Four Biomarkers Combined Sharply Boost Heart Disease Risk Prediction

Coronary artery disease remains the leading cause of death globally, yet standard risk calculators often miss high-risk individuals — particularly younger adults — who could benefit most from early intervention. This large prospective study set out to determine whether integrating genomic data with established lipid and inflammatory biomarkers could improve on conventional risk prediction tools.

Using data from 215,695 UK Biobank participants aged 40–69, researchers tracked incident CAD over 12 years. Four biomarkers were assessed at baseline: a CAD polygenic risk score (PRS), LDL cholesterol (LDL-C), lipoprotein(a) (Lp(a)), and high-sensitivity C-reactive protein (hsCRP). Multivariable Cox regression models, C-statistics, and net reclassification indices were calculated across age and sex subgroups.

All four biomarkers independently predicted CAD. Hazard ratios per elevation were 1.79 for PRS, 1.64 for hsCRP, 1.60 for LDL-C, and 1.20 for Lp(a). Individuals with all four biomarkers elevated had a 4.65-fold increased CAD risk. The combined four-biomarker model achieved a C-statistic of 0.753 versus 0.740 for the Pooled Cohort Equations, and delivered a 32% continuous net reclassification improvement. Risk associations were strongest in younger participants regardless of sex, and PRS showed a stronger effect in men than women.

These findings carry real clinical implications. Midlife — when intervention is most impactful — is precisely where this combined model excels. Measuring PRS alongside standard lipid and inflammatory panels could identify high-risk individuals years earlier than current practice allows.

Caveats include the study's predominantly European-ancestry UK Biobank population, which may limit generalizability. Additionally, the abstract-level data available does not detail how biomarker thresholds were defined or how treatment during follow-up may have influenced outcomes.