Metformin is associated with lower risks of incident cancers compared to sulphonylurea use: a population-based study with competing risk analyses

Abstract Objective To investigate the effects of sulphonylurea compared to metformin on the risks new-onset cancers and mortality in patients with type 2 diabetes mellitus. Methods This is a retrospective population-based cohort study of patients with type 2 diabetes mellitus between 1st January and 31st December 2009 with follow-up until 31st December 2019. Propensity score matching (1:2 ratio) between metformin and sulphonylurea users was performed. The primary outcome is new diagnosis of any cancer. The secondary outcomes are cancer, cardiovascular, or all-cause mortality. Cox regression was used to identify significant predictors. Competing risks were considered using cause-specific or subdistribution hazard models. Propensity score stratification, inverse probability weighting, and high-dimensional propensity score adjustment were also used. Results In the matched cohort, 36487 sulphonylurea and 71306 metformin users were included. Univariable Cox regression showed that compared to metformin users, sulphonylurea users had a higher risks of any new cancer diagnosis (hazard ratio [HR]: 2.26, 95% confidence interval [CI]: 2.18-2.34, P<0.0001), including pancreatic (HR: 4.03 [3.37, 4.82], P<0.0001), colorectal (HR: 2.12 [1.97, 2.27], P <0.0001), genitourinary (HR: 1.73 [1.57, 1.90], P<0.0001), breast (HR: 1.36 [1.22, 1.51], P<0.0001) and lung cancer (HR: 3.37 [3.11, 3.65], P<0.0001) cancers. Sulphonylurea users also had higher risks of cancer-related (HR: 2.38, 95% CI: 1.87-3.03; P<0.0001) and all-cause mortality (HR: 1.92, 95% CI: 1.78 -2.08; P<0.0001). Conclusion Metformin use was associated with lower risks of new diagnosis of any cancer, cancer-related and all-cause mortality in patients with type 2 diabetes mellitus.Figure 1.Study flow diagram.FIgure 2.New diagnosis of cancer.