The safety of standardized management of suspected cardiovascular immune-related events in patients treated by immunecheckpoints inhibitors

Abstract Background Immune checkpoint inhibitors (ICIs) treatment carries a risk of immune-related adverse events (irAEs), including cardiovascular (CV) events. Current guidelines recommend troponin testing to detect early CV irAEs, especially myocarditis, prompting hospital admission. However, high volumes might hinder hospital admission for all suspected CV irAEs patients. Purpose To assess the safety of standardized management in outpatients with suspected irAEs. Methods A prospective single-center cohort study was conducted, including all adults referred to our unit for suspected CV irAEs. Patients exhibiting CV symptoms, receiving double ICIs, or showing ECG changes were referred to cardiology or intensive care based on severity. Patients with no CV symptoms or ECG changes, on single ICI therapy, and no other immune disorder were assessed in a day stay within 8 working days. Patients underwent serial serum testing, ECG, transthoracic echocardiogram, and cardiac magnetic resonance imaging (cMRI) and were assessed by a multidisciplinary team (MDT) including cardiology and internal medicine specialists. Coronary artery imaging and endomyocardial biopsy were performed when deemed necessary (Figure 1). The primary endpoint was CV mortality at 30 days after referral. Secondary endpoints included identifying factors associated with the diagnosis of CV irAEs at referral. Events were adjudicated by a panel of experts (FC, MM, SH) who reviewed the MDT's weekly meeting conclusions. Results 175 patients were enrolled from March 2022 to October 2023; 40 (22.89%) were male, median age 61 (IQR 48.0-72.0) years. 146 (83.4%) were seen in the outpatient clinic. No CV deaths occurred within the initial 30 days. 95 patients (54.1%) had CV irAEs at referral, including 72 (41.1%) with myocarditis; 10 (5.7%) with acute coronary syndromes; 9 (5.1%) with pericarditis; 7 (4.0%) with heart failure; 4 (2.3%) with high-degree conduction or ventricular arrhythmias; and 1 (0.6%) with Tako-Tsubo. 8 (4.6%) had a combination of CV irAEs. The majority of patients admitted to the hospital had CV irAEs (24, 82.7%). On multivariable regression analysis(Figure 2), CV or muscle symptoms(OR 2.677 [CI 1.207-5.937], p=0.015), prior CV disease history(OR 3.668 [CI 1.486-9.057], p=0.0048), and abnormal ECG (OR 2.855 [CI 1.074-7.589], p=0.035) remained associated with CV irAEs at referral. Including global longitudinal strain (GLS) in the model, we identified prior CV disease history. (OR 7.011 [CI 2.217-22.167], p=0.0009) and GLS <16% (OR 2.774 [CI 1.07-7.17], p=0.035) as significant. Conclusions We demonstrate the safety of a standardized outpatient management approach for most patients referred for suspected CV irAEs, mainly triggered by asymptomatic elevation of troponin. No patient experienced CV death at 30 days. When suspecting a CV irAEs, CV or muscle symptoms, previous history of CV disease, and abnormal ECG should prompt speedy assessment given the higher likelihood of CV irAEsFigure 1