Abstract
Aim: To present an update of a prospective study evaluating an accelerated hypofractionated whole breast irradiation (WBI) schedule of the START A trial plus hypofractionated boost in breast cancer patients. Patients and Methods: One hundred and forty consecutive patients ≥55 years were included in this study. Patients received postoperative WBI with 13×3.2 Gy to 41.6 Gy plus a boost of 3.0 Gy/fraction to 9-12 Gy applied in <3.5 weeks, depending on the resection margin. Prospectively planned follow-up (FU) visits, including objective and subjective assessment of treatment tolerance, were performed at 0 and 8 weeks, as well as one, two, four or more years following radiotherapy (RT). Results: The 3-year rates of local control, nodal control, disease-free and overall survival were 99%, 100%, 96% and 91%, respectively. Cosmetic outcome was very good with 99% (n=110/111), 98% (n=99/101) and 100% (n=59/59) of the patients being satisfied or very satisfied one, two and four years after RT, respectively. Conclusion: Acceleration of the START A regime with 41.6 Gy WBI plus additional boost of 9-12 Gy remained effective and well-tolerated.
Compared to conventional fractionation with 50 Gy in 25 fractions or 50.4 Gy in 28 fractions, hypofractionation is gaining importance after long-term results of four large randomized trials that included more than 7,000 patients were published. Equal findings for conventionally fractionated vs. hypofractionated whole breast irradiation (WBI) in terms of both treatment outcome and toxicity were found (1-5). Despite these results, the clinical implication of hypofractionated WBI is still limited (6). While patients in the Canadian trial and START B trial were treated five times a week with 15-16 fractions of 2.66/2.67 Gy up to total doses of 40.05 Gy and 42.5 Gy, patients in the START A trial received total doses of 39 Gy or 41.6 Gy in 13 fractions on three days per week (doses per fraction=3.0 or 3.2 Gy) given over 5 weeks, in order to keep the overall treatment time comparable to the standard arm of 50 Gy in 5 weeks. A boost to the tumor bed was administered in 61%/43% of the patients using conventional fractionation with 10-14 Gy (5 fractions of 2.0 Gy per week) in START A/B. Thus, in the START A trial, the cumulative overall treatment time was longer than 5 weeks, i.e. not considerably shorter than with conventional WBI.
In order to provide a significantly shorter overall treatment time with expectedly equal outcome and tolerance, we introduced an accelerated postoperative hypofractionation radiation regimen of 41.6 Gy in 13 fractions plus 3-4×3 Gy boosts in our clinical routine of which first results have previously been reported (7). The recent analysis includes a larger cohort with a longer follow-up (FU) period, in order to allow a more valid estimation of the efficacy and tolerance of this hypofractionation approach.
In addition, the use of hypofractionation schedules for the boost and for ductal carcinoma in situ (DCIS) is discussed.
Patients and Methods
From 03/2009-12/2013, 140 patients >55 years with invasive breast cancer (n=134) or DCIS (n=6) were postoperatively treated in our Institution. The lower age limit of 55 years was decided due to the uncertainty regarding the biological iso-efficacy of the hypofractionation WBI with a total dose with 50 Gy in 2.0 Gy per fraction for younger patients. As the total dose to the tumor bed is even more important in younger patients (8), patients aged <55 were excluded to avoid potential under-dosage in younger women. Patient-related parameters are summarized in Table I.
Treatment and patients'-related parameters.
The boost schedule was individually based on the resection margins as follows:
R0 resection: 41.6 Gy in 13 fractions given 4 times a week and 9 Gy boost in 3 fractions (on day 5 each week); cumulative dose: 50.6 Gy in 3.2 weeks.
R1 resection: 41.6 Gy in 13 fractions given 4 times a week and 12 Gy boost in 4 fractions (on day 5 each week, afterwards daily); cumulative dose: 53.6 Gy in 3.4 weeks.
