Global amyloid burden enhances network efficiency of tau propagation in the brain

J Alzheimers Dis. 2024 Dec 16:13872877241294084. doi: 10.1177/13872877241294084. Online ahead of print.

Abstract

Background: Amyloid-β (Aβ) and hyperphosphorylated tau are crucial biomarkers in Alzheimer's disease (AD) pathogenesis, interacting synergistically to accelerate disease progression. While Aβ initiates cascades leading to tau hyperphosphorylation and neurofibrillary tangles, PET imaging studies suggest a sequential progression from amyloidosis to tauopathy, closely linked with neurocognitive symptoms.

Objective: To analyze the complex interactions between Aβ and tau in AD using probabilistic graphical models, assessing how regional tau accumulation is influenced by Aβ burden.

Methods: Data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) and Anti-Aβ Treatment in Asymptomatic Alzheimer's (A4) study were utilized, involving participants across various cognitive stages and employing both Florbetapir and Flortaucipir as tracers. Tau standardized uptake value ratio values were harmonized across studies, and participants were stratified into quantile groups based on Aβ levels. A LASSO regularized Gaussian graphical model analyzed partial correlations among brain regions to discern patterns of tau accumulation across different Aβ levels.

Results: Statistical analyses revealed significant differences in tau structure among low, medium, and high Aβ groups in both ADNI and A4 cohorts, with graph metrics, such as small-world coefficient, indicating increased tau efficiency as Aβ burden increased.

Conclusions: Our findings indicate that tau accumulates more efficiently with increasing Aβ burden, highlighting an interplay that could inform development of dual-targeting therapies in AD. This study underscores the importance of Aβ and tau interactions in AD progression and supports the hypothesis that targeting both pathologies could be crucial for therapeutic interventions.

Keywords: Alzheimer's disease; PET imaging; amyloidosis; probabilistic graphical models; tauopathy.