Testing for association between exonic glucagon-like peptide 1 receptor mutation with physical and brain health traits in UK Biobank.

Abstract

A low-frequency exonic single nucleotide variation (SNV) in the GLP-1R gene, rs10305492, has been associated with differences in cardiometabolic health. There are well-established links between cardiometabolic and cognitive health, e.g. on neuropsychological tests, brain structure and/or risk of incident dementia. There is however a lack of data regarding rs10305492 and brain health metrics; this study aimed to test for potential associations. We tested for association between rs10305492 A allele presence, vs. cognitive, brain imaging and dementia-related outcomes based on self-report and incident cases from hospital episode statistics, in UK Biobank (n range=47,920 to 395,475). We controlled for covariates of age, sex, apolipoprotein (APOE) e4 genotype, 8 principal components (PCs) for population stratification, genotyping chip, and smoking history. We used linear and logistic regressions for continuous/binary outcomes (standardized betas and odds ratios [OR] respectively; A allele presence vs. absence). Despite significant associations with better physical health traits, there were generally no rs10305492-A associations with scores on cognitive tests, structural brain phenotype metrics or all-cause dementia, Alzheimer's disease or vascular dementia (P>0.05). There was a nominally significant protective association with lower log white matter hyperintensity volumes (standardized beta = -0.07, 95% confidence intervals = -0.13 to -0.01, P=0.028). These findings support clinical trial data in providing orthogonal, genetics-based evidence suggesting efficacy for GLP-1R agonists on physical health; but not necessarily directly for improved brain health including protection against dementia. Ongoing clinical trials are needed to provide more robust evidence. This article is protected by copyright. All rights reserved.