Blood-based transcriptional signatures at hospital admission are associated with cardiac magnetic resonance markers of STEMI prognosis

Abstract Funding Acknowledgements Type of funding sources: Public Institution(s). Main funding source(s): Italian Ministry of Health. Background While mortality following ST-segment elevation myocardial infarction (STEMI) is declining, the number of patients developing heart failure due to STEMI is on the rise. Cardiovascular magnetic resonance (CMR) has emerged as an important imaging modality for assessing parameters of the short-term outcome, which add incremental prognostic value above traditional outcome markers alone in acute reperfused STEMI. Prognostic relevant pathways of leukocyte involvement in STEMI outcome are largely unknown. Aim We sought to identify a set of circulating cellular transcripts measured on hospital admission that predict the short-term outcome of STEMI patients as assessed by CMR markers. Methods Thirty consecutive patients (24 males, 6 females, age 61 ± 10 years) admitted with STEMI at our Centre between 2012 and 2015 were enrolled in this retrospective pilot study. The whole-blood transcriptome was analysed by RNA-sequencing using total RNA isolated from peripheral blood samples drawn on hospital admission. Patients were studied with a 1.5T MR scanner within 1 week after primary PCI, and late gadolinium enhancement (LGE) mass and myocardial salvage index (MSI) were measured. K-means clustering was performed to group the samples according to the distribution of values of each CMR variable (i.e., high, medium, and low levels). We used the generalized negative binomial linear model approach with the edgeR package to perform gene-level differential expression analysis among the 3 groups of each CMR variable. To infer the biological functions of genes associated with CMR parameters, we used Gene Ontology Biological Processes terms for Gene Set Enrichment Analysis. To identify the smallest set of genes that can discriminate between groups of patients, we exploited a class of adaptive heuristic search algorithms using the GARS package. Results We identified specific gene expression patterns at baseline associated with high, medium, or low LGE and MSI values at 1-week follow-up. The most representative processes associated with high values of LGE were suggestive of adaptive immune response mediated by T- and B-cells, while innate immune response pathways were associated with medium-lower LGE values. We observed an association of low MSI values with adaptive immune response processes; high-medium values of MSI were, instead, associated with inflammatory-related functions. As for predictors of CMR surrogate markers of patient’s outcome, we identified a set of 13 genes that classified the three LGE mass groups with an accuracy of 91%. Similarly, we identified 13 specific genes that discriminated the three MSI groups with a 100% accuracy. Conclusions We unveiled through RNA-sequencing data mining a set of informative transcriptional features that predict CMR phenotypes after STEMI. Overall, our results could pave the way for the identification of novel blood-based biomarkers to improve early prognosis and therapeutic decision-making in STEMI.