Improved multimodal prediction of progression from MCI to Alzheimer's disease combining genetics with quantitative brain MRI and cognitive measures
Emilie T. Reas, Alexey Shadrin, Oleksandr Frei, Ehsan Motazedi, Linda K. McEvoy, Shahram Bahrami, Dennis van der Meer, Carolina Makowski, Robert Loughnan, Xin Wang, Iris Broce, Sarah J. Banks, Vera Fominykh, Weiqiu Cheng, Dominic Holland, Olav B. Smeland, Tyler M. Seibert, Geir Selbæk, James Brewer, Chun Chieh Fan, Ole A. Andreassen, Anders M. Dale, for the Alzheimer's Disease Neuroimaging Initiative
Abstract
INTRODUCTION: There is a pressing need for non-invasive, cost-effective tools for early detection of Alzheimer's disease (AD). METHODS: Using data from the Alzheimer's Disease Neuroimaging Initiative (ADNI), Cox proportional models were conducted to develop a multimodal hazard score (MHS) combining age, a polygenic hazard score (PHS), brain atrophy, and memory to predict conversion from mild cognitive impairment (MCI) to dementia. Power calculations estimated required clinical trial sample sizes after hypothetical enrichment using the MHS. Cox regression determined predicted age of onset for AD pathology from the PHS. RESULTS: The MHS predicted conversion from MCI to dementia (hazard ratio for 80th versus 20th percentile: 27.03). Models suggest that application of the MHS could reduce clinical trial sample sizes by 67%. The PHS alone predicted age of onset of amyloid and tau. DISCUSSION: The MHS may improve early detection of AD for use in memory clinics or for clinical trial enrichment. HIGHLIGHTS: A multimodal hazard score (MHS) combined age, genetics, brain atrophy, and memory. The MHS predicted time to conversion from mild cognitive impairment to dementia. MHS reduced hypothetical Alzheimer's disease (AD) clinical trial sample sizes by 67%. A polygenic hazard score predicted age of onset of AD neuropathology.