Target Trial of Semaglutide Versus Dulaglutide or Liraglutide in Patients Receiving Hemodialysis

OBJECTIVE: To evaluate the comparative effectiveness of semaglutide versus dulaglutide or liraglutide in patients with diabetes receiving maintenance hemodialysis. RESEARCH DESIGN AND METHODS: Using 2011-2022 U.S. Renal Data System and Medicare data, we emulated a target trial evaluating semaglutide versus dulaglutide or liraglutide in patients receiving hemodialysis. Overlap weights were used to adjust for confounders. Fine-Gray models were used to assess death on dialysis and hospitalization. A generalized estimating equation was used for repeated measures of body weight. RESULTS: We identified 1,312 semaglutide and 2,601 dulaglutide or liraglutide users. After overlap weighting, the effective sample size was 3,200 patients (mean age 61; 63% White; 28% Black; 52% male; median [interquartile range] BMI 34 [29-40] kg/m2; median follow-up 1.1 [0.6-1.9] years). Semaglutide versus dulaglutide or liraglutide was associated with a mortality on dialysis subdistribution hazard ratio (sHR) of 0.82 (95% CI 0.70-0.97; 1-year cumulative incidence 11% vs. 13%) and all-cause hospitalization sHR of 0.91 (95% CI 0.83-0.99; 1-year incidence 57% vs. 62%). Body weight loss was higher with semaglutide (-4.8%, 95% CI -5.3% to -4.2%) at 1 year vs. -3.4%, 95% CI -3.9% to -3.1%). Direct comparisons of semaglutide versus dulaglutide showed similar results for mortality (sHR 0.82, 95% CI 0.69-0.97) and hospitalization (sHR 0.88, 95% CI 0.81-0.97), but semaglutide versus liraglutide showed no statistically significant benefit for mortality or hospitalization. CONCLUSIONS: Among patients with diabetes receiving hemodialysis and equipoise between semaglutide and dulaglutide or liraglutide, semaglutide may be the preferred glucagon-like peptide 1 agent compared with dulaglutide.