- Email: [email protected]
Adaptive radiation therapy for breast IMRT-simultaneously integrated boost: Three-year clinical experience
Radiotherapy and Oncology 103 (2012) 183–187
Contents lists available at SciVerse ScienceDirect
Radiotherapy and Oncology journal homepage: www.thegreenjournal.com
Adaptive radiotherapy for breast cancer
Adaptive radiation therapy for breast IMRT-simultaneously integrated boost: Three-year clinical experience Coen W. Hurkmans a,⇑, Ingrid Dijckmans a, Miranda Reijnen a, Jorien van der Leer a, Corine van Vliet-Vroegindeweij b, Maurice van der Sangen a a
Department of Radiation Oncology, Catharina Hospital, Eindhoven; and b Department of Radiation Oncology, The Netherlands Cancer Institute – Antoni van Leeuwenhoek Hospital, Amsterdam, The Netherlands
a r t i c l e
i n f o
Article history: Received 29 March 2011 Received in revised form 15 December 2011 Accepted 19 December 2011 Available online 24 January 2012 Keywords: Breast cancer Post-operative seroma volume change Adaptive image guided radiation therapy Simultaneously integrated boost Tumor bed volume
a b s t r a c t Purpose: It has been shown that seroma volumes decrease during breast conserving radiotherapy in a significant percentage of patients. We report on our experience with an adaptive radiation therapy (ART) strategy involving rescanning and replanning patients to take this reduction into account during a course of intensity-modulated radiation therapy with simultaneously integrated boost (IMRT-SIB). Materials: From April 2007 till December 2009, 1274 patients eligible for SIB treatment were enrolled into this protocol. Patients for which the time between the initial planning CT (CT1) and lumpectomy was less than 30 days and who had an initial seroma volume >30 cm3 were rescanned at day 10 of treatment (CT2) and replanned when significant changes were observed by the radiation oncologist. Patients received 28 fractions of 1.81 Gy to the breast and 2.30 Gy to the boost volume. Results: Nine percent (n = 113) of the 1274 patients enrolled met the criteria and were rescanned. Of this group, 77% (n = 87) of treatment plans were adapted. Time between surgery and CT1 (20 days versus 20 days for adapted and non-adapted plans, p = 0.89) and time between CT1 and CT2 (21 days versus 22 days for adapted and non-adapted plans, p = 0.43) revealed no procedural differences which might have biased our results. In the adapted plans, seroma decreased significantly from 60 to 27 cm3 (p < 0.001), TBV from 70 to 45 cm3 (p < 0.001) and PTVboost from 277 to 220 cm3 (p < 0.001). The volume receiving more than 95% of the boost dose (V95%(total-dose)) could be reduced by 19% (linear fit, R2 = 0.73) from on average 360 to 292 cm3 (p < 0.001). Delay in treatment and the use of a prolonged treatment schedule with different fractionation for patients with seroma could thus be prevented. Conclusion: The adaptive radiation therapy IMRT-SIB procedure has proven to be efficient and effective, leading to a clinically significant reduction of the high dose volume. Seroma present in a subgroup of patients referred for breast radiation therapy does not hamper the introduction of highly conformal IMRT-SIB techniques. Ó 2012 Elsevier Ireland Ltd. All rights reserved. Radiotherapy and Oncology 103 (2012) 183–187
The breast tumor bed after breast conserving surgery generally does not change significantly during radiation therapy when no seroma is present [1]. However, post-operative seroma is sometimes present and seroma reductions are seen between the acquisition of the planning CT scan and end of radiation therapy in a substantial percentage of the patients [2–5]. These reductions can be substantial and have hampered the introduction of simultaneously integrated boost (SIB) techniques for breast cancer patients, as it was not clear how much these variations would effect SIB plans. SIB techniques have been shown to provide more conformal treatment plans than conventional sequential boost planning in the absence of volume changes [6,7]. Furthermore, SIB enables a reduction in the overall treatment time by 1 week ⇑ Corresponding author. Address: Department of Radiation Oncology, Catharina Hospital, Michelangelolaan 1, Eindhoven, The Netherlands. E-mail address: [email protected] (C.W. Hurkmans). 0167-8140/$ - see front matter Ó 2012 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.radonc.2011.12.014
when the conventional fractionation scheme of 25 fractions of 2 Gy followed by 8 fractions of 2 Gy is replaced by 28 fractions of 1.81 Gy to the breast and 2.30 Gy to the boost volume using a SIB technique. Both schedules result in the same biologically effective dose using an a/b of 10. Moreover, only one treatment plan has to be made (the SIB plan) instead of a breast treatment plan and a boost treatment plan. Recently it has been shown that even for patients with an initial seroma, the SIB techniques can lead to improved dose conformity if midway during the treatment an adapted treatment plan is made [8]. Generating adapted treatment plans for all patients with seroma visible on the CT scan made for treatment planning would, however, lead to a significant loss of the efficiency gained by the introduction of the SIB technique. At the Catharina Hospital in Eindhoven we therefore introduced an adaptive radiation therapy (ART) strategy involving rescanning and replanning a specific group of patients. We report here on our
184
Clinical SIB-ART
experience with this ART protocol and provide data for further optimization of such adaptive strategies. Materials and methods Patient data From April 2007 till December 2009, 1274 patients who received breast conserving surgery and who were referred for radiotherapy including a boost dose were enrolled into this ART protocol. All patients underwent a planning CT (CT1) as part of the standard planning for radiotherapy with a slice thickness of 3 mm. Patients were scanned in supine orientation using arm and knee supports with a lead wire placed around the breast as an aid in generating the breast contour. Patients who had an initial seroma volume of more than 30 cm3 and for whom the time between the initial planning CT1 and the lumpectomy was less than 30 days were rescanned between treatment fraction 10 and 11 (CT2). The criterion of 30 days is based on several publications in which the seroma decrease as a function of time has been reported [2–5]. Although inter-observer differences in the delineation of the boost volume are known to exist, the boost volume changes have been shown to be significant even in the presence of such inter-observer variations [4,9]. The criterion of 30 cc was based on our initial clinical experience and was chosen to be relatively small in order to detect as much clinically significant changes as possible without having to scan all patients a second time. CT2 was matched to CT1 using rigid registration (translations only) on a user defined region of interest comprising the breast and a part of the surrounding anatomy (ribs). The contours from CT1 were then copied to CT2. The treating radiation oncologist reviewed these contours. If deemed valuable by the radiation oncologist, the contours were adapted where needed based on the CT2 images. The decision to adapt the contours was left to the treating physician. Although this decision was taken considering among others the clarity and extent of the observed volume difference, no strict quantitative limits were defined. A new treatment plan was generated based on these new contours. Fig. 1 shows an overview of all the steps in this adaptive protocol. All CT images were imported in our treatment planning system (Pinnacle 8.0, Philips Radiation Oncology Systems, Milpitas, CA). The lungs were delineated using an automatic contouring tool and visually verified. The heart contours were manually delineated by specially trained radiation therapy technologists. The breast clinical target volume (CTVbreast), tumor bed volume (TBV) and seroma were delineated by the treating radiation oncologist. The TBV corresponded with the visible surgical resection site, and in some patients could include part of the seroma. Whether or not to include part of the seroma depended on the pre- and post-operative images available, the resection margins and surgical report. Occasionally, the seroma was clearly not at the original tumor side but e.g., along the surgical incision. Surgical clips were used as an aid, although it was not mandatory to always include all clips. The breast planning target volume (PTVbreast) was generated by expanding the CTVbreast by 5 mm uniformly. In order to use this PTV for inverse plan optimization the first 7 mm from the skin inwards was excluded. All delineations were approved by a radiation oncologist. To obtain the boost clinical target volume (CTVboost), the delineated TBV was expanded by 10 mm, subsequently ribs and muscles were excluded and restricting it to lie within the breast CTV. Finally, a PTVboost suitable for inverse optimization was obtained by an additional 5 mm expansion followed by a correction to exclude air and external with a 7 mm margin.
