GLP-1 Receptor Agonists vs. SGLT2 Inhibitors and the Risk of Alzheimer’s Disease in Type 2 Diabetes: A Systematic Review and Meta-Analysis
Levi Silva1, Carolina Moura3, Maria Valverde4, Anderson Silva5, Luana Braz2, Kauã Silva2, Emanuel Aguiar2, Vitor Volta2, Clara de Oliveira2, Filipi Andreão2, Henrique Cal6, Diogo Santos7
1Neurology, Federal University of Rio de Janeiro, 2Federal University of Rio de Janeiro, 3Neurology, Santa Casa de São Paulo, 4Neurology, Faculty of Medical Sciences and Health of Juiz de Fora- FCMS, 5Physiology and Pharmacology, Federal University of Pernambuco, 6Neurology, Faculty of Medical Sciences, Faculdade Mar Atlântico, 7Neurology, University of São Paulo, USP
Objective:
To evaluate whether the use of GLP-1 receptor agonists (GLP-1RAs) and SGLT2 inhibitors (SGLT2i) reduces the risk of Alzheimer’s disease (AD) and dementia in patients with type 2 diabetes (DM2).
Background:
People with DM2 have an increased risk of AD. GLP-1RAs and SGLT2i have emerged as potential interventions to prevent AD and dementia and help improve the knowledge of its physiopathology.
Design/Methods:
PubMed, Embase and Cochrane were searched for randomized controlled trials (RCTs), cohort, and case-control studies between January 1, 2006, and April 15, 2025, comparing GLP-1RA and/or SGLT2i in adults with DM2 without cognitive impairment at baseline. A total of 15 studies met the inclusion criteria (1 RCT and 14 cohort studies). The primary outcome was AD and dementia, assessed through cognitive scores and effect measures such as hazard ratio (HR). Secondary outcomes were HbA1C levels and their relation to cognitive protection. A regression analysis was performed to assess the variables that could explain the result obtained.
Results:
GLP-1RA and SGLT2i consistently demonstrated a reduced risk of AD (HR = 0.84; 95% CI: 0.75–0.94) and dementia (HR = 0.80; 95% CI: 0.72–0.87) compared to other glucose-lowering therapies, with moderate/high heterogeneity and moderate certainty. Also, no statistically difference was assessed between a direct comparison of GLP-1RA and SGLT2i (HR =1.13; 95% CI, 0.76–1.68), and higher baseline HbA1C levels could suggest a lower HR for AD risk.
Conclusions:
These findings reinforce the potential of GLP-1RAs and SGLT2i as therapeutic options for slowing cognitive decline in DM2, depending on glycemic modulation.
Disclaimer: Abstracts were not reviewed by Neurology® and do not reflect the views of Neurology® editors or staff.