Free-water: A promising structural biomarker for cognitive decline in aging and mild cognitive impairment

Aditi Sathe, Yisu Yang, Kurt G. Schilling, Niranjana Shashikumar, Elizabeth E. Moore, Logan Dumitrescu, Kimberly R. Pechman, Bennett A. Landman, Katherine A. Gifford, Timothy J. Hohman, Angela L. Jefferson, Derek B. ArcherImaging 2024 · DOI

TL;DR

This study introduces "free-water" (FW) diffusion MRI metrics as a more sensitive biomarker for predicting cognitive decline in aging and Alzheimer's disease (AD) compared to traditional structural MRI measures like hippocampal volume. FW, particularly in the hippocampus, significantly explains unique variance in both cross-sectional and longitudinal cognitive performance, especially for memory decline.

What Problem Does This Paper Address?

Alzheimer's disease (AD) is a major cause of dementia, and identifying individuals at risk or experiencing early decline is crucial for timely intervention. Traditional neuroimaging, like measuring hippocampal volume, has been valuable but may not capture the earliest microstructural changes. Diffusion MRI (dMRI) offers a way to probe these microstructural changes, but conventional metrics can be confounded by factors like cerebrospinal fluid (CSF) contamination. This paper addresses the need for more sensitive and specific biomarkers for early detection and monitoring of cognitive decline in aging and AD by exploring the utility of "free-water" (FW) corrected dMRI metrics.

Methods

The researchers utilized a bi-tensor modeling approach to derive "free-water" (FW) corrected diffusion MRI metrics. This method separates the signal into contributions from intracellular water and extracellular free water, aiming to provide a more biologically specific measure of tissue microstructure. Specifically, they calculated:

  • Hippocampal FW: Free-water content within the hippocampus.
  • FW-corrected Fractional Anisotropy (FAFWcorr): A measure of water diffusion directionality, corrected for free-water contamination.
  • AD Signatures (Schwarz and McEvoy): Composite scores derived from cortical thickness and/or hippocampal volume in regions known to be affected by AD. FW and FAFWcorr were also quantified within these AD signature regions.

They then used linear regression for cross-sectional associations between these biomarkers and cognitive performance, and linear mixed-effects regression for longitudinal associations. Competitive model analyses were performed to determine the unique variance explained by each biomarker beyond standard covariates (age, sex, education, etc.).

Datasets

The study used data from the Vanderbilt Memory & Aging Project (VMAP). This was a longitudinal observational study including:

  • Participants: 296 older adults (aged 60+) with either cognitively unimpaired (CU) status or mild cognitive impairment (MCI).
  • Observations: 870 MRI and neuropsychological assessment sessions across up to four time points (baseline, 18 months, 3 years, 5 years).
  • Demographics: Baseline age of 73 ± 7 years, with 40% diagnosed with MCI.

Key Insights and Ideas

  • FW as a Superior Biomarker: The study strongly suggests that free-water (FW) metrics, particularly hippocampal FW, are more sensitive to cognitive decline than traditional structural measures like hippocampal volume.
  • Microstructural Sensitivity: FW appears to capture subtle microstructural changes that precede overt atrophy, making it a promising early biomarker for AD.
  • Beyond Traditional Metrics: The FW correction method helps to disentangle tissue microstructure from CSF contamination, leading to more biologically specific and interpretable results.
  • AD Signatures Enhanced by FW: Incorporating FW into AD signature regions (Schwarz and McEvoy) also improved their predictive power for cognitive performance.

Key Results

  • Cross-sectional Memory Performance: Hippocampal volume, hippocampal FW, and both Schwarz and McEvoy AD signatures (including their FW components) were all significantly associated with memory performance at baseline.
  • Unique Variance Explained (Baseline):
    • For memory, the McEvoy AD Signature explained the most unique variance beyond covariates (ΔR<sub>adj</sub><sup>2</sup> = 3.47 ± 1.65%).
  • For executive function, the Schwarz FW metric explained the most unique variance (ΔR<sub>adj</sub><sup>2</sup> = 2.43 ± 1.63%).
  • Longitudinal Memory Decline: Hippocampal FW was the most significant predictor of memory decline over time, explaining substantial unique variance (ΔR<sub>adj</sub><sup>2</sup> = 8.13 ± 1.25%) and outperforming all other biomarkers examined (p<sub>FDR</sub> = 1.95 x 10<sup>-11</sup>).
  • Longitudinal Executive Function: Hippocampal FW also showed the lowest FDR-corrected p-values as a predictor for executive function composite scores.
  • Interaction with Age: Most neuroimaging biomarkers showed significant interactions with aging in predicting longitudinal cognitive scores, with hippocampal FW × age interactions yielding the lowest FDR-corrected p-values for several cognitive domains.

