Associations of pathway‐specific polygenic risk scores with clinical status and Alzheimer’s‐related biomarkers

Abstract

AbstractBackgroundThe APOE gene represents the largest genetic risk factor in sporadic Alzheimer’s disease (AD), but there is a substantial polygenic component as well. Polygenic risk scores (PRS) represent a useful approach to summarize small effects across the genome but may obscure which pathways are affected. Variability across individuals at the genetic level may contribute to the extensive phenotypic heterogeneity of AD. Therefore, we calculated polygenic risk in multiple pathways associated with AD and examined its relationship with clinical status and AD biomarkers.MethodA total of 1,411 participants from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) with genotyping data were included. Top gene sets from a pathway analysis of AD GWAS summary statistics were combined into clusters based on proportion of overlapping genes. Pathway‐specific PRSs (ps‐PRS) were calculated for each participant by restricting to SNPs annotated to genes in each cluster. We tested associations of ps‐PRSs with diagnostic status (AD vs cognitively normal, MCI vs cognitively normal) and rate of progression to AD among those with no dementia at baseline. Additional analyses examined associations with amyloid and tau positivity, and hippocampal volume. The APOE region was excluded from all ps‐PRSs and the number of APOE‐ε4 alleles was included as a covariate along with age, gender, and the first 3 principal components. All analyses were FDR‐corrected.ResultThirteen pathway clusters were identified relating to categories such as immune response, amyloid precursor processing, protein localization, lipid transport and binding, tyrosine kinase, and endocytosis. Nine ps‐PRSs were significantly higher in AD and MCI compared to controls. A large subset of these were associated with both faster progression to AD and hippocampal volume. Amyloid and tau‐positivity were both related to endocytosis ps‐PRSs. Amyloid‐positivity was additionally related to microglial activation, lipid processing, fibril formation and amyloid precursor processing, whereas tau‐positivity was related to protein localization.ConclusionAlthough genetic risk for AD is widely distributed, AD‐phenotypes may be preferentially associated with risk in specific pathways. Defining genetic risk along multiple dimensions at the individual level may help clarify the etiology of heterogeneity in AD.