Sensitivity of NIH Toolbox‐Cognition Battery task performance to amyloid positivity status in cognitively normal older adults

Abstract

AbstractBackgroundAmong cognitively normal older adults, high brain amyloid burden may signal the presence of early Alzheimer’s disease processes. While research has shown subtle neuropsychological performance differences among cognitively intact older adults with and without elevated brain amyloid, early investigation has not found similar between‐group differences on the NIH Toolbox‐Cognition Battery (NIHTB‐CB; Snitz et al., 2020). Here, we utilize data from a longitudinal observational driving study to evaluate potential differences in NIHTB‐CB task performance among older adults determined to be cognitively normal via clinical research consensus, based on brain amyloid status.MethodData from cognitively normal adults age 65+ who completed an amyloid Positron Emission Tomography scan, NIHTB‐CB tasks, and additional neuropsychological testing through the Michigan Alzheimer’s Disease Research Center were analyzed. Amyloid positivity was determined based on a centiloid scale (≤10 indicating amyloid negativity; ≥20 amyloid positivity). Thirty participants were amyloid negative and 17 were amyloid positive. Nine individuals with intermediate amyloid values were excluded. Group differences in NIHTB‐CB performance were analyzed via independent sample t‐tests. Linear regressions were used to assess demographic and NIHTB‐CB task/neuropsychological performance predictors of amyloid positivity.ResultThere were no demographic differences between the groups. Overall, amyloid negative participants outperformed their amyloid positive peers on NIHTB‐CB List Sorting and backward Digit Span tasks (p’s<0.05), both of which are measures of working memory. There were no other group differences in cognitive performance, including episodic memory. The final NIHTB‐CB regression model only included NIHTB List Sorting, which accounted for 16.2% of the variance in amyloid status (F = 8.69, p = 0.005). Digit Span backward similarly accounted for 16.6% of the variance in amyloid status (F = 6.39, p = 0.017).ConclusionAmong cognitively normal older adults with known amyloid status, working memory performance, but not episodic memory performance, differentiated amyloid positive and negative groups. This is consistent with research demonstrating an association between elevated amyloid and executive dysfunction, but not memory impairment, in cognitively normal adults (Tideman et al., 2022). NIHTB‐CB’s List Sorting Working Memory is a complex, challenging sequencing task that appears sensitive to amyloid positivity in older adults and task performance may serve as a marker for early prediction of Alzheimer’s disease.