RESUMO
BACKGROUND: PIK3CA mutations frequently occur in breast cancer patients. This study was conducted to evaluate the relationship between PIK3CA mutations and neoadjuvant treatment response and to analyze the clinical implications. METHODS: PubMed, Embase, and the Cochrane database were searched for relevant studies in September 2017. The pooled risk ratio (RR) was estimated using fixed effects or random effects models according to heterogeneity among studies. RESULTS: This meta-analysis included 20 studies with 4392 patients. The pooled RR showed that PIK3CA mutation is correlated to lower pathological complete response (pCR) in unselected HER2+ patients (RR = 0.73; 95% confidence interval [CI] 0.66-0.81), thus the predictive value of PIK3CA status may be stronger in HER2+/HR+ patients (RR = 0.50; 95% CI 0.27-0.93) and those administered dual-targeting treatment (RR = 0.55; 95% CI 0.39-0.78). In contrast with wild type, either exon 9 (RR = 0.55; 95% CI 0.39-0.78) or exon 20 (RR = 0.71; 95% CI 0.58-0.89) mutations were significantly associated with lower pCR. The predictive value of exon 9 mutations was not significantly greater than exon 20 mutations (RR = 0.76; 95% CI 0.51-1.13). CONCLUSION: In early breast cancer, PIK3CA mutations seem to identify HER2+ patients who are less likely to reach pCR. The clinical implications of PIK3CA mutations tend to vary between exon 9 and exon 20. This mechanism should be explored in further studies.
