Multimodal neuroimaging biomarkers and subtle cognitive decline in a population-based cohort without dementia.

Abstract

Background: The relationship between subtle cognitive decline and Alzheimer's disease (AD) pathology as measured by biomarkers in settings outside of specialty memory clinics is not well characterized.

Objective: To investigate how subtle longitudinal cognitive decline relates to neuroimaging biomarkers in individuals drawn from a population-based study in an economically depressed, small-town area in southwestern Pennsylvania, USA.

Methods: A subset of participants without dementia (N = 115, age 76.53 years ± 6.25) from the Monongahela Youghiogheny Healthy Aging Team (MYHAT) study completed neuroimaging including magnetic resonance imaging (MRI) measures of AD-signature region cortical thickness and white matter hyperintensities (WMH), Pittsburgh compound B (PiB)-positron emission tomography (PET) for amyloid-β (Aβ) deposition, and [¹⁸F]AV-1451-PET for tau deposition. Neuropsychological evaluations were completed at multiple timepoints up to 11 years prior to neuroimaging. Aβ positivity was determined using a regional approach. We used linear mixed models to examine neuroimaging biomarker associations with retrospective cognitive slopes in five domains and a global cognitive composite.

Results: Among Aβ(+) participants (38%), there were associations between (i) tau Braak III/IV and language decline (p < 0.05), (ii) cortical thickness and both memory decline (p < 0.001) and global cognitive decline (p < 0.01), and (iii) WMH and decline in executive function (p < 0.05) and global cognition (p < 0.05). Among Aβ(-) participants, there was an association between tau Braak III/IV and decline on tests of attention/psychomotor speed (p < 0.05).

Conclusions: These findings confirm an Aβ-dependent early AD biomarker pathway, and suggest a possible Aβ-independent, non-AD process underlying subtle cognitive decline in a population-based sample of older adults without dementia.