SIB technique The SIB technique has been described previously [6]. In short, this mono-isocentric technique consists of two tangential IMRT
beams to irradiate the breast and three additional beams to irradiate the boost volume. Two of these added beams are tangential, with gantry angles optimized according to the position of the TBV. The third beam is a beam with a gantry angle approximately inbetween the two tangential boost field angles. The fractionation scheme was 28 fractions of 1.81 Gy for the breast and an additional 0.49 Gy to the boost volume. This schedule was considered biologically equivalent to 50 Gy in 25 fractions to the breast combined with a sequential boost of 16 Gy in 8 fractions [11].
Plan optimization and evaluation No or minimal radiation of the contralateral breast was ensured by visual inspection during setup of the tangential beams. Initial optimization of the treatment plans was performed on the basis of the inverse planning objectives. Afterward, each plan was further optimized if possible. A table with the planning objectives and a figure with the volumes defined for inverse plan optimization and evaluation can be found in the Supplementary material on the web. DVHs were calculated for all delineated volumes. To evaluate breast target coverage, an additional volume was created by excluding the PTVboost from the PTVbreast. This volume was named PTVeval. Target coverage was considered clinically acceptable if at least 95% of the PTVboost received 95% of the prescribed total dose and at least 97% of the PTVeval received 95% of the prescribed breast dose. Results All generated treatment plans adhered to the target volume coverage requirements. Nine percent (n = 113) of the 1274 patients enrolled into the adaptive protocol met the criteria for acquisition of a second CT and were rescanned between treatment fraction 10 and 11. Of this group, 77% (n = 87) of treatment plans were adapted and 23% (n = 26) continued treatment based on the initial treatment plan. Adapted plans were implemented from fraction 16 onwards. No statistical differences were observed between the adapted group and non-adapted group concerning timing of CT1 and CT2 (Table 1). Furthermore, PTVbreast size was not significantly different between the two groups. In the adapted plans, seroma reduced significantly from 60 to 27 cm3, TBV from 70 to 45 cm3 and PTVboost from 277 to 220 cm3 (Table 2). The volume receiving more than 95% of the boost dose (V95%(total-dose)) could be reduced from on average 360 to 292 cm3. This is a reduction of 68 cm3. A linear correlation was found between the initial and adapted plan V95%(total-dose) (R2 = 0.73, Fig. 2). V95%(total-dose) was on average reduced by 19%. Heart and lung mean doses were not statistically significantly different between initial plans that were adapted and initial plans that were rescanned but not adapted. Between the adapted and initial plan of the same patients no statistical differences were found concerning the mean dose to the heart. The difference in the mean dose to the lungs was very small (4.8 Gy versus 4.7 Gy) but statistically significant (p = 0.02). In Fig. 3, the reduction of V95%(total-dose) is presented as function of initial TBV and seroma volumes for all patients that were rescanned. Three categories are defined: (1) patients with no plan adaptation, (2) patients with adapted plans resulting in less than 50 cm3 reduction of V95%(total-dose) and (3) patients with adapted plans resulting in more than 50 cm3 reduction of V95%(total-dose). In our retrospective analysis of the data, the volume of 50 cm3 was considered to be a clinically significant reduction [11].
C.W. Hurkmans et al. / Radiotherapy and Oncology 103 (2012) 183–187
185
Patient receiving breast conserving surgery
No: No SIB treatment
Eligible for SIB? Yes
Adaptive protocol
Time between surgery and CT1 (planning CT) < 30 days and seroma 3 volume > 30 cm ?
No: Standard SIB treatment
Yes Second CT between fraction 10 and 11 Match CT2 with CT1 and copy contours Significant change in TBV?
No: Continue SIB treatment without adaptation
Yes Adapt contours and plan and implement at fraction 16
Fig. 1. Adaptive protocol scheme.
Table 1 Characteristics of the rescanned patient group including volume data from CT1. Twosided Student’s T-tests assuming equal variance were used to determine p-values.
TBV (cm3) Seroma (cm3) PTV breast (cm3) Lungs mean dose (Gy) Heart mean dose (Gy)a Days between surgery and CT1 Days between CT1 and CT2 a
Adapted: n = 87 (range)
Non-adapted: n = 26 (range)
pvalue
70 (7–241) 60 (30–241) 923 (243–2057) 4.8 Gy (2.8–8.6) 4.5 Gy (1.4–7.5) 20 (11–29)
51 (5–104) 80 (30–471) 853 (78–1483) 4.7 (2.0–6.6) 4.2 (1.7–6.7) 20 (10–30)
0.02 0.08 0.43 0.56 0.56 0.89
21 (15–36)
22 (15–48)
0.43
Left sided breast cancer patients only (n = 49).
Table 2 Comparison of initial and adapted plan characteristics. Two-sided pair-wise Student’s T-tests were used to determine p-values.
