(A-304) Glucagon-like Peptide-1 Receptor Agonists vs. Sodium-glucose Cotransporter 2 Inhibitors and Risk for Dementia in Individuals with Diabetes: A Propensity-weighted, Population-based Cohort Study.

Background: Several observational studies and systematic reviews suggest that both GLP-1 receptor agonists (GLP-1 RA) and sodium-glucose cotransporter 2 inhibitors (SGLT-2i) on the risk of GLP-1 RA and SGLT-2i are associated with a reduced risk of dementia and Alzheimer's disease when compared to other antidiabetic drugs. However, the relative risk between these two drugs groups is unknown.

Objectives: We aimed to evaluate the comparative risk of dementia and Alzheimer’s disease in patients treated with GLP-1 RA or SGLT-2i.

Methods: Population-based, propensity-weighted, new user, active comparator design study including patients with Type 2 diabetes aged 60 years old and over initiating treatment with GLP-1 RA or the comparator group sodium-glucose cotransporter 2 inhibitors (SGLT-2i) in a south European region with 5 million inhabitants (Valencia, Spain), from 2015 to 2021. We carried out Cox regression modelling using propensity scores, and we provided main, per protocol risk estimates. Secondary analyses included intention-to-treat estimates for the main analysis, analyses including only patients with obesity at baseline, as well as analyses for Alzheimer’s disease and vascular dementia outcomes.

Results: We included 21,071 new users of SGLT-2i and 1,577 new users of GLP-1 RA in the main analysis. Mean follow-up was 430 days for GLP-1 RA users and 343 days for SGLT-2i users in per protocol analyses, and 788 and 796 days in intention-to-treat analyses, respectively. In adjusted, per protocol main analysis, no risk differences were found for dementia in GLP-1 RA vs SGLT 2i patients (HR: 1.67; IC95%: 0.71, 3.92). Secondary analyses yielded similar, non-significant results.

Conclusions: Our results suggest that GLP-1 RA are associated with a similar risk of dementia when compared to SGLT-2i in older individuals with Type 2 diabetes. Further research is needed to validate our findings.