Can glucagon like peptide-1 receptor agonists be beneficial for brain health?
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background
GLP-1 receptor agonists (GLP-1 RAs) are widely used for treatment of type 2 diabetes and obesity and have demonstrated cardiovascular benefits, but their effect on dementia risk is uncertain. We used a drug-target Mendelian randomisation (MR) framework to estimate the causal effect of GLP-1 receptor (GLP-1R) agonism on dementia risk and brain structure.
Methods
We used single nucleotide polymorphisms (SNPs) in GLP1R to recapitulate the effect of GLP1-RAs against circulating markers of glucose, glycated haemoglobin and outcomes of type 2 diabetes and obesity, to understand if we can reliably proxy agonism at GLP1R using gene variants. We then considered clinical outcomes, including all-cause dementia, vascular dementia, Alzheimer’s disease, and neuroimaging outcomes. Analyses were conducted in UK Biobank and replicated in FinnGen and All of Us and results were combined using meta-analysis.
Findings
Genetically proxied GLP-1R agonism was associated with a 17% lower risk of vascular dementia, with directional consistency across the two largest cohorts. No consistent association was observed for all-cause dementia. An elevated risk of Alzheimer’s disease was observed in UK Biobank but was not replicated in FinnGen or All of Us, and the pooled estimate across cohorts was null. There was no consistent evidence of an effect on neuroimaging outcomes, though higher GLP-1R agonism was associated with greater total brain volume.
Interpretation
Genetic evidence supports a potential protective effect of GLP-1R agonism on vascular dementia, consistent with the established cardio-metabolic benefits of this drug class. The null pooled finding for Alzheimer’s disease suggests that GLP-1R agonism does not meaningfully modify neurodegenerative pathways specific to this disease, though results from ongoing clinical trials are awaited. These findings highlight the importance of distinguishing between dementia subtypes when evaluating the cognitive effects of cardiometabolic therapies.
Funding
This work is funded by Diabetes Research & Wellness Foundation, Professor David Matthews Non-Clinical Fellowship to VG (ref: SCA/01/NCF/22). RS is an NIHR Research Professor, NIHR303160 is funded by the NIHR for this research project. GB is funded by an MRC grant.
Panel: Research in context
Evidence before this study
Evidence of a protective effect of GLP1R agonism in dementia to date is limited to observational studies and early-phase clinical trials. A large US-based target trial emulation study reported an association between semaglutide (a GLP-1 RA) use and reduced first-time Alzheimer’s disease diagnosis in patients with Type 2 diabetes, compared to insulin. Phase 2 randomised trials of GLP-1 RAs in Alzheimer’s disease have shown limited efficacy, while the phase 3 evoke and evoke+ trials are yet to report. No prior Mendelian Randomisation study has examined the causal effect of GLP-1R agonism on dementia risk or brain structure.
Added value of this study
This study provides the first genetic evidence on the potential causal relationship between GLP-1R agonism and dementia. By distinguishing between dementia subtypes and replicating findings across three large independent cohorts, we show that GLP-1R agonism may have a causal effect on vascular dementia but there is no consistent evidence of an effect on Alzheimer’s disease. The use of neuroimaging outcomes alongside clinical dementia outcomes strengthens the investigation of potential biological mechanisms.
Implications of all the available evidence
The available evidence suggests that GLP-1R agonism may reduce vascular dementia risk through its established cardio-metabolic effects, and that this dementia subtype should be prioritised in future clinical trials. The consistent null finding for Alzheimer’s disease across all studies suggests that neurodegenerative pathways specific to this dementia subtype are unlikely to be modified by GLP-1R agonism.