TBV (cm3) Seroma (cm3) PTVboost (cm3) V95%(total-dose) (cm3) Lungs mean dose (Gy) Heart mean dose (Gy)a a
Initial plan
Adapted plan
p-value
70 (7–241) 60 (30–241) 277 (56–622) 360 (93–916) 4.8 Gy (2.8–8.6) 4.5 Gy (1.4–7.5)
45 (9–184) 27 (0–179) 220 (61–564) 292 (95–687) 4.7 (1.8–6.6) 4.3 (1.7–7.0)
<0.001 <0.001 <0.001 <0.001 0.02 0.18
Left sided breast cancer patients only (n = 49).
To see if further optimization of our adaptive protocol is possible, the data shown in Fig. 3 were further analyzed.
There are four patients replanned with an initial TBV below 35 cm3. All four had less than 50 cm3 reduction of their V95%(total-dose). For the two patients with initial TBVs of 14.3 and 24.5 cm3 and respective seroma volumes of 56.6 and 62.8 cm3 the TBVs (including their positions) derived from CT2 were identical. Thus, target coverage would not have been compromised if no replanning would have taken place. For the patient with an initial TBV of only 6.5 cm3 the seroma volume decreased from 112.1 to 16.3 cm3, causing a shift of the TBV (Figure given in web supplement). For the patient with an initial TBV of 29.7 cm3 the seroma volume decreased from 102.5 to 60.0 cm3, also causing a shift of the TBV. Of the latter two patients, for which the seroma was in the vicinity of the TBV, TBV coverage would have been compromised if their plans would not have been adapted. Three patients with an initial TBV below 35 cm3 but a seroma volume above 100 cm3 (145.5 cm3, 163.7 and 471.1 cm3) were not replanned. It appeared for two of these patients the seroma was located in the axilla and seroma shrinkage did not influence the TBV or position as they were located more caudally in the breast. For one patient, the seroma volume was located in the breast but, although the volume reduced from 471.1 to 295.0 cm3 the TBV did not change in volume or position (datapoint not shown in Fig. 3). Based on Fig. 3 and the data given in the previous paragraph, a subgroup might be defined for which rescanning does not lead to a clinically relevant plan adaptation. This is the group of patients with an initial TBV below 35 cm3 and seroma volume below 100 cm3 or a seroma located in the axilla. If these patients would be excluded in our adaptive radiotherapy protocol, only 7.6% (n = 97) of our patients would have been rescanned. Out of this group, 80% (n = 78) would have been replanned. The V95%(total-dose)
186
Clinical SIB-ART
3
V95%(total-dose) adapted plan (cm )
1000 900 800 700
V95%(total-dose) adapted plan = 0.81 x V95%(total-dose) initial plan
600
R2 = 0.73
500 400 300 200 100 0 0
100
200
300
400
500
600
700
800
900
1000
3
V95%(total-dose) initial plan (cm ) Fig. 2. V95%(total-dose) in the adapted plan as a function of V95%(total-dose) in the initial plan.
250
<50 cm3 reduction no plan adaptation >50 cm3 reduction
Initial seroma (cm3 )
200
150
100
50
0 0
50
100
150
200
250
3
Initial Tumour Bed Volume (cm )
Fig. 3. Reduction of V95%(total-dose) as a function of initial TBV and seroma volume for all rescanned patients. Red squares represent non-adapted plans. Black squares represent adapted plans with less than 50 cm3 reduction of V95%(total-dose). Green squares represent adapted plans with more than 50 cm3 reduction of V95%(total-dose).
would have been reduced by 74 cm3 from on average 376 to on average 302 cm3. In contrast to what may be expected, no strong correlations were found between V95%(total-dose) reduction and initial seroma volume (R2 = 0.10) or between V95%(total-dose) reduction and seroma volume decrease (R2 = 0.37). In five patients the seroma volume decreased slightly. For one additional patient, the seroma volume increased with 76.0 cm3 from 101.4 to 179.4 cm3. No clinical features were found that could predict this increase. Also, total breast volume decrease did not predict for seroma volume decrease (R2 = 0.07, data not shown).
Discussion The aim of this study was to review our experience with an adaptive radiation therapy protocol. The efficiency in terms of
percentage of patients that needed rescanning and the efficacy in terms of reduction of the high dose volume (V95%(total-dose)) was quantified. In addition, further improvement of our adaptive radiotherapy protocol was shown to be possible. This would result in an effective protocol: on average a 74 cm3 reduction of V95%(total-dose) could be reached. It would also be efficient: 80% of the patients which would be rescanned would be replanned. One might see the 20% of patients that would not be replanned as false positive results of our selection criteria. Inherently, little data are available about possible false negative results, i.e., patients that would have shown a clinically significant decrease in V95%(total-dose) but are not included in our new criteria. However, the somewhat broader criteria used for our current study suggest that, e.g., including patients with smaller initial seroma will not detect more patients which benefit from replanning. Alderliesten et al. recently already showed that the adaptive SIB technique is superior to sequential treatment planning based on data from 21 patients with breast seroma [8]. They concluded that patients with a larger initial seroma or a larger seroma reduction benefit mostly from re-planning and derived two thresholds: patients with an initial seroma volume P40 cm3 should be monitored and patients with seroma reduction of P20 cm3 in the first 4 weeks after acquisition of the initial planning CT should be replanned. Based on our data from 113 patients, we suggested to rescan (monitor) patients with an initial breast seroma volume P30 cm3 which is completely or partly part of the TBV and an initial TBV of P35 cm3. The specific distinction between breast seroma which is completely or partly part of the TBV and seroma is of importance as often axilla seroma was detected which did not influence TBV or position. The specific distinction between breast seroma and TBV is of importance as the breast seroma is not always part of the TBV. As three patients with a breast seroma volume between 35 and 40 cm3 in our cohort had a clinically relevant reduction of V95%(total-dose) (reductions of 55, 86 and 100 cm3), we suggested the lower limit of 35 cm3 instead of 40 cm3 of initial breast seroma volume. Patients with a breast seroma volume P100 cm3 in the vicinity of the TBV should also be rescanned, as seroma volume changes might in those cases lead to a displacement of the TBV. Our recommendation is based on the same question as brought forward by Alderliesten et al. [8]: ‘‘what specific dosimetric gain is clinically relevant enough to spend the additional time associated with re-planning?’’ In a previous study by Borger et al., for each 100 cm3 increase in irradiated boost volume a fourfold increase in risk of fibrosis was
