RESUMO
PURPOSE: Pathological complete response(pCR) during neoadjuvant chemotherapy(NAC) has been proposed as a predictor for better prognosis in breast cancer. However, few studies compare the outcomes of patients receiving NAC and adjuvant chemotherapy(AC). METHODS: We retrospectively matched the patients who received NAC(N = 462) and AC(N = 462) by age, time of diagnosis, and primary clinical stage using the propensity score match in breast cancer patients treated in Sir Run Run Shaw Hospital with the median follow up of 67 months. Death from breast cancer and recurrence were used as endpoints. A multivariable Cox models were used to estimate the hazard ratios for breast-cancer specific survival (BCSS) and DFS. A multivariable logistic regression model was simulated to predict pCR. RESULTS: In patients who received NAC, 18.0%(83/462) patients achieved pCR, while the rest of the patients did not. pCR subgroup demonstrated significant better BCSS and DFS than patients receiving AC(BCSS: HR = 0.39, 95% CI:0.12-0.93, P = 0.03; DFS: HR = 0.16, 95%CI 0.009-0.73, P = 0.013) and non-pCR patients(BCSS: HR = 0.32, 95%CI 0.10-0.77, P = 0.008; DFS: HR = 0.12, 95%CI 0.007-0.55, P = 0.002). Patients who received AC demonstrated insignificant survival compared to non-pCR patients(BCSS: HR= 0.82, 95%CI 0.62-1.10, P = 0.19; DFS: HR = 0.75, 95%CI 0.53-1.07, P = 0.12). Patients with AC had significant better DFS than non-pCR patients(HR = 0.33, 95% CI 0.10-0.94, P = 0.04) in luminal B Her2+ patients. More NAC cycles(>2), TNBC, lower cT stage, and mixed histology indicate higher possibility of pCR(AUC = 0.89). CONCLUSION: pCR patients with NAC indicated better prognosis than patients receiving AC or non-pCR patients from NAC. The timing of chemotherapy may need carefully pondering in luminal B Her2+ patients.