Immune checkpoint inhibitors induce immune-mediated myocarditis and inflammasome activation in a tumor-bearing mouse model

Introduction Immune checkpoint inhibitor–associated myocarditis (ICI-M) is the most severe complication produced by ICIs. Although rare (1–2% of patients), ICI-M carries a high mortality rate (25–40%) and frequently presents with life-threatening arrhythmias and conduction disorders. ICI-M is characterized by myocardial infiltration of CD4⁺/CD8⁺ T lymphocytes and macrophages, leading to myocardial necrosis, with strong activation of interferon-γ signaling and the NLRP3 inflammasome. Mechanisms driving immune cell recruitment to the myocardium remain poorly understood, and current in vivo models only partially address disease complexity, particularly the contribution of myocardial endothelial dysfunction. Objective This study aims to investigate immune checkpoint inhibitor–associated myocarditis in vivo and to dissect the cellular and molecular mechanisms driving immune-mediated myocardial injury using a murine model involving tumor burden and ICI treatment. Method The ICI-M model was generated by subcutaneous injection of murine melanocytic cells into BALB/c mice and subsequent combined anti–PD-1/anti–CTLA-4 therapy. Four groups were formed: saline-treated controls (PHY), tumor-bearing mice treated with saline (K+PHY), tumor-bearing mice treated with IgG (K+IgG), and tumor-bearing mice treated with ICIs (K+ICI). Electrocardiograms were recorded and cardiac tissues collected for immunohistological and molecular analyses. Results Compared with IgG controls, ICI treatment reduced tumor volume by 85% (p<0.01) and induced marked myocardial immune infiltration, with increased CD3⁺ T cells (3.1% vs 0.4%, p<0.001) and CD68⁺ macrophages (2.9% vs 2.1%, p<0.05), together with PD-L1 overexpression (2.5% vs 1.5%, p<0.05). PD-L1 expression was also increased in IgG-treated animals but immune infiltration was absent, suggesting non-specific immune activation. ICI treatment induced pronounced electrophysiological alterations with reduced QRS and T wave amplitudes, QT interval shortening, cardiac axis deviation (Lewis index), and QRS prolongation. Molecular analyses revealed activation of the NLRP3 inflammasome pathway in tumor-bearing ICI-treated mice vs control mice, with increased GBP5 (3.0% vs 1.4%, p<0.001) and CASPASE-1 (3,1% vs 1.2%, p<0.001) expression. Aim2 expression was increased by 9.1% in tumor-bearing mice (p<0.001) while endothelial activation, monitored using CD31 overexpression (11.4%, p<0.01), was solely observed in ICI-treated mice. Conclusion Immune checkpoint inhibitors induce myocardial immune infiltration associated with specific electrophysiological abnormalities of ICI-associated myocarditis. In addition to confirming NLRP3 inflammasome involvement, these findings identify myocardial endothelial injury, likely driven by AIM2 inflammasome activation as a key mechanism of immune-mediated myocarditis. These results indicate that our tumor-bearing model may help to address the pathophysiology of ICI-related cardiotoxicity.