C.W. Hurkmans et al. / Radiotherapy and Oncology 103 (2012) 183–187
observed [10]. Therefore, a reduction of 50 cm3 in volume of excess dose during RT can be considered clinically significant and should be taken into consideration. Seroma reduction was not found to correlate strongly with high dose volume reduction as suggested by Alderliesten et al. [16]. However, the initial V95%(total-dose) did correlate with V95%(total-dose) reduction. Although this might be used to derive a more objective criterion for re-planning, the correlation is not very strong. We believe the decision to re-plan treatment of an individual patient should be taken considering all the relevant patient information. Our analysis was based on data derived from CT1 and CT2 without recalculation of all treatment plans based on CT1 on the CT2 datasets. As no relevant systematic changes in breast shape were seen, we hypothesize that recalculation would not alter our results. We expect morbidity to correlate stronger with the total V95%(total-dose) rather than the undesired V95%(total-dose), i.e., outside of the TBVPTV (Vexcess-dose). Thus no attempt was made to include Vexcess-dose in our analyzes. A recent update of the Canadian phase III trial comparing standard fractionation to hypofractionation (16 2.66 Gy) for breast cancer, together with new ASTRO guidelines on hypofractionation will lead to a broader use of hypofractionated schedules [11,12]. The ASTRO task force agreed that the use of hypofractionated whole breast radiotherapy alone (without a boost) is not appropriate when a TBV boost is thought to be indicated and the majority of the taskforce thought that there were sufficient data showing the safety of HF-WBI followed by a TBV boost to recommend its use. In the Netherlands, consensus has been reached that hypofractionation including a boost can be safely given, either sequentially or as a SIB. The SIB schedule is 21 2.17 Gy to the PTVbreast and an additional 0.49 Gy to the PTVboost, calculated based on an a/b-ratio of 4.6 Gy for tumor control [13]. For an adaptive hypofractionated SIB schedule to be approximately as effective as the adaptive schedule investigated here, the second CT scan could be acquired between fraction 8 and 9 and clinically implemented from fraction 11 onwards. Yang and co-workers have recently shown that seroma volume changes, for the patients for which such changes are expected, might also be monitored using CBCT [14]. The discrepancies between CBCT and CT were shown to be minor, with low interobserver variation, based on data from 10 patients. However, not all institutions possess CBCT at present. Moreover, as no strong correlation exists between seroma volume decrease and reduction of V95%(total-dose), CBCT monitoring of all patients with seroma without the additional criteria for scanning and replanning as presented in this article would substantially increase the workload. Conclusions We reported on our 3 year experience with an Adaptive Radiation Therapy strategy involving rescanning and replanning patients receiving breast radiotherapy using a Simultaneously Integrated Boost. The strategy has proven to be efficient with, based on the selection criteria used within this study, only 9% of patients needing a second CT scan. It was also effective, with 77% of these patients
187
being replanned, leading to a clinically significant reduction of the high dose volume by on average 19% (68 cm3). Conflict of interest The authors of this manuscript have no conflicts of interest to disclose. Acknowledgments The authors thank everyone within the Department of Radiation Oncology of the Catharina Hospital who contributed to the treatment of the patients of which the data are presented in this manuscript. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.radonc.2011.12.014. References [1] Penninkhof J, Quint S, Baaijens M, et al. Practical use of the extended no action level (eNAL) correction protocol for breast cancer patients with implanted surgical clips. Int J Radiat Oncol Biol Phys 2012;82:1031–7. [2] Yang TJ, Elkhuizen PH, Minkema D, et al. Clinical factors associated with seroma volume reduction in breast-Conserving Therapy for early-stage breast cancer: a multi-institutional analysis. Int J Radiat Oncol Biol Phys 2010;76:1325–32. [3] Jacobson G, Betts V, Smith B. Change in volume of lumpectomy cavity during external-beam irradiation of the intact breast. Int J Radiat Oncol Biol Phys 2006;65:1161–4. [4] Hurkmans C, Admiraal M, van der SM, et al. Significance of breast boost volume changes during radiotherapy in relation to current clinical interobserver variations. Radiother Oncol 2009;90:60–5. [5] Oh KS, Kong FM, Griffith KA, et al. Planning the breast tumor bed boost: changes in the excision cavity volume and surgical scar location after breastconserving surgery and whole-breast irradiation. Int J Radiat Oncol Biol Phys 2006;66:680–6. [6] Hurkmans CW, Meijer GJ, van Vliet-Vroegindeweij C, et al. High-dose simultaneously integrated breast boost using intensity-modulated radiotherapy and inverse optimization. Int J Radiat Oncol Biol Phys 2006;66:923–30. [7] van der Laan HP, Dolsma WV, Maduro JH, et al. Three-dimensional conformal simultaneously integrated boost technique for breast-conserving radiotherapy. Int J Radiat Oncol Biol Phys 2007;68:1018–23. [8] Alderliesten T, den HS, Yang TI, et al. Dosimetric impact of post-operative seroma reduction during radiotherapy after breast-conserving surgery. Radiother Oncol 2011;100:265–70. [9] van Mourik AM, Elkhuizen PH, Minkema D, et al. Multiinstitutional study on target volume delineation variation in breast radiotherapy in the presence of guidelines. Radiother Oncol 2010;94:286–91. [10] Borger JH, Kemperman H, Smitt HS, et al. Dose and volume effects on fibrosis after breast conservation therapy. Int J Radiat Oncol Biol Phys 1994;30:1073–81. [11] Whelan TJ, Pignol JP, Levine MN, et al. Long-term results of hypofractionated radiation therapy for breast cancer. N Engl J Med 2010;362:513–20. [12] Smith BD, Bentzen SM, Correa CR, et al. Fractionation for whole breast irradiation: An American Society for Radiation Oncology (ASTRO) evidencebased guideline. Int J Radiat Oncol Biol Phys 2011;81:59–68. [13] Bentzen SM, Agrawal RK, Aird EG, et al. The UK Standardisation of Breast Radiotherapy (START) Trial A of radiotherapy hypofractionation for treatment of early breast cancer: a randomised trial. Lancet Oncol 2008;9:331–41. [14] Yang TJ, Minkema D, Elkhuizen PH, et al. Clinical applicability of cone-beam computed tomography in monitoring seroma volume change during breast irradiation. Int J Radiat Oncol Biol Phys 2010;78:119–26.
