Eigenvector centrality functional connectivity patterns indicate age and APOE4 carrier status.
Poster No:
1535
Submission Type:
Abstract Submission
Authors:
Aïda Fall1, Maria Preti1, Mohamed Eshmawey2, Sonja Kagerer3, Dimitri Van De Ville4, Paul Unschuld5
Institutions:
1University of Geneva, Geneva, Geneva, 2Geneva university hospitals, Geneva , Geneva, 3Institute for Regenerative Medicine (IREM), Zurich , Zurich, 4École Polytechnique Fédérale de Lausanne, Lausanne, Switzerland, 5Geneva university hospitals, Geneva , Geneva
First Author:
Co-Author(s):
Introduction:
Age and Apolipoprotein E4 (APOE4) are the greatest risk factors for Alzheimer's disease (AD) (Yin et al., 2018) but the fundamental processes underlying the onset of the pathology remains unclear. Functional brain change in AD, as revealed by functional magnetic resonance imaging (fMRI), is characterized by disruption of brain networks (Sperling et al., 2010). Carrier-status of the ApolipoproteinE4 (ApoE4) allele is associated with faster progression of AD, and therefore a potentially more pronounced disintegration of brain functional connectivity (FC).
Here, we investigated functional connectivity changes associated with aging and the presence of APOE4 allele, to link individual brain network properties with AD risk.
Here, we investigated functional connectivity changes associated with aging and the presence of APOE4 allele, to link individual brain network properties with AD risk.
Methods:
We included in the study 128 individuals from the Alzheimer's Disease Neuroimaging Initiative database (ADNI). This sample includes 54 APOE4 carriers and 74 non-carriers. 76 were non-symptomatic (39 Cognitive Normal (CN), 37 with Subjective Memory Complaint (SMC)) and 52 were symptomatic (39 with Mild Cognitive Impairment (MCI), 13 with AD).
Resting-state fMRI images were preprocessed with conventional in-house pipelines including realignment, regression of nuisance signals (motion, cerebrospinalfluid and white matter signals), spatial smoothing. Regional time courses were then obtained by parcellation into 379 brain regions (360 from Glasser parcellation (Glasser et al., 2016) and 19 subcortical ones (Fischl et al., 2002)), and band-pass filtered within the range [0.01-0.15 Hz]. Functional connectomes (FC) were obtained as pairwise Pearson's correlation between regional timecourses, and eigenvector centrality was computed as the first eigenvector of FC.
We investigated the relationship between eigenvector centrality, aging and the presence of APOE4 with Partial Least Square Correlation (PLSC) analysis (Krishnan et al., 2011), where the brain data matrix X included eigenvector centrality for each subject, and behavioral matrix Y included z-scored age of each subject and symptomatic index (1 for symptomatic and 0 for non-symptomatic). By eigendecomposition of the covariance matrix between these two sets of variables, we are able to extrapolate sets of variables that maximize the correlation between brain and behavior.
Resting-state fMRI images were preprocessed with conventional in-house pipelines including realignment, regression of nuisance signals (motion, cerebrospinalfluid and white matter signals), spatial smoothing. Regional time courses were then obtained by parcellation into 379 brain regions (360 from Glasser parcellation (Glasser et al., 2016) and 19 subcortical ones (Fischl et al., 2002)), and band-pass filtered within the range [0.01-0.15 Hz]. Functional connectomes (FC) were obtained as pairwise Pearson's correlation between regional timecourses, and eigenvector centrality was computed as the first eigenvector of FC.
We investigated the relationship between eigenvector centrality, aging and the presence of APOE4 with Partial Least Square Correlation (PLSC) analysis (Krishnan et al., 2011), where the brain data matrix X included eigenvector centrality for each subject, and behavioral matrix Y included z-scored age of each subject and symptomatic index (1 for symptomatic and 0 for non-symptomatic). By eigendecomposition of the covariance matrix between these two sets of variables, we are able to extrapolate sets of variables that maximize the correlation between brain and behavior.
