Abstract INTRODUCTION Tau positron emission tomography (PET) probes Alzheimer's disease (AD) severity via regional tau spread but is not widely available. We tested whether multiregion structural magnetic resonance imaging (MRI) could approximate individual tau burden. METHODS We studied 378 Alzheimer's Disease Neuroimaging Initiative (ADNI) participants with mild cognitive impairment (MCI)‐AD or AD dementia with paired T1‐MRI and [18F]flortaucipir tau‐PET (≤6 months apart). Regional cortical thickness and volume were extracted with FreeSurfer. Principal component analysis and multivariable linear regression yielded MRI signatures of tau‐PET standardized uptake value ratio (SUVR) in Braak composite regions (I, III–IV, V–VI), a meta‐temporal region of interest (ROI), and a global neocortical meta‐ROI. Performance and high/low tau classification were evaluated by leave‐one‐out cross‐validation against published cut‐offs. RESULTS MRI signatures were significantly associated with tau‐PET burden across all regions (p < 0.001). High‐versus‐low tau discrimination varied: AUC ≈ 0.70 in Braak I, ≈0.89 in Braak V–VI, and ≈0.90 globally. Discussion Here we provide proof‐of‐concept evidence that multiregion T1‐MRI patterns can inform on tau‐PET burden in AD and may support approximate tau staging when tau‐PET is unavailable, especially in subjects with more advanced tau burden.
Martina Pulze, S. Garbarino, L. Lorenzini et al.· Alzheimer's & Dementia· 0 citations
Introduction Cognitive reserve (CR) has been proposed as a key factor explaining inter-individual variability in cognitive performance despite comparable neuropathology. However, its role across the Alzheimer’s disease (AD) continuum remains unclear. This study investigates stage-dependent effects of CR on the relationship between memory performance and brain structural network integrity across healthy subjects (HS), individuals with subjective cognitive decline (SCD), and patients with amnestic mild cognitive impairment (a-MCI), and AD dementia. Materials and methods A total of 209 participants underwent a comprehensive neuropsychological assessment and 3T MRI. Source-based morphometry identified three grey matter structural covariance networks, involving orbitofrontal-temporal-insular regions (OTIN), precuneus-posterior cingulate cortex (PreCiN), and cingulate-hippocampal regions (CHiN). A composite memory score was derived using factor analysis. Regression and moderation models examined the predictive and moderating effects of CR (operationalized as years of education) and network integrity on cognitive performance within each group. Results OTIN and PreCiN showed progressive structural vulnerability along the AD continuum, whereas CHiN showed no significant between-group differences. Across the sample, OTIN and PreCiN integrity significantly predicted cognitive performance. In HS, CR was positively associated with memory performance independently of structural network integrity, suggesting an additive protective role of cognitive reserve in healthy aging. In the SCD group, CR was not directly associated with memory, and only limited effects emerged, indicating early alterations in reserve-related processes. In a-MCI patients, the significant interaction between CR and OTIN integrity suggested patterns consistent with compensatory mechanisms, with higher reserve supporting memory despite structural decline. In AD patients, CR and its interaction with structural networks no longer predicted cognitive outcomes, suggesting a possible exhaustion of reserve capacity. Conclusion These findings support a stage-dependent model of CR, characterized by an additive protective role in healthy aging, patterns consistent with compensatory recruitment in early cognitive decline, and a possible loss of reserve effectiveness beyond a critical neuropathological threshold. Distinct network vulnerabilities and stage-specific CR effects highlight potential windows for reserve-enhancing interventions across the AD continuum.
Laura Serra, Sabrina Bonarota, Giulia Caruso et al.· Frontiers in Human Neuroscie...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
