corresponding values for
gated patients was significantly less with V95%>93.2% (p < 0.001), V40Gy/35Gy of 0.25, 0.38, and 0.16. Table 3 summarizes the 50th
and even worse for left-sided patients treated in free breathing (median), 90th, and 95th percentiles for lung, heart and breast
(FB), where V95%>88.8% (p < 0.001) (Fig. 1a). In the 40 Gy treatment dosimetric parameters.
arm, non-compliance to the constraint V105%-V107%2% was found
in 68 treatment plans of which 47 and 21 patients had large
(CTVp_breast > 600 mL) and small breasts (CTVp_breast  600 mL), 4. Discussion

This paper presents dosimetric parameters from the individual

Table 1 CT-based treatment plans of all patients treated in the DBCG HYPO
Characteristics of the 1854 patients treated in the DBCG HYPO trial.
trial. Compliance with the pre-defined dose constraints was high
Fractionation schedule for all prospectively reported parameters, except the under-
50 Gy 40 Gy Total dosed volume of CTVp_breast (1-V95%). For 95% of the right-sided
Number of patients (n) 937 917 1854 patients, CTVp_breast V95% was>94.3% whereas the corresponding
Age [y] mean (min–max) 59 (42–83) 59 (41–82) 59 (41–83) values for left-sided patients were 93.2% (gated) and 88.8% (FB),
CTVp_breast [mL] mean 709 (644, 721 (635, 715 (640,
showing that the heart constraints had higher priority in the pro-
(median, IQR) 502) 531) 520)
Laterality tocol than CTVp_breast coverage. This reflects the focus on the risk
Right (n) 455 445 900 of late cardiac morbidity in this group of early breast cancer
Left (n) 482 472 954 patients [11–13] and a recent analysis of cardiac substructures in
Gating (left-sided patients) (n) the Danish HYPO cohort showed decreasing mean heart and
Yes 231 226 457
No 251 246 497
LADCA doses with treatment year [14]. However, the excellent
Boost (n) clinical results obtained in this study with nine-year loco-
No 721 703 1424 regional recurrence risks of 3.3% (95% Cl, 2.0% to 5.0%) for the
10 Gy/5 fr 183 185 368 50 Gy arm and 3.0% (95% Cl, 1.9% to 4.5%) for the 40 Gy arm indi-
16 Gy/8 fr 33 29 62
cate that this trade-off was feasible [7]. No statistically significant

Table 2
Compliance with dose constraints in the DBCG HYPO trial for volume of CTVp_breast receiving>95% of prescription dose (V95%), volume of CTVp_breast receiving 105–107% (40 Gy
plans) and 107–110% of prescription dose (V105%-V107%, V107%-V110%), maximum dose to CTVp_breast (Dmax), volumes of lung and heart receiving 20 Gy (50 Gy plans) or 17 Gy
(40 Gy plans) (V20Gy/V17Gy), heart receiving 40 Gy (50 Gy plans) or 35 Gy (40 Gy plans) (V40Gy/V35Gy), and maximum dose to left anterior descending coronary artery (LADCA).

50 Gy 40 Gy
Protocol constraint Compliance Non compliance Missing data Compliance Non compliance Missing data
V95%  95% 767 (81.9%) 166 (17.7%) 4 (0.4%) 730 (79.6%) 185 (20.2%) 2 (0.2%)
V105%-V107% 2% – – – 846 (92.3%) 68 (7.4%) 3 (0.3%)
V107%-V110% <2cm3 899 (95.9%) 35 (3.7%) 3 (0.3%) 907 (98.9%) 9 (1.0%) 1 (0.1%)
Dmax  110% 927 (98.9%) 7 (0.7%) 3 (0.3%) 911 (99.3%) 5 (0.5%) 1 (0.1%)
Lung V20Gy/V17Gy  25% 924 (98.6%) 8 (0.9%) 5 (0.5%) 912 (99.5%) 3 (0.3%) 2 (0.2%)
Heart V20Gy/V17Gy  10% Right 431 (94.7%) 0 (0.0%) 24 (5.3%) 412 (92.6%) 0 (0.0%) 33 (7.4%)
Left 480 (99.6%) 0 (0.0%) 2 (0.4%) 470 (99.6%) 1 (0.2%) 1 (0.2%)
Heart V40Gy/V35Gy  5% Right 431 (94.7%) 0 (0.0%) 24 (5.3%) 412 (92.6%) 0 (0.0%) 33 (7.4%)
Left 480 (99.6%) 0 (0.0%) 2 (0.4%) 470 (99.6%) 1 (0.2%) 1 (0.2%)
LADCA Dmax  20 Gy/17 Gy Right 404 (88.8%) 0 (0.0%) 51 (11.2%) 395 (88.8%) 0 (0.0%) 50 (11.2%)
Left 411 (85.3%) 29 (6.0%) 42 (8.7%) 382 (79.3%) 47 (9.8%) 43 (8.9%)

