Predicting sporadic Alzheimer's disease progression via inherited Alzheimer's disease‐informed machine‐learning
Nicolai Franzmeier, Nikolaos Koutsouleris, Tammie L.S. Benzinger, Alison Goate, Celeste M. Karch, Anne M. Fagan, Eric McDade, Marco Duering, Martin Dichgans, Johannes Levin, Brian A. Gordon, Yen Ying Lim, Colin L. Masters, Martin N. Rossor, Nick C. Fox, Antoinette O’Connor, Jasmeer P. Chhatwal, Stephen Salloway, Adrian Danek, Jason Hassenstab, Peter R. Schofield, John C. Morris, Randall J. Bateman, the Alzheimer's disease neuroimaging initiative (ADNI), the Dominantly Inherited Alzheimer Network (DIAN), Michael Ewers
Abstract
INTRODUCTION: Developing cross-validated multi-biomarker models for the prediction of the rate of cognitive decline in Alzheimer's disease (AD) is a critical yet unmet clinical challenge. METHODS: We applied support vector regression to AD biomarkers derived from cerebrospinal fluid, structural magnetic resonance imaging (MRI), amyloid-PET and fluorodeoxyglucose positron-emission tomography (FDG-PET) to predict rates of cognitive decline. Prediction models were trained in autosomal-dominant Alzheimer's disease (ADAD, n = 121) and subsequently cross-validated in sporadic prodromal AD (n = 216). The sample size needed to detect treatment effects when using model-based risk enrichment was estimated. RESULTS: = 25%) in sporadic AD. Model-based risk-enrichment reduced the sample size required for detecting simulated intervention effects by 50%-75%. DISCUSSION: Our independently validated machine-learning model predicted cognitive decline in sporadic prodromal AD and may substantially reduce sample size needed in clinical trials in AD.