Genome-Wide Association Analysis across Endophenotypes in Alzheimer's Disease: Main Effects and Disease Stage-Specific Interactions

Abstract

The underlying genetic susceptibility for Alzheimer's disease (AD) is not yet fully understood. The heterogeneous nature of the disease challenges genetic association studies. Endophenotype approaches can help to address this challenge by more direct interrogation of biological traits related to the disease. AD endophenotypes based on amyloid-β, tau, and neurodegeneration (A/T/N) biomarkers and cognitive performance were selected from the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort (N = 1565). A genome-wide association study (GWAS) of quantitative phenotypes was performed using an SNP main effect and an SNP by Diagnosis interaction (SNP × DX) model to identify disease stage-specific genetic effects. Nine loci were identified as study-wide significant with one or more A/T/N endophenotypes in the main effect model, as well as additional findings significantly associated with cognitive measures. These nine loci include SNPs in or near the genes APOE, SRSF10, HLA-DQB1, XKR3, and KIAA1671. The SNP × DX model identified three study-wide significant genetic loci (BACH2, EP300, and PACRG-AS1) with a neuroprotective effect in later AD stage endophenotypes. An endophenotype approach identified novel genetic associations and provided insight into the molecular mechanisms underlying the genetic associations that may otherwise be missed using conventional case-control study designs.

Keywords: APOE; FDG-PET; GWAS; amyloid-PET; cerebrospinal fluid biomarkers; endophenotype; genetic interaction; genetics; magnetic resonance imaging.

Figure 1

Matrix of main effect analysis results. Each row indicates the top SNP for a genetic region after LD trimming, and each column represents an AD endophenotype. Ordered based on minimum p-value across the row. The asterisks represent the p-value of the association with [***] indicating meeting the study-wide significant threshold (p ≤ 8.33 × 10⁻⁹), [**] the conventional genome-wide threshold (p ≤ 5 × 10⁻⁸), and [*] the suggestive association threshold (p ≤ 1 × 10⁻⁵). The color and box size relate to the β value effect size for a given association, with larger box size relating to distance from zero in either positive (blue, suggesting neuroprotective) or negative (red, suggesting neuropathological effect) direction. ^† SNP retains significance when including DX as a covariate.

Figure 2

Matrix of SNP x Diagnosis analysis results. Each row indicates the top SNP for a genetic region after LD trimming, and each column represents an AD endophenotype. Endophenotypes showing no level of significance were removed for clarity. The asterisks represent the p-value of the association with [***] indicating meeting the study-wide significant threshold (p ≤ 8.33 × 10⁻⁹), [**] the conventional genome-wide threshold (p ≤ 5 × 10⁻⁸), and [*] the suggestive threshold (p ≤ 1 × 10⁻⁵). The color and box size relate to the β value effect size for a given association, with larger box size relating to distance from zero in either positive (blue, suggesting neuroprotective) or negative (red, suggesting neuropathological effect) direction.

Figure 3

Violin and boxplot distribution of Parietal Lobe Cortical Thickness, stratified by (A) rs1065272 SNP, (B) Diagnosis, and (C) SNP and Diagnosis. (A) represents the main effect analysis, (B) the association with Diagnosis, and (C) the SNP × DX interaction association, with DX codes as an ordinal logistic variable of distinct categories with known ordinal relation (CN < EMCI < LMCI < AD). The asterisks represent the p-value of the association with [***] indicating meeting the study-wide significant threshold (p ≤ 8.33 × 10⁻⁹).