Contents lists available at SciVerse ScienceDirect
Radiotherapy and Oncology journal homepage: www.thegreenjournal.com
Adaptive radiotherapy for breast cancer
Adaptive radiation therapy for breast IMRT-simultaneously integrated boost: Three-year clinical experience Coen W. Hurkmans a,⇑, Ingrid Dijckmans a, Miranda Reijnen a, Jorien van der Leer a, Corine van Vliet-Vroegindeweij b, Maurice van der Sangen a a
Department of Radiation Oncology, Catharina Hospital, Eindhoven; and b Department of Radiation Oncology, The Netherlands Cancer Institute – Antoni van Leeuwenhoek Hospital, Amsterdam, The Netherlands
a r t i c l e
i n f o
Article history: Received 29 March 2011 Received in revised form 15 December 2011 Accepted 19 December 2011 Available online 24 January 2012 Keywords: Breast cancer Post-operative seroma volume change Adaptive image guided radiation therapy Simultaneously integrated boost Tumor bed volume
a b s t r a c t Purpose: It has been shown that seroma volumes decrease during breast conserving radiotherapy in a significant percentage of patients. We report on our experience with an adaptive radiation therapy (ART) strategy involving rescanning and replanning patients to take this reduction into account during a course of intensity-modulated radiation therapy with simultaneously integrated boost (IMRT-SIB). Materials: From April 2007 till December 2009, 1274 patients eligible for SIB treatment were enrolled into this protocol. Patients for which the time between the initial planning CT (CT1) and lumpectomy was less than 30 days and who had an initial seroma volume >30 cm3 were rescanned at day 10 of treatment (CT2) and replanned when significant changes were observed by the radiation oncologist. Patients received 28 fractions of 1.81 Gy to the breast and 2.30 Gy to the boost volume. Results: Nine percent (n = 113) of the 1274 patients enrolled met the criteria and were rescanned. Of this group, 77% (n = 87) of treatment plans were adapted. Time between surgery and CT1 (20 days versus 20 days for adapted and non-adapted plans, p = 0.89) and time between CT1 and CT2 (21 days versus 22 days for adapted and non-adapted plans, p = 0.43) revealed no procedural differences which might have biased our results. In the adapted plans, seroma decreased significantly from 60 to 27 cm3 (p < 0.001), TBV from 70 to 45 cm3 (p < 0.001) and PTVboost from 277 to 220 cm3 (p < 0.001). The volume receiving more than 95% of the boost dose (V95%(total-dose)) could be reduced by 19% (linear fit, R2 = 0.73) from on average 360 to 292 cm3 (p < 0.001). Delay in treatment and the use of a prolonged treatment schedule with different fractionation for patients with seroma could thus be prevented. Conclusion: The adaptive radiation therapy IMRT-SIB procedure has proven to be efficient and effective, leading to a clinically significant reduction of the high dose volume. Seroma present in a subgroup of patients referred for breast radiation therapy does not hamper the introduction of highly conformal IMRT-SIB techniques. Ó 2012 Elsevier Ireland Ltd. All rights reserved. Radiotherapy and Oncology 103 (2012) 183–187
The breast tumor bed after breast conserving surgery generally does not change significantly during radiation therapy when no seroma is present [1]. However, post-operative seroma is sometimes present and seroma reductions are seen between the acquisition of the planning CT scan and end of radiation therapy in a substantial percentage of the patients [2–5]. These reductions can be substantial and have hampered the introduction of simultaneously integrated boost (SIB) techniques for breast cancer patients, as it was not clear how much these variations would effect SIB plans. SIB techniques have been shown to provide more conformal treatment plans than conventional sequential boost planning in the absence of volume changes [6,7]. Furthermore, SIB enables a reduction in the overall treatment time by 1 week ⇑ Corresponding author. Address: Department of Radiation Oncology, Catharina Hospital, Michelangelolaan 1, Eindhoven, The Netherlands. E-mail address: [email protected] (C.W. Hurkmans). 0167-8140/$ - see front matter Ó 2012 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.radonc.2011.12.014
when the conventional fractionation scheme of 25 fractions of 2 Gy followed by 8 fractions of 2 Gy is replaced by 28 fractions of 1.81 Gy to the breast and 2.30 Gy to the boost volume using a SIB technique. Both schedules result in the same biologically effective dose using an a/b of 10. Moreover, only one treatment plan has to be made (the SIB plan) instead of a breast treatment plan and a boost treatment plan. Recently it has been shown that even for patients with an initial seroma, the SIB techniques can lead to improved dose conformity if midway during the treatment an adapted treatment plan is made [8]. Generating adapted treatment plans for all patients with seroma visible on the CT scan made for treatment planning would, however, lead to a significant loss of the efficiency gained by the introduction of the SIB technique. At the Catharina Hospital in Eindhoven we therefore introduced an adaptive radiation therapy (ART) strategy involving rescanning and replanning a specific group of patients. We report here on our
184
Clinical SIB-ART
experience with this ART protocol and provide data for further optimization of such adaptive strategies. Materials and methods Patient data From April 2007 till December 2009, 1274 patients who received breast conserving surgery and who were referred for radiotherapy including a boost dose were enrolled into this ART protocol. All patients underwent a planning CT (CT1) as part of the standard planning for radiotherapy with a slice thickness of 3 mm. Patients were scanned in supine orientation using arm and knee supports with a lead wire placed around the breast as an aid in generating the breast contour. Patients who had an initial seroma volume of more than 30 cm3 and for whom the time between the initial planning CT1 and the lumpectomy was less than 30 days were rescanned between treatment fraction 10 and 11 (CT2). The criterion of 30 days is based on several publications in which the seroma decrease as a function of time has been reported [2–5]. Although inter-observer differences in the delineation of the boost volume are known to exist, the boost volume changes have been shown to be significant even in the presence of such inter-observer variations [4,9]. The criterion of 30 cc was based on our initial clinical experience and was chosen to be relatively small in order to detect as much clinically significant changes as possible without having to scan all patients a second time. CT2 was matched to CT1 using rigid registration (translations only) on a user defined region of interest comprising the breast and a part of the surrounding anatomy (ribs). The contours from CT1 were then copied to CT2. The treating radiation oncologist reviewed these contours. If deemed valuable by the radiation oncologist, the contours were adapted where needed based on the CT2 images. The decision to adapt the contours was left to the treating physician. Although this decision was taken considering among others the clarity and extent of the observed volume difference, no strict quantitative limits were defined. A new treatment plan was generated based on these new contours. Fig. 1 shows an overview of all the steps in this adaptive protocol. All CT images were imported in our treatment planning system (Pinnacle 8.0, Philips Radiation Oncology Systems, Milpitas, CA). The lungs were delineated using an automatic contouring tool and visually verified. The heart contours were manually delineated by specially trained radiation therapy technologists. The breast clinical target volume (CTVbreast), tumor bed volume (TBV) and seroma were delineated by the treating radiation oncologist. The TBV corresponded with the visible surgical resection site, and in some patients could include part of the seroma. Whether or not to include part of the seroma depended on the pre- and post-operative images available, the resection margins and surgical report. Occasionally, the seroma was clearly not at the original tumor side but e.g., along the surgical incision. Surgical clips were used as an aid, although it was not mandatory to always include all clips. The breast planning target volume (PTVbreast) was generated by expanding the CTVbreast by 5 mm uniformly. In order to use this PTV for inverse plan optimization the first 7 mm from the skin inwards was excluded. All delineations were approved by a radiation oncologist. To obtain the boost clinical target volume (CTVboost), the delineated TBV was expanded by 10 mm, subsequently ribs and muscles were excluded and restricting it to lie within the breast CTV. Finally, a PTVboost suitable for inverse optimization was obtained by an additional 5 mm expansion followed by a correction to exclude air and external with a 7 mm margin.
