Comparative dementia risk with GLP1 receptor agonists, SGLT2 inhibitors, or DPP4 inhibitors: a population-based cohort study

Abstract

Abstract Background We aim to investigate comparative dementia risk associated with glucagon-like peptide-1 receptor agonists (GLP1-RAs), sodium-glucose cotransporter-2 inhibitors (SGLT2i), and dipeptidyl peptidase-4 inhibitors (DPP4i) in adults aged ≥ 60 years with type 2 diabetes. Methods This target trial emulation cohort study used primary care electronic health records from the UK Clinical Practice Research Datalink. Initiators of GLP1-RAs, SGLT2i, or DPP4i aged ≥ 60 years with type 2 diabetes and without cognitive impairment were compared pairwise in three separate analyses. After propensity scores overlap weighting, 13,965 pairs of GLP1-RA versus DPP4i initiators (cohort entry: 2006–2022; mean age: 66.9 years), 25,533 pairs of SGLT2i versus DPP4i initiators (cohort entry: 2013–2022; mean age: 69.0 years), and 14,214 pairs of GLP1-RA versus SGLT2i initiators (cohort entry: 2013–2022; mean age: 67.9 years) were analyzed. The primary outcome was incident all-cause dementia. The primary analysis was an intention-to-treat analysis. A secondary as-treated analysis for continuous use was performed. Results In the intention-to-treat analysis, dementia risk was not different between GLP1-RA and DPP4i initiators (hazard ratio [HR] 0.95, 95% confidence interval [CI] 0.87–1.04; rates: 6.29 versus 6.64 per 1000 person-years; mean follow-up: 6.54 years); however, continuous GLP1-RA versus DPP4i use was associated with a 21% lower risk (HR 0.79, 95% CI 0.64–0.97; rates: 3.64 versus 4.82; mean follow-up: 2.71 years). SGLT2i versus DPP4i initiation was associated with a 14% lower dementia risk in the intention-to-treat analysis (HR 0.86, 95% CI 0.79–0.94; rates: 4.83 versus 5.60; mean follow-up: 4.91 years), and the as-treated analysis showed greater risk reduction (HR 0.70, 95% CI 0.60–0.82; rates: 3.82 versus 5.46; mean follow-up: 2.43 years). Dementia risk was comparable between GLP1-RA versus SGLT2i in both intention-to-treat (HR 0.98, 95% CI 0.87–1.11; rates: 4.85 versus 4.95; mean follow-up: 5.09 years) and as-treated (HR 1.07, 95% CI 0.85–1.36; rates: 3.71 versus 3.56; mean follow-up: 2.40 years) analyses. Conclusions In people with type 2 diabetes aged ≥ 60 years, SGLT2i are associated with reduced dementia risk, but dementia risk reduction associated with the GLP1-RAs studied is less certain.