Schizophrenia Linked to a Fragile Cardiac Regulatory State in Donor Hearts

People with schizophrenia die of cardiovascular disease at rates far exceeding the general population, yet the heart itself has remained an understudied piece of that puzzle. A new study published in the Journal of Translational Medicine argues that part of the answer may lie not in how heart cells look or what they produce, but in how their gene-regulatory networks are organised. Using donated human heart tissue, researchers describe a recurring cardiomyocyte state in schizophrenia defined by preserved molecular output but broadly weakened regulatory coherence.

The team, led by Bing Xia, Tianqi Guo, Yiming Bo and colleagues with senior author Liliang Li, applied a donor-resolved, multiscale framework to 190,590 left-ventricular nuclei from 25 donors. Because the donors were death-matched, the analysis could adjust for the myocardial background shared across sudden-death cases and isolate what distinguished schizophrenia samples from the rest.

Cell-state reconstruction, local differential-abundance analysis and covariate-aware latent-factor modelling converged on the same observation: a recurrent cardiomyocyte state in which the cells’ molecular output was intact while the coordination among regulatory programmes was diminished. A stable 71-gene programme spanning contraction, energetics and calcium handling captured this state, and the output–coherence decoupling held up under exact donor-level inference, with a P value of 0.0156, and leave-one-donor-out stability analysis.

The regulatory programme did not exist in isolation. It intersected a schizophrenia-associated network previously mapped in the anterior cingulate cortex, showed its strongest cardiovascular genetic enrichment for ventricular repolarisation, and occupied disease networks for arrhythmia and cardiomyopathy. In other words, the same cardiac network space that carries schizophrenia biology also carries known markers of electrical instability in the heart.

Antipsychotic drug targets mapped onto that same network space, raising the possibility that the myocardium in schizophrenia is not simply exposed to more cardiac stress but is differently prepared to absorb it. Supporting this concern, excess arrhythmia reporting replicated across non-overlapping pharmacovigilance periods, with an overall reporting odds ratio of 4.24.

To test what weakened coordination might actually do, the researchers ran in silico perturbations of cardiomyocytes, progressively disrupting physiological coordination among selected calcium-control axes while preserving marginal effects. Under drug stress, these disruptions broadened calcium-response distributions and produced larger abnormal electrophysiological tails, with stronger effects within the calcium-control axis family. The modelling suggests that robustness under pharmacological challenge may emerge from coordinated system organisation rather than from any single component.

A striking methodological finding emerged from the perturbational analysis itself. Across 13,205 perturbagens, expression response was essentially unrelated to regulatory-topology response, with a Spearman correlation of 0.007 and an AUROC of 0.495. If that result generalises, it implies that drugs producing similar transcriptional signatures could nonetheless reshape regulatory networks in very different ways, a caution for approaches that rely on expression matching alone to predict cardiac side effects.

The authors are careful about scope. This is a study of donated tissue from a limited number of donors, and the perturbation experiments are computational rather than clinical. The work identifies an association between schizophrenia and a myocardial regulatory state; it does not demonstrate that the state causes arrhythmias or sudden death in patients. Still, the framework connects schizophrenia biology, ventricular repolarisation, cardiovascular disease networks and antipsychotic drug-target space within a single analytical structure.

The conceptual takeaway is the paper’s most provocative claim: cardiac drug safety may depend not only on the perturbation itself but on the regulatory state of the myocardium that receives it. A heart whose regulatory networks are loosely coordinated, the study suggests, may respond to the same antipsychotic exposure differently from one whose networks are tightly organised, even when the individual genes and proteins involved are unchanged.

The article was published on 27 October 2026 as an open-access research paper, and the authors note that the shared early version is subject to further edits before the final version of record.

Subject of Research: Myocardial gene-regulatory network organisation in schizophrenia and its relation to cardiac vulnerability under antipsychotic drug stress

Article Title: Network medicine reveals myocardial regulatory vulnerability in schizophrenia

Article References: Xia, B., Guo, T., Bo, Y., Min, X., Cui, Q., Ma, L., Liu, W., Zhu, C., Qin, H., & Li, L. (2026). Network medicine reveals myocardial regulatory vulnerability in schizophrenia. Journal of Translational Medicine. https://doi.org/10.1186/s12967-026-09154-2

Image Credits: AI Generated

DOI: 10.1186/s12967-026-09154-2

Keywords: schizophrenia, cardiomyocytes, network medicine, gene regulation, arrhythmia, antipsychotics, ventricular repolarisation, calcium handling, single-nucleus transcriptomics, cardiovascular mortality, pharmacovigilance, drug safety