Basic Science and Pathogenesis.

Background: Several studies have indicated sex-specific genetic risk for Alzheimer's disease (AD), but these were centered on non-Hispanic White individuals of European ancestry. We sought to identify sex-specific genetic variants for AD in non-Hispanic and Hispanic subjects of admixed African ancestry.

Method: Participants were ages 60+, of African ancestry (≥25%), and diagnosed as cases or controls. Genetic data were available from SNP arrays imputed to TOPMed or whole genome sequencing (WGS). Genome-wide association studies (GWAS) were performed per data type, sex, and Hispanic/non-Hispanic status (Table-1), requiring genotyping rate ≥80% and minor allele frequency ≥1%, followed by meta-analysis (Plink v2.0; GWAMA v2.2.2). A sex-by-SNP interaction model was also evaluated. GWAS performed multiple linear regression on an AD-age score (Le Guen & Belloy et al. 2021; Figure-1), adjusting for APOE*4/APOE*2 dosage, the first five genetic principal components (PC-AiR; GENESIS; R v3.6), and array/sequencing center. Variants were filtered to be covered on at least half of the individuals per stratum.

Result: We found 3 genome-wide significant loci in men (Figure-2) with lead variants 1) rs74853649 ∼85kb downstream of FHIT, 2) rs73411425 in the PILRA/ZCWPW1/NYAP1 AD risk locus, and 3) rs3736450 intronic on WWOX. Despite that rs73411425 is within a known locus, the variant is novel. Notably, all three variants showed significant sex-by-SNP interactions and were specific to African rather than European ancestry (Table-2). Further, in men, we also observed a suggestive association intronic on RBFOX1, which was previously implicated as a risk gene harboring variants associated with higher amyloid burden in brain (Raghavan et al. 2020), with strongest effect sizes in African-American individuals (those samples were independent of the current data).

Conclusion: The study is an initial exploration of sex-specific genetic effects in African ancestry individuals and warrants replication in larger data. Importantly, we show the relevance of expanding beyond non-Hispanic white samples for sex dimorphism research, identifying male-specific risk variants that have elevated frequencies in African ancestry, while being rare in other ancestries. In ongoing work, we are performing cross-ancestry meta-analyses of sex-stratified GWAS followed by sex-stratified omics integration to increase power to uncover sex-specific genes relevant to AD.