Modeling the Impact of Extracellular Vesicle Cargoes in the Diagnosis of Coronary Artery Disease

: We aimed to assess the relationship among circulating extracellular vesicles (EVs), hypoxia-related proteins, and the conventional risk factors of life-threatening coronary artery disease (CAD) to find more precise novel biomarkers. : Patients were categorized based on coronary CT angiography. Patients with a Segment Involvement Score > 5 were identified as CAD patients. Individuals with a Segment Involvement Score < 5 were considered control subjects. The characterization of EVs and analysis of the plasma concentration of growth differentiation factor-15 were performed using multicolor or bead-based flow cytometry. The plasma protein levels of glycogen phosphorylase, muscle form, clusterin, and carboxypeptidase N subunit 1 were determined using an enzyme-linked immunosorbent assay. Multiple logistic regression was used to determine the association of the biomarkers with the CAD outcome after accounting for established risk factors. The analysis was built in three steps: first, we included the basic clinical and laboratory variables (Model 1), then we integrated the plasma protein values (Model 2), and finally, we complemented it with the circulating EV pattern (Model 3). To assess the discrimination value of the models, an area under (AUC) the receiver operating curve was calculated and compared across the three models. : The area under the curve (AUC) values were 0.68, 0.77, and 0.84 in Models 1, 2, and 3, respectively. The variables with the greatest impact on the AUC values were hemoglobin (0.2 (0.16-0.26)) in Model 1, carboxypeptidase N subunit 1 (0.12 (0.09-0.14)) in Model 2, and circulating CD41+/CD61+ EVs (0.31 (0.15-0.5)) in Model 3. A correlation analysis showed a significant impact of circulating CD41+/CD61+ platelet-derived EVs ( = 0.03, r = -0.4176) in Model 3. : Based on our results, the circulating EV profile can be used as a supportive biomarker, along with the conventional laboratory markers of CAD, and it enables a more sensitive, non-invasive diagnostic analysis of CAD.