Effect of different glucose-lowering therapies on cancer incidence in type 2 diabetes: An observational population-based study

Abstract Aim To assess the effect of metformin on cancer incidence in type 2 diabetes (T2DM), considering possible interactions with other glucose-lowering drugs and diabetes duration. Methods Study cohort included diabetes patients aged 20–84 on December 2009, still alive and resident in Reggio Emilia province as of December 2011. Drug exposure was assessed for 2009–2011; subjects taking metformin continuously, with or without other hypoglycaemic drugs, were compared to subjects on diet-only therapy. The cohort was followed up from 2012 to 2014 through the cancer registry. Age- and sex-adjusted incidence rate ratios (IRRs) were computed using Poisson regression models for all sites, lung, breast, liver, colorectal, prostate and pancreatic cancer. Results The cohort includes 17,026 people with T2DM, 7460 taking metformin. 887 cancers occurred during follow-up, 348 among metformin users. Cancer risk was similar in T2DM patients using metformin and those on diet-only. The risk for prostate (IRR = 0.65; 95%CI:0.36; 1.17), liver (IRR = 0.82; 95%CI:0.36; 1.85) and breast (IRR = 0.77; 95%CI:0.43; 1.40) cancers only was slightly reduced; for lung (IRR = 1.52; 95%CI:0.92; 2.50), pancreas (IRR = 1.51; 95%CI:0.59:3.89) and colon-rectum (IRR = 1.71; 95%CI:0.94; 3.08) the risk was slightly increased. Conclusions There is no evidence of antitumor effect of metformin. A possible decrease only for breast, liver and prostate cancer, is compatible with random fluctuations.