SIB technique The SIB technique has been described previously [6]. In short, this mono-isocentric technique consists of two tangential IMRT
beams to irradiate the breast and three additional beams to irradiate the boost volume. Two of these added beams are tangential, with gantry angles optimized according to the position of the TBV. The third beam is a beam with a gantry angle approximately inbetween the two tangential boost field angles. The fractionation scheme was 28 fractions of 1.81 Gy for the breast and an additional 0.49 Gy to the boost volume. This schedule was considered biologically equivalent to 50 Gy in 25 fractions to the breast combined with a sequential boost of 16 Gy in 8 fractions [11].
Plan optimization and evaluation No or minimal radiation of the contralateral breast was ensured by visual inspection during setup of the tangential beams. Initial optimization of the treatment plans was performed on the basis of the inverse planning objectives. Afterward, each plan was further optimized if possible. A table with the planning objectives and a figure with the volumes defined for inverse plan optimization and evaluation can be found in the Supplementary material on the web. DVHs were calculated for all delineated volumes. To evaluate breast target coverage, an additional volume was created by excluding the PTVboost from the PTVbreast. This volume was named PTVeval. Target coverage was considered clinically acceptable if at least 95% of the PTVboost received 95% of the prescribed total dose and at least 97% of the PTVeval received 95% of the prescribed breast dose. Results All generated treatment plans adhered to the target volume coverage requirements. Nine percent (n = 113) of the 1274 patients enrolled into the adaptive protocol met the criteria for acquisition of a second CT and were rescanned between treatment fraction 10 and 11. Of this group, 77% (n = 87) of treatment plans were adapted and 23% (n = 26) continued treatment based on the initial treatment plan. Adapted plans were implemented from fraction 16 onwards. No statistical differences were observed between the adapted group and non-adapted group concerning timing of CT1 and CT2 (Table 1). Furthermore, PTVbreast size was not significantly different between the two groups. In the adapted plans, seroma reduced significantly from 60 to 27 cm3, TBV from 70 to 45 cm3 and PTVboost from 277 to 220 cm3 (Table 2). The volume receiving more than 95% of the boost dose (V95%(total-dose)) could be reduced from on average 360 to 292 cm3. This is a reduction of 68 cm3. A linear correlation was found between the initial and adapted plan V95%(total-dose) (R2 = 0.73, Fig. 2). V95%(total-dose) was on average reduced by 19%. Heart and lung mean doses were not statistically significantly different between initial plans that were adapted and initial plans that were rescanned but not adapted. Between the adapted and initial plan of the same patients no statistical differences were found concerning the mean dose to the heart. The difference in the mean dose to the lungs was very small (4.8 Gy versus 4.7 Gy) but statistically significant (p = 0.02). In Fig. 3, the reduction of V95%(total-dose) is presented as function of initial TBV and seroma volumes for all patients that were rescanned. Three categories are defined: (1) patients with no plan adaptation, (2) patients with adapted plans resulting in less than 50 cm3 reduction of V95%(total-dose) and (3) patients with adapted plans resulting in more than 50 cm3 reduction of V95%(total-dose). In our retrospective analysis of the data, the volume of 50 cm3 was considered to be a clinically significant reduction [11].
C.W. Hurkmans et al. / Radiotherapy and Oncology 103 (2012) 183–187
185
Patient receiving breast conserving surgery
No: No SIB treatment
Eligible for SIB? Yes
Adaptive protocol
Time between surgery and CT1 (planning CT) < 30 days and seroma 3 volume > 30 cm ?
No: Standard SIB treatment
Yes Second CT between fraction 10 and 11 Match CT2 with CT1 and copy contours Significant change in TBV?
No: Continue SIB treatment without adaptation
Yes Adapt contours and plan and implement at fraction 16
Fig. 1. Adaptive protocol scheme.
Table 1 Characteristics of the rescanned patient group including volume data from CT1. Twosided Student’s T-tests assuming equal variance were used to determine p-values.
TBV (cm3) Seroma (cm3) PTV breast (cm3) Lungs mean dose (Gy) Heart mean dose (Gy)a Days between surgery and CT1 Days between CT1 and CT2 a
Adapted: n = 87 (range)
Non-adapted: n = 26 (range)
pvalue
70 (7–241) 60 (30–241) 923 (243–2057) 4.8 Gy (2.8–8.6) 4.5 Gy (1.4–7.5) 20 (11–29)
51 (5–104) 80 (30–471) 853 (78–1483) 4.7 (2.0–6.6) 4.2 (1.7–6.7) 20 (10–30)
0.02 0.08 0.43 0.56 0.56 0.89
21 (15–36)
22 (15–48)
0.43
Left sided breast cancer patients only (n = 49).
Table 2 Comparison of initial and adapted plan characteristics. Two-sided pair-wise Student’s T-tests were used to determine p-values.
