Association of Plasma Biomarkers of Alzheimer Disease and Neurodegeneration With Longitudinal Intra-Network Functional Brain Connectivity

Alzheimer disease (AD) is defined by cortical β-amyloid (Aβ), tau, and neurodegeneration, which contribute to cognitive decline, in part, by altering large-scale functional brain networks. While cortical Aβ and tau have been associated with changes in functional brain connectivity, it is unknown whether plasma biomarkers relate to such changes. In a healthy community sample of cognitively unimpaired adults free from major CNS disease from the Baltimore Longitudinal Study of Aging, we examined whether plasma biomarkers of AD pathology (Aβ42/40, phosphorylated tau [pTau-181]), astrogliosis (glial fibrillary acidic protein [GFAP]), and neuronal injury (neurofilament light chain [NfL]) were associated with longitudinal changes in functional connectivity and whether changes in functional connectivity were related to longitudinal cognition. Plasma biomarkers were measured using the Quanterix SIMOA assays. Intranetwork connectivity (3T resting-state fMRI) from 7 functional networks was derived using a predefined cortical parcellation mask for each participant visit. Cognitive performance was assessed concurrently with fMRI scan. Covariate-adjusted linear mixed-effect models were used to determine (1) whether plasma biomarkers were associated with longitudinal changes in connectivity, (2) whether the magnitude of the biomarker-connectivity relationships differed by amyloid status, and (3) whether changes in connectivity co-occurred with longitudinal changes in cognition. Our primary findings (n = 486; age = 65.5 ± 16.2 years; 54% female; mean follow-up time = 4.3 ± 1.7 years) showed that higher baseline GFAP was associated with faster declines in somatomotor (β = -0.04, p = 0.01, 95% CI -0.06 to -0.01), limbic (β = -0.03, p = 0.02, 95% CI -0.06 to -0.005), and frontoparietal (β = -0.04, p = 0.02, 95% CI -0.07 to -0.01) network connectivity. Amyloid status moderated several biomarker-connectivity associations. For instance, higher baseline NfL was related to faster declines in visual and limbic network connectivity, but only among amyloid-positive participants. Among 421 participants with ≥2 fMRI visits (age = 71.7 ± 11.4 years; 55% female; follow-up time = 3.9 ± 1.6 years), longitudinal changes in connectivity were associated with concurrent declines in cognition; however, these results did not survive multiple comparison correction. Among cognitively unimpaired participants, plasma biomarkers of amyloidosis, astrogliosis, and neuronal injury are associated with declines in network connectivity, particularly among amyloid-positive participants. Major limitations include the lack of inclusion of the sensitive pTau-217 and pTau-231 isoforms and comparative PET biomarkers.