Large-scale, comprehensive plasma metabolomic analyses reveal potential biomarkers for the diagnosis of early-stage coronary atherosclerosis

Background: Coronary atherosclerosis (CAS) is a prevalent and chronic life-threatening disease. However, the detection of CAS at an early stage is difficult because of the lack of effective noninvasive diagnostic methods. The present study aimed to characterize the plasma metabolome of early-stage CAS patients to discover metabolomic biomarkers, develop a novel metabolite-based model for accurate noninvasive diagnosis of early-stage CAS, and explore the underlying metabolic mechanisms involved. Methods: A total of 100 patients with early-stage CAS and 120 age- and sex-matched control subjects were recruited from the Chinese Han population and further randomly divided into training (n = 120) and test sets (n = 100). The metabolomic profiles of the plasma samples were analyzed by an integrated untargeted liquid chromatography-mass spectrometry approach, including two separation modes and two ionization modes. Univariate and multivariate statistical analyses were employed to identify potential biomarkers and construct an early-stage CAS diagnostic model. Results: The integrated analytical method established herein improved metabolite coverage compared with single chromatographic separation and MS ionization mode. A total of 80 metabolites were identified as potential biomarkers of early-stage CAS, and these metabolites were mainly involved in glycerophospholipid, fatty acid, sphingolipid, and amino acid metabolism. An effective diagnostic model for early-stage CAS was established, incorporating 11 metabolites and achieving areas under the receiver operating characteristic curve (AUCs) of 0.984 and 0.908 in the training and test sets, respectively. Conclusions: Our study not only successfully developed an effective noninvasive diagnostic model for identifying early-stage CAS but also provided novel insights into the pathogenesis of CAS.