Impact of margin status and re-excision on local control in patients undergoing breast-conservation therapy for ductal carcinoma in situ.

Abstract

57 Background: The purpose of this study was to identify the impact of final surgical margin (SM) status, SM width, and re-excision on outcomes in patients with ductal carcinoma in situ (DCIS) undergoing breast conservation therapy (BCT). Methods: The study population consisted of women diagnosed with DCIS undergoing BCT between 1989-2014. All women received adjuvant whole breast radiation plus a boost. The primary endpoint was local control (LC) defined as an ipsilateral breast failure. A negative SM was defined as > 2 mm, close SM was defined as > 0 to < 2 mm, and a positive SM was defined as tumor at the inked SM. Cox proportional hazards model was used to determine predictors of outcomes on multivariate analysis (MVA). Actuarial incidence of LC was estimated using the Kaplan-Meier method. Results: A total of 498 patients were included. The median age was 58 (range 30-91) and the median follow-up was 8.3 years (3 months-27 years). A total of 400 patients had a final negative SM, 87 had a close SM, and 11 had a positive SM. A total of 172 patients received adjuvant hormonal therapy, 265 patients required at least one re-excision. Patients with positive or close SMs were more likely to receive a radiation dose > 60 Gy (p < 0.001) and undergo re-excision (p < 0.01). The 10-year LC rates were not significantly different between patients with a negative (93.5%), close (91.8%), or positive (100%) SM (p = 0.57). There was no difference in 10-year LC rates according to a SM width of 0-1 mm (100%), > 1 to 2 mm (88.5%), or > 2 mm (93.5%) (p = 0.85). On univariate analysis, there was no significant difference in LC when comparing negative versus close or positive (p = 1.0) SMs. There was no difference in LC in patients undergoing re-excision for initial close or positive SMs (p = 0.55). On MVA, after controlling for age, dose, hormonal therapy, comedo subtype, and grade, there were no factors associated with LC. Conclusions: This large single-institution experience demonstrates that risks of local failure remain poorly characterized. Re-excision and whole breast radiation plus boost resulted in excellent LC for women with DCIS. Our data suggests that trials aimed at personalized de-intensified local therapy are warranted.