Integrated Non-Contrast CT Biomarkers for Enhanced Screening of Significant Coronary Artery Disease

Abstract Background Coronary artery calcium (CAC) scoring is widely used for cardiovascular risk assessment. However, its limited sensitivity for detecting non-calcified and inflammatory coronary artery disease (CAD) has prompted interest in additional biomarkers from non-contrast CT to improve screening accuracy. Purpose To evaluate the diagnostic performance of a novel multiparametric non-contrast CT-based scoring model- NOMICAD (Non-Contrast Multiparametric Imaging Score for Coronary Artery Disease) to screen for significant coronary artery disease . Materials and Methods This retrospective single-center observational study included patients who underwent CACS and coronary CT angiography (CCTA) between December 2021 and December 2025. Significant CAD was defined as CAD-RADS ≥3. Additional non-contrast CT calcium score-derived biomarkers, including pericoronary fat thickness (PCFT), pericoronary fat attenuation (PCFA), aortic root calcification, and liver attenuation were evaluated. PCFT and PCFA measurements from non-contrast calcium score CT were compared with CCTA-derived measurements to assess correlation, agreement, and feasibility as surrogate markers of coronary inflammation and atherosclerotic burden. These parameters were subsequently integrated with CACS to derive Scoring System (NOMICAD) using multivariable logistic regression analysis. Diagnostic performance of the scoring system was evaluated using receiver operating characteristic (ROC) analysis and decision curve analysis, with CAD-RADS classification on CCTA used as the reference standard. Results A total of 98 patients were included. CACS alone demonstrated good discrimination for significant CAD (AUC 0.79; 95% CI 0.70–0.88). PCFA and PCFT demonstrated significant association with CAD-RADS ≥3. PCFA and PCFT measured on non-contrast CACS demonstrated strong agreement to CCTA in predicting significant coronary artery disease (CAD-RADS ≥3). Strong positive correlation with strong agreement between the two modalities suggest that parameters obtained from CT calcium scoring may serve as reliable surrogate markers for those measured on CT coronary angiography. The derived multiparametric NOMICAD model demonstrated robust diagnostic performance with an AUC of 0.97 (95% CI 0.94–0.99), significantly outperforming calcium scoring alone. Decision curve analysis demonstrated superior net clinical benefit across clinically relevant thresholds. Conclusion NOMICAD, integrating coronary calcium scoring with additional non-contrast CT-derived inflammatory biomarkers, significantly improves screening performance for clinically significant coronary artery disease compared with calcium scoring alone. Multiparametric CT-based risk stratification may improve early identification of high-risk patients.