HAI Book 2025 - Flipbook - Page 421
Hall, Brandon
105
Cortical freewater increases with tau tangle aggregation
Brandon Hall1,2, Etienne Aumont1,2, Nesrine Rahmouni1,2, Joseph Therriault1,2, Seyyed-Ali
Hosseini1,2, Lydia Trudel1,2, Jaime Fernandez-Arias1,2, Arthur Macedo1,2, Stijn Servaes1,2, Gleb
Bezgin1,2, Yi-Ting Wang1,2, Yansheng Zheng1,2, Jenna Stevenson1,2, Sulantha Sanjeewa1,2,
Andrea Benedet5, Gallen Triana-Baltzer6, Hartmuth Kolb6, Nicholas Ashton7, Henrik
Zetterberg5, Kaj Blennow8, Tharick Pascoal4, Jesse Klostranec1,9, Pedro Rosa-Neto1,2,3
1
McGill University, Montreal, QC, CA
Translational Neuroimaging Laboratory, The McGill University Research Centre for Studies in Aging, Montreal, QC,
CA
3
Douglas Hospital Research Centre, Verdun, QC, CA
4
University of Pittsburg, Pittsburg, PA, US
5
Department of Psychiatry and Neurochemistry, Institute of Neuroscience and Physiology, The Sahlgrenska
Academy, Gothenburg, SE
6
Neuroscience Biomarkers, Janssen Research & Development, La Jolla, CA, US
7
Wallenberg Centre for Molecular and Translational Medicine, University of Gothenburg, Gothenburg, SE
8
Clinical Neurochemistry Laboratory, Sahlgrenska University Hospital, Mölnda, SE
9
Division of Diagnostic and Interventional Neuroradiology, Montreal Neurological Institute and Hospital, Montreal, QC,
CA
2
It is unclear how cortical microstructure changes across biological stages of Alzheimer9s disease (AD), and how
this occurs in response to amyloid plaques, tauopathy, and inflammation. We hypothesize that tauopathy is
chiefly responsible. To test this, we investigated how different measures of tauopathy influence microstructural
integrity in the whole brain, temporal meta-ROI, and hippocampus.
We sampled 301 participants (59% female, age 67±10) from the TRIAD cohort with T1 MRI, diffusion-weighted MRI,
amyloid PET (F18-NAV4694), tau PET (F18-MK6240), and plasma biomarkers. We used the NODDI MATLAB toolbox to
derive voxelwise freewater images, wherein higher cortical values represent inflammation and decreased
microstructural integrity. We used FreeSurfer 7 to derive temporal meta-ROI) and hippocampal masks, and
Standard Uptake Value Ratio (SUVR) to calculate amyloid and tau PET images. Plasma biomarkers were measured
by Simoa assay. We stratified participants by amyloid SUVR positivity, plasma ptau 217 positivity (T1), and tau
SUVR positivity (T2). We performed voxelwise correlations between freewater and tau SUVR via VoxelStats
MATLAB toolbox. Voxelwise regressions between freewater and plasma biomarkers and the mediation analyses
performed in R 4.4.1. Models covaried for amyloid SUVR, age, sex, and APOE4 genotype status. Multiple
comparisons correction performed by random field theory method.
Tau PET and plasma p-tau217 positively correlated with freewater in temporal and occipital regions (Fig1A, Fig2A).
However, the effect of p-tau217 is only present in participants positive for tau tangles (Fig 2C). Freewater
increased only in the A+T2+ group in the meta-ROI and hippocampus (Fig1B,C). Tau SUVR in meta-ROI mediated
the correlations between both GFAP and p-tau217 with freewater in the A+T+ group (Fig3).
In neocortical and allocortical AD-relevant regions, increased freewater correlates with tau tangle accumulation.
Our results suggest that while astrogliosis is involved with inflammation and decreasing microstructural integrity,
HAI2025 - 421
