Survival following locoregional recurrence of breast cancer: univariate and multivariate analysis.

Although prognostic variables for locoregional recurrence of breast cancer have been evaluated by univariate analysis, multifactorial analysis has not been previously performed. In the present study, survival following chest wall and/or regional lymphatic recurrence was determined in 230 patients with locoregionally recurrent breast cancer without evidence of distant metastases treated at the Radiation Oncology Center, Mallinckrodt Institute of Radiology and affiliated hospitals. Multifactorial analysis demonstrated that the site of recurrences correlated most strongly with overall survival (p = 0.001). The 5-year actuarial overall survival was 44-49% for patients with isolated chest wall, axillary, and internal mammary lymph node recurrence. Patients with either supraclavicular, multiple lymphatic, or concomitant chest wall and lymphatic recurrence had an 21-24% 5-year overall survival. The 5-year disease-free survival was 28-37% for patients with chest wall, axillary, or internal mammary recurrences compared to 4-13% for those with supraclavicular, chest wall and lymphatic, or those with multiple sites of lymphatic recurrence. Disease-free interval from mastectomy to recurrence was also found to be a significant prognostic factor for overall survival (p = 0.005). Fifty percent of patients with a disease-free interval of at least 2 years survived 5 years following locoregional relapse, compared to 35% for those with disease-free interval of less than 2 years. In the subset of patients with small chest wall recurrences (excised or less than 3 cm) and a disease-free interval of at least 2 years, the 5-year overall and disease-free survivals were 67% and 54%, respectively. These results suggest that subsets of patients with locoregional recurrence of breast cancer can survive for long periods of time. The conventional wisdom that chest wall and/or regional nodal recurrence following mastectomy uniformly confers a dismal prognosis is not necessarily true.