An Integrated Germline and Somatic Genomic Model Improves Risk Prediction for Coronary Artery Disease

Multiple germline and somatic genomic factors are associated with risk of coronary artery disease (CAD), but there is no single measure of risk that integrates all information from a DNA sample, limiting clinical use of genomic information. To address this gap, we developed an integrated genomic model (IGM), analogous to a clinical risk calculator that combines various clinical risk factors into a unified risk estimate. The IGM includes six genetic drivers for CAD, including germline factors (familial hypercholesterolemia [FH] variants, CAD polygenic risk score [PRS], proteome PRS, metabolome PRS) and somatic factors (clonal hematopoiesis of indeterminate potential [CHIP], and leukocyte telomere length [LTL]). We evaluated the IGM on CAD risk prediction in the UK Biobank (N=391,536), and validated it in the Trans-Omics for Precision Medicine (TOPMed) program (N=34,177). The 10-year CAD risk based on the IGM profile ranged from 1.1% to 15.5% in the UK Biobank and from 3.8% to 33.0% in TOPMed, with a more pronounced gradient in males than females. IGM captured the cumulative effect of multiple genetic drivers, identifying individuals at high risk for CAD despite lacking obvious high risk genetic factors, or individuals at low risk for CAD despite having known genetic risk variants such as FH and CHIP. The IGM had the highest performance in younger individuals (C-statistic 0.805 [95% CI, 0.699-0.913] for age [≤] 45 years). In middle age, IGM augmented the performance of the Pooled Cohort Equations (PCE), a clinical risk calculator for CAD. Adding IGM to PCE resulted in a continuous net reclassification index of 33.45% (95% CI, 32.11%-34.76%). We present the first model that integrates all currently available information from a single "DNA biopsy" to translate complex genetic information into a single risk estimate.