Aim: To review the predictive values of Ki-67 before neoadjuvant chemotherapy (NAC) for breast cancer patients. Methods: PubMed and EMBASE were searched. Random-effect model meta-analysis was conducted using Revman software. Results: High Ki-67 was associated with more pathological complete responses (pCRs) events (odds ratio: 3.10; 95% CI: 2.52-3.81; 53 studies, 10,848 patients) regardless of HR+, HER2(+) and triple-negative breast cancer types, the definitions of pCR and cut-off points for Ki-67. Ki-67 could predict pCR in those who received anthracyclines plus taxanes, and anthracyclines only, and those from Asia and Europe. Conclusion: High Ki-67 before NAC was a predictor for pCR in neoadjuvant setting for breast cancer patients.

Aim: To assess the prognostic values of Ki-67 in neoadjuvant setting for breast cancer patients. Methods: PubMed and EMBASE were searched. Revman software was used to conduct random-effect model meta-analysis. Results: 49 studies (14,076 patients) were included. High Ki-67 before and after neoadjuvant chemotherapy were associated with worse overall survival (OS; before: hazard ratio [HR]: 2.29; 95% CI: 1.42-3.69; after: HR: 2.24; 95% CI: 1.82-2.75) and disease-free survival (DFS; before: HR: 1.54; 95% CI: 1.23-1.95; after: HR: 2.08; 95% CI: 1.83-2.37). Low/no reduction or increase might be associated with worse DFS (HR: 2.13; 95% CI: 1.51-3.02) and OS. Conclusion: Ki-67 before and after neoadjuvant chemotherapy, as well as the change could predict the prognosis for breast cancer patients.