Semaglutide users show 12% lower incident diabetes risk versus liraglutide in real-world data

Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) like semaglutide and liraglutide are widely used for chronic weight management and have known benefits in reducing the risk of type 2 diabetes and cardiovascular disease (CVD) compared to placebo. However, the comparative effectiveness of these two specific GLP-1 RAs against each other for preventing these critical outcomes has remained unclear. Understanding which agent offers superior protection is crucial for optimizing treatment strategies in patients without existing diabetes or CVD, particularly given their increasing use in broader populations.

Researchers conducted a target trial emulation using an active comparator, new user design with data from MarketScan claims databases. They identified enrollees free of diabetes and CVD who were prescribed either semaglutide or liraglutide between 2021-2023. Patients were matched 1:1, resulting in 57,456 GLP-1 RA users (mean age 45; 83% female). Primary endpoints were incident diabetes and incident CVD (defined as myocardial infarction, heart failure, or stroke). Cox regression, adjusted for a propensity score reflecting comorbidities and medication use, was employed to assess comparative risks.

Over a mean 1-year follow-up, the study observed 1104 incident diabetes events and 57 incident CVD events. After regression adjustment, semaglutide users demonstrated a 12% lower risk of incident diabetes compared to liraglutide users (hazard ratio [HR]: 0.88; 95% CI: 0.78, 0.99). The proportional hazards assumption for diabetes was violated (p < .0001), leading to stratified results by follow-up time. During the initial 6 months, semaglutide use was not significantly associated with diabetes risk compared to liraglutide (HR: 0.99 [0.82-1.18]).

However, a significant association emerged after 6 months of follow-up, where semaglutide use was linked to a lower risk of diabetes (HR: 0.80 [0.68, 0.94]).

No significant difference was detected in the risk of CVD events between the two drugs (HR: 1.25 [0.74-2.11]), suggesting similar cardiovascular protection or insufficient power/follow-up to detect a difference.