Development and Validation of a Clinical Polygenic Risk Report in U.S.-Based Health Systems for 8 Cardiovascular Conditions.
Development and Validation of a Clinical Polygenic Risk Report in U.S.-Based Health Systems for 8 Cardiovascular Conditions.
👥 作者
Misra Anika
(Program in Medical and Population Genetics and the Cardiovascular Disease Initiative)
Jowell Amanda
(Broad Institute of MIT and Harvard)
Haidermota Sara
(Cambridge)
Perez Emma
(Massachusetts)
Mahanta Lisa
(USA; Center for Genomic Medicine)
O'Brien Kimberly J
(Massachusetts General Hospital)
Nagy Anna
(Boston)
Hao Limin
(Massachusetts)
Truong Buu
(USA.; Program in Medical and Population Genetics and the Cardiovascular Disease Initiative)
Aragam Krishna
(Broad Institute of MIT and Harvard)
Fahed Akl C
(Cambridge)
Khurshid Shaan
(Massachusetts)
Ellinor Patrick T
(USA; Center for Genomic Medicine)
Lewis Anna C F
(Massachusetts General Hospital)
Lennon Niall
(Boston)
Hornsby Whitney
(Massachusetts)
Lebo Matthew S
(USA; Mass General Brigham Department of Medicine)
Karlson Elizabeth W
(Boston)
Natarajan Pradeep
(Massachusetts)
Patel Aniruddh P
(USA.; Program in Medical and Population Genetics and the Cardiovascular Disease Initiative)
📝 摘要
Polygenic risk scores (PRS) stratify inherited cardiovascular risk, but their path to clinical implementation remains unclear. We aimed to develop and validate integrated PRS for 8 cardiovascular conditions and outline a framework for their clinical reporting. We analyzed genotype and clinical data from 245,394 All of Us Research Program participants. Publicly available PRS for 8 traits-coronary artery disease, atrial fibrillation, type 2 diabetes, venous thromboembolism (VTE), thoracic aortic aneurysm (TAA), extreme hypertension, severe hypercholesterolemia, and elevated lipoprotein(a)-were combined using PRSmix, an elastic-net approach. Integrated PRS were externally validated in 53,306 Mass General Brigham Biobank participants using logistic regression, adjusting for age, sex, and ancestry. Of 53,306 genotyped Mass General Brigham Biobank participants (55.6% women, mean age 53 ± 17 years), integrated PRS demonstrated robust discrimination and appropriate calibration across 8 cardiovascular traits. Comparing high genetic risk (top 10% of PRS distribution, or top 20% for rarer TAA and VTE) vs average risk (26th-75th percentiles, or 21st-80th percentiles for TAA and VTE) yielded ORs: coronary artery disease (3.7 [95% CI: 3.4-4.1]), type 2 diabetes (3.1 [95% CI: 2.8-3.3]), atrial fibrillation (3.0 [95% CI: 2.7-3.3]), VTE (1.9 [95% CI: 1.6-2.0]), TAA (1.7 [95% CI: 1.5-1.9]), hypertension (2.1 [95% CI: 1.8-2.3]), hypercholesterolemia (4.1 [95% CI: 3.7-4.5]), and lipoprotein(a) (41.0 [95% CI: 27.0-62.2]). Incorporating integrated PRS into clinical models improved risk classification, while prospective analyses confirmed significant associations with incident cardiovascular outcomes. Integrated PRS offer an implementable framework for genetic risk reporting, and are now available as a clinically orderable test. Broader prospective validation studies are needed to further establish clinical utility.