Profiling and predicting distinct tau progression patterns: An unsupervised data-driven approach to flortaucipir...

Area of Science:

• Neuroimaging
• Alzheimer's Disease Research
• Biomarker Discovery

Background:

• Detecting tau burden and accumulation patterns in Alzheimer's disease (AD) remains a challenge.
• Longitudinal tau positron emission tomography (PET) imaging offers insights into disease progression.

Purpose of the Study:

• To identify distinct tau accumulation profiles using an unsupervised, data-driven approach.
• To develop predictive models for tau accumulation types based on baseline markers.

Main Methods:

• Utilized unsupervised, data-driven whole-brain pattern analysis of longitudinal tau PET data.
• Analyzed flortaucipir PET data from the Alzheimer's Disease Neuroimaging Initiative, Avid Pharmaceuticals, and Harvard Aging Brain Study.
• Developed baseline models incorporating flortaucipir levels, amyloid beta (Aβ) positivity, and clinical variables.

Main Results:

• Identified three distinct flortaucipir progression profiles: stable, moderate accumulator, and fast accumulator.
• Baseline markers predicted moderate and fast accumulators with 81% and 95% positive predictive values, respectively.
• Screening for fast tau accumulation and Aβ positivity reduced required sample size by 46-77% for clinical trial power.

Conclusions:

• Predicting tau progression using baseline imaging and clinical markers can facilitate targeted screening of high-risk individuals.
• This approach may optimize patient selection for specific treatment regimens in Alzheimer's disease clinical trials.