Limitations and Future Work

  • Cohort Specificity: The VMAP cohort was well-educated and predominantly non-Hispanic White. Future studies should incorporate more diverse cohorts to ensure generalizability.
  • Multi-cohort Studies: While the VMAP cohort is large, multi-cohort studies would enhance statistical robustness.
  • Single-Shell dMRI: The study used single-shell dMRI data. Future work could leverage multi-shell dMRI acquisitions for more advanced multi-compartment analyses.
  • Gray Matter Complexity: The study examined FAFWcorr in gray matter, which has a complex architecture. Future research may clarify the interpretation of these findings.
  • Clinical Staging: Further work is needed to examine changes in global FW values and their association with AD across different stages of MCI (mild, moderate, severe) to assess FW-DTI's utility in screening and diagnosis.
  • Genetic and White Matter Tracts: Ongoing research by the team is exploring the role of specific white matter tracts and genetic factors in conjunction with gray matter changes.

Summary updated

Abstract

Diffusion MRI derived free-water (FW) metrics show promise in predicting cognitive impairment and decline in aging and Alzheimer's disease (AD). FW is sensitive to subtle changes in brain microstructure, so it is possible these measures may be more sensitive than traditional structural neuroimaging biomarkers. In this study, we examined the associations among FW metrics (measured in the hippocampus and two AD signature meta-ROIs) with cognitive performance, and compared FW findings to those from more traditional neuroimaging biomarkers of AD. We leveraged data from a longitudinal cohort (n<sub>participants</sub> = 296, n<sub>observations</sub> = 870, age at baseline: 73 ± 7 years, 40% mild cognitive impairment [MCI]) of older adults who underwent serial neuropsychological assessment (episodic memory, information processing speed, executive function, language, and visuospatial skills) and brain MRI over a maximum of four time points, including baseline (n = 284), 18-month (n = 246), 3-year (n = 215), and 5-year (n = 125) visits. The mean follow-up period was 2.8 ± 1.3 years. Structural MRI was used to quantify hippocampal volume, in addition to Schwarz and McEvoy AD Signatures. FW and FW-corrected fractional anisotropy (FAFWcorr) were quantified in the hippocampus (hippocampal FW) and the AD signature areas (Schwarz<sub>FW</sub>, McEvoy<sub>FW</sub>) from diffusion-weighted (dMRI) images using bi-tensor modeling (FW elimination and mapping method). Linear regression assessed the association of each biomarker with baseline cognitive performance. Additionally, linear mixed-effects regression assessed the association between baseline biomarker values and longitudinal cognitive performance. A subsequent competitive model analysis was conducted on both baseline and longitudinal data to determine how much additional variance in cognitive performance was explained by each biomarker compared to the covariate only model, which included age, sex, race/ethnicity, apolipoprotein-ε4 status, cognitive status, and modified Framingham Stroke Risk Profile scores. All analyses were corrected for multiple comparisons using an FDR procedure. Cross-sectional results indicate that hippocampal volume, hippocampal FW, Schwarz and McEvoy AD Signatures, and the Schwarz<sub>FW</sub> and McEvoy<sub>FW</sub> metrics are all significantly associated with memory performance. Baseline competitive model analyses showed that the McEvoy AD Signature and Schwarz<sub>FW</sub> explain the most unique variance beyond covariates for memory (ΔR<sub>adj</sub> <sup>2</sup> = 3.47 ± 1.65%) and executive function (ΔR<sub>adj</sub> <sup>2</sup> = 2.43 ± 1.63%), respectively. Longitudinal models revealed that hippocampal FW explained substantial unique variance for memory performance (ΔR<sub>adj</sub> <sup>2</sup> = 8.13 ± 1.25%), and outperformed all other biomarkers examined in predicting memory decline (p<sub>FDR</sub> = 1.95 x 10<sup>-11</sup>). This study shows that hippocampal FW is a sensitive biomarker for cognitive impairment and decline, and provides strong evidence for further exploration of this measure in aging and AD.
BibTeX
@inproceedings{sathe2024free,
  author = {Aditi Sathe and Yisu Yang and Kurt G. Schilling and Niranjana Shashikumar and Elizabeth E. Moore and Logan Dumitrescu and Kimberly R. Pechman and Bennett A. Landman and Katherine A. Gifford and Timothy J. Hohman and Angela L. Jefferson and Derek B. Archer},
  title = {{Free-water: A promising structural biomarker for cognitive decline in aging and mild cognitive impairment}},
  booktitle = {Imaging},
  year = {2024},
  doi = {10.1162/imag\_a\_00293},
}
Figure