In Vivo Cortical Microstructure - A Proxy for Tauopathy & Cognitive Impairment

Aggregation of hyperphosphorylated tau protein is currently one of the most reliable indicators of Alzheimers pathology and cognitive impairment in older adults. However, it would be useful to have a non-invasive, accessible proxy measure that does not rely on Positron Emission Tomography (PET). We used data from multi-shell diffusion-weighted imaging (DWI) to assess indices from the Neurite Orientation Dispersion and Density Imaging (NODDI) model to determine possible proxies for tau and relationship with cognitive impairment. After controlling for age, sex, and the time difference between the scan acquisitions (DWI vs. PET), we used multiple factor analysis (MFA) to assess the fit between NODDI indices (orientation dispersion [ODI], neurite density [NDI], and free-water [fISO]), cortical thickness, and tau binding (via PET). We used data from 80 participants from the ADNI-3 sample who had a multi-shell DWI and an [18F]AV-1451 (tau) PET scan. Of these 80, 49 individuals were considered cognitively normal older adults (age ~74 years), 26 individuals had a diagnosis of mild cognitive impairment (age ~75 years), and five individuals had Alzheimers dementia (age ~78 years). fISO and tau shared a large amount of spatial overlap, and both strongly correlated with the first MFA dimension. Macrostructural features (such as cortical thickness and subcortical volume) introduced in a follow-up analysis were less related to this first MFA dimension than fISO and eight percent less than tau. Subsequent mediation analyses demonstrated that fISO mediated the relationship between cortical thickness and tau, explaining all of the variance. Microstructural features derived from advanced DWI acquisitions such as fISO may be useful proxies for tau. Cortical fISO, rather than cortical thickness, may represent the impact of tau on the brain (and, by extension, cognition).