Chronic coronary syndrome (CCS) is a newly proposed concept and is hallmarked by more long-term major adverse cardiovascular events (MACEs), calling for accurate prognostic biomarkers for initial risk stratification.
Data-independent acquisition liquid chromatography tandem mass spectrometry (DIA LC-MS/MS) quantitative proteomics was performed on 38 patients with CCS; 19 in the CCS events group and 19 in the non-events group as the controls. We also developed a machine-learning-based pipeline to identify proteins as potential biomarkers and validated the target proteins by enzyme-linked immunosorbent assay in an independent prospective cohort.
Fifty-seven differentially expressed proteins were identified by quantitative proteomics and three final biomarkers were preliminarily selected from the machine-learning-based pipeline. Further validation with the prospective cohort showed that endothelial protein C receptor (EPCR) and cholesteryl ester transfer protein (CETP) levels at admission were significantly higher in the CCS events group than they were in the non-events group, whereas the carboxypeptidase B2 (CPB2) level was similar in the two groups. In the Cox survival analysis, EPCR and CETP were independent risk factors for MACEs. We constructed a new prognostic model by combining the Framingham coronary heart disease (CHD) risk model with EPCR and CETP levels. This new model significantly improved the C-statistics for MACE prediction compared with that of the Framingham CHD risk model alone.
Plasma proteomics was used to find biomarkers of predicting MACEs in patients with CCS. EPCR and CETP were identified as promising prognostic biomarkers for CCS.