120
M.S. Thomsen, M. Berg, S. Zimmermann et al. Clinical and Translational Radiation Oncology 28 (2021) 118–123

Fig. 2. a. Lung V20Gy (50 Gy plans) and V17Gy (40 Gy plans) b. Heart V20Gy (50 Gy
plans) and V17Gy (40 Gy plans), and c. Heart V40Gy (50 Gy plans) and V35Gy (40 Gy
plans) versus percentile of patients. The constraints in the trial protocol are shown
Fig. 1. a: Volume of CTVp_breast covered with 95% of prescription dose (V95%), b: with horizontal dashed-dotted lines, whereas the vertical lines show the 90th (solid
Volume of CTVp_breast with a hot spot dose between 105% and 107% of the line) and 95th (dashed line) percentiles.
prescription dose (V105%-V107%) for 40 Gy plans, and c: Volume of CTVp_breast with
a hot spot dose between 107% and 110% of the prescription dose (V107%-V110%) for
both 40 Gy and 50 Gy plans versus percentile of patients. In b and c results are
high energy photon beam in the treatment plan, thereby increasing
shown for patients with a CTVp_breast smaller than or equal to 600 mL and
CTVp_breast>600 mL. Constraints are shown with horizontal dashed-dotted lines,
the build-up zone in the breast. However, this was not reflected in
whereas the vertical lines show the 90th (solid line) and 95th (dashed line) the under-dosed volumes of CTVp_breast, where no statistically
percentiles. In c, two (27.0 and 48.6 mL) and one (24.3 mL) data points are outside significant difference was observed between the 50 Gy and 40 Gy
the scale for CTVp_breast > 600 mL (50 Gy plans) and CTVp_breast > 600 mL (40 Gy plans. Thus, the hot-spot constraint V105%-V107%2% used in the
plans), respectively.
protocol is recommended as a hot-spot constraint for whole breast
RT plans in the future.
differences were found between the results for 40 Gy and 50 Gy From a clinical perspective, no patients were hospitalized with
plans except for CTVp_breast hot-spot volume (V107%-V110%). radiation pneumonitis in the DBCG HYPO trial, and only very few
The statistically significant larger proportion of 50 Gy patients patients had died from heart or lung disease with no indication
with a higher hot-spot breast volume compared to 40 Gy patients of an excess risk [7]. For the lung and heart, the 95th percentile val-
was a result of the additional dose constraint of V105%-V107%2% for ues for lung and heart presented in table 3 are therefore suggested
40 Gy plans. As the patients in the two arms had similar breast as new OAR constraints for whole-breast RT planning. These con-
sizes, it was expected, that more 50 Gy plans could have had a straints are to our knowledge the first constraints for doses to
smaller hot-spot volume if the additional 40 Gy hot-spot constraint OAR based on dosimetric data collected from all treatment plans
had been applied to the 50 Gy patients. The lower hot spot volumes in a randomized trial, and they represent constraints depending
for 40 Gy patients may be due to the use of a higher proportion of on laterality and use of respiratory gated technique.

121
M.S. Thomsen, M. Berg, S. Zimmermann et al. Clinical and Translational Radiation Oncology 28 (2021) 118–123

Table 3
50th (median), 90th, and 95th percentile values for lung V20Gy/V17Gy, heart V20Gy/V17Gy and V40Gy/V35Gy, and volume of CTVp_breast covered with 95% of prescription dose (V95%)
for right-sided and left-sided patients treated with respiratory gating and in free breathing, respectively. The same figures are given for volume of CTVp_breast with doses
between 105% and 107% of prescription dose (V105%-V107%) (40 Gy) and between 107% and 110% of prescription dose (V107%-V110%) for 40 Gy plans, 50 Gy plans, and pooled.

