Glucagon-like peptide-1 receptor activation and mental health: a drug-target mendelian randomization study

Concerns have been raised about the psychiatric safety of glucagon-like peptide-1 receptor (GLP-1R) agonists, but trial evidence suggests that they ameliorate depressive symptoms. We aimed to assess the associations of GLP-1R activation with mental health well-being and the risk of mental health disorders and substance use disorders. We performed drug-target Mendelian randomization and colocalization analyses using the largest relevant genome-wide association studies and replicated in FinnGen. After correcting for multiple comparisons, genetically predicted lower body mass index (BMI) via GLP-1R activation was associated with a better well-being spectrum (0.06 standard deviation [95% confidence interval 0.03-0.08]), lower risk of major depressive disorder (odds ratio 0.82 [0.75-0.88]), and lower risk of bipolar disorder (odds ratio 0.61 [0.47-0.79]) per 1-kg/m² decrease in BMI. There was also suggestive evidence that genetically predicted lower BMI via GLP-1R activation was associated with lower risk of substance use disorders. These associations were stronger than the associations for genetically predicted lower BMI and lower glycated hemoglobin (HbA1c) based on genome-wide variants. The posterior probabilities of colocalization of BMI and each significant outcome within 1 Mb of the GLP1R gene were 59.3% for the well-being spectrum, 76.9% for major depressive disorder, and 45.9% for bipolar disorder. The posterior probabilities of colocalization were > 80% for the well-being spectrum and bipolar disorder when conditioning on the presence of a variant associated with the outcome. This study provides genetic evidence that GLP-1R activation is associated with better mental health well-being, lower risk of major depressive disorder, and lower risk of bipolar disorder, possibly beyond its effect on BMI and HbA1c.