Alzheimer’s disease polygenic risk score associates with hippocampal subfield atrophy and immune-related genetic mechanism
TL;DR
Specific AD-SNPs enriched in immune-brain axis pathways rather than Aβ related processes, were associated with reduced volumes of CA1 and subiculum prior to dementia onset, which supports that AD-PRS may capture genetic susceptibility to concomitant neurodegenerative diseases frequently misdiagnosed as AD.
Abstract
Introduction Hippocampal atrophy is frequently observed in neurodegenerative diseases such as Alzheimer’s disease (AD) or hippocampal sclerosis of aging (HS-aging). Volume loss in the hippocampus is described as prodromal stage of dementia and has been associated with AD polygenic risk score (PRS). CA1 and subiculum atrophy have been suggested to be a promising in vivo biomarker for HS-aging. Recent studies suggest that some loci associated with AD may be more related to other brain diseases concomitant with AD. We aimed to find which significant single nucleotide polymorphisms (SNPs) in the latest AD genome wide association studies (GWAS) could be potentially related with early atrophy of specific hippocampal subregions related to HS-aging. Materials and methods We used regression models to assess the relation of the AD-PRS, and genome-wide significant AD variants, with CA1 and subiculum volumes assessed by magnetic resonance imaging (MRI) in 1,859 participants without dementia (with mild cognitive impairment or cognitively healthy). Co-regulatory network analyses and over-representation enrichment analyses were conducted to identify biological pathways enriched with co-regulatory networks of genes associated with hippocampal subregion volumes. We meta-analyzed data from seven cohorts to associate their AD-PRS with AD in the presence of concomitant brain pathologies. Results Reduced volumes of CA1 and subiculum show association with higher levels of AD-PRS and with variant rs5848 in GRN. This variant was enriched in immune-related pathways. AD-PRS showed no significant association with AD pathology alone, but was strongly associated with AD in the presence of concomitant neurodegenerative pathologies, with HS-aging showing the largest effect size. Discussion Specific AD-SNPs enriched in immune-brain axis pathways rather than Aβ related processes, were associated with reduced volumes of CA1 and subiculum prior to dementia onset. This supports that AD-PRS may capture genetic susceptibility to concomitant neurodegenerative diseases frequently misdiagnosed as AD. Given that AD-PRS is most strongly associated with AD in cases with HS-aging, and that CA1 and subiculum are promising in vivo biomarker for HS-aging also linked with rs5848 in GRN, our findings are consistent with HS-aging related vulnerability. This insight links early hippocampal subfields atrophy to shared genetic mechanisms and biological pathways of interest.