A Multimodal Study of Biomarkers of Alzheimer’s Disease In Adults with Down Syndrome: Diffusion Tensor Imaging and Tau-PET
Abstract
Purpose
Individuals with Down syndrome (DS) are at high risk of developing Alzheimer's Disease, partially driven by trisomy 21-related overproduction of amyloid precursor protein and plaque (Aβ) development. Tau neurofibrillary tangles typically follow Aβ and obey signature hierarchical patterns, yet their relationship to neurodegeneration in DS remains understudied. Diffusion magnetic resonance imaging (dMRI) may be used to assess white matter (WM) microstructure. It probes the diffusion of water molecules using the diffusion tensor imaging (DTI) model. DTI measures may be modulated by neurodegeneration. For instance, higher fractional anisotropy (FA) has been linked to greater tract coherence. This study applied linear mixed-effects (LME) modeling to assess relationships between whole-brain WM DTI and tau-positron emission tomography (PET), both globally and across tau NFT regions-of-interest (ROIs: NFT1-2, NFT3-4, and NFT5-6).
Methods
99 adults with DS underwent DTI and Aβ-([C-11]PiB) and tau-([F-18]AV-1451) PET imaging; 53 had repeated scans, yielding 152 image sets. Multisite data from the Alzheimer’s Biomarker Consortium-DS were harmonized using longCombat for inter-site variability. Whole-brain and ROI-specific PET tau and Aβ burden were calculated using standard uptake value ratios. dMRIs were preprocessed for motion and artifacts, aligned to their respective T1-weighted images, fit to DTI via an in-house pipeline, and segmented using FreeSurfer ROIs. LME models were fit for each DTI parameter (FA, mean, axial, and radial diffusivities (MD, AxD, RD)) with tau-PET measures, adjusting for age, sex, Aβ, and repeated measures: DTI~Age+Sex+Tau-PET-ROI+Aβ+(1|Subject)
Results
All DTI parameters were significantly associated with whole-brain, NFT3-4, and NFT5-6 tau-PET (p<0.05). Only AxD was significantly associated with NFT1-2 tau-PET. Furthermore, increased tau burden was associated with decreased FA and increased diffusivities.
Conclusion
Multimodal imaging reveals an association of WM microstructural degradation with tau accumulation, particularly in the limbic and neocortical regions. Whole-brain WM DTI may provide a sensitive, non-ionizing marker for studying tau-related neurodegeneration in DS.