Tirzepatide versus semaglutide for the prevention of mild cognitive impairment, dementia, and Alzheimer's disease in type 2 diabetes: A real-world, retrospective cohort study.

BACKGROUND

Glucagon-like peptide-1 (GLP-1) receptor agonists have shown promise in managing type 2 diabetes mellitus (T2DM) and provide metabolic and cardiovascular benefits. Their associations with neurocognitive outcomes in clinical populations remain uncertain.

METHODS

We conducted a retrospective, population-based cohort study using de-identified electronic health records from the global TriNetX research network, predominantly comprising U.S.-based healthcare organizations. Adults with T2DM who initiated tirzepatide or semaglutide were included. Propensity score matching (1:1) was applied to balance baseline characteristics, including age, sex, race, and cardiometabolic and psychiatric comorbidities. Incident diagnoses of mild cognitive impairment (MCI), dementia, and Alzheimer's disease (AD) occurring at least 12 months after treatment initiation were assessed using risk ratios (RRs) and time-to-event analyses with Kaplan-Meier estimators.

FINDINGS: A total of 290,606 patients initiated semaglutide and 44,471 initiated tirzepatide. After propensity score matching, each group included 44,470 patients. Compared with semaglutide, initiation of tirzepatide was associated with a lower risk of MCI (RR 0.12; 95% CI 0.06-0.22), dementia (RR 0.15; 95% CI 0.09-0.26), and AD (RR 0.48; 95% CI 0.22-1.01). Absolute risks were low for all outcomes, and separation of survival curves was more consistent for MCI than for dementia or AD.

INTERPRETATION: In this real-world observational analysis of adults with T2DM, initiation of tirzepatide was associated with a lower incidence of MCI compared with semaglutide, whereas associations for dementia and AD were less consistent. These findings should be interpreted descriptively and as hypothesis-generating. Prospective randomized trials with longer follow-up and systematic neurocognitive assessment are needed to clarify whether, and under which conditions, incretin-based therapies influence neurocognitive outcomes.