Aims and background: Triple-negative breast cancer (TNBC) has fluctuating pathological complete response (pCR) rates to neoadjuvant chemotherapy (NAC) according to published reports. Biomarkers predicting pCR rates of NAC would improve TNBC patients' outcomes. We conducted a meta-analysis to estimate the prognostic function of Ki-67 in relation to pCR rates of NAC in TNBC.
Methods and study design: Relevant publications in the literature from January 2006 to March 2013 were selected by searching PubMed, SpringerLink, Web of Science, Scopus and the Cochrane Library. The quality of prognostic studies was evaluated according to the standard reported by Hayden et al. Relative risk (RR) and 95% confidence interval (CI) were used to estimate the prognostic function of Ki-67 for pCR rates in TNBC. The fail-safe number was used to detect possible publication bias. Review Manager and MIX software was used to merge extracted data.
Results: The pCR rate of TNBC with high Ki-67 expression was 3.36 times that of low Ki-67 expression TNBC. The merged RR was 3.36 (95% CI: 1.61-7.02) and the fail-safe number was 34. No obvious publication bias but heterogeneity of the case series was detected.
Conclusions: Ki-67 was a predictor of pCR rates to NAC in TNBC.