Cardiotoxicity Induced by Immune Checkpoint Inhibitors: A Pharmacovigilance Study From 2014 to 2019 Based on FAERS

Abstract

This study was to scientifically and systematically explore the association between cardiotoxicity and immune checkpoint inhibitors (ICIs) and also to characterize the spectrum of ICI-related cardiac complications. From the first quarter of 2014 to the fourth quarter of 2019, data from the FDA Adverse Event Reporting System database were selected to conduct the disproportionality analysis. Reporting odds ratios and information components were used to evaluate the signal after statistical shrinkage transformation. In total, 7,443,137 cases and 36,326,611 drug-adverse event pairs were collected, among which 9,271 cases were identified to be related to ICI-induced cardiotoxicities. The number of male patients was much higher than that of females (5,579 vs. 3,031) and males presented a slightly higher reporting frequency than females in general, which was statistically significant (ROR = 1.04, 95%CI: 0.99-1.09, p < 0.001). Simultaneously, the proportion of serious or life-threatening outcomes in males was significantly higher than in females (ROR = 1.05, 95%CI: 0.96-1.15, p < 0.001). Importantly, ICIs were associated with over-reporting frequencies of cardiotoxicities in general (ROR025 = 1.06, IC025 = 0.08). PD-1 and PD-L1 were found to be related to cardiac adverse events, corresponding to ROR025 = 1.06, IC025 = 0.08, and ROR025 = 1.06, IC025 = 0.08, respectively, while anti-CTLA-4 (cytotoxic T-lymphocyte-associated protein 4) was significantly associated with some specific adverse events rather than common adverse events. The spectrum of cardiotoxicities induced by ICIs mostly differed among individual agents, but also demonstrated some common features. Dyspnea (N = 2,527, 21.25%), myocarditis (N = 614, 5.16%), atrial fibrillation (N = 576, 4.84%), cardiac failure (N = 476, 4.00%), and pericardial effusion (N = 423, 3.56%) were the top five cardiac adverse events reported in the database. Among them, myocarditis was the only one caused by all ICIs with strong signal value and high risk, warranting further attention. Overall, this investigation mainly showed the profile of cardiotoxicities caused by ICIs, which varied between different ICI therapies, but also shared some similarities in specific symptoms such as myocarditis. Therefore, it is vital and urgent to recognize and manage ICI-related cardiotoxicities, known to frequently occur in clinical practice, at the earliest point.

Keywords: FAERS; adverse drug reaction; cardiotoxicity; disproportionality analysis; immune checkpoint inhibitors; information component; pharmacovigilance; reporting odds ratio.

FIGURE 1

Cardiotoxicity Signal Profiles of Different ICI Strategies.