Details on treatment planning have been previously described (7). FU: Acute toxicity was assessed weekly during radiotherapy (RT), at the end of RT and at 8 weeks following RT. Prospectively planned FU visits, including both objective and subjective assessments of treatment tolerance and satisfaction with the cosmetic result (questionnaires for physicians/patients), were performed at one (n=111), at two (n=101) and at four years (n=59) following RT. Acute toxicity and late toxicity were assessed according to Common Terminology Criteria for Adverse Events (CTCAE) version 4.0.
Diesease-free survival (DFS) and overall survival (OAS).
Informed consent. Informed consent was obtained from all patients for prospective data collection, including electronic photo documentation and retrospective data evaluation.
Statistical analysis. Survival analyses were performed using SPSS version 21 (http://www.ibm.com). Ordinal and binary logistic regression analyses were carried out using commercially available Minitab software version 16.2.4.3 (http://www.advanceduninstaller.com). p-Values of less than 0.05 were considered statistically significant.
Results
Disease control. The mean/median FU time was 40/36 months (range=2-81). No isolated local or locoregional recurrence was observed. The 3-year local control, nodal control and disease-free survival (DFS) rates were 99%, 100% and 96%, respectively. The overall survival (OAS) rate at 3 years was 91% (Figure 1: DFS and OAS). At the time of analysis, 5 patients had died after a FU of 15, 25, 30, 42 and 52, months, respectively: three patients died due to distant metastases (meningeosis, n=1; disseminated metastases, n=1; liver metastases, n=1) and two patients died from non-breast cancer-related causes (pulmonary embolism, n=1; second primary in the lung with distant metastases, n=1). Seven patients were lost in FU after 8 weeks (n=4) and 2 years (n=3) after RT completion.
Acute and sub-acute side effects. Grade 1 acute skin toxicity was observed in 82% directly after RT and in 30% at 8 weeks following RT, grade 2 acute skin toxicity occurred in 5% and 0% of patients, respectively. No grade 3 acute dermatitis was observed (Figure 2). Pain within the treated breast was reported in 19% of patients prior to RT (grade 1 of 10: 7%, grade ≥2: 12%) and in 29% (grade 1: 11%, ≥grade 2: 18%) and 20% of patients (grade 1: 11%, ≥grade 2: 9%) directly and 8 weeks following RT. Early edema of the breast was seen in 9% of patients prior to postoperative RT and in 13% and 12% of patients directly and 8 weeks following RT, respectively. In 6/12 (50%) pre-RT edema patients, edema was resolved at treatment completion.
Acute skin toxicity (grade 0-3), pain (scale 0-10) and edema (grade 0-1) pre-radiotherapy, directly post-radiotherapy, and 8 weeks after completion of RT. Assessment of pain and edema additionally 1 year, 2 years and 4 years late.
A-C: Late toxicity: A=pigmentation (grade 0-1), teleangiectasia (grade 0-1), fibrosis (grade 0-2) 1 year, 2 years and 4 years after completion of radiotherapy. Additional evaluation of the boost regions seperately within the same time periods.
Cosmetic satisfaction 1 year, 2- and 4 years after completing radiotherapy (four grades of satisfaction: very satisfied to not satisfied).
Late side-effects (whole breast). Grade 1 edema of the breast was observed in 9% (10/111), 0% (0/80) and 0% (0/50) of patients after one, two and four years, respectively. Hyperpigmentation of the whole breast was observed in 23%, 14% and 6% of patients after one, two and four years, respectively. After one, two and four years, telangiectasia of the irradiated breast was present in 2%, 4% and 10% of patients, respectively (Figure 3), and fibrosis was found in 19%, 4% and 2% of patients, respectively (Figure 3).
Late side-effects (boost region). Hyperpigmentation of the boost region was observed in 29%, 20% and 12% of patients after one, two and four years, respectively. Telangiectasia of the boost region was present in 5%, 16% and 16% of patients (Figure 3), fibrosis in 35%, 20% and 12% of patients after one, two and four years, respectively (Figure 3).
