Published Exchange in Brain Disputes Whether a Genetic Marker Predicts GLP-1 Drugs' Effect on Parkinson's

The neurology journal Brain published a cohort study and an accompanying scientific exchange in its October 2026 print issue (dated October 5, 2026) that together test, and partly undercut, a genetic argument for why GLP-1 receptor agonists might protect against Parkinson's disease.

The original study, led by Kyung-A Woo of Harvard Medical School and Seoul National University Hospital, used the National Institutes of Health's All of Us Research Program to build a cohort of 7,039 adults with type 2 diabetes, age 50 or older, who started a GLP-1 receptor agonist between 2005 and 2023 and had no prior diagnosis of Parkinson's disease or related disorders, or of Alzheimer's disease and related dementias. Rather than measuring GLP-1 receptor expression directly, the researchers used a genetic proxy: a 15-variant cis-expression quantitative trait locus (cis-eQTL) genetic risk score associated with higher systemic expression of GLP1R, the gene encoding the GLP-1 receptor, dichotomized into high- and low-score groups of roughly equal size (3,520 versus 3,519 participants). Over an average follow-up of 3.8 years, the high-score group had a lower rate of incident Parkinson's disease and related disorders than the low-score group, with a hazard ratio of 0.78 (95% confidence interval 0.62 to 0.98) and an estimated rate difference of -1.36 cases per 1,000 person-years (95% CI -2.51 to -0.20). The association held up in sensitivity analyses that excluded early outcome events and short drug exposure, but there was no corresponding association for incident Alzheimer's disease and related dementias (hazard ratio 1.02, 95% CI 0.81 to 1.28).

In the same issue, a team led by Rebecca Gurney at UCL's Queen Square Institute of Neurology, which ran the Exenatide-PD3 trial, published a letter testing whether that same genetic score could predict clinical response to a GLP-1 drug in people who already had Parkinson's disease. Exenatide-PD3 (ClinicalTrials.gov identifier NCT04232969) was a completed, randomized, placebo-controlled trial that assigned 194 participants with diagnosed Parkinson's disease to once-weekly exenatide or placebo for 96 weeks and, as previously reported, found no significant difference between the two groups on its primary motor-symptom outcome or on secondary clinical outcomes. Applying the identical 15-variant GLP1R expression genetic instrument, using the same expression quantitative trait locus weights from the GTEx v8 reference panel that Woo's group had used, to the 119 European-ancestry trial participants with complete genetic and clinical data, Gurney and colleagues found no significant interaction between genetic risk score group and treatment assignment, meaning the score did not identify which participants were more likely to benefit from exenatide.

Woo and colleagues published a reply in the same issue arguing the two findings are not necessarily in conflict. They noted that their own cohort consisted of people with type 2 diabetes who had not yet developed Parkinson's disease, while Exenatide-PD3 enrolled people with clinically established disease, and wrote that the absence of a genetic-score-by-treatment interaction in the trial "is not in tension" with their own observation. They proposed that if GLP-1 receptor agonists provide any neuroprotective effect through metabolic stabilization, that effect could be strongest in metabolically vulnerable people before or during early, prodromal disease, and weaker once Parkinson's has become clinically manifest, which would make a systemic GLP1R expression proxy like theirs "less informative" for predicting treatment response in already-diagnosed patients.

Both sides frame their work as hypothesis-generating rather than conclusive. The original cohort study is an observational genetic-epidemiology analysis, not a randomized trial, and cannot establish that GLP-1 receptor agonists prevent Parkinson's disease; its authors describe their genetic risk score as a tool that may help design future prevention trials, not as evidence ready to guide patient care. The Exenatide-PD3 trial itself found no clinical benefit of exenatide for people with already-diagnosed Parkinson's disease, a result this exchange does not change. Exenatide is approved by the FDA, marketed as Byetta and Bydureon, for the treatment of type 2 diabetes; it is not approved by the FDA, the European Medicines Agency, or Australia's Therapeutic Goods Administration for Parkinson's disease or any neurological condition. This article describes published research findings and scientific debate, not treatment or dosing guidance; any decisions about diabetes or Parkinson's disease care should be made with a qualified clinician.