Abstract Body (Do not enter title and authors here): Heart failure is a leading cause of morbidity and mortality worldwide, arising from a range of underlying cardiomyopathies, including myocardial infarction, hypertrophic, dilated, and arrhythmogenic forms. Despite their clinical heterogeneity, the cellular and molecular mechanisms driving heart failure remain incompletely understood. To address this, we constructed a unified single-cell transcriptomic atlas by integrating human heart tissue data across these conditions, comprising 1.8 million nuclei from 195 individuals.
Our atlas reveals disease-specific cellular and transcriptional landscapes, including ischemia-associated cell states in myocardial infarction that might reflect a conserved stress response across cardiac cell types. Furthermore, we identified divergent fibroblast subpopulations across heart failures, each exhibiting distinct ligand-receptor interactions with immune cells and differential cytokine responsiveness, indicating disease-specific intercellular communication.
Leveraging the large cohort of the atlas, we stratified patients based on their transcriptional signatures, identifying six molecularly defined patient clusters. We then mapped gene expression profiles to a comprehensive compendium of 51 genome-wide association studies, including cardiomyopathy-specific and related cardiovascular traits. These transcriptionally defined patient clusters were enriched for distinct genetic traits, underscoring the potential of single-cell RNA sequencing for molecularly guided patient stratification.
Finally, gene prioritization analyses highlighted pathways essential for maintaining cardiomyocyte structural and functional integrity and lead to the identification of targeted therapeutic strategy aimed at restoring cardiomyocyte performance.