Is a higher boost dose of radiation necessary after breast-conserving therapy for patients with breast cancer with final close or positive margins?

To determine rates of loco-regional recurrence (LRR), distant failure and overall survival for patients with breast cancer treated with breast-conserving therapy (BCT) with a close or positive surgical margin (C/PM) treated with standard dose boost radiation compared with a higher boost of radiation. We retrospectively studied 1476 patients with T1-T3 invasive breast cancer treated with BCT between 1992 and 2009. Median age was 57 years. Patients were divided into three groups: Group I included 1197 patients (81 %) with negative margins who received a standard boost (median 60 Gy) total dose to the lumpectomy cavity; Group II included 116 patients (8 %) with C/PM who received a standard boost (median 60 Gy); and Group III included 163 patients (11 %) with C/PM who received a higher boost (median 68 Gy). Biological subtypes (e.g., ER, PR, HER2/neu) were available for 858 patients (58 %) and were also assessed for any relationship to LRR rate. The Kaplan-Meier, Cox-regression, and log-rank tests were used to estimate rates of LRR and the significance of risk factors. Median follow-up was 8.6 years. The overall 5- and 10-year cumulative incidences of LRR were 2.1 % (95 % CI 0.8-2.1 %) and 4.5 % (95 % CI 3.4-6.0 %), respectively. The 5- and 10-year cumulative incidences of LRR for Group I (negative margins + standard boost) were 1.9 and 4.4 %; for Group II (C/PM + standard boost) were 3.9 and 7.0 %; and for Group III (C/PM + higher boost) were 2.9 and 3.8 %, respectively. No statistically significant differences in LRR rates were found among the three groups (p = 0.4). Similar results were obtained for distant failure (p = 0.3) and overall survival (p = 0.4). On multivariate analysis, tumor grade (p = 0.03), systemic-therapy (p = 0.005), node positivity (p = 0.05), young age (p = 0.001), and biological subtype (p = 0.04) were statistically significantly associated with higher LRR. Higher boost dose and margin positivity were not significant. Our data suggest that the 10-year risk of local recurrence for patients with close or positive margins receiving a standard boost was 7 % compared to 3.8 % for those receiving a higher boost; however, this difference was not significant. A higher boost dose did not significantly improve local control, nor did margins impact LRR risk in our cohort of patients.