GLP-1-based therapy and ICD-10-documented heart failure or respiratory failure events in non-diabetic adults with rheumatoid arthritis and obesity: a TriNetX federated cohort study.

INTRODUCTION/OBJECTIVES: Glucagon-like peptide-1 (GLP-1)-based therapies have emerged as a major advance in cardiometabolic care; however, no prospective outcomes data exist for these agents in non-diabetic adults with rheumatoid arthritis (RA) and obesity. We examined whether GLP-1-based therapy was associated with first post-landmark ICD-10-documented heart failure (HF) or respiratory failure (RF) events. METHODS: We conducted a retrospective cohort study in the TriNetX US Collaborative Network. Adults with RA, body mass index ≥ 30 kg/m, and baseline-year disease-modifying antirheumatic drug (DMARD) therapy were included; patients with diabetes and overlapping systemic autoimmune diseases were excluded. Exposure was semaglutide or tirzepatide documented within 0-90 days after the index BMI, and comparators were strict never-users. Cohorts were propensity score-matched 1:1 on 68 covariates. The primary endpoint was the first post-landmark ICD-10-documented HF or RF during days 91-365 among patients without pre-index documentation of the corresponding endpoint. RESULTS: After matching, 3483 patients remained per cohort, with all standardized mean differences < 0.10. During days 91-365, the primary endpoint occurred in 23/3176 GLP-1 users (0.7%) and 57/3144 never-users (1.8%) (hazard ratio (HR): 0.48; 95% confidence interval (CI): 0.30-0.78; p = 0.002; absolute risk difference: - 1.1 percentage points). The HF and RF components showed directionally similar associations. Findings were directionally similar at extended follow-up and, where estimable, in the calendar-time-restricted analysis. Bias probes showed no differential utilization. CONCLUSIONS: GLP-1-based therapy was associated with substantially lower hazards of first post-landmark ICD-10-documented HF or RF events. These observations, while compelling, are hypothesis-generating and require prospective validation before informing clinical use. Key Points • In a propensity score-matched TriNetX cohort of non-diabetic adults with rheumatoid arthritis and obesity, GLP-1-based therapy was associated with a lower hazard of first post-landmark ICD-10-documented heart failure or respiratory failure events. • In absolute terms, the primary composite occurred in 0.7% of GLP-1 users and 1.8% of never-users during days 91-365, corresponding to approximately one fewer event per 100 patients. • Heart failure and respiratory failure, analyzed separately, showed directionally consistent lower hazards, although event counts were small. • The findings provide preliminary RA-specific evidence for a prospective study, but they should not be used to guide treatment decisions.