Can residual proliferative cancer burden predict long-term outcomes following neoadjuvant chemotherapy in breast cancer?

Abstract

Residual proliferative cancer burden (RPCB) has been suggested as a strong predictor model of long-term outcomes in breast cancer undergoing neoadjuvant chemotherapy (NACT). In our study, we aimed to compare the prognostic value of multiple post-NACT classifications for assessing residual disease. Archival surgical specimens of 97 patients with primary breast cancer who underwent NACT were evaluated for residual cancer burden (RCB). The post-operative Ki-67 proliferation index was quantified using immunohistochemistry on post-treatment surgical excision specimens with residual disease. Then, we calculated the RPCB scores by combining the anatomical RCB index with the biological post-therapeutic Ki-67 using the Cox proportional hazard model for each parameter. Using the Kaplan-Meier method, RCBIII showed an unfavourable prognosis with worse relapse-free survival (RFS) (estimated 5-year RFS rate of 38%) than RCBI, which displayed a similarly good prognosis as pathological complete response (equal to RCB0) (estimated 5-year RFS rates of 80% and 100%, respectively) (p=0.012). The RCBII showed an intermediate prognosis (estimated 5-year RFS rate of 79%). A higher post-NACT Ki-67 (greater than cut-off 20%) had a negative impact on the overall survival and RFS (p<0.0001 for both) using the Kaplan-Meier method. In multivariate analysis, the histological residual tumour size, number of affected lymph nodes, and RCB index remained independent prognostic factors for RFS. In addition, RPCBIII showed the worst prognosis (with an estimated 5-year RFS rate of 38%) compared to RPCBI (estimated 5-year RFS rate of 83%) (p=0.039) by the Kaplan-Meier method. The area under the curve of the RCB index was 0.82 compared to 0.62 for the RPCB model in terms of RFS prediction. Our study highlighted the potential stratification of RCBII cases based on the RPCB classification. Further studies with larger cohorts will be needed to validate whether the RCPB adds value to residual disease assessment.