Routine application of cardiac magnetic resonance imaging in patients with suspected myocarditis from immune checkpoint inhibitor therapy

Abstract Background Cardiovascular adverse events including myocarditis associated with immune checkpoint inhibitor (ICI) therapy remain a challenge in clinical practice. Diagnostic assessment of ICI-related myocarditis (ICI-M) is challenging due to a variable phenotype, confounding effects and limited sensitivity of diagnostic tools. Cardiac magnetic resonance imaging (CMR) is used to diagnose non-ICI myocarditis, but evidence on diagnostic sensitivity is low, particularly in patients with a discrete phenotype. Here, we aim to assess the impact of CMR in patients with suspected ICI-related myocarditis and relate to final diagnosis including endomyocardial biopsy findings. Methods All consecutive patients receiving CMR for suspected ICI-M between September 2019 and January 2024 were included and retrospectively analysed. CMR parameters were correlated with clinical, laboratory and echocardiographic parameters and stratified for presence or absence of myocarditis as per final diagnosis. Results A total of 55 patients who received CMR for suspected ICI-related myocarditis were analysed, including 25 patients with confirmed ICI-M as per final diagnosis and 30 patients with cardiotoxicity other than myocarditis or no cardiotoxicity (ICI-O). The mean age (ICI-M versus ICI-O) was 65.7±13.6 versus (vs.) 67.3±9.9 (p=0.61) years, 32.0% vs. 26.7% (p=0.67) were female, and 40.0% vs. 26.7% (p=0.29) had pre-existing coronary heart disease. Cardiac biomarkers and echocardiographic data did not differ between the groups. In CMR analysis, presence of late gadolinium enhancement (LGE) was associated with ICI-M (56.0% vs. 26.7%, p=0.027). Myocardial oedema was generally rare and not associated with ICI-M. Endomyocardial biopsy (EMB) was used in 6 out of 55 patients (10.9%) with suspected ICI-M. In 4 of these 6 patients (66.7%) EMB confirmed ICI-M diagnosis. Of note, only 2/4 patients with EMB-confirmed myocarditis showed abnormal CMR findings. Conclusion The diagnostic assessment of ICI-M is challenged by low sensitivity of common diagnostic measures, often requiring a multimodal approach. Presence of LGE in CMR was associated with ICI-M, but sensitivity and specificity are low. Our data emphasize the high value of EMB in patients with clinically suspected ICI-M despite inconspicuous CMR. Prospective data to improve diagnostic criteria are needed.