IC‐P‐125: Location, location, location: The distributed effects of amyloid on functional connectivity in the harvard aging brain study
Year: 2015
Type: article
Abstract: As we move to treat Alzheimer's disease in the preclinical stages, more sensitive methods are needed to detect subtle differences between “normal” aging and preclinical AD. Here we compare DMN connectivity results between PiB-/PiB+ subjects using two methods: (1) a traditional Cluster Significance (CS) approach, and (2) a novel Cluster Distribution Significance (CDS) approach that quantifies the deviance of clusters from a uniformly distributed pattern. We included 267 individuals from the 284 subject Harvard Aging Brain Study cohort with baseline PiB-PET and resting-state fMRI. Individuals with a PiB-PET value greater than 1.19, based on a Gaussian mixture model, were classified as PiB+. A two-sample t-test (PiB+ versus PiB-) with mean motion, SNR, age, and gender as covariates was used to evaluate the effect of amyloid on DMN connectivity as measured with TBR[1]. With CS, we used a traditional cluster-based significance threshold (p<0.05 corrected). Next, we developed the CDS approach to quantify the deviance of clusters from that of a uniformly distributed pattern across resting-state networks using the top (PiB->PiB+) or bottom (PiB+>PiB-) 5% of voxel-wise connectivity differences. The number of supra-threshold voxels in each of 11 principle resting state networks were compared to a simulated uniform distribution across networks using chi-square statistics (p<0.05). Using CS, one cluster on the border of the primary visual and dorsal attention networks was less connected in PiB+. Using CDS, the spatial distribution of connectivity differences between PiB- and PiB+ individuals was significantly different than chance. Decreased connectivity in PiB+ individuals was most prominent in the default-mode, primary visual, dorsal attention, and precuneus; while increased connectivity in PiB+ individuals was most prominent in language and subcortical networks.
Source: Alzheimer s & Dementia
Language: English
Subfield: Cognitive Neuroscience
Field: Neuroscience
Domain: Life Sciences
Open Access status: bronze
OpenAlex ID: W2302169413