Observational study reveals that regional amyloid-β and functional connectivity align with subtype-specific tau spreading, suggesting pre-existing brain network vulnerabilities.
Spatiotemporal patterns of tau accumulation differ between Alzheimer disease (AD) subtypes in brain networks related to clinical manifestations. Converging evidence suggests that accumulation may be driven by spreading through network connections and/or interactions with amyloid-β (Aβ). Persons with atypical AD presentations might carry pre-existing network-specific vulnerabilities, which might predispose these areas to tau pathology. However, it is unknown if regional patterns of Aβ and network connectivity differ between tau-defined subtypes and whether these patterns precede tau accumulation. We used Subtype-and-Stage-Inference (SuStaIn) to identify subtypes in 919 tau-PET scans from 709 participants in the Knight ADRC. We tested group differences between tau-defined subtypes in regional Aβ-PET and resting-state-functional-connectivity (RSFC) seed maps to early-stage tau epicenter regions using retrospective imaging data preceding the tau-PET scans by up to 14 years, including across stages of AD progression. Most participants were tau-negative ( N = 522). Among tau-positive participants, we identified three tau-PET subtypes. Limbic tau was the most common subtype ( N = 136, 73.9%), resembling amnestic AD. Participants with the posterior tau subtype ( N = 25, 13.6%) had specific cognitive deficits on visuospatial-dominant tasks, resembling posterior cortical atrophy. Participants with the medial-temporal-lobe-sparing (MTL-sparing) tau subtype ( N = 23, 12.5%) were younger and more severely impaired, resembling early-onset dysexecutive AD. Subtype-specific Aβ-PET elevations colocalized with regions of tau accumulation, including in preliminary retrospective analyses of scans preceding tau positivity. RSFC was elevated between subtype-specific regions of tau spreading. Specifically, connectivity was elevated between occipital regions in the posterior tau subtype and was elevated between temporal and frontal regions in the MTL-sparing subtype. In preliminary retrospective analyses, these patterns were detectable in small subsets of scans acquired when participants were Aβ-negative or cognitively unimpaired, before they would later exhibit atypical tau patterns, and extended with disease progression. Regional patterns of Aβ, tau, and functional connectivity align in subtype-specific brain networks. Elevated connectivity between tau spreading regions might indicate network-specific vulnerabilities that predispose these regions to tau aggregation, possibly preceding Aβ positivity. These results suggest a temporal sequence of disease progression in brain networks, which might inform explanations of clinico-pathological heterogeneity and the development of precision medicine tools for early detection, prognosis, and treatment of atypical AD.
No takes yet. Share an insight, caveat, or question.
Millar et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: