Coronary computed tomography angiography (CTA) describes several features of coronary plaques, i.e. location, severity, and composition. Integrated CTA scores are able to identify individual patterns of higher risk. We sought to test whether circulating biomarkers related with metabolism and inflammation could predict high risk coronary anatomy at CTA in patients with stable chest pain.We evaluated a panel of 17 biomarkers in 429 patients (60.3 ± 0.4 years, 268 males) with stable chest pain who underwent coronary CTA having been enrolled in the Evaluation of Integrated Cardiac Imaging (EVINCI) study. The individual CTA risk score was calculated combining plaque extent, severity, composition, and location. The presence and distribution of non-calcified, mixed and calcified plaques were analyzed in each patient.After adjustment for age, sex and medical treatment, high-density lipoprotein (HDL) cholesterol, leptin, and interleukin-6 (IL-6) were independent predictors of CTA risk score at multivariate analysis (P = 0.050, 0.002, and 0.007, respectively). Integrating these biomarkers with common clinical variables, a model was developed which showed a better discriminating ability than the Framingham Risk Score and the Euro-SCORE in identifying the patients with higher CTA risk score (area under the receiver-operating characteristics curve = 0.81, 0.63 and 0.71, respectively, P < 0.001). These three biomarkers were significantly altered in patients with mixed or non-calcified plaques.In patients with stable chest pain, low HDL cholesterol, low leptin and high IL-6 are independent predictors of high risk coronary anatomy as defined by an integrated CTA risk score.