Association Between Polygenic Scores for Brain Structure and Alzheimer's Disease

Alzheimer’s disease (AD) is a progressive neurodegenerative condition and the most common type of dementia. Research has shown that volumetric changes in subcortical brain structures, which make up roughly 25% of the brain's total volume and affect many physiological functions. This research project aims to investigate the genetic association between brain structures and AD. GWAS data of eight different brain structures (ICV, accumbens, amygdala, caudate, hippocampus, pallidum, putamen, and thalamus) were used to create polygenic scores (PGSs) in a discovery (cases/controls = 2,651/2,768) and replication dataset (cases/controls = 553/644). We utilized PRSice-2 software to generate PGSs and assess genetic correlations using two models: one adjusted for principal components (PC1 and PC2), and another incorporating sex, age, and APOE e4 allele count. Meta-analyses were conducted for discovery and replication samples using random and fixed-effect models with p<0.05 being significant. Mendelian Randomization (MR) analyses were conducted between brain structures and AD. The PGSs of three brain structures–amygdala, hippocampus, and pallidum–were significantly associated (p<0.0133) with AD in the fixed effects model, showing that a higher volume of these structures is inversely correlated with AD. However, the association did not remain significant for the hippocampus after being adjusted for covariates, suggesting that the association between the hippocampus and AD may depend on these covariates. MR analysis indicated hippocampal volume had an inverse causal effect on the development of AD, underscoring the critical role of the hippocampus in the disease's progression.