It has been demonstrated that long non-coding RNAs (lncRNAs) are important in the gene regulatory network and their dysregulated expression has been implicated in cardiovascular disease. However, little is known regarding lncRNA expression patterns and their function in the progression of acute coronary syndromes (ACSs). In the present study, the expression profiles of lncRNAs from 52 patients with ACS were analyzed by re-annotating existing microarray data. The lncRNA expression profiles in the two distinct clinical entities of ACS, myocardial infarction (MI) and unstable angina (UA), were examined. Out of the 2,332 lncRNAs assessed, it was identified that 18 lncRNAs were upregulated and 35 lncRNAs were downregulated in patients with MI compared to those with UA. Furthermore, the expression profiles of patients with ACS were compared at different time points and significantly altered lncRNA expression was observed during the progression of ACS. A total of 7 candidate lncRNA biomarkers were identified and an lncRNA-based classifier was developed to predict MI risk based on the expression data of the 7 lncRNAs using random forest and support vector machine strategies. This achieved a classification accuracy of 90.38% with a sensitivity of 100% and a specificity of 68.75%. Additionally, functional analysis suggested that these 7 lncRNAs may be involved in known MI-associated biological processes and pathways.
Keywords: acute coronary syndrome; diagnosis; expression profiles; long non-coding RNAs.