On transcriptomic aging heterogeneity in human heart disease

Abstract

Cardiovascular disease may arise when specific cardiac cell types undergo accelerated biological aging, yet a cell type–resolved map of aging heterogeneity in the human heart is lacking. Here, we quantify cell type–specific biological age using single-nucleus transcriptomic clocks trained on healthy human hearts and applied to diverse disease states. Across major conditions, we observe pronounced, diagnosis- and lineage-specific discrepancies between chronological and predicted age: the strongest acceleration localizes to endocardial endothelium in dilated cardiomyopathy, whereas myocardial infarction shows negative Δ-age across multiple lineages consistent with acute dedifferentiated/reparative programs rather than aging. These findings nominate endothelial/vascular compartments as potential drivers of risk and argue that resolving biological age at cell-type resolution can prioritize mechanisms for follow-up and guide therapeutic target selection.