Percentile Right Left, gating Left free breathing

Lung V20Gy/V17Gy [%] 50th 16.6 13.5 14.3
90th 22.3 18.3 20.9
95th 23.8 19.6 22.3
Heart V20Gy/V17Gy [%] 50th 0.0 0.1 0.7
90th 0.0 1.1 4.5
95th 0.0 2.0 6.4
Heart V40Gy/V35Gy [%] 50th 0.0 0.0 0.0
90th 0.0 0.1 1.9
95th 0.0 0.4 3.2
CTVp_breast V95% [%] 50th 97.6 97.0 95.5
90th 95.2 94.1 90.4
95th 94.3 93.2 88.8
Percentile 40 Gy/15fx 50 Gy/25fx All
CTVp_breast V105% -V107% [%] 50th 0.0 NA NA
90th 1.8 NA NA
95th 2.7 NA NA
CTVp_breast V107% -V110% [mL] 50th 0.0 0.0 0.0
90th 0.0 0.1 0.0
95th 0.0 1.2 0.3

The high degree of compliance (>99%) with the heart protocol also prospectively collected in a national storage facility of DICOM
constraints was due to the fact that in the treatment planning data (plan data bank), where values for analysis are not limited to a
the heart had the highest priority after the tumour bed. Shortly pre-defined set. Due to legislation issues at the time of recruit-
after the end of recruitment in the HYPO trial an initial analysis ment, it was not possible for the Norwegian and German centres
of the heart doses from the Danish HYPO patients was presented to export treatment plans to the Danish data bank. Thus, only
to the DBCG RT Committee, and it was decided to change the heart results for the predefined and prospectively reported parameters
constraints in the RT guidelines to V35Gy < 1% and V17Gy < 5% for can be presented for the total number of treatment plans in the
whole-breast planning. These values may now be reduced even trial.
further by the results presented here. For loco-regional treatment Brink et al. demonstrated that the data quality in a treatment
the corresponding DBCG constraints are  5% and  10%, plan data bank was superior to manually reported data [23]. There-
respectively. fore, quality assurance of the manually entered dosimetric param-
In accordance with other large trials (IMPORT LOW, FAST, and eters was carried out prior to the data analysis. If a dosimetric
FAST Forward), this trial recommended two tangential fields with value deviated much from the expected range, the responsible
field-in-field segments to obtain a homogeneous dose distribution physicist at the centre was asked to confirm the value. Dose cover-
[5,15,16]. However, 3D CRT forward field-in-field planning can be age of the CTVp_breast (V95%) was the dosimetric parameter with
tedious. Therefore, several planning studies with a limited number most need of corrections during the data validation, confirming
of patients have investigated alternative field arrangements using the findings in [23]. This was probably because the parameter to
either two inversed planned tangential IMRT fields, IMRT with be prospectively reported was 1-V95% while the value most easily
more than two fields or volumetric modulated arc therapy (VMAT) read from the treatment planning systems was V95%. Thus, for man-
[17–19]. The advantage of a tangential field technique is that irra- ually recording and reporting dose values it is recommended to
diation of the contralateral side of the patient is minimised. Fur- choose parameters which are easily read from dose volume
thermore, Aznar et al. showed in a systematic review of lung histograms.
doses from breast cancer RT that tangential fields spared both During the trial period, the fraction of left-sided patients treated
the ipsilateral and contralateral lung better than IMRT for whole with respiratory gating increased from 7% in 2009–2010 to 85% in
breast RT without nodal irradiation [20]. With breathing adaption, 2013–2014, thus in total 457 (48%) of the left-sided patients were
3D-CRT and tangential IMRT fields also had higher potential for treated with gating, whereas the remaining 497 left-sided patients
sparing the heart compared to a partial VMAT technique [21]. were treated in FB. Results from treatment plans for 684 left-sided
These conclusions were recently validated in another study show- Danish patients in the DBCG HYPO trial submitted to the plan data
ing that the 3D-CRT in many situations is to be preferred [28]. bank presented by Berg et al were in harmony with other studies
Formation of fibrosis is dependent on several factors including reporting that respiratory gated technique can improve target, lung
hot-spot volumes in the breast [22]. This study has shown that and heart dose [24–26]. This was also observed in the complete
breast hot-spot volumes also for large breasts can be minimized data set from this trial. Apart from a lower heart exposure for gated
with tangential fields. Thus, this technique is far from an outdated patients, also lower lung V20Gy/17Gy values were obtained, poten-
method for whole breast RT. tially resulting in a decreased risk of a second lung cancer [27].
This report was limited to the dosimetric parameters defined In conclusion, the prospectively collected treatment data from
prior to the start of the trial and manually reported to the DBCG all treated breast cancer patients in the DBCG HYPO trial demon-
database. In this way it was possible to collect data from all strated a high degree of compliance with pre-specified constraints
patients treated. However, at that time the DBCG RT committee for treatment planning. This suggests that more strict dose con-
decided to report values related to CTVp_breast, whereas straints for organs at risk and for high dose breast volumes can
PTVp_breast values were not included in the list of manually be applied and are recommended for future whole breast RT
reported values. All treatment plans from the Danish centres were planning.

