Risk score roulette: A cautionary tale of polygenic risk score reliability

Authors: XUE, DIANE - University Of Pennsylvania; BLUE, ELIZABETH - University Of Washington; WOOD, ALEXIS - Children'S Nutrition Research Center (CNRC); ROTTER, JEROME - Harbor-Ucla Medical Center; FOHNER, ALISON - University Of Washington

Submitted to: Journal of Alzheimer's Disease
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/23/2025
Publication Date: 9/9/2025
Citation: Xue, D., Blue, E.E., Wood, A.C., Rotter, J.I., Fohner, A.E. 2025. Risk score roulette: A cautionary tale of polygenic risk score reliability. Journal of Alzheimer's Disease. 108(1):65-70. https://doi.org/10.1177/13872877251375098.
DOI: https://doi.org/10.1177/13872877251375098

Interpretive Summary: Obesity increases the risk of developing Alzheimer's disease (AD), and accelerates the cognitive decline seen in AD. This study looked at how well different genetic models can predict a person's risk of developing AD. While these models, called polygenic risk scores (PRS), seemed to perform similarly at the population level, they often gave inconsistent results when used to predict risk for an individual. In other words, the same person might be classified as high-risk by one model and low-risk by another. This is a problem because unreliable predictions could misguide personal health decisions or future treatments. Since PRS are increasingly also being developed for obesity and related conditions, this raises similar concerns about their reliability for guiding prevention or intervention across a host of conditions. For patients, families, and healthcare providers, the findings highlight the need for caution when using genetic risk scores in medical decisions. For researchers, public health officials, and policy makers, this study underscores the importance of testing the reliability of these tools before applying them widely.

Technical Abstract: Genetic risk prediction for Alzheimer's disease (AD) has high potential impact, yet few studies have assessed the reliability of various polygenic risk score (PRS) methods at the individual level. Here, we evaluated the reliability of AD PRS estimates among 6338 participants from the Multi-Ethnic Study of Atherosclerosis. We compared four PRS models that have been previously associated with dementia risk. Despite similar population-level performance metrics, inter-model reliability of individual-level risk assessment was low, even among individuals classified in the top and bottom deciles. These findings raise serious concerns about the downstream application of PRS for guiding interventions for AD.