Abstract
Background
Immune checkpoint inhibitor therapy has significantly improved treatment of advanced malignant diseases. However, patients receiving immune checkpoint inhibitor therapy with programmed death 1 (PD1) blocking agents are at risk for cardiotoxicity with high mortality. The underlying pathomechanisms have not yet been elucidated.
Purpose
This study aims to evaluate the cardiotoxic effect of PD1-blocking agents and its underlying mechanism with focus on myocardial inflammation and metabolism.
Methods
A transplantable melanoma mouse model was used to study PD1 blocking therapy in a preclinical setting. In brief, mice were subcutaneously transplanted with a melanoma cell line and treated with anti-PD1 antibodies or non-specific immunoglobulin control for 14 days. Murine transthoracic echocardiography including strain analysis was conducted to assess left ventricular (LV) function. Pressure/volume analysis was performed using a micro-tip catheter introduced into the LV via the right commune carotid artery. Inotropic stress was induced by dobutamine. Myocardial immune cell infiltration and expression of PD1/PD-L1 was assessed using flow cytometry. A combined approach for mass spectrometry-guided profiling of proteome, lipids and metabolites was applied to evaluate changes in cardiomyocyte function and metabolism.
Results
Reduced tumor size in anti-PD1-treated animals confirmed response to treatment (n=7; p=0.018). Echocardiographic examination revealed reduced LV ejection fraction (EF) (n=7–8; p=0.001) and reduced global radial strain in anti-PD1-treated mice compared to control littermates (n=3–4; p=0.004). Remarkably, pressure/volume catheterization indicated reduced EF, stroke volume and stroke work under dobutamine stress in anti-PD1-treated mice (p=0.013; n=3–4). Anti-PD1 treatment was associated with a 2-fold elevated level of CD4+ and CD8+ T-cells in murine hearts (n=8; p=0.009 and p=0.049). CD44 expression was upregulated in CD8+ T-cells of anti-PD1-treated animals (n=8; p=0.024). Proteomics revealed downregulation of proteins critical for cardiomyocyte contraction, e.g. ryanodine receptor 2 and L-type calcium channel beta 2 (n=4; p<0.05). Analysis of metabolites and lipids indicated dysfunctional energy metabolism. To identify a potential underlying mechanism, expression of PD1 and its ligand PD-L1 on cardiac cell populations was examined. PD-L1 was mainly expressed on cardiac endothelial cells while PD1 was expressed on 10–20% of murine cardiomyocytes (n=12; p<0.001 and p=0.004).
Conclusion
The obtained results point towards a cardiotoxic effect of PD1 blocking therapy with severely disturbed cardiac function and disrupted cardiomyocyte functional integrity. Myocardial expression of the PD1 receptor could mediate the observed effect. This could potentially promote the development of PD1 immune checkpoint inhibitor-associated myocarditis in patients.
Acknowledgement/Funding
IFORES research grant of the Medical Faculty, University Duisburg-Essen, Essen, Germany
A Lack of Biomarkers for Cardiac Complications of Immune Checkpoint Inhibitor Therapy