Late side-effects (patient satisfaction, pain). The proportions of the patients who were satisfied or very satisfied were 99%, 98% and 100% after one, two and four years, respectively (Figure 4).
One year after completion of RT, 23% of patients (25/111) complained about pain (grade 1 of 10: 14%, ≥grade 2: 9%; Figure 3). After two and four years, pain was still present in 11% (9/80) (grade 1: 9%, ≥grade 2: 2%) and 6% (3/50) of patients, respectively. No pain grade 2 or higher was described after four years.
Predictive value of early symptoms on later tolerance. Nine out of 28 patients complaining about pain before RT start complained about persistent pain during further treatment (no analgesic medication required). Pain status before, after RT and 8 weeks later did not correlate with the presence of pain in further FU period. Also, edema of the breast immediately after RT did not correlate with its persistence during FU.
Volumetric considerations. The mean pre-RT volume of the breast planning target volume (PTV) was 689 cm3 (range=256-1,612). The mean volume of the boost PTV region was 98 cm3 (range=20-328). Additionally, we assessed the volume of 105%, 110% and 115% isodoses outside the boost PTV. The mean volume of 43.7 Gy (105%), 45.8 Gy (110%) and 47.8 Gy (115%) isodose was 158 cm3 (range=0-1349), 81 cm3 (range=0-815) and 48 cm3 (range=0-512), respectively.
No continuous correlation in logistic regression analyses was found in acute and late toxicity and dose-volume parameters like breast- and boost-volume, boost dose or different isodose levels.
Discussion
An accelerated regimen of the hypofractionated arm of the START A regimen with a hypofractionated boost has been clinically implemented and evaluated.
Early results were previously evaluated (7). Herein, we present an update of the prospectively assessed data of our single-institution accelerated hypofractionated WBR schedule with 3.2 Gy per fraction with a sequential resection margin-adapted hypofractionated boost. We found that the vast majority of patients were satisfied or very satisfied with cosmetic outcomes. The used regimen was well-tolerated and resulted in high locoregional control rates (Tables II and III). Late tolerance and patients' cosmetic results. Late tissue tolerance is a major concern in the context of hypofractionated radiation therapy. In terms of intermediate to late tolerance, our results were comparable to the results of the four randomized trials and to other studies that used hypofractionated boost schedules or a simultaneous integrated boost (SIB) (Tables II and III).
Literature review: randomized studies with different hypofractionation schedules in breast cancer patients, LR, local recurrence; CTx, chemotherapy; RT ln, radiotherapy periclavicular of lymph nodes; ccb, concomitant boost dose escalation; PB, partial breast.
In the present study, most patients considered the cosmetic results very satisfying or satisfying, agreeing well with the results of Scorsetti et al. and Kyrgias et al. (9,10) and compare favorably with most other studies reporting on cosmesis after hypofractionated WBI (2, 11-14).
Early side-effects. In the present study, no grade 3 acute toxicity was observed. Grade 2 skin reactions of 5% did compare well with Chadha et al. and Franco et al. (13, 15) and with most other hypofractionation studies supporting those results (14, 16-18). The above mentioned randomized trials mainly focused on late tissue effects (1-4).
Ciammella et al. found a statistically not-significant trend towards development of more late skin toxicity in patients with increased acute toxicity (14). In the own cohort, early treatment tolerance was high (no grade 3).
In line with Hannan et al. showing hypofractionated treatment of large-sized breast (PTV >1,500 cm3) feasible in terms of outcome and toxicity, we did not find a correlation between the breast size and outcome either (19).
Local control. In the four large prospective hypofractionation trials with 10-year FU, local recurrence rates ranged from 3.8%-14.8% (1-4). In the studies with different boost regimes, shown in Table IIB with significantly shorter FU, the local recurrence rates ranged from 0-4%. Our local recurrence rate of a more aged cohort (mean=67 years, range=55-92) was 1% at 3 years. This is comparable with the corresponding age adjusted collective in the EORTC study after the same FU period (8).
Hypofractionation in patients with DCIS. In our study, six patients with DCIS were included who did receive accelerated hypofractionation following breast conserving surgery. A recently published study with a mean FU of 9.2 years showed similar outcomes for hypofractionation in patients with DCIS after breast conserving therapy with local recurrence-free survival rates of 86% for conventional fractionation and 89% for hypofractionation (20). In the hypofractionation group (n=638), 54% patients received an additional boost. Cante et al. showed boost application in a hypofractionated regime in patients with DCIS to be safe delivering a daily boost of 0.25 Gy in each patient summing up to a total dose of 50 Gy (45 Gy WBI) (21). In a recent review article, Nilsson et al. stated hypofractionated RT to be a safe option in patients with DCIS after breast-conserving surgery, while the addition of a boost to the tumor bed reduced the risk of a local recurrence in case of positive resection margins (22).
Boost application in hypofractionated RT schedules. Recently, Bartelink et al. published the 20-year FU data of the EORTC 22881-10882 trial “boost versus no boost” confirming improvement of local control with the addition of a sequential boost to the tumor bed after conventionally fractionated WBI (8). Since hypofractionation gains importance, the question arises whether and how to integrate a higher dose to the tumor bed in hypofractionation schedules. The four randomized hypofractionation trials applied either no boost or a conventionally fractionated boost of 5-7×2 Gy in 43-75% of the patients (Table II). This procedure led to a prolongation of the overall treatment time of at least one week and, therefore, reduced the time sparing effect.
Both SIB and sequential boost have been shown to be feasible and effective in hypofractionation treatment of breast cancer patients (Table III). The published regimes using a SIB were heterogeneous with respect to the doses per fraction administered to the whole breast and to the tumor bed. Doses per fraction of WBI ranged between 2.2 Gy and 2.8 Gy, with doses per fraction of the boost irradiation between 2.5 Gy and 3.4 Gy. In the regimens that included a sequential boost, the doses per fraction ranged from 2.6 Gy to 3.4 Gy for the whole breast and from 2.65 Gy to 8.0 Gy in the boost region. The two approaches were compared in one retrospective study, which showed statistically significant slightly worse cosmetic results in those patients who received a SIB (23). However, since the two approaches have not yet been compared in a randomized trial, it is not yet clear whether the regimen using a SIB is inferior to the regimen including a sequential boost. Lacking high-level of evidence, the American Society for Radiation Oncology (ASTRO) stated that the boost indication is not clearly defined in hypofractionation (24). The IMPORT HIGH and LOW trials may provide us with more information regarding the efficacy and the tolerance of SIB using hypofractionated regimens (Table II).
Our regimen represents a further acceleration of the START A regimen that includes a shorter overall treatment time than most other hypofractionation schedules (Table II). Boost doses in the own cohort were defined according to the resection margin (<2 mm vs. ≥2 mm) (25).
Logistic regression analyses did not reveal any correlation of acute or late effects and isodose levels, PTVs of breast and/or boost or boost dose (9 Gy vs. 12 Gy).
In line with Kouloulias et al. and Ciamella et al., we did not find application of chemotherapy to be a risk factor for an enhanced skin reaction (14, 25).
Conclusion
According to this intermediate-time update of our prospective study, acceleration of the START A regime plus hypofractionated boost was effective, well-tolerated and associated with a high patient satisfaction regarding the cosmetic results.
Different boost schedules in breast cancer treatment with hypofractionated RT.
Acknowledgements
Dres Michelle Brown, Yousef Najafi, Edna Holz and Oliver Riesterer contributed to the clinical follow-up of the assessed patients.
Footnotes
Competing Interests
The Authors declare that they have no competing interests.
- Received February 4, 2016.
- Revision received March 17, 2016.
- Accepted April 2, 2016.
- Copyright© 2016 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved