Transcriptomic Predictors of Pulmonary Hypertension Development in Heart Failure Patients

Introduction: Over half of heart failure with preserved ejection fraction (HFpEF) patients develop pulmonary hypertension (PH), which has a five-year mortality rate of about 50%. Evidence-based treatment for these patients is needed, but no approved treatments currently exist. The aim of this study was to use RNA-sequencing (RNA-seq) data to identify transcriptomic predictors of isolated post-capillary pulmonary hypertension (IpcPH) and combined pre- and post-capillary pulmonary hypertension (CpcPH) pathogenesis in HFpEF patients. Methods: Whole blood samples were selected from two biobanks, from which peripheral blood mononuclear cells (PBMCs) were extracted. RNA was isolated from PBMCs and RNA-seq was performed. Adjusted pairwise gene expression values were compared among three groups (n=10/group): HFpEF without PH; IpcPH; and CpcPH. The following additional analyses were also performed: gene set enrichment analysis, network analysis, in silico expression quantitative trait loci (eQTL) analysis, and genetic association analysis. Results: After controlling for age, sex, race and smoking status, 152 genes/transcripts were significantly downregulated, and 194 genes/transcripts were significantly upregulated in CpcPH patients compared with HFpEF without PH patients. No significant differentially expressed genes were found in other pairwise comparisons. Based on the enrichment analysis of all 22,418 genes/transcripts, 11 pathways were significantly upregulated in CpcPH compared to HFpEF without PH, one of which, the cell-cycle, maintained significance after correction for multiple comparisons. Network analysis revealed a high level of co-expression among the candidate genes (false discovery rate < 0.05 and log2-fold change > 1.5 or < -1.5) and 6 genes had eQTLs in one of the tissues of interest. Eighty six unique candidate eQTLs were identified from the 6 genes; of these, genotype array data was only available for 15 SNPs, none of which showed significance with PH in the larger dataset. Conclusion: This study confirms that there is a complex network of molecular pathways that modulate the intricate CpcPH phenotype involving inflammation, cell proliferation and mitochondrial dysfunction. In addition, the results suggest some overlap between CpcPH phenotype and cancer etiology, as well as between CpcPH and other PH subtypes. Future work should focus on the translational confirmation of these molecular signatures.