Immune-related adverse events associated with immune checkpoint inhibitors: An updated comprehensive disproportionality analysis of the FDA adverse event reporting system

Highlights

• It was the most comprehensive study to analyze irAEs based on FAERS.

• Lung toxicities and endocrine toxicities generated signals in all ICI regimens.

• Pneumonitis and interstitial lung disease were the most frequently reported irAEs.

Abstract

Backgrounds

Immune-related adverse events were reported in patients treated with immune checkpoint inhibitors (ICIs). However, with the increasing number of immune-related adverse events (irAEs), the differences of each immune checkpoint inhibitor regimen had not been fully assessed.

Methods

Disproportionality analysis was used in data mining of the suspected adverse events after ICIs administration based on the Food and Drug Administration Adverse Event Reporting System (FAERS) from January 2004 to December 2019. The onset time and fatality proportion of ICI-associated irAEs were further evaluated.

Results

A total of 32,441 reports of ICI-associated irAEs were gathered. This study showed that all ICI regimens generated lung toxicity and endocrine toxicity signals. Colitis, pneumonitis and interstitial lung disease were the most common ICI-associated irAEs. Five regimens including durvalumab monotherapy, ipilimumab monotherapy, ipilimumab plus nivolumab, ipilimumab plus pembrolizumab, durvalumab plus tremelimumab were associated with irAEs. Anti-PD-1 agents generated more signals of ocular toxicities than anti-PD-L1 agents, while anti-PD-L1 agents reported more signals of hematologic toxicities. Anti-CTLA-4 agents showed more signals of gastrointestinal toxicities compared with anti-PD-1 or anti-PD-L1 agents. The highest fatality proportion of lung toxicities with durvalumab monotherapy, hematological toxicities with avelumab monotherapy, renal and skin toxicities with cemiplimab monotherapy were found.

Conclusion

Our results demonstrated that each ICI regimen had different characteristics of irAEs. Pembrolizumab had the highest fatality proportion. Ipilimumab plus pembrolizumab had the shortest median time to onset irAEs. Further studies were expected to assess whether there were clinically relevant differences exist among ICIs.

Introduction

Immune checkpoint inhibitors (ICIs) were a novel class of cancer treatment drugs including cytotoxic T-lymphocyte antigen 4 (CTLA-4: ipilimumab, tremelimumab), programmed cell death 1 (PD-1: nivolumab, cemiplimab, pembrolizumab), and programmed cell death ligand 1 (PD-L1: atezolizