Alzheimer's Imaging Consortium.

Area of Science:

• Neuroscience
• Radiology
• Gerontology

Background:

• Alzheimer's disease (AD) spectrum exhibits variability in tau accumulation, but the drivers of this heterogeneity remain unclear.
• Previous research suggests differences in tau deposition patterns across individuals, necessitating further investigation into their origins and clinical relevance.

Purpose of the Study:

• To investigate heterogeneity in spatial patterns of tau accumulation using a data-driven approach.
• To examine how these spatial tau patterns relate to participant characteristics across the normal aging and clinical AD spectrum.

Main Methods:

• Generated whole-brain voxel-wise flortaucipir (FTP) SUVR slope maps from 648 participants across four independent studies.
• Employed spatial Independent Component Analysis (ICA) to decompose tau accumulation patterns and k-means clustering to identify distinct participant groups.

Main Results:

• Identified four distinct tau accumulation patterns: non-accumulators, early accumulators, temporoparietal accumulators, and widespread neocortical accumulators.
• Linked tau accumulation patterns to clinical diagnosis (cognitively normal, MCI, AD dementia), Aβ status, age, and APOE genotype.
• Widespread accumulators showed younger age, higher baseline Aβ, greater tau burden in specific regions, and worse cognition compared to other groups.

Conclusions:

• A data-driven approach revealed four distinct tau accumulation patterns associated with age, clinical diagnosis, baseline AD pathology, and APOE genotype.
• More extensive tau accumulation patterns within clinical groups correlated with higher baseline AD pathology.
• Findings support the hypothesis that tau accumulation heterogeneity in Alzheimer's disease is influenced by multiple between-participant factors.