Neuropathology-basedAPOEgenetic risk score better quantifies Alzheimer’s risk

Abstract

INTRODUCTION:APOEϵ4-carrier status or ϵ4 allele count are included in analyses to account for theAPOEgenetic effect on Alzheimer's disease (AD); however, this does not account for protective effects ofAPOEϵ2 or heterogeneous effect of ϵ2, ϵ3, ϵ4 haplotypes. METHODS: We leveraged results from an autopsy-confirmed AD study to generate a weighted risk score forAPOE(APOE-npscore). We regressed cerebrospinal fluid (CSF) amyloid and tau biomarkers onAPOEvariables from the Wisconsin Registry for Alzheimer's Prevention, Wisconsin Alzheimer's Disease Research Center, and Alzheimer's Disease Neuroimaging Initiative (ADNI). RESULTS: TheAPOE-npscore explained more variance and provided a better model fit for all three CSF measures thanAPOEϵ4-carrier status and ϵ4 allele count. These findings were replicated in ADNI and observed in subsets of cognitively unimpaired participants. DISCUSSION: TheAPOE-npscore reflects the genetic effect on neuropathology and provides an improved method to account for APOE in AD-related analyses.