122
M.S. Thomsen, M. Berg, S. Zimmermann et al. Clinical and Translational Radiation Oncology 28 (2021) 118–123

Declaration of Competing Interest [12] Darby SC, Ewertz M, McGale P, Bennet AM, Blom-Goldman U, Brønnum D, et al.
Risk of ischemic heart disease in women after radiotherapy for breast cancer.
N Engl J Med 2013;368(11):987–98.
The authors declare that they have no known competing finan- [13] Laugaard Lorenzen E, Christian Rehammar J, Jensen M-B, Ewertz M, Brink C.
cial interests or personal relationships that could have appeared Radiation-induced risk of ischemic heart disease following breast cancer
radiotherapy in Denmark, 1977–2005. Radiother Oncol 2020;152:103–10.
to influence the work reported in this paper.
[14] Finnegan R, Lorenzen EL, Dowling J, Jensen I, Berg M, Thomsen MS, et al.
Analysis of Cardiac Substructure Dose in a Large, Multi-Centre Danish Breast
Acknowledgments Cancer Cohort (the DBCG HYPO trial): Trends and Predictive Modelling.
Radiother Oncol 2020;153:130–8.
[15] Yarnold JR. First results of the randomised UK FAST Trial of radiotherapy
We thank the patients who participated in this trial and the hypofractionation for treatment of early breast cancer (CRUKE/04/015).
staff at participating centres and in the DBCG office. BVO was sup- Radiother Oncol 2011;100:93–100.
ported by the Danish Cancer Society and CIRRO (Centre for Inter- [16] Brunt AM, Haviland JS, Wheatley DA, Sydenham MA, Alhasso A, Bloomfield DJ.
FAST-Forward phase III multi-centre non-inferiority randomised controlled
ventional Research in Radiation Oncology) and the DCCC RT trial of 1- versus 3-week hypofractionated breast radiotherapy: 5-year results
(Danish Comprehensive Cancer Centre RadioTherapy Group). for efficacy and late normal tissue effects. Lancet 2020;395:1613–26.
[17] Penninkhof J, Spadola S, Breedveld S, Baaijens M, Lanconelli N, Heijmen B.
Individualized Selection of Beam Angles and Treatment Isocenter in Tangential
References Breast Intensity Modulated Radiation Therapy. Int J Rad Oncol Biol Phys
2017;98(2):447–53.
[1] Overgaard M, Bentzen SM, Juul Christensen J, Hjøllund Madsen E. The value of [18] Virén T, Heikkilä J, Myllyoja K, Koskela K, Lahtinen T, Seppälä J. Tangential
the NSD formula in equation of acute and late radiation complications in volumetric modulated arc therapy technique for left-sided breast cancer
normal tissue following 2 and 5 fractions per week in breast cancer patients radiotherapy. Radiat Oncol 2015;10:79.
treated with postmastectomy irradiation. Radiother Oncol 1987;9(1):1–11. [19] Karpf D, Sakka M, Metzger M, Grabenbauer GG. Left breast irradiation with
[2] Friberg S, Rudén B-I. Hypofractionation in radiotherapy: An investigation of tangential intensity modulated radiotherapy (t-IMRT) versus tangential
injured Swedish women, treated for cancer of the breast. Acta Oncol 2009;48 volumetric modulated arc therapy (t-VMAT): trade-offs between secondary
(6):822–31. cancer induction risk and optimal target coverage. Radiat Oncol 2019;14:156.
[3] Yarnold J, Ashton A, Bliss J, Homewood J, Harper C, Hanson J, et al. [20] Aznar MC, Duane FK, Darby SC, Wang Z, Taylor CW. Exposure of the lungs in
Fractionation sensitivity and dose response of late adverse effects in the breast cancer radiotherapy: A systematic review of lung doses published
breast after radiotherapy for early breast cancer: long-term results of a 2010–2015. Radiother Oncol 2018;126(1):148–54.
randomised trial. Radiother Oncol 2005;75(1):9–17. [21] Vikström J, Hjelstuen MH, Wasbø E, Mjaaland I, Dybvik KI. A comparison of
[4] Whelan TJ, Pignol J-P, Levine MN, Julian JA, MacKenzie R, Parpia S, et al. Long- conventional and dynamic radiotherapy planning techniques for early-stage
