AIMS: To perform active safety monitoring of newer glucose-lowering therapies.
METHODS: Using Danish nationwide registries, we conducted an active comparator new user cohort study comparing adverse events among initiators of sodium-glucose co-transporter 2 inhibitors (SGLT2is), glucagon-like peptide 1 receptor agonists (GLP1-RAs), and dipeptidyl peptidase 4 inhibitors (DPP4is).
RESULTS: We identified 49 365 SGLT2i, 33 018 GLP1-RA, and 50 531 DPP4i initiators. Of 43 associations identified, we disregarded obesity-related signals (N = 2) and observed beneficial effects (N = 7). SGLT2i initiators had higher risks of phimosis (vs. GLP1-RA: hazard ratio [HR] 1.69, 95% compatibility intervals [CI] 1.42-2.02; vs. DPP4i: HR 2.19, 95% CI 1.84-2.59) and secondary polycythaemia (vs. GLP1-RA: HR 2.67, 95% CI 1.62-4.38; vs. DPP4i: HR 1.63, 95% CI 1.14-2.32). DPP4i initiators had increased risk of bacterial intestinal infections (vs. SGLT2i: HR 1.56, 95% CI 1.18-2.08; vs. GLP1-RA: HR 2.08, 95% CI 1.49-2.94) and anaemia (vs. SGLT2i: HR 2.00, 95% CI 1.22-3.23; vs. GLP1-RA: HR 1.92, 95% CI 1.18-3.23). GLP1-RA initiators had a near-immediate elevated risk of uterine cancer (vs. SGLT2i: HR 1.82, 95% CI 1.20-2.78; vs. DPP4i: HR 1.94, 95% CI 1.19-3.16).
CONCLUSION: While some signals likely reflect residual confounding, several signals - particularly those related to infections and cancer - warrant further investigation.