TBV (cm3) Seroma (cm3) PTVboost (cm3) V95%(total-dose) (cm3) Lungs mean dose (Gy) Heart mean dose (Gy)a a
Initial plan
Adapted plan
p-value
70 (7–241) 60 (30–241) 277 (56–622) 360 (93–916) 4.8 Gy (2.8–8.6) 4.5 Gy (1.4–7.5)
45 (9–184) 27 (0–179) 220 (61–564) 292 (95–687) 4.7 (1.8–6.6) 4.3 (1.7–7.0)
<0.001 <0.001 <0.001 <0.001 0.02 0.18
Left sided breast cancer patients only (n = 49).
To see if further optimization of our adaptive protocol is possible, the data shown in Fig. 3 were further analyzed.
There are four patients replanned with an initial TBV below 35 cm3. All four had less than 50 cm3 reduction of their V95%(total-dose). For the two patients with initial TBVs of 14.3 and 24.5 cm3 and respective seroma volumes of 56.6 and 62.8 cm3 the TBVs (including their positions) derived from CT2 were identical. Thus, target coverage would not have been compromised if no replanning would have taken place. For the patient with an initial TBV of only 6.5 cm3 the seroma volume decreased from 112.1 to 16.3 cm3, causing a shift of the TBV (Figure given in web supplement). For the patient with an initial TBV of 29.7 cm3 the seroma volume decreased from 102.5 to 60.0 cm3, also causing a shift of the TBV. Of the latter two patients, for which the seroma was in the vicinity of the TBV, TBV coverage would have been compromised if their plans would not have been adapted. Three patients with an initial TBV below 35 cm3 but a seroma volume above 100 cm3 (145.5 cm3, 163.7 and 471.1 cm3) were not replanned. It appeared for two of these patients the seroma was located in the axilla and seroma shrinkage did not influence the TBV or position as they were located more caudally in the breast. For one patient, the seroma volume was located in the breast but, although the volume reduced from 471.1 to 295.0 cm3 the TBV did not change in volume or position (datapoint not shown in Fig. 3). Based on Fig. 3 and the data given in the previous paragraph, a subgroup might be defined for which rescanning does not lead to a clinically relevant plan adaptation. This is the group of patients with an initial TBV below 35 cm3 and seroma volume below 100 cm3 or a seroma located in the axilla. If these patients would be excluded in our adaptive radiotherapy protocol, only 7.6% (n = 97) of our patients would have been rescanned. Out of this group, 80% (n = 78) would have been replanned. The V95%(total-dose)
186
Clinical SIB-ART
3
V95%(total-dose) adapted plan (cm )
1000 900 800 700
V95%(total-dose) adapted plan = 0.81 x V95%(total-dose) initial plan
600
R2 = 0.73
500 400 300 200 100 0 0
100
200
300
400
500
600
700
800
900
1000
3
V95%(total-dose) initial plan (cm ) Fig. 2. V95%(total-dose) in the adapted plan as a function of V95%(total-dose) in the initial plan.
250
<50 cm3 reduction no plan adaptation >50 cm3 reduction
Initial seroma (cm3 )
200
150
100
50
0 0
50
100
150
200
250
3
Initial Tumour Bed Volume (cm )
Fig. 3. Reduction of V95%(total-dose) as a function of initial TBV and seroma volume for all rescanned patients. Red squares represent non-adapted plans. Black squares represent adapted plans with less than 50 cm3 reduction of V95%(total-dose). Green squares represent adapted plans with more than 50 cm3 reduction of V95%(total-dose).
would have been reduced by 74 cm3 from on average 376 to on average 302 cm3. In contrast to what may be expected, no strong correlations were found between V95%(total-dose) reduction and initial seroma volume (R2 = 0.10) or between V95%(total-dose) reduction and seroma volume decrease (R2 = 0.37). In five patients the seroma volume decreased slightly. For one additional patient, the seroma volume increased with 76.0 cm3 from 101.4 to 179.4 cm3. No clinical features were found that could predict this increase. Also, total breast volume decrease did not predict for seroma volume decrease (R2 = 0.07, data not shown).
Discussion The aim of this study was to review our experience with an adaptive radiation therapy protocol. The efficiency in terms of
percentage of patients that needed rescanning and the efficacy in terms of reduction of the high dose volume (V95%(total-dose)) was quantified. In addition, further improvement of our adaptive radiotherapy protocol was shown to be possible. This would result in an effective protocol: on average a 74 cm3 reduction of V95%(total-dose) could be reached. It would also be efficient: 80% of the patients which would be rescanned would be replanned. One might see the 20% of patients that would not be replanned as false positive results of our selection criteria. Inherently, little data are available about possible false negative results, i.e., patients that would have shown a clinically significant decrease in V95%(total-dose) but are not included in our new criteria. However, the somewhat broader criteria used for our current study suggest that, e.g., including patients with smaller initial seroma will not detect more patients which benefit from replanning. Alderliesten et al. recently already showed that the adaptive SIB technique is superior to sequential treatment planning based on data from 21 patients with breast seroma [8]. They concluded that patients with a larger initial seroma or a larger seroma reduction benefit mostly from re-planning and derived two thresholds: patients with an initial seroma volume P40 cm3 should be monitored and patients with seroma reduction of P20 cm3 in the first 4 weeks after acquisition of the initial planning CT should be replanned. Based on our data from 113 patients, we suggested to rescan (monitor) patients with an initial breast seroma volume P30 cm3 which is completely or partly part of the TBV and an initial TBV of P35 cm3. The specific distinction between breast seroma which is completely or partly part of the TBV and seroma is of importance as often axilla seroma was detected which did not influence TBV or position. The specific distinction between breast seroma and TBV is of importance as the breast seroma is not always part of the TBV. As three patients with a breast seroma volume between 35 and 40 cm3 in our cohort had a clinically relevant reduction of V95%(total-dose) (reductions of 55, 86 and 100 cm3), we suggested the lower limit of 35 cm3 instead of 40 cm3 of initial breast seroma volume. Patients with a breast seroma volume P100 cm3 in the vicinity of the TBV should also be rescanned, as seroma volume changes might in those cases lead to a displacement of the TBV. Our recommendation is based on the same question as brought forward by Alderliesten et al. [8]: ‘‘what specific dosimetric gain is clinically relevant enough to spend the additional time associated with re-planning?’’ In a previous study by Borger et al., for each 100 cm3 increase in irradiated boost volume a fourfold increase in risk of fibrosis was
