Abstract / Summary
Individual brain aging is shaped by both genetic and environmental factors, yet their combined influence remains poorly understood. Using multimodal MRI, genomic data, and lifestyle measures from 46,843 participants in the UK Biobank, we developed and validated a multimodal brain age prediction model (R² = 0.72-0.75). Genome-wide association analyses identified 19 independent loci and 122 mapped genes associated with brain aging. Through cross-sectional and longitudinal analyses, we found that higher genetic liability and unhealthy lifestyle predicted the acceleration of brain aging. Moreover, linkage disequilibrium score regression revealed a significant negative genetic correlation between lifestyle and brain aging. A genome-wide gene–lifestyle interaction analysis identified a significant locus at rs62287096 upstream of HRASLS , where risk-allele carriers showed amplified effects of poor lifestyle but reduced brain aging under healthy lifestyle conditions. Supporting a regulatory role for this locus, dual-luciferase reporter assays demonstrated significant enhancer activity with allele-specific effects at rs62287096. These findings highlight modifiable pathways that may offset genetic vulnerability and inform precision strategies for healthy brain aging.