RESUMO
We aimed to assess the efficacy of neoadjuvant chemotherapy (NACT) in a cohort of 213 triple-negative breast cancer (TNBC) patients treated in real-world practice at eight Italian cancer centers. We computed descriptive statistics for all the variable of interest. Factors testing significant in univariate analysis were included in multivariate models. Survival data were compared by Kaplan-Meier curves and log-rank test. The median follow-up was 45 months. We observed 60 (28.2%) pathological complete response (pCR). The sequential anthracyclines-taxanes-based regimens produced the highest rate of pCR (42.6%), followed by concomitant anthracycline-taxane (24.2%), and other regimens (15.6%) (p = 0.008). When analyzing the role of baseline Ki-67, a 50% cut-off was the optimal threshold value for pCR prediction (p = 0.0005). The 5-year disease-free survival (DFS) was 57.3% and the 5-year overall survival (OS) was 70.8%. In patients not achieving pCR, the optimal Ki-67 variation between biopsy and surgical specimen with prognostic relevance on long-term outcomes was 13% (p = 0.04). Patients with a Ki-67 reduction (rKi-67)<13% had worse outcomes compared to those who experienced pCR or a rKi-67≥13%. The number of NACT cycles also affected long-term outcomes (5-year DFS 65.7% vs 51.6% in patients having received >6 cycles compared with their counterparts, p = 0.02). In multivariate analysis, node status, grading, and bio-pathological treatment response (including pCR and rKi-67) impacted DFS and OS. Our results confirmed the advantage conferred by more than 6 cycles of a sequential antracycline-taxane-based NACT. Higher baseline Ki-67 values shows greater predictive significance on pathogical response, while the rKi-67 plays a prognostic role on long-term outcomes.