Abstract
IMPORTANCE: APOE*4 is the strongest, common genetic risk factor for late-onset Alzheimer's disease (AD) with widespread and cell-type-specific impact on disease pathogenesis. Genetics and omics studies can help identify genes that counteract the effects of APOE*4, but so far have remained relatively small and crucially did not assess genetic findings through a cell-type-specific framework. OBJECTIVES: Perform a large-scale APOE*4 stratified genome-wide association study (GWAS) of AD integrated with genetically tethered cell-type-specific multi-omics data. DESIGN: Meta-analysis of APOE*4 stratified AD GWAS in case-control, family-based, population-based, and longitudinal AD-related cohorts from the Alzheimer's Disease Genetics Consortium, Alzheimer's Disease Sequencing Project, and UK Biobank. Integration of GWAS with brain cell-type-specific genetic regulation of gene expression data, from the Religious Orders Study and Memory and Aging Project, to identify APOE*4 and cell-type-specific AD genes. Cell-type-specific multi-omics gene prioritization followed by compound and drug repurposing. Data analyzed between January 2023 and September 2025. SETTING: Genetic data available from high-density single-nucleotide polymorphism (SNP) microarrays and whole-genome sequencing (WGS). Single-nucleus (sn) RNA-seq data from dorsolateral prefrontal cortex. PARTICIPANTS: 567,521 eligible participants for AD genetic association studies were selected from referred and volunteer samples, of which 119,852 were excluded for analysis exclusion criteria. MAIN OUTCOME AND MEASURES: APOE*4 and cell-type-specific gene Z-scores and FDR-corrected P-values. Gene prioritization scores and APOE*4 stratified enriched compounds. RESULTS: 67 and 17 significant cell-type-gene pairs were identified in APOE*4 non-carriers (APOE*4-) and carriers (APOE*4+) respectively. Oligodendrocytes displayed the largest proportion of APOE*4+ genes. 46 cell-type-gene pairs were supported by at least half of the gene prioritization analyses. Several prioritized genes were druggable and displayed enrichment of APOE*4 stratified drugs or compounds. Top APOE*4+ genes with connections to enriched drugs or compounds included TNS3 (astrocytes), CISD1 and SLC23A2 (oligodendrocytes), and UBXN4 (excitatory neurons). CONCLUSION AND RELEVANCE: We identified a set of APOE*4 stratified genes that may be causal for AD through brain cell-type-specific mechanisms and prioritized top genes for further interrogation. We additionally identified compounds that may be repurposed or shed light on therapeutic avenues for treating AD based on an individual's APOE*4 status. Top identified compounds such as Hydrocortisone and Trolox implicate oxidative stress and neuroinflammation as potential biological targets in APOE*4+ individuals.