Comparative cardiovascular safety of glucagon-like peptide-1 receptor agonists versus other antidiabetic drugs in routine care: a cohort study

Aims: To evaluate the comparative cardiovascular disease (CVD) safety of glucagon-like peptide-1 receptor agonists (GLP-1 RAs) in head-to-head comparisons with dipeptidyl peptidase-4 (DPP-4) inhibitors, sulphonylureas or insulin, when added to metformin, as used in 'real-world' patients with type 2 diabetes mellitus (T2DM). Methods: Within a large US commercial health plan database linked to laboratory test results, we identified three pairwise 1 : 1 propensity-score-matched cohorts of patients with T2DM aged >= 18 years treated with metformin who initiated a GLP-1 RA or a comparator, i.e. DPP-4 inhibitor (n=35 534), second-generation sulphonylureas (n=28 138) or insulin (n=47 068), between 2005 and 2013. We examined the association between drug initiation and a composite CVD endpoint, comprising hospitalizations for acute myocardial infarction, unstable angina, stroke or coronary revascularization. Results: During the course of 1 year, there were 13.9 and 13.7 CVD events per 1000 person-years among propensity-score-matched initiators of GLP-1 RAs versus DPP-4 inhibitors [hazard ratio (HR) 1.02; 95% confidence interval (CI) 0.84-1.24]; and 12.1 versus 14.0 events among initiators of GLP-1 RAs versus sulphonylureas (HR 0.86; 95% CI 0.69-1.08). The effect estimates for GLP-1 RAs versus insulin were sensitive to the adjustment for glycated haemoglobin, after which the HR was 1.01 (95% CI 0.73-1.41). Results were robust across several sensitivity analyses, including an as-treated analysis considering up to 8.7 years of follow-up. Conclusions: This large study, performing head-to-head comparisons of GLP-1 RAs with other antidiabetic agents in real-world patients, provides estimates of relative safety precise enough to exclude large differences in CVD risk and adds further understanding to results from recent clinical trials.