TY - JOUR
T1 - MRI predictors of amyloid pathology
T2 - results from the EMIF-AD Multimodal Biomarker Discovery study
AU - ten Kate, Mara
AU - Redolfi, Alberto
AU - Peira, Enrico
AU - Bos, Isabelle
AU - Vos, Stephanie J.
AU - Vandenberghe, Rik
AU - Gabel, Silvy
AU - Schaeverbeke, Jolien
AU - Scheltens, Philip
AU - Blin, Olivier
AU - Richardson, Jill C.
AU - Bordet, Regis
AU - Wallin, Anders
AU - Eckerstrom, Carl
AU - Molinuevo, Jose Luis
AU - Engelborghs, Sebastiaan
AU - Van Broeckhoven, Christine
AU - Martinez-Lage, Pablo
AU - Popp, Julius
AU - Tsolaki, Magdalini
AU - Verhey, Frans R. J.
AU - Baird, Alison L.
AU - Legido-Quigley, Cristina
AU - Bertram, Lars
AU - Dobricic, Valerija
AU - Zetterberg, Henrik
AU - Lovestone, Simon
AU - Streffer, Johannes
AU - Bianchetti, Silvia
AU - Novak, Gerald P.
AU - Revillard, Jerome
AU - Gordon, Mark F.
AU - Xie, Zhiyong
AU - Wottschel, Viktor
AU - Frisoni, Giovanni
AU - Visser, Pieter Jelle
AU - Barkhof, Frederik
PY - 2018/9/27
Y1 - 2018/9/27
N2 - Background: With the shift of research focus towards the pre-dementia stage of Alzheimer's disease (AD), there is an urgent need for reliable, non-invasive biomarkers to predict amyloid pathology. The aim of this study was to assess whether easily obtainable measures from structural MRI, combined with demographic data, cognitive data and apolipoprotein E (APOE) epsilon 4 genotype, can be used to predict amyloid pathology using machine-learning classification. Methods: We examined 810 subjects with structural MRI data and amyloid markers from the European Medical Information Framework for Alzheimer's Disease Multimodal Biomarker Discovery study, including subjects with normal cognition (CN, n = 337, age 66.5 +/- 72, 50% female, 27% amyloid positive), mild cognitive impairment (MCI, n = 375, age 69. 1 +/- 7.5, 53% female, 63% amyloid positive) and AD dementia (n = 98, age 67.0 +/- 7.7, 48% female, 97% amyloid positive). Structural MRI scans were visually assessed and Freesurfer was used to obtain subcortical volumes, cortical thickness and surface area measures. We first assessed univariate associations between MRI measures and amyloid pathology using mixed models. Next, we developed and tested an automated classifier using demographic, cognitive, MRI and APOE epsilon 4 information to predict amyloid pathology. A support vector machine (SVM) with nested 10-fold cross-validation was applied to identify a set of markers best discriminating between amyloid positive and amyloid negative subjects. Results: In univariate associations, amyloid pathology was associated with lower subcortical volumes and thinner cortex in AD-signature regions in CN and MCI. The multi-variable SVM classifier provided an area under the curve (AUC) of 0.81 +/- O. 07 in MCI and an AUC of 0.74 +/- 0.08 in CN. In CN, selected features for the classifier included APOE epsilon 4, age, memory scores and several MRI measures such as hippocampus, amygdala and accumbens volumes and cortical thickness in temporal and parahippocampal regions. In MCI, the classifier including demographic and APOE epsilon 4 information did not improve after additionally adding imaging measures. Conclusions: Amyloid pathology is associated with changes in structural MRI measures in CN and MCI. An automated classifier based on clinical, imaging and APOE epsilon 4 data can identify the presence of amyloid pathology with a moderate level of accuracy. These results could be used in clinical trials to pre-screen subjects for anti-amyloid therapies.
AB - Background: With the shift of research focus towards the pre-dementia stage of Alzheimer's disease (AD), there is an urgent need for reliable, non-invasive biomarkers to predict amyloid pathology. The aim of this study was to assess whether easily obtainable measures from structural MRI, combined with demographic data, cognitive data and apolipoprotein E (APOE) epsilon 4 genotype, can be used to predict amyloid pathology using machine-learning classification. Methods: We examined 810 subjects with structural MRI data and amyloid markers from the European Medical Information Framework for Alzheimer's Disease Multimodal Biomarker Discovery study, including subjects with normal cognition (CN, n = 337, age 66.5 +/- 72, 50% female, 27% amyloid positive), mild cognitive impairment (MCI, n = 375, age 69. 1 +/- 7.5, 53% female, 63% amyloid positive) and AD dementia (n = 98, age 67.0 +/- 7.7, 48% female, 97% amyloid positive). Structural MRI scans were visually assessed and Freesurfer was used to obtain subcortical volumes, cortical thickness and surface area measures. We first assessed univariate associations between MRI measures and amyloid pathology using mixed models. Next, we developed and tested an automated classifier using demographic, cognitive, MRI and APOE epsilon 4 information to predict amyloid pathology. A support vector machine (SVM) with nested 10-fold cross-validation was applied to identify a set of markers best discriminating between amyloid positive and amyloid negative subjects. Results: In univariate associations, amyloid pathology was associated with lower subcortical volumes and thinner cortex in AD-signature regions in CN and MCI. The multi-variable SVM classifier provided an area under the curve (AUC) of 0.81 +/- O. 07 in MCI and an AUC of 0.74 +/- 0.08 in CN. In CN, selected features for the classifier included APOE epsilon 4, age, memory scores and several MRI measures such as hippocampus, amygdala and accumbens volumes and cortical thickness in temporal and parahippocampal regions. In MCI, the classifier including demographic and APOE epsilon 4 information did not improve after additionally adding imaging measures. Conclusions: Amyloid pathology is associated with changes in structural MRI measures in CN and MCI. An automated classifier based on clinical, imaging and APOE epsilon 4 data can identify the presence of amyloid pathology with a moderate level of accuracy. These results could be used in clinical trials to pre-screen subjects for anti-amyloid therapies.
KW - Alzheimer's disease
KW - Mild cognitive impairment
KW - Biomarkers
KW - Magnetic resonance imaging
KW - Amyloid
KW - Machine learning
KW - Support vector machine
KW - European Medical Information Framework for Alzheimer's Disease
KW - MILD COGNITIVE IMPAIRMENT
KW - ALZHEIMERS-DISEASE
KW - BETA DEPOSITION
KW - SELECTION BIAS
KW - DEMENTIA
KW - ATROPHY
KW - PREVENTION
KW - MACHINE
KW - DECLINE
KW - PROJECT
U2 - 10.1186/s13195-018-0428-1
DO - 10.1186/s13195-018-0428-1
M3 - Article
C2 - 30261928
SN - 1758-9193
VL - 10
JO - Alzheimer's Research & Therapy
JF - Alzheimer's Research & Therapy
M1 - 100
ER -