Disrupted graph theory properties of resting-state networks in young APOEe4 carriers
Poster No:
1519
Submission Type:
Abstract Submission
Authors:
Ludmila Kucikova1, Adam Brass1, Jianmin Zeng2, Graciela Muniz-Terrera3,4, Craig Ritchie4,5, John O'Brien6, Li Su1,6
Institutions:
1University of Sheffield, Sheffield, United Kingdom, 2Southwest University, Chongqing, China, 3Ohio University, Athens, OH, 4University of Edinburgh, Edinburgh, United Kingdom, 5Scottish Brain Sciences, Edinburgh, United Kingdom, 6University of Cambridge, Cambridge, United Kingdom
First Author:
Co-Author(s):
Craig Ritchie
University of Edinburgh|Scottish Brain Sciences
Edinburgh, United Kingdom|Edinburgh, United Kingdom
University of Edinburgh|Scottish Brain Sciences
Edinburgh, United Kingdom|Edinburgh, United Kingdom
Li Su
University of Sheffield|University of Cambridge
Sheffield, United Kingdom|Cambridge, United Kingdom
University of Sheffield|University of Cambridge
Sheffield, United Kingdom|Cambridge, United Kingdom
Introduction:
Alzheimer's Disease (AD) is defined by a prolonged preclinical period during which brain changes occur without producing overt clinical symptoms. Understanding the relationship between biomarkers, potential risk factors, and related clinical symptoms is crucial to capture different stages of this phase reliably. Middle-aged carriers of risk factors for AD show disturbances in brain structure and microstructure (reviewed in Mak et al., 2017) and brain function (reviewed in Habib et al., 2017), specifically in functional connectivity of large-scale networks of APOEe4-carriers (reviewed in Kucikova et al., 2021). Mapping the trajectory of connectivity differences in young adulthood is still inconclusive. However, our recent research indicates decreased connectivity in the youngest sample of adults with APOEe4 gene that has been previously looked at (Kucikova et al., 2023). The aim of present study is to extend these findings by examining the graph theoretical properties of functional connectivity.
Methods:
We analysed the effects of one of the most important genetic risk factors for AD, specifically the APOEe4, on resting-state functional connectivity in a sample of 129 cognitively healthy young adults (aged 17-22). Seven resting-state networks of interest included the DMN, executive, salience, sensorimotor, dorsoattentional, visual, and language networks. Graphs were constructed from single subject correlation matrix. To threshold connectivity within the graphs, we initially used a standard z-scores method. However, recognising the impact of methodological choices on outcomes, we sought to comprehensively investigate the sensitivity of different analytical pipelines. Hence, the second part of this study involved calculating graph properties using alternative thresholding methods (i.e., correlation coefficient, cost) and employing three different types of brain parcellation (i.e., ICA, Oxford-Harvard atlas, Power's atlas) to compare the reliability of the findings across different methodological choices.
Results:
The APOEe4-carriers consistently showed decreased functional connectivity in multiple resting-state networks compared to non-carriers. Specifically, we observed decreased degree (T=-3.3, pFDR<0.01; Fig1A1) and global efficiency (T=-2.85, pFDR<0.01; Fig1A2) and increased betweenness centrality (T=2.92, pFDR<0.05; Fig1A3) within the salience network; decreased global efficiency (T=-2,73, pFDR<0.05; Fig1B1) and decreased average path length (T=-529, p<0.0001; Fig1B2) within the sensorimotor network; and increased betweenness centrality (T=2.6, pFDR<0.05; Fig1C1), degree (T=2.82, pFDR=0.01; Fig1C2), and global efficiency (T=3.08, pFDR<0.01; Fig1C3) within dorsoattentional network. Findings in other networks were more sensitive to parcellation and/or thresholding methods used.
·The 3D maps of significant differences between APOEe4-carriers and non-carriers.
Conclusions:
Our results demonstrate that APOEe4 allele alters functional connectivity of multiple networks in cognitively healthy young adults. Functional connectivity was consistently disturbed in salience, sensorimotor, and dorsoattentional networks, reflecting alterations in graph theory properties crucial for communication and information flow within these networks. These results carry implications for our understanding of brain vulnerability related to genetic risk for AD in young adulthood.
While our findings were mostly reproducible across different thresholding methods which emphasises the internal validity, the choice of parcellation methods yielded less consistent results. This may be explained by the brain atlas concordance problem (Bohland et al., 2009). In the realm of clinical applications, our study underscores the critical role of methodological choices, particularly in the selection of a brain parcellation method, which can enhance the interpretability and applicability of neuroimaging findings in clinical contexts, potentially informing early diagnostic strategies or intervention approaches for individuals at risk of AD.
While our findings were mostly reproducible across different thresholding methods which emphasises the internal validity, the choice of parcellation methods yielded less consistent results. This may be explained by the brain atlas concordance problem (Bohland et al., 2009). In the realm of clinical applications, our study underscores the critical role of methodological choices, particularly in the selection of a brain parcellation method, which can enhance the interpretability and applicability of neuroimaging findings in clinical contexts, potentially informing early diagnostic strategies or intervention approaches for individuals at risk of AD.
Disorders of the Nervous System:
Neurodegenerative/ Late Life (eg. Parkinson’s, Alzheimer’s) 2
Genetics:
Genetic Association Studies
Lifespan Development:
Early life, Adolescence, Aging
Modeling and Analysis Methods:
Connectivity (eg. functional, effective, structural) 1
Neuroinformatics and Data Sharing:
Brain Atlases
Keywords:
Aging
Atlasing
1|2Indicates the priority used for review
Provide references using author date format
Habib, M., Mak, E., Gabel, S., Su, L., Williams, G., Waldman, A., ... & O’Brien, J. T. (2017). Functional neuroimaging findings in healthy middle-aged adults at risk of Alzheimer’s disease. Ageing research reviews, 36, 88-104.
Mak, E., Gabel, S., Mirette, H., Su, L., Williams, G. B., Waldman, A., ... & O’Brien, J. (2017). Structural neuroimaging in preclinical dementia: from microstructural deficits and grey matter atrophy to macroscale connectomic changes. Ageing research reviews, 35, 250-264.
Kucikova, L., Goerdten, J., Dounavi, M. E., Mak, E., Su, L., Waldman, A. D., ... & Ritchie, C. W. (2021). Resting-state brain connectivity in healthy young and middle-aged adults at risk of progressive Alzheimer’s disease. Neuroscience & Biobehavioral Reviews, 129, 142-153.
Kucikova, L., Zeng, J., Muñoz-Neira, C., Muniz-Terrera, G., Huang, W., Gregory, S., ... & Su, L. (2023). Genetic risk factors of Alzheimer’s Disease disrupt resting-state functional connectivity in cognitively intact young individuals. Journal of Neurology, 1-10.