Identification of novel candidate biomarkers for acute myocardial infarction by the Olink proteomics platform

Both pathological and normal processes depend on proteins. In this study, plasma protein profiles were analyzed by a novel proximity extension assay (PEA) to identify potential pathogenic mechanisms and diagnostic biomarkers in patients diagnosed with acute myocardial infarction (AMI).In this study, we identified a total of 92 plasma proteins using the Olink Target 96 Cardiovascular III panel in a cohort consisting of 30 healthy controls (HC), 28 patients with unstable angina (UA) and 30 patients with AMI. Subsequently, we conducted a differential expression analysis to identify protein molecules that were specifically expressed in patients with AMI. To gain insights into the potential functional mechanisms of these differentially expressed molecules, we performed Gene Ontology (GO) enrichment and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses. Following that, the utilization of least absolute shrinkage and selection operator (LASSO) regression facilitated the identification of potential protein biomarkers, enabling the differentiation between AMI and UA. A diagnostic model was subsequently developed through logistic regression, and the effectiveness of these markers was assessed using receiver operating characteristic (ROC) analysis. Ultimately, the diagnostic capabilities of these potential biomarkers were validated in an independent validation cohort consisting of 30 UA cases and 30 AMI cases.In this study, a comprehensive analysis of plasma proteins identified a total of 92 proteins. Further analysis using analysis of variance revealed that 25 proteins exhibited specific expression in the AMI group compared to the HC and UA groups. Additionally, KEGG enrichment analysis indicated that these differentially expressed proteins were primarily associated with the activation of cytokine-cytokine receptor interaction, PI3K-Akt signaling pathway, and GnRH signaling pathway. AGRP, TGM2, IL6, GH1, and CA5A were identified through LASSO regression as prospective protein biomarkers for distinguishing between UA and AMI. The diagnostic model comprising these five proteins exhibited exceptional performance in both the discovery and validation datasets, surpassing AUC values of 0.9.The findings of our study provide additional insights into the involvement of the inflammatory response and AKT cascade response in the development of AMI. Moreover, we have identified potential protein markers that could be utilized for the accurate diagnosis of AMI. These results offer a fresh perspective for clinical decision-making in the context of AMI.