Abstract
INTRODUCTION: Pathologic complete response (pCR) after neoadjuvant chemotherapy has a demonstrated survival advantage; however, outcomes for non-pCR by receptor status are less understood. We sought to evaluate survival and distant recurrence by receptor status for patients with residual stage II/III breast cancer. METHODS: A stage-stratified random sample of 11,366 patients with stage II-III breast cancer in 2006-2007 was selected from 1217 facilities in the National Cancer Database for a Commission on Cancer Special Study. We identified patients with residual pathologic stage II/III cancer who received standard of care therapy based on receptor status. Distant recurrence and 5-year survival were abstracted and Kaplan-Meier curves were generated by receptor status. Multivariable Cox regression was used to estimate hazard ratios for death and distant recurrence. RESULTS: A total of 734 patients had residual disease; 58%, 28%, and 14% were ER or PR+/Her2neu-, ER and PR-/Her2neu-, and Her2neu+ (any ER/PR), respectively. ER and PR-/Her2neu- cancers had the poorest 5-year overall (52% vs. 82% for Her2neu+ and ER or PR+/Her2neu-, p < 0.0001) and distant recurrence-free survival (57% vs. 72% Her2neu+ and 77% ER or PR+/Her2neu, p < 0.0001). Cox regression models demonstrated a higher likelihood of distant recurrence and death for patients with ER and PR-/Her2neu- disease (HR 2.25, 95% CI 1.56-3.24 and HR 3.19, 95% CI 2.20-4.64 respectively) compared with ER or PR+/Her2neu-. CONCLUSIONS: Patients with residual ER and PR-/Her2neu- cancer have a significant risk of distant recurrence and mortality compared with other breast cancer types, supporting the consideration for additional adjuvant therapy and novel clinical trials in this cohort. Trial registry number ClinicalTrials.gov identifier NCT02171078.