Poster No:
2121
Submission Type:
Abstract Submission
Authors:
Baptiste Couvy-Duchesne1, Vinecnt Frouin2, Vincent Bouteloup3, Pierrick Bourgeat4, Nikitas Koussis5, Bryan Paton5, Julia Sidorenko6, Wei Wen7, Alle Meije Wink8, Luigi Lorenzini9, Karen Mather10, Julian Trollor11, Jean-François Mangin12, Carole Dufouil13, Michelle Lupton14, Jurgen Fripp4, Michael Breakspear15, Peter Visscher6, olivier colliot16
Institutions:
1The University of Queensland, Brisbane, QLD, 2Paris-Saclay University, CEA, CNRS, Neurospin, Baobab, Saclay, 3Univ. Bordeaux, Inserm, Bordeaux Population Health, UMR1219, Bordeaux, France, 4Australian e-Health Research Centre, CSIRO Health and security, Brisbane, Australia, 5University of Newcastle, New Lambton Heights, New South Wales, 6Institute for Molecular Bioscience, the University of Queensland, Brisbane, Australia, 7Centre for Healthy Brain Ageing, Discipline of Psychiatry & Mental Health, School of Clinical Medici, Sydney, Australia, 8Amsterdam University Medical Centre, Amsterdam, Noord-Holland, 9Amsterdam UMC, Amsterdam, Netherlands, 10Centre for Healthy Brain Ageing (CHeBA), University of New South Wales, Sydney, New South Wales, 11Department of Developmental Disability Neuropsychiatry, UNSW Sydney, Sydney, Australia, 12Université Paris-Saclay, CEA, CNRS, Neurospin, Gif-sur-Yvette, France, 13Univ. Bordeaux, Inserm, Bordeaux Population Health, UMR1219, Bordeaux, France CIC 1401 EC, Pôle Sant, Bordeaux, Australia, 14QIMR Berghofer Medical Research Institute, Brisbane, Australia, 15University of Newcastle, Newcastle, N/A, 16CNRS, paris, paris
First Author:
Co-Author(s):
Vinecnt Frouin
Paris-Saclay University, CEA, CNRS, Neurospin, Baobab
Saclay
Vincent Bouteloup
Univ. Bordeaux, Inserm, Bordeaux Population Health, UMR1219
Bordeaux, France
Pierrick Bourgeat
Australian e-Health Research Centre, CSIRO Health and security
Brisbane, Australia
Nikitas Koussis
University of Newcastle
New Lambton Heights, New South Wales
Bryan Paton
University of Newcastle
New Lambton Heights, New South Wales
Julia Sidorenko
Institute for Molecular Bioscience, the University of Queensland
Brisbane, Australia
Wei Wen
Centre for Healthy Brain Ageing, Discipline of Psychiatry & Mental Health, School of Clinical Medici
Sydney, Australia
Alle Meije Wink
Amsterdam University Medical Centre
Amsterdam, Noord-Holland
Karen Mather
Centre for Healthy Brain Ageing (CHeBA), University of New South Wales
Sydney, New South Wales
Julian Trollor
Department of Developmental Disability Neuropsychiatry, UNSW Sydney
Sydney, Australia
Carole Dufouil
Univ. Bordeaux, Inserm, Bordeaux Population Health, UMR1219, Bordeaux, France CIC 1401 EC, Pôle Sant
Bordeaux, Australia
Michelle Lupton
QIMR Berghofer Medical Research Institute
Brisbane, Australia
Jurgen Fripp
Australian e-Health Research Centre, CSIRO Health and security
Brisbane, Australia
Peter Visscher
Institute for Molecular Bioscience, the University of Queensland
Brisbane, Australia
Introduction:
Grey-matter atrophy in Alzheimer's disease (AD) maps the neuronal and synaptic loss associated with neurodegeneration. Atrophy patterns may be used to predict the disease risk or serve as secondary endpoints in clinical trials. We sought to extend our knowledge about grey-matter regions (thickness or surface) associated with AD status, as well as those associated with Mild Cognitive Impairment, Alzheimer's conversion, parental history of dementia, and cognitive or functional scores. All of these can inform on the progression, and nature of the grey-matter atrophy across the disease stages.
Methods:
We gathered T1w MRI brain images and associated data from 10 neuroimaging cohorts (N=9,140) and from the population-based UK Biobank (N=37,664). We performed Region of Interest (ROI; Desikan-Killiani atlas) as well as vertex-wise analyses, using grey-matter measurements generated using pipelines from the ENIGMA consortium (based on FreeSurfer 6.0). Analyses included estimation of whole-brain morphometricity (fraction of explained variance by all brain [ROI or vertex-wise] measurements), as well as brain wide association studies. We considered 24 traits of interest, that focus on different disease stages (e.g. conversion within 1-5 years, MCI, AD status), disease risk (parental history) and scales (e.g. MMSE, CDR, GDS, RAVLT) that capture more specific cognitive, psychiatric or functional dimensions. To boost statistical power, we meta-analysed results from 8 cohorts, and validated the findings using replication and out-of-sample prediction in 2 independent cohorts. We contrasted the grey-matter regions associated with the different traits, or those that we identified using ROI and vertex-based approach.
Results:
We found significant morphometricity between our traits and the grey-matter measurements, except for parental history of AD, and the Geriatric Depression Scale. Vertex-wise morphometricity was 3 to 21 times larger than the ROI based one, indicating that vertex-wise data captures more signal of interest. We identified 94 trait-ROI significant associations, and 307 distinct clusters of trait-vertex associations, partly overlapping with the ROI findings. For AD vs. controls (N=796 cases, 2752 controls), our results confirm atrophy of the hippocampus, amygdala and of the medial temporal lobe (fusiform and parahippocampal gyri) and our vertex-wise results provide a precise localisation of the atrophied regions. In addition, we identified replicable atrophy in several subcortical (putamen, accumbens) and cortical regions (inferior parietal, postcentral, middle temporal, transverse temporal, inferior temporal, paracentral, superior frontal), some of which have rarely been reported. The analysis of AD conversion, MCI status and cognitive/functional scores yielded fewer associated regions, that partly overlapped with the regions associated with AD. We combined the significant ROI or vertices to build interpretable predictors, which achieved statistically significant out of sample prediction (AUC in 0.53-0.70). Lastly, we found that the AD grey-matter score could predict cognition, MCI status, conversion, genetic risk, or tau concentration from CSF in non-diseased individuals (AUC in 0.54-0.70).
Conclusions:
Our large sample size, systematic replication and out-of-sample prediction provides robust maps of grey-matter atrophy in AD. Our joint analyses of AD status, conversion and cognitive/functional scores help shed light on the evolution of atrophy across the disease stages, and its relationship with cognitive or functional impairment. The vertex-wise analysis complements the ROI based approach in identifying additional brain regions and offering a localised description of the atrophied regions. All our significant findings explain a fraction of the morphometricity, suggesting that more grey-matter regions remain to be identified using larger samples, to improve prediction and unveil the full map of atrophy in AD.
Disorders of the Nervous System:
Neurodegenerative/ Late Life (eg. Parkinson’s, Alzheimer’s) 2
Learning and Memory:
Long-Term Memory (Episodic and Semantic)
Lifespan Development:
Aging
Neuroanatomy, Physiology, Metabolism and Neurotransmission:
Cortical Anatomy and Brain Mapping 1
Novel Imaging Acquisition Methods:
Anatomical MRI
Keywords:
Cognition
Computational Neuroscience
Degenerative Disease
Meta- Analysis
STRUCTURAL MRI
Sub-Cortical
Other - Grey-matter
1|2Indicates the priority used for review
Provide references using author date format
NA