PurposeTo assess the effect of metformin intake on cancer incidence and mortality. MethodsOriginal articles in English published until June 15, 2012 were searched for in electronic databases (MEDLINE, ISI Web of Science and EMBASE databases) and relevant reviews were examined. Meta-analysis was applied to calculate the summary relative risk (SRR) and their 95% confidence intervals (95% CI). Sensitivity analysis was conducted to assess the robustness of the pooled estimator. The risk of publication bias was assessed by the Egger regression asymmetry test. ResultsAccording to the eligibility criteria, 37 studies comprising 1,535,636 participants, were selected in terms of intervention and data of cancer incidence or mortality. Among metformin users compared with non-users, the SRR for overall-cancer incidence was 0.73 (95% CI, 0.64–0.83) and that for mortality was 0.82 (95% CI, 0.76–0.89). The risk reductions for liver, pancreatic, colorectal and breast cancer incidence were 78%, 46%, 23% and 6%, respectively. Also, metformin can reduce the mortality of liver cancer (SRR, 0.23; 95% CI, 0.09–0.60) and breast cancer (SRR, 0.63; 95% CI, 0.40–0.99). No statistically significant association between metformin and prostate cancer incidence was found. ConclusionsMetformin can reduce the incidence of overall cancer, liver cancer, pancreatic cancer, colorectal cancer and breast cancer as well as the mortality of overall cancer, liver cancer and breast cancer. No beneficial effect on prostate cancer incidence was found for meformin intake in the meta-analysis.