A polygenic risk score predicts mosaic loss of chromosome Y in circulating blood cells

Cell & Bioscience

Moeen Riaz¹  Andrew Bakshi¹  Paul Lacaze¹  John McNeil¹  Galina Polekhina¹  Jonatan Halvardson²  Jonas Mattisson²  Adam Ameur²  Marcus Danielsson²  Lars A. Forsberg³ 

[1] Department of Epidemiology and Preventive Medicine, School of Public Health and Preventive Medicine, Monash University, Melbourne, Australia;Department of Immunology, Genetics and Pathology, Science for Life Laboratory, Uppsala University, Uppsala, Sweden;Department of Immunology, Genetics and Pathology, Science for Life Laboratory, Uppsala University, Uppsala, Sweden;The Beijer Laboratory, Uppsala University, Uppsala, Sweden;

关键词: Mosaic loss of chromosome Y;    LOY;    mLOY;    Polygenic risk score;    PRS;    ASPREE;   

DOI  :  10.1186/s13578-021-00716-z

来源: Springer

【 摘 要 】

BackgroundMosaic loss of Y chromosome (LOY) is the most common somatic change that occurs in circulating white blood cells of older men. LOY in leukocytes is associated with increased risk for all-cause mortality and a range of common disease such as hematological and non-hematological cancer, Alzheimer’s disease, and cardiovascular events. Recent genome-wide association studies identified up to 156 germline variants associated with risk of LOY. The objective of this study was to use these variants to calculate a novel polygenic risk score (PRS) for LOY, and to assess the predictive performance of this score in a large independent population of older men.ResultsWe calculated a PRS for LOY in 5131 men aged 70 years and older. Levels of LOY were estimated using microarrays and validated by whole genome sequencing. After adjusting for covariates, the PRS was a significant predictor of LOY (odds ratio [OR] = 1.74 per standard deviation of the PRS, 95% confidence intervals [CI] 1.62–1.86, p < 0.001). Men in the highest quintile of the PRS distribution had > fivefold higher risk of LOY than the lowest (OR = 5.05, 95% CI 4.05–6.32, p < 0.001). Adding the PRS to a LOY prediction model comprised of age, smoking and alcohol consumption significantly improved prediction (AUC = 0.628 [CI 0.61–0.64] to 0.695 [CI 0.67–0.71], p < 0.001).ConclusionsOur results suggest that a PRS for LOY could become a useful tool for risk prediction and targeted intervention for common disease in men.

【 授权许可】

CC BY