Associating Cognition With Amyloid Status Using Partially Ordered Set Analysis.

Area of Science:

• Neuroscience
• Cognitive Psychology
• Biomarker Research

Background:

• Brain amyloid-beta positivity is a known risk factor for cognitive impairment and dementia.
• The precise cognitive functions most affected by amyloid-beta accumulation remain unclear.
• Neuropsychological tests often engage multiple cognitive domains, complicating direct associations.

Purpose of the Study:

• To investigate the association between specific cognitive functions and brain amyloid-beta positivity.
• To explore these associations across different age groups in cognitively normal and MCI subjects.
• To identify cognitive and demographic predictors of amyloid-beta positivity.

Main Methods:

• Utilized partially ordered set (POSET) models to classify detailed cognitive profiles from ADNI2 and AIBL datasets.
• Assessed episodic memory, cognitive flexibility, verbal fluency, attention, and perceptual-motor speed.
• Employed random forest analysis, stratified by age, to predict amyloid positivity using cognitive, demographic, and APOE4 data.

Main Results:

• Episodic memory and attention differences linked to amyloid positivity were observed in younger age groups (<70 and 70-80 years) in ADNI2.
• Episodic memory differences were found in the 70-80 years group in the AIBL study; no cognitive differences were noted in the oldest group (81-90 years).
• Random forest models highlighted intermediate episodic memory, delayed recall, and APOE4 allele count as key predictors of amyloid positivity, with age-dependent variable importance.

Conclusions:

• The relationship between cognitive abilities and brain amyloid-beta positivity is age-dependent.
• Episodic memory consistently emerged as the most significant cognitive predictor of amyloid-beta positivity across studies.
• Cognitive testing, APOE4 genotyping, and demographic data effectively predict brain amyloid-beta presence.