LAUSR

BACKGROUND At present, the antitumor effect of metformin is controversial. Previous meta-analyses included observational studies, of which the results can be influenced by many confounders, affecting the result of meta-analyses and weakening the strength of evidence. Therefore, we conducted a meta-analysis to confirm the effect of metformin use on patients with advanced or unresectable cancers, including randomized clinical trials (RCTs). METHODS We searched for RCTs in accordance with the inclusion and exclusion criteria. A meta-analysis was conducted to combine hazard ratios (HRs) or risk ratios (RRs) and their 95% confidence intervals (CIs), using a random-effects model. RESULTS Finally, 7 eligible RCTs were included in meta-analysis. Overall, the combined results revealed that treatment with metformin did not improve the overall survival (OS) of patients (HR, 1.12; 95%CI, 0.91-1.37, P>0.05), and there was no clear evidence that metformin use was related to improved progression-free survival (PFS) (HR,1.17; 95%CI, 0.97-1.40; P>0.05). The pooled RR for grade III or IV adverse events was 0.92 (95%CI, 0.52-1.60; P>0.05), indicating that the use of metformin was not significantly related to increased toxicity. CONCLUSION Metformin does not significantly improve the survival of patients with advanced or unresectable cancer, regardless of cancer type and region. Open Science Framework： DOI 10.17605/OSF.IO/SPKE8.