(093) Effectiveness of GLP-1 Receptor Agonists on Bodyweight Among Patients with Type-2 Diabetes Mellitus: Emulation of Five Randomized Clinical Trials Using Electronic Medical Records

Background: Weight loss provides a key element in managing type-2 diabetes mellitus (T2DM) patients, given that over 90% are overweight or obese. Glucagon-like peptide-1 receptor agonists (GLP-1RAs) have demonstrated efficacy in promoting weight loss compared to placebo among T2DM patients with a mean difference of -4.2% (95%CI -5.6, -2.9) estimated by a recent meta-analysis on 10 randomized clinical trials (RCTs). However, the effectiveness observed in routine clinical practice may be different, and is unknown.

Objectives: To address the unmet need for real-world evidence on the effectiveness of GLP-1RAs in weight management among patients with T2DM, we emulated an RCT for each of the five commonly prescribed GLP-1RAs.

Methods: We applied a new-user cohort design with a real-world placebo (i.e., non-user comparator) group using the IQVIA Medical Research Data (IMRD) incorporating data from THIN, A Cegedim Database. Cohort entry was defined as the date of a BMI measurement >=25 kg/m2 recorded among patients aged >=18 years between Jan 1, 2006 and June 30, 2023. The index date of GLP-1RA users was the date of their first GLP-1RA prescription, whereas the same date was used as the index date for their 1:4 propensity score (PS)-matched non-user comparator patients. All patients were hard-matched on prior use of anti-diabetic medications. Intention-to-treat analyses were used to assess percentage changes in bodyweight and estimated treatment differences from baseline to 52 weeks using ANCOVA.

Results: We included 14,831 new users PS-matched to 39,733 comparators distributed across the treatment cohorts: semaglutide (n=2,467 vs n=6,709), liraglutide (n=5,151 vs n=13,204), exenatide (n=3,841 vs n=10,452), dulaglutide (n=2,808 vs n=7,639), and lixisenatide (n=564 vs n=1,729). Estimated mean changes in bodyweight were -5.1% (SE 0.1), -3.4% (SE 0.1), -4.0% (SE 0.1), -3.0% (SE 0.1), and -3.2% (SE 0.2) for semaglutide, liraglutide, exenatide, dulaglutide, and lixisenatide, respectively. Corresponding estimated treatment differences were determined to -2.6% (95%CI -3.2, -1.9; p< 0.01), -1.5% (-1.7, -1.2; p< 0.01), -2.6% (-2.9, -2.3; p< 0.01), -0.4% (-0.8, -0.04; p< 0.05), and -0.9% (-1.5, -0.2; p< 0.05).

Conclusions: Among overweight and obese patients with T2DM, all GLP-1RAs were associated with a significant mean reduction in bodyweight compared to their matched non-user comparators. Weight loss was most pronounced among users of semaglutide and exenatide. Compared to most RCTs, the magnitude of weight loss estimated in our study was attenuated, which may be explained by the lower adherence, and frequent treatment discontinuation observed in real-world patients.