Abstract
BACKGROUND: Achieving a pathologic complete response (pCR) has been associated with improved long-term outcomes in clinical trials. However, the benefit of achieving pCR across subtypes and its prognostic effect on real-world outcomes has not been well described. METHODS: A retrospective analysis of the Breast Cancer Outcomes Unit database was undertaken to identify patients with stage I-III breast cancer treated with neoadjuvant chemotherapy from 2005 to 2010 in British Columbia. Patients were separated into two groups: those with pCR and those with residual invasive disease in the breast/axillary lymph nodes (RD). The primary endpoint was relapse-free survival (RFS). Key secondary endpoints included breast cancer-specific survival (BCSS) and overall survival (OS). RESULTS: Of 267 patients identified, 74 patients (28%) achieved pCR and 193 patients (72%) had RD. Median follow-up was 7.5 years. The 5-year RFS was higher in the pCR group compared to the RD group (84% vs 70%; HR 0.45, p = 0.011). The 5-year BCSS was also higher in the pCR group than in the RD group (90% vs 77%; HR 0.39, p = 0.014). In multivariable analyses, pCR was associated with improved RFS (HR 0.39, p = 0.0077) and BCSS (HR 0.35, p = 0.015), whereas traditional pathological prognostic factors were not. Patients with TNBC who achieved pCR had improved RFS and BCSS compared to those with RD (HR 0.26, p = 0.020 and HR 0.35, p = 0.090, respectively). A similar but non-statistically significant trend was seen in the HER-2-positive and ER + subtypes. CONCLUSIONS: Achieving pCR after neoadjuvant chemotherapy was associated with clinically meaningful improvements in survival parameters in a real-world setting. The cumulative data support pCR as a valid surrogate endpoint in both clinical trials and population-based settings.