Risk of Sight-Threatening Diabetic Retinopathy with Glucagon-Like Peptide-1 Receptor Agonist Use in Routine Clinical Practice: Comparative Effectiveness of Semaglutide, Dulaglutide, Liraglutide, and Exenatide.

PURPOSE: To investigate whether glucagon-like peptide-1 receptor agonist (GLP-1 RA) agents differ with respect to the risk of developing sight-threatening diabetic retinopathy complications in patients with type 2 diabetes at moderate cardiovascular risk. DESIGN: Retrospective observational study under the target trial emulation framework involving adult (≥21 years) enrollees in United States commercial, Medicare Advantage, and Medicare fee-for-service plans from January 1, 2014 to December 31, 2022. The study population included adults with type 2 diabetes, moderate cardiovascular disease risk, no baseline advanced diabetic retinopathy, and minimum 1 year enrollment before GLP-1 RA treatment. METHODS: Because of differences in GLP-1 RA agent use over time, we performed both a 3-way comparison of patients initiating liraglutide, dulaglutide, or exenatide from January 1, 2015, to December 31, 2021, and a 2-way comparison of patients initiating semaglutide or dulaglutide between January 1, 2019, and December 31, 2021. We assessed differences in complications using inverse propensity score weighted Cox proportional hazards models. MAIN OUTCOME MEASURES: Treatment for diabetic macular edema (DME) or proliferative diabetic retinopathy (PDR). RESULTS: When comparing patients who initiated treatment with exenatide (n = 14 076, median follow-up 969 days; interquartile range [IQR] 578-1444) to those starting dulaglutide (n = 54 787, median follow-up 948 days; IQR 551-1457) or liraglutide (n = 25 562, median follow-up 1007 days; IQR 575-1494), no differences were found in the hazard (hazard ratio [HR]) of composite treatment for DME or PDR: exenatide versus dulaglutide (HR 0.90; 95% confidence interval (CI): 0.73-1.12), liraglutide versus dulaglutide (HR: 0.97; 95% CI: 0.79-1.19), or liraglutide versus exenatide (HR: 1.08; 95% CI: 0.83-1.38), nor were differences found in any comparisons for treatment of DME (HR: 0.93 [95% CI: 0.74-1.18]; HR 0.99 [95% CI: 0.79-1.24]; HR: 1.06 [95% CI: 0.80-1.40]) or PDR individually (HR: 1.16 [95% CI: 0.81-1.67]; HR: 1.05 [95% CI: 0.73-1.51]; HR: 0.91 [95% CI: 0.58-1.40). Likewise, when comparing patients initiating semaglutide (n = 30 911, median follow-up 625 days [IQR: 455-850]) versus dulaglutide (n = 32 844, median follow-up 639 days [IQR: 459-878]), the hazards for treatment of DME or PDR (HR: 0.88; 95% CI: 0.70-1.11), DME (HR: 0.89; 95% CI: 0.69-1.14), and PDR (HR: 0.70; 95% CI: 0.45-1.09) all found no difference between drugs. CONCLUSIONS: Despite the differences in systemic efficacy among the examined GLP-1 RAs, we identified no difference in the risk of sight-threatening diabetic retinopathy after initiation of different GLP-1 RA agents among adults with type 2 diabetes at moderate cardiovascular risk. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.