Exploring the cardiotoxic effects of AZD5991 in human living myocardial slices

Cesare will use his award to generate pilot data using an ex vivo human myocardial slice model, previously developed with NC3Rs funding, as a tool to detect early indicators of cardiotoxicity. Working in partnership with AstraZeneca, the team will assess how the model responds to AZD5991, a molecule that was halted in phase 1 clinical trials due to cardiotoxicity of uncertain clinical significance, helping to support the use of the human slice model for mechanistic toxicity purposes.

Preclinical cardiovascular safety assessment in drug development relies on complementary in vivo and in vitro approaches. Whilst animal models offer a holistic view of cardiovascular perturbations, species differences in physiology limit quantitative translational understanding to humans. Despite advances in 3D models incorporating induced pluripotent stem cell-derived cardiomyocytes, cardiac fibroblasts and endothelial cells, these systems do not fully reproduce complex multicellular tissue architecture, native extracellular matrix and cellular functional and structural maturity of the adult human heart. To address these limitations, we developed a living human myocardial slice platform at Imperial College London that preserves native tissue architecture, cellular composition, electrophysiology, and mechanical properties, enabling measurement of contractile force, viability, calcium handling, and injury biomarkers under controlled conditions. We propose a pilot study to evaluate AZD5991, a potent and selective MCL‑1 inhibitor, that advanced to a phase 1 clinical trial in patients with hematologic malignancies, but showed a high incidence of troponin elevation across dose levels, attributed to cardiotoxicity of uncertain clinical significance (PMID-39167622), leading to discontinuation. This history makes AZD5991 and its inactive enantiomer, an ideal test case to determine whether human myocardial slices can reveal preclinical cardiotoxic liability and provide mechanistic insight linking clinical troponin signals to measurable effects on contractility, calcium dynamics, viability, and troponin T release. Successful detection and mechanistic characterization of AZD5991‑associated signals would validate the translational relevance of the myocardial slice platform, strengthen decision‑making for cardiovascular safety, inform safer first-inhuman (FiH) dosing, and enable broader academic–industry adoption, including collaboration and outsourcing to support quantitative therapeutic predictions while reducing reliance on animal models. Ultimately, this project aims to enhance the predictive power and clinical translatability of preclinical cardiotoxicity assessment, accelerate safety evaluations, and support the development of safer therapies for patients with urgent medical needs.