term results of hypofractionated radiation therapy for breast cancer. N Engl J breast cancer utilizing deep inspiration breath hold. Acta Oncol 2018;57
Med 2010;362(6):513–20. (10):1325–30.
[5] Bentzen SM, Agrawal RK, Aird EGA, Barrett JM, Barrett-Lee PJ, Bliss JM, et al. [22] Donovan E, Bleakley N, Denholm E, Evans P, Gothard L, Hanson J, et al.
The UK Standardisation of Breast Radiotherapy (START) Trial B of radiotherapy Randomised trial of standard 2D radiotherapy (RT) versus intensity modulated
hypofractionation for treatment of early breast cancer: a randomised trial. radiotherapy (IMRT) in patients prescribed breast radiotherapy. Radiother
Lancet 2008;371:1098–107. Oncol 2007;82(3):254–64.
[6] Haviland JS, Owen JR, Dewar JA, Agrawal RK, Barrett J, Barrett-Lee PJ, et al. The [23] Brink C, Lorenzen EL, Krogh SL, Westberg J, Berg M, Jensen I, et al. DBCG hypo
UK Standardisation of Breast Radiotherapy (START) trials of radiotherapy trial validation of radiotherapy parameters from a national databank versus
hypofractionation for treatment of early breast cancer: 10-year follow-up manual reporting. Acta Oncol 2018;57(1):107–12.
results of two randomised controlled trials. Lancet Oncol 2013;14 [24] Berg M, Lorenzen EL, Jensen I, Thomsen MS, Lutz CM, Refsgaard L, et al. The
(11):1086–94. potential benefits from respiratory gating for breast cancer patients regarding
[7] Offersen BV, Alsner J, Nielsen HM, Jakobsen EH, Nielsen MH, Krause M et al. target coverage and dose to organs at risk when applying strict dose limits to
Hypofractionated versus standard fractionated radiotherapy of 1882 patients the heart: results from the DBCG HYPO trial. Acta Oncol 2018;57(1):113–9.
with early breast cancer or ductal carcinoma in situ in the randomized phase [25] Vikström J, Hjelstuen MHB, Mjaaland I, Dybvik KI. Cardiac and pulmonary dose
III trial: the DBCG HYPO trial. J Clin Oncol 2020;38:JCO.20.01363-3625. reduction for tangentially irradiated breast cancer, utilizing deep inspiration
[8] Nielsen MH, Berg M, Pedersen AN, Andersen K, Glavicic V, Jakobsen EH, et al. breath-hold with audio-visual guidance, without compromising target
Delineation of target volumes and organs at risk in adjuvant radiotherapy of coverage. Acta Oncol 2011;50(1):42–50.
early breast cancer: national guidelines and contouring atlas by the Danish [26] Nissen HD, Appelt AL. Improved heart, lung and target dose with deep
Breast Cancer Cooperative Group. Acta Oncol 2013;52(4):703–10. inspiration breath hold in a large clinical series of breast cancer patients.
[9] Offersen BV, Boersma LJ, Kirkove C, Hol S, Aznar MC, Biete Sola A, et al. ESTRO Radiother Oncol 2013;106(1):28–32.
consensus guideline on target volume delineation for elective radiation [27] Grantzau T, Thomsen MS, Væth M, Overgaard J. Risk of second primary lung
therapy of early stage breast cancer. Radiother Oncol 2015;114(1):3–10. cancer in women after radiotherapy for breast cancer. Radiother Oncol
[10] Feng M, Moran JM, Koelling T, Chughtai A, Chan JL, Freedman L, et al. 2014;111(3):366–73.
Development and validation of a heart atlas to study cardiac exposure to [28] Stick LB, Lorenzen EL, Yates ES, Anandadas C, Andersen K, Aristei C, et al.
radiation following treatment for breast cancer. Int J Radiat Oncol Biol Phys Selection criteria for early breast cancer patients in the DBCG proton trial - The
2011;79(1):10–8. randomised phase III trial strategy. CTRO 2021;27:126–31. https://doi.org/
[11] Taylor CW, Brønnum D, Darby SC, Gagliardi G, Hall P, Jensen M-B, et al. Cardiac 10.1016/j.ctro.2021.01.012.
dose estimates from Danish and Swedish breast cancer radiotherapy during
1977–2001. Radiother Oncol 2011;100(2):176–83.

123