Results:
PLSC analysis yielded one significant latent component (p-val<0.001). By looking at brain and behavioral scores (Fig. 1 and 2, respectively), we can highlight which brain / behavioral variables contribute the most to the multivariate correlation. We found a significant relationship between centrality, aging, the presence of APOE4 and disease progression in brain regions in both parietal and frontal lobes. In particular, decreased centrality in the dorsal somatomotor and visuospatial attention networks (Fig. 1, blue nodes) is associated with aging, cognitive impairment, and the presence of the genetic risk factor for AD, APOE4.
The presence of FC alterations in aging is largely explored in literature, with results concordant with our findings (Dennis et al., 2014). However, our analysis suggests that the same pattern of FC alterations which is associated with aging is acerbated by the presence of not only cognitive impairment but also APOE4, possibly accelerating the degeneration process in carrier individuals.
This opens a window for further insights into the exploration of pathological mechanisms in APOE4 carriers, with the perspective of an early diagnosis.
The presence of FC alterations in aging is largely explored in literature, with results concordant with our findings (Dennis et al., 2014). However, our analysis suggests that the same pattern of FC alterations which is associated with aging is acerbated by the presence of not only cognitive impairment but also APOE4, possibly accelerating the degeneration process in carrier individuals.
This opens a window for further insights into the exploration of pathological mechanisms in APOE4 carriers, with the perspective of an early diagnosis.
Conclusions:
In this work, we explored the impact of aging on brain functional connectivity and the potential mediation of this effect by the presence of the genetic risk factor for AD. We found a signature brain pattern associated with aging, but acerbated by the presence of APOE4, corroborating evidence of earlier degeneration in these subjects. Future longitudinal studies will allow to further validate these findings and advance towards an early diagnosis of AD based on multiple markers.
Disorders of the Nervous System:
Neurodegenerative/ Late Life (eg. Parkinson’s, Alzheimer’s) 2
Lifespan Development:
Aging
Modeling and Analysis Methods:
Connectivity (eg. functional, effective, structural) 1
Novel Imaging Acquisition Methods:
BOLD fMRI
Keywords:
Aging
Data analysis
Degenerative Disease
FUNCTIONAL MRI
Other - Alzheimer's disease
1|2Indicates the priority used for review
Provide references using author date format
Sperling RA, Dickerson BC, Pihlajamaki M, Vannini P, LaViolette PS, Vitolo OV, Hedden T, Becker JA, Rentz DM, Selkoe DJ, Johnson KA. (2010) Functional alterations in memory networks in early Alzheimer's disease. Neuromolecular Med. 2010 Mar;12(1):27-43. doi: 10.1007/s12017-009-8109-7. PMID: 20069392; PMCID: PMC3036844.
Glasser MF, Coalson TS, Robinson EC, Hacker CD, Harwell J, Yacoub E, Ugurbil K, Andersson J, Beckmann CF, Jenkinson M, Smith SM, Van Essen DC. (2016) A multi-modal parcellation of human cerebral cortex. Nature. 2016 Aug 11;536(7615):171-178. doi: 10.1038/nature18933. Epub 2016 Jul 20. PMID: 27437579; PMCID: PMC4990127.
Fischl B, Salat DH, Busa E, Albert M, Dieterich M, Haselgrove C, van der Kouwe A, Killiany R, Kennedy D, Klaveness S, Montillo A, Makris N, Rosen B, Dale AM. (2002) Whole brain segmentation: automated labeling of neuroanatomical structures in the human brain. Neuron. 2002 Jan 31;33(3):341-55. doi: 10.1016/s0896-6273(02)00569-x. PMID: 11832223.
Krishnan A, Williams LJ, McIntosh AR, Abdi H. Partial Least Squares (PLS) methods for neuroimaging: a tutorial and review. Neuroimage. 2011 May 15;56(2):455-75. doi: 10.1016/j.neuroimage.2010.07.034. Epub 2010 Jul 23. PMID: 20656037.
Dennis EL, Thompson PM. Functional brain connectivity using fMRI in aging and Alzheimer's disease. Neuropsychol Rev. 2014 Mar;24(1):49-62. doi: 10.1007/s11065-014-9249-6. Epub 2014 Feb 23. PMID: 24562737; PMCID: PMC4109887.