C.W. Hurkmans et al. / Radiotherapy and Oncology 103 (2012) 183–187
observed [10]. Therefore, a reduction of 50 cm3 in volume of excess dose during RT can be considered clinically significant and should be taken into consideration. Seroma reduction was not found to correlate strongly with high dose volume reduction as suggested by Alderliesten et al. [16]. However, the initial V95%(total-dose) did correlate with V95%(total-dose) reduction. Although this might be used to derive a more objective criterion for re-planning, the correlation is not very strong. We believe the decision to re-plan treatment of an individual patient should be taken considering all the relevant patient information. Our analysis was based on data derived from CT1 and CT2 without recalculation of all treatment plans based on CT1 on the CT2 datasets. As no relevant systematic changes in breast shape were seen, we hypothesize that recalculation would not alter our results. We expect morbidity to correlate stronger with the total V95%(total-dose) rather than the undesired V95%(total-dose), i.e., outside of the TBVPTV (Vexcess-dose). Thus no attempt was made to include Vexcess-dose in our analyzes. A recent update of the Canadian phase III trial comparing standard fractionation to hypofractionation (16 2.66 Gy) for breast cancer, together with new ASTRO guidelines on hypofractionation will lead to a broader use of hypofractionated schedules [11,12]. The ASTRO task force agreed that the use of hypofractionated whole breast radiotherapy alone (without a boost) is not appropriate when a TBV boost is thought to be indicated and the majority of the taskforce thought that there were sufficient data showing the safety of HF-WBI followed by a TBV boost to recommend its use. In the Netherlands, consensus has been reached that hypofractionation including a boost can be safely given, either sequentially or as a SIB. The SIB schedule is 21 2.17 Gy to the PTVbreast and an additional 0.49 Gy to the PTVboost, calculated based on an a/b-ratio of 4.6 Gy for tumor control [13]. For an adaptive hypofractionated SIB schedule to be approximately as effective as the adaptive schedule investigated here, the second CT scan could be acquired between fraction 8 and 9 and clinically implemented from fraction 11 onwards. Yang and co-workers have recently shown that seroma volume changes, for the patients for which such changes are expected, might also be monitored using CBCT [14]. The discrepancies between CBCT and CT were shown to be minor, with low interobserver variation, based on data from 10 patients. However, not all institutions possess CBCT at present. Moreover, as no strong correlation exists between seroma volume decrease and reduction of V95%(total-dose), CBCT monitoring of all patients with seroma without the additional criteria for scanning and replanning as presented in this article would substantially increase the workload. Conclusions We reported on our 3 year experience with an Adaptive Radiation Therapy strategy involving rescanning and replanning patients receiving breast radiotherapy using a Simultaneously Integrated Boost. The strategy has proven to be efficient with, based on the selection criteria used within this study, only 9% of patients needing a second CT scan. It was also effective, with 77% of these patients
187
being replanned, leading to a clinically significant reduction of the high dose volume by on average 19% (68 cm3). Conflict of interest The authors of this manuscript have no conflicts of interest to disclose. Acknowledgments The authors thank everyone within the Department of Radiation Oncology of the Catharina Hospital who contributed to the treatment of the patients of which the data are presented in this manuscript. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.radonc.2011.12.014. References [1] Penninkhof J, Quint S, Baaijens M, et al. Practical use of the extended no action level (eNAL) correction protocol for breast cancer patients with implanted surgical clips. Int J Radiat Oncol Biol Phys 2012;82:1031–7. [2] Yang TJ, Elkhuizen PH, Minkema D, et al. Clinical factors associated with seroma volume reduction in breast-Conserving Therapy for early-stage breast cancer: a multi-institutional analysis. Int J Radiat Oncol Biol Phys 2010;76:1325–32. [3] Jacobson G, Betts V, Smith B. Change in volume of lumpectomy cavity during external-beam irradiation of the intact breast. Int J Radiat Oncol Biol Phys 2006;65:1161–4. [4] Hurkmans C, Admiraal M, van der SM, et al. Significance of breast boost volume changes during radiotherapy in relation to current clinical interobserver variations. Radiother Oncol 2009;90:60–5. [5] Oh KS, Kong FM, Griffith KA, et al. Planning the breast tumor bed boost: changes in the excision cavity volume and surgical scar location after breastconserving surgery and whole-breast irradiation. Int J Radiat Oncol Biol Phys 2006;66:680–6. [6] Hurkmans CW, Meijer GJ, van Vliet-Vroegindeweij C, et al. High-dose simultaneously integrated breast boost using intensity-modulated radiotherapy and inverse optimization. Int J Radiat Oncol Biol Phys 2006;66:923–30. [7] van der Laan HP, Dolsma WV, Maduro JH, et al. Three-dimensional conformal simultaneously integrated boost technique for breast-conserving radiotherapy. Int J Radiat Oncol Biol Phys 2007;68:1018–23. [8] Alderliesten T, den HS, Yang TI, et al. Dosimetric impact of post-operative seroma reduction during radiotherapy after breast-conserving surgery. Radiother Oncol 2011;100:265–70. [9] van Mourik AM, Elkhuizen PH, Minkema D, et al. Multiinstitutional study on target volume delineation variation in breast radiotherapy in the presence of guidelines. Radiother Oncol 2010;94:286–91. [10] Borger JH, Kemperman H, Smitt HS, et al. Dose and volume effects on fibrosis after breast conservation therapy. Int J Radiat Oncol Biol Phys 1994;30:1073–81. [11] Whelan TJ, Pignol JP, Levine MN, et al. Long-term results of hypofractionated radiation therapy for breast cancer. N Engl J Med 2010;362:513–20. [12] Smith BD, Bentzen SM, Correa CR, et al. Fractionation for whole breast irradiation: An American Society for Radiation Oncology (ASTRO) evidencebased guideline. Int J Radiat Oncol Biol Phys 2011;81:59–68. [13] Bentzen SM, Agrawal RK, Aird EG, et al. The UK Standardisation of Breast Radiotherapy (START) Trial A of radiotherapy hypofractionation for treatment of early breast cancer: a randomised trial. Lancet Oncol 2008;9:331–41. [14] Yang TJ, Minkema D, Elkhuizen PH, et al. Clinical applicability of cone-beam computed tomography in monitoring seroma volume change during breast irradiation. Int J Radiat Oncol Biol Phys 2010;78:119–26.