TL;DR
This study reveals how specific Alzheimer's disease (AD) genetic risk factors influence the microstructure of limbic white matter (WM) tracts, which are crucial for memory. By integrating advanced diffusion MRI techniques with genetic data from a large cohort, the researchers identified key genes and their interactions with cognitive status that impact WM integrity, offering new insights into early AD pathogenesis.
What Problem Does This Paper Address?
Alzheimer's disease (AD) is a devastating neurodegenerative disorder characterized by the deterioration of brain structures, including white matter (WM) tracts essential for neural communication. While AD has a significant genetic component, the precise mechanisms by which these genetic risk factors affect WM microstructure, particularly in the limbic system (critical for memory), remain incompletely understood. Traditional diffusion MRI (dMRI) metrics have limitations in accurately quantifying WM integrity due to factors like free water (FW). This paper aims to bridge this gap by using advanced dMRI techniques and large-scale genetic data to investigate how specific AD genetic risk variants and polygenic risk scores (PGSs) influence limbic WM microstructure, and how these effects are modulated by cognitive status.
Methods
The researchers employed a multi-cohort approach, harmonizing dMRI data from seven well-established studies (ADNI, BIOCARD, BLSA, NACC, ROSMAP, VMAP, and WRAP) comprising 2,614 non-Hispanic White aging adults. A key methodological advancement was the use of free water (FW)-corrected diffusion tensor imaging (DTI) metrics. This bi-tensor model separates extracellular water (FW) from intracellular space, providing a more biologically accurate quantification of WM microstructure. Conventional DTI metrics like fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AxD), and radial diffusivity (RD) were calculated, along with their FW-corrected counterparts (FAFWcorr, MDFWcorr, AXDFWcorr, RDFWcorr).
Genetic data included:
- AD risk variant analysis: 36 specific AD-associated single nucleotide polymorphisms (SNPs) identified from previous Genome-Wide Association Studies (GWASs) were analyzed.
- AD polygenic score (PGS) analysis: A PGS for AD was calculated using GWAS summary statistics, representing an individual's overall genetic predisposition to AD.
Statistical analyses involved linear regression models to assess the main effects of AD risk variants and PGSs on WM microstructure metrics, controlling for age, sex, and principal components of ancestry. Interaction analyses were conducted to examine how cognitive status (cognitively unimpaired vs. cognitively impaired) modulated the effects of genetic factors.
Datasets
The study utilized data from seven large, well-established cohorts:
- Alzheimer's Disease Neuroimaging Initiative (ADNI)
- Biomarkers of Cognitive Decline Among Normal Individuals (BIOCARD)
- Baltimore Longitudinal Study of Aging (BLSA)
- National Alzheimer's Coordinating Center (NACC)
- Religious Orders Study and Rush Memory and Aging Project (ROSMAP)
- Vanderbilt Memory and Aging Project (VMAP)
- Wisconsin Registry for Alzheimer's Prevention (WRAP)
This combined dataset, totaling 2,614 participants, represents a significant effort in harmonizing and leveraging existing longitudinal aging and AD research data. No new datasets were introduced.
Key Insights and Ideas
- Specific genes matter: Not all AD genetic risk variants have the same impact. The study pinpointed specific genes (TMEM106B, PTK2B, WNT3, and APOE) whose variants are distinctly associated with alterations in limbic WM microstructure.
- Cognitive status is a moderator: The effect of certain genetic variants (specifically in MS4A6A) on WM microstructure is not uniform; it depends on an individual's cognitive status. This suggests that the genetic influence on WM might change as cognitive decline progresses.
- Polygenic risk matters too: Beyond individual variants, an individual's overall genetic predisposition to AD (PGS) is linked to WM integrity, particularly affecting free water content in the limbic system.
- Early impact: The findings suggest that these genetic influences on WM microstructure may occur early in the disease process, potentially preceding overt clinical AD symptoms, as indicated by the low percentage of participants with an AD diagnosis at the time of imaging.
Key Results
- Six AD risk variants associated with genes TMEM106B, PTK2B, WNT3, and APOE showed significant associations with limbic WM microstructure.
- Variants in TMEM106B were generally associated with more intact WM microstructure.
- Variants in PTK2B and WNT3 were associated with less intact WM microstructure.
- The APOE variant rs429358 showed mixed effects, impacting free water and radial diffusivity in specific tracts.
- MS4A6A variants showed significant interaction effects with cognitive status on STG MDFWcorr. Specifically, the alternative alleles were associated with reduced MDFWcorr in cognitively impaired individuals but increased MDFWcorr in cognitively unimpaired individuals.
- AD polygenic risk scores (PGSs) were significantly associated with higher free water (FW) and FAFWcorr measures, particularly in the fornix.
- When considering the interaction between AD PGS and cognitive status, cognitively impaired individuals with higher AD PGS showed decreased RDFWcorr but increased AXDFWcorr and FAFWcorr, while cognitively unimpaired individuals showed no significant differences.
Limitations and Future Work
The authors acknowledge several limitations:
- Limited diversity: The study cohort consisted solely of non-Hispanic White individuals of European ancestry, limiting the generalizability of findings to other populations.
- Cross-sectional design: The analyses are correlational and do not establish causality.
- Single-shell dMRI: While enabling a larger cohort, single-shell dMRI is less detailed than multi-shell approaches (like NODDI) for deconvolving diffusion signals.
Future work suggested includes:
- Replicating findings in diverse cohorts to ensure broader generalizability.
- Utilizing multi-shell diffusion models for a more detailed understanding of microstructural features.
- Exploring novel deep learning techniques for improved FW estimation.
- Investigating the interplay between neuroinflammation and WM microstructure using advanced DWI signal analysis.
- Integrating WM neurodegeneration measures into the AT(N) framework for a more comprehensive AD definition.
Summary updated