Browse by author
Lookup NU author(s): Dr Judith HarrisonORCiD
This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0).
Copyright © 2025 Harrison, Foley, Simmonds, Bracher-Smith, Holmans, Stergiakouli, Caseras, Escott-Price and Jones.Introduction: Alzheimer’s disease involves progressive white matter microstructural degeneration that may precede clinical symptoms by decades. While polygenic risk scores (PRS) quantify cumulative genetic liability for AD, genome-wide PRS lack mechanistic specificity. We tested whether pathway-specific PRS, targeting areas of biology including tau binding, lipid metabolism, and immune response, are differentially associated with diffusion MRI measures across the lifespan. Methods: We analyzed two population-based cohorts: the Avon Longitudinal Study of Parents and Children (ALSPAC; mean age = 19.8 years, n = 517) and UK Biobank (mean age = 64.2 years, n = 18,172). Genome-wide and nine pathway-specific PRS for Alzheimer’s disease were constructed using GWAS summary statistics and a clumping threshold of r2 < 0.2 at p < 0.001. Diffusion MRI data were processed separately within each cohort: in ALSPAC, tract-based fractional anisotropy (FA) and mean diffusivity (MD) were extracted using probabilistic tractography from native-space regions of interest; in UK Biobank, diffusion metrics were derived from TBSS-aligned skeletons and standard atlas-based ROIs. Analyses focused on three tracts vulnerable to early AD pathology: the dorsal cingulum, parahippocampal cingulum, and fornix. Multiple linear regression models were used to assess PRS associations with FA and MD, adjusting for demographic, scanner, and genetic ancestry covariates. False discovery rate correction addressed multiple comparisons, and sensitivity analyses were performed excluding the APOE region. Results: In UK Biobank, higher PRS for protein–lipid complex assembly and tau protein binding were robustly associated with lower fractional anisotropy and higher mean diffusivity in both dorsal and parahippocampal cingulum segments (False discovery rate-corrected p < 0.05), explaining more variance than APOE alone; no significant effects emerged in the fornix. Genome-wide PRS showed weaker, non-significant associations. In ALSPAC, no PRS metric survived FDR correction, though nominal trends appeared in the dorsal cingulum. Sensitivity analyses confirmed that key cingulum associations in older adults persisted after omitting APOE. Conclusion: Pathway-specific polygenic risk for Alzheimer’s disease manifests in white matter microstructure by mid- to late adulthood but not in early adulthood, suggesting an age-dependent emergence of genetic effects. dMRI phenotypes may thus serve as intermediate biomarkers for dissecting mechanistic pathways of preclinical Alzheimer’s disease vulnerability.
Author(s): Harrison JR, Foley SF, Simmonds E, Bracher-Smith M, Holmans P, Stergiakouli E, Caseras X, Escott-Price V, Jones DK
Publication type: Article
Publication status: Published
Journal: Frontiers in Neuroscience
Year: 2025
Volume: 19
Online publication date: 28/10/2025
Acceptance date: 18/09/2025
Date deposited: 25/11/2025
ISSN (print): 1662-4548
ISSN (electronic): 1662-453X
Publisher: Frontiers Media SA
URL: https://doi.org/10.3389/fnins.2025.1638503
DOI: 10.3389/fnins.2025.1638503
Data Access Statement: Publicly available datasets were analyzed in this study. This data can be found here: The datasets analyzed in this study are accessible upon request from ALSPAC and UK Biobank. Both studies provide comprehensive, searchable data dictionaries and variable search tools on their respective websites to support data discovery (http://www.bristol.ac.uk/alspac/researchers/ourdata/, https://www.ukbiobank.ac.uk/)
Altmetrics provided by Altmetric
