Long-Term Toxicity and Outcomes of Hypofractionated Radiation Therapy with an Incorporated Boost for Early-Stage Breast Cancer
L. Wang,C. T. Murphy,E. Handorf,G. M. Freedman,M. E. Johnson,T. Shaikh,S. B. Hayes,R. J. Bleicher,L. Goldstein,P. R. Anderson
DOI: https://doi.org/10.1016/j.ijrobp.2015.07.142
2015-01-01
Abstract:To report the 8-year results of hypofractionated radiation therapy (HRT) with an incorporated boost and to compare toxicities and outcomes to propensity matched conventional fractionation (CRT) with sequential boost for early-stage breast cancer. The HRT cohort included 75 patients (pts) treated on a prospective Phase II study and 172 pts treated off trial in a similar manner. Inclusions were breast conservation, pts >18 years, Tis-T2, stage 0–II, and whole-breast IMRT with incorporated boost. The whole-breast received 2.25 Gy per fraction (fx) to 45 Gy with an incorporated tumor bed boost to 2.8 Gy per fx for a total of 56 Gy in 20 fx over 4 weeks. Women treated with CRT (whole-breast IMRT to 50 Gy in 2 Gy fx plus a sequential 10–16 Gy boost) were matched to HRT pts in a 2:1 ratio without replacement using a propensity score algorithm for age, T stage, systemic therapy use, bra cup size, type of boost (photon vs electron), grade, lymphovascular invasion, and ER/PR status. Adverse events during RT were considered acute, while events >1.5months after RT were considered late. All side effects were graded based on CTCAE v3. Patient reported cosmesis was scored using the Harvard scale (1 = excellent, 2 = good, 3 = fair, 4 = poor). Kaplan-Meier curves and Cox proportional hazard models with robust standard errors were used to estimate local control (LC). Proportions were evaluated using Mantel-Haenszel tests to account for the effect of matching. Median follow-up was 52 months (range, 3.5–130). Median age was 56 years (range, 31–88). The HRT patients included 19% Tis, 65% T1, and 16% T2 tumors; 86% were node negative. Eighty percent had negative margins, 18% had close (>0 and <2 mm) margins, and 2% had positive margins. The majority of pts received systemic therapy (45% endocrine therapy, 10% chemotherapy, and 22% both). We matched 241 of the 247 HRT pts to 482 CRT pts with no significant differences between groups except more unknown grade in CRT group. Eight-year LC was 97.5% for HRT versus 96.6% in the CRT group (P = .78). There was no significant difference in any grade acute skin erythema in HRT versus CRT (91.3% vs 90.9%, P = .89). Grade 2+ acute side effects were more frequent in CRT versus HRT (47.1% vs 18.7%, P < 0.01). There was no difference in any grade late side effects experienced in the CRT versus HRT group (47.7% vs 41.5%, P = .12); 79% of late side effects in both groups were grade 1. The most common late side effects were edema, hyperpigmentation, erythema, and fibrosis. Mean pt reported cosmesis scores were similar at 3 years for HRT versus CRT (1.9 vs 1.9). The 4-week course of hypofractionated whole-breast radiation with incorporated boost was associated with excellent long-term LC, less acute side effects, similar cosmesis, and late side effects compared to conventional fractionation with sequential boost.