1309-OR: Dementia Progression with GLP-1 Receptor Agonists in People with Type 2 Diabetes and Mild Cognitive Impairment: Retrospective Cohort Study

Introduction and Objective: Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) were proposed to exert neuroprotective effects. Preliminary data from the EVOKE placebo-controlled phase-3 trials indicate that 2-year oral semaglutide therapy, although improving Alzheimer's-related biomarkers, did not mitigate disease progression in people with early Alzheimer's, mostly without type 2 diabetes (T2D). It remains unclear whether GLP-1 RAs therapy may reduce the risk of progression to dementia in people with type 2 diabetes and mild cognitive impairment. Methods: In a retrospective cohort study using the TriNetX global network, we included adults with T2D, mild cognitive impairment, and HbA1c 10.5% without dementia (2010-2021). We assessed the incidence of dementia or Alzheimer's disease among propensity-score-matched (1:1) individuals initiating GLP-1 RA versus dipeptidyl peptidase-4 inhibitors (DPP4i) during a 5-year potential follow-up. In a sensitivity analysis, we compared those initiating GLP-1 RA versus basal insulin. Results: We included 1320 matched individuals (702 women) who initiated GLP-1 RAs or DPP4i, with a mean age 67.2 years and HbA1c 7.6%. Onset of dementia or Alzheimer's occurred among 101 and 132 participants in the GLP-1 RA and DPP4i groups, respectively (Cox proportional hazards ratio HR 0.74 95% CI, 0.57-0.95; E-value 2.06 lower CI limit 1.28). At a mean follow-up of 3.9 years, the respective Kaplan-Meier-estimated incidences were 15.5% and 20.4%, representing an absolute risk difference of -4.9% 95% CI -9.4 - -0.4 with GLP-1 RAs versus DPP4i and a number needed to treat of 21 11 - 234 individuals to prevent one event. Similar findings were observed when comparing those starting GLP-1 RA versus basal insulin (HR 0.63 0.46 - 0.87). Conclusion: In a multinational cohort of people with T2D and mild cognitive impairment, GLP-1 RA initiation versus DPP4i was associated with a lower dementia incidence over 3.9 years, supporting the need for dedicated clinical trials. Disclosure M. Schechter: None. D.R. Sehtman-Shachar: None. A. Fishkin: None. O. Mosenzon: Employee; Current; Regeneron Pharmaceuticals Inc. Stock/Shareholder; Current; Regeneron Pharmaceuticals Inc. T. Cukierman-Yaffe: Speaker's Bureau; Ended; AstraZeneca, Boehringer Ingelheim International GmbH. Advisory Panel; Ended; Amgen Inc. Speaker's Bureau; Ended; Novo Nordisk. Research Support; Current; Novo Nordisk. Speaker's Bureau; Ended; Medtronic. Research Support; Current; Medtronic. Advisory Panel; Ended; Abbott. Speaker's Bureau; Ended; Sanofi. Research Support; Current; Sanofi. Speaker's Bureau; Ended; Lilly. R. Sharon: None. G. Leibowitz: None. G. Aharon-Hananel: Other - PI on clinical studies, speakers bureau; Current; Nordic Bioscience A/S. Other - PI on clinical studies, advisory, speakers bureau; Current; Eli Lilly and Company, AstraZeneca. Research Support; Current; Sanofi. Research Support; Ended; Bayer AG. Speaker's Bureau; Ended; Abbott.