Abstract Details

Title Graph Theoretical Analysis of Resting-State Functional Connectivity MRI in Alzheimer's Disease Detects Network Dysfunction before Clinical Symptoms

Topic Aging and Dementia

Presentation(s) P06 - (-)

Poster/Presentation
Number 022

Objective

Background BACKGROUND: Previous research has demonstrated that AD exhibits dramatic effects on rs-fcMRI during the symptomatic course of clinical disease. However, AD pathology accumulates and exhibits a pathological influence many years prior to onset of symptoms. The impact of this pathology on rs-fcMRI and large scale neural function is unclear.

Design/Methods DESIGN/METHODS: We collected rs-fcMRI and cerebrospinal fluid (CSF) on 205 CDR 0 patients, 90 CDR 0.5 patients, and 31 CDR 1 patients. CDR 0 patients were further classified as 132 showing no signs of AD pathology, 46 showed signs of CSF amyloid pathology, and 13 showed signs of CSF amyloid pathology and neurodegeneration (tau). We calculated three graph theoretical measures, path length, clustering coefficient and modularity, which were compared based on both clinical and pathological status. We also assessed the classifying potential of these measures using linear discriminant analysis (LDA).

Results RESULTS: We observed declines in clustering coefficient and modularity, but not path length, as a function of increasing AD (indexed by CDR). LDA was able to classify patients with symptomatic AD (CDR>0) compared to those without symptomatic AD (CDR=0). The CDR 0 group with AD pathology had graph theory metrics that were between CDR 0 with no pathology and mild symptomatic AD phenotype (CDR=0.5). While the accuracy of the LDA was weaker than symptomatic vs non-symptomatic AD, this classifier still performed significantly better than chance.

Conclusions CONCLUSIONS: This work extends previous work using graph theory in AD by showing that changes in graph measures is stepwise with respect to disease severity. These data suggest that the presence of AD pathology is sufficient to cause rs-fcMRI dysfunction similar to that seen in symptomatic AD.

Authors/Disclosures
Matthew Brier, MD, PhD (Washington University) PRESENTER	Dr. Brier has received personal compensation in the range of $5,000-$9,999 for serving as a Consultant for Genentech. The institution of Dr. Brier has received research support from NIH.
Jewell Thomas	No disclosure on file
	No disclosure on file
John C. Morris, MD, FAAN (Washington University)	Dr. Morris has received personal compensation in the range of $500-$4,999 for serving on a Scientific Advisory or Data Safety Monitoring board for CBR International Advisory Board. Dr. Morris has received personal compensation in the range of $500-$4,999 for serving on a Scientific Advisory or Data Safety Monitoring board for Cure Alzheimers Fund. Dr. Morris has received personal compensation in the range of $500-$4,999 for serving on a Scientific Advisory or Data Safety Monitoring board for LEADS Steering Commitee. The institution of Dr. Morris has received research support from NIH grants. Dr. Morris has received intellectual property interests from a discovery or technology relating to health care. Dr. Morris has received intellectual property interests from a discovery or technology relating to health care.
Beau M. Ances, MD, PhD, MS, FAAN (Washington University in Saint Louis)	Dr. Ances has received personal compensation in the range of $0-$499 for serving on a Scientific Advisory or Data Safety Monitoring board for NIH.
	No disclosure on file

��ɫ��

��ɫ��