Abstract 10637: A Pooled Multi-National Validation Study of a Machine Learning, High-Sensitivity Troponin-Based Multi-Proteomic Model to Predict the Presence of Obstructive Coronary Artery Disease as Compared to High Sensitivity Troponin Alone in a Subset of Patients with Diabetes

Abstract

Introduction: Diabetes Mellitus (DM) is a major risk factor for coronary artery disease (CAD) associated with a two-fold increase in mortality. We aimed to validate the performance of a machine learning based multi-biomarker diagnostic panel to predict obstructive coronary artery disease (oCAD) compared to high-sensitivity cardiac troponin-I (hs-cTnI) alone in a subset of patients with DM. Methods: A previously developed multiple biomarker scoring model was utilized. The pooled cohort included 132 patients at Inova Fairfax Hospital, 69 patients at Massachusetts General Hospital and 46 patients at the University Hospital Hamburg-Eppendorf with a mixture of acute and lesser acute presentations. Three clinical factors (sex, age, and previous coronary percutaneous intervention) and three biomarkers (hs-cTnI, Adiponectin, and Kidney Injury Molecule-1) were combined. oCAD was defined as >50% coronary obstruction in at least one coronary artery (for the University Hospital Hamburg-Eppendorf cohort) or >70% coronary obstruction in at least one coronary artery (for the other two cohorts). The multiple biomarker diagnostic panel’s performance to predict oCAD was also compared to hs-cTnI alone. Results: The multiple protein panel had an area under the receiver-operating characteristic curve of 0.77 (95% CI, 0.70, 0.84, p <0.001) for the presence of oCAD (Figure 1). At optimal cutoff, the score had 79% sensitivity, 57% specificity, and a positive predictive value of 81% for oCAD. The multiple biomarker panel had a diagnostic odds ratio of 4.82 (95% CI 2.68, 8.67, p<0.01). In comparison, in patients without an acute MI, hs-cTnI alone had an area under the receiver-operating characteristic curve of 0.53 (95% CI, 0.49, 0.57, p = 0.253) for oCAD (Figure 1). Conclusions: In this multinational pooled cohort, a previously described novel machine learning, multiple protein biomarker panel provided high accuracy to diagnose patients for oCAD in a subset of patients with DM.