Impact of semaglutide on cognitive function in patients with type 2 diabetes and mild cognitive impairment: a 24-month observational study.

Type 2 diabetes mellitus (T2DM) is associated with an increased risk of mild cognitive impairment (MCI) and dementia. Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) provide cardiovascular and metabolic benefits, and emerging data suggest neurocognitive effects. We conducted a 24-month prospective observational study of 72 adults (≥ 50 years) with T2DM and MCI treated with semaglutide (subcutaneous [SC] or oral) and assessed at baseline (T0) and every 6 months (T1-T4). Outcomes included anthropometrics, fasting glucose, HbA1c, LDL-c, and neuropsychological tests (Montreal Cognitive Assessment [MoCA], Digit Symbol Substitution Test [DSST], Beck Depression Inventory [BDI]). A control cohort of 60 patients with T2DM and MCI not receiving semaglutide was evaluated with the same protocol. Semaglutide was associated with significant reductions in body weight, BMI, waist circumference, fasting glucose, and HbA1c (all p < 0.05). Cognitive performance improved, as shown by MoCA scores (median Δ≈ + 4 points, p < 0.001) at T4 compared with T0; at T4, semaglutide recipients also had higher MoCA scores than controls (+ 4.0 vs - 1.5, p < 0.001). DSST and BDI scores also improved (p = 0.001 and p = 0.044, respectively). MoCA improved similarly with both formulations, whereas DSST improvement reached significance only with SC semaglutide. In multivariable regression, semaglutide independently predicted MoCA improvement (+ 4.76 points, p < 0.001). No semaglutide-treated patient progressed to dementia versus 14 in controls (p = 0.002). In adults with T2DM and MCI, semaglutide was associated with improved cognition and a lower risk of progression to dementia. Differences between formulations may reflect distinct central nervous system engagement. Randomized trials are warranted.