Predictors of local recurrence following accelerated partial breast irradiation: a pooled analysis.

PURPOSE: To analyze a pooled set of nearly 2,000 patients treated on the American Society of Breast Surgeons (ASBS) Mammosite Registry Trial and at William Beaumont Hospital (WBH) to identify factors associated with local recurrence following accelerated partial breast irradiation (APBI).

METHODS AND MATERIALS: A total of 1,961 women underwent partial breast irradiation between April 1993 and November 2010 as part of the ASBS Registry Trial or at WBH. Rates of ipsilateral breast tumor recurrence (IBTR), regional recurrence (RR), distant metastases (DM), disease-free survival (DFS), cause-specific survival (CSS), and overall survival (OS) were analyzed for each group and for the pooled cohort. Clinical, pathologic, and treatment-related variables were analyzed including age, tumor stage/size, estrogen receptor status, surgical margins, and lymph node status to determine their association with IBTR.

RESULTS: The two groups weres similar, but WBH patients were more frequently node positive, had positive margins, and were less likely to be within the American Society for Radiation Oncology-unsuitable group. At 5 years, the rates of IBTR, RR, DM, DFS, CSS, and OS for the pooled group of patients were 2.9%, 0.5%, 2.4%, 89.1%, 98.5%, and 91.8%, respectively. The 5-year rate of true recurrence/marginal miss was 0.8%. Univariate analysis of IBTR found that negative estrogen receptor status (odds ratio [OR], 2.83, 95% confidence interval 1.55-5.13, p = 0.0007) was the only factor significantly associated with IBTR, while a trend was seen for age less than 50 (OR 1.80, 95% confidence interval 0.90-3.58, p = 0.10).

CONCLUSIONS: Excellent 5-year outcomes were seen following APBI in over 1,900 patients. Estrogen receptor negativity was the only factor associated with IBTR, while a trend for age less than 50 was noted. Significant differences in factors associated with IBTR were noted between cohorts, suggesting that factors driving IBTR may be predicated based on the risk stratification of the patients being treated.