MRI predictors of amyloid pathology: results from the EMIF-AD Multimodal Biomarker Discovery study

Abstract

BackgroundWith 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) ε4 genotype, can be used to predict amyloid pathology using machine-learning classification.MethodsWe 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 ± 7.2, 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 ε4 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.ResultsIn 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 ± 0.07 in MCI and an AUC of 0.74 ± 0.08 in CN. In CN, selected features for the classifier included APOE ε4, 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 ε4 information did not improve after additionally adding imaging measures.ConclusionsAmyloid pathology is associated with changes in structural MRI measures in CN and MCI. An automated classifier based on clinical, imaging and APOE ε4 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.Electronic supplementary materialThe online version of this article (10.1186/s13195-018-0428-1) contains supplementary material, which is available to authorized users.

Reduction of recruitment costs in preclinical AD trials: validation of automatic pre-screening algorithm for brain amyloidosis

“…The AUC we obtain on the MCI cohort is comparable to the ones obtained in other studies or slightly higher [43,4,42]. Ten Kate et al [42] obtain a slightly Statistical Methods in Medical Research. 2019. https://doi.org/10.1177/0962280218823036 better AUC for the prediction in CN subjects.…”

“…In future studies, different neuroimaging features could be used to test this hypothesis that cognitive changes are anterior to substantial structural changes, in line with previous studies on optimal neuroimaging feature selection in pre-clinical AD [29]. Alternatively, a more advanced feature selection algorithm might be able to identify the most informative MRI features and therefore improve their performance, as proposed in other methods [42].…”

“…Apostolova et al [4] also include MRI features by using hippocampus volume and cognitive, APOE and peripheral blood protein information on MCI subjects using an SVM. Ten Kate et al [42] use an SVM and tree-based feature selection to predict amyloidosis in CN and MCI subjects using cognitive, socio-demographic, APOE and MRI features. In this paper, we propose to take a cost-effective approach of the amyloidosis prediction, by comparing different methods in terms of cost reduction.…”

Detecting Amyloid Positivity in Elderly With Increased Risk of Cognitive Decline

“…Very few studies have focused specifically on cognitively normal populations, despite the key importance of this group who could potentially benefit from interventions that are started early, before the onset of cognitive impairment. Lower performance has been reported for models developed in cognitively normal individuals, with AUC values up to 0.74-0.77 (Mielke et al, 2012;Insel et al, 2016;ten Kate et al, 2018;Ansart et al, 2020). Models for detecting Aβ pathology have most often been developed based on demographic data, cognitive performance, and apolipoprotein E (APOE) genotype.…”

Multimodal magnetic resonance imaging predicts regional amyloid- β burden in the brain

“…A limited number of prior studies have used MR imaging to predict amyloid- β positivity. Ten et al (2018) used a combination of features as predictors for amyloid- β status prediction (Ten Kate et al, 2018). They used subject demographics, cognitive variables, regional estimates of volume and cortical thickness from MRI, and APOE ε4 information along with machine learning classifier called support vector machine (SVM) with nested 10-fold crossvalidation to identify the best discriminating features between amyloid- β positive and amyloid- β negative groups.…”