OBJECTIVE: GLP-1 receptor agonists (GLP-1RAs) and SGLT2 inhibitors (SGLT2is) facilitate weight loss and exhibit immunomodulatory effects, but their impact on the risk of developing autoimmune rheumatic diseases (ARDs) is unclear. We compared ARD incidence following initiation of GLP-1RAs, or SGLT2is, vs. a weight-neutral comparator (DPP4 inhibitors, DPP4is).
METHODS: We performed a population-based cohort study using administrative health data from a Canadian province with universal healthcare. We included adults with type 2 diabetes (T2D) and no prior ARD initiating a GLP-1RA, SGLT2i, or DPP4i between 2014/01/01 and 2022/12/31. Incident ARD cases, including rheumatoid arthritis, psoriatic disease, axial spondyloarthritis, and systemic ARDs (SARDs; systemic lupus erythematosus, systemic sclerosis, Sjögren's disease, idiopathic inflammatory myopathies, and systemic vasculitides), were identified using validated algorithms. Propensity score (PS) weighting was used to balance cohorts at treatment initiation, then hazard ratios (HRs) were estimated using Cox regression.
RESULTS: Among 229,300 adults, 49,514 initiated GLP-1RAs, 101,925 initiated SGLT2is, and 77,861 initiated DPP4is. After PS weighting, ARD incidence per 10,000 person-years was 29.1 (95% CI, 23.5-35.5) with GLP-1RAs, 24.4 (19.8-29.7) with SGLT2is, and 27.3 (22.1-33.4) with DPP4is. Mean follow-up was 1.3-1.6 years. Relative to DPP4is, adjusted HRs of ARD were 1.04 (0.81-1.33) with GLP-1RAs and 0.93 (0.75-1.16) with SGLT2is. Risk of SARDs, but not other diseases, was lower with SGLT2is vs. DPP4is (aHR, 0.51 [0.31-0.84]).
CONCLUSION: Neither GLP-1RA nor SGLT2i treatment was associated with lower ARD risk vs. DPP4is in adults with T2D; however, SGLT2i use was associated with significantly lower risk of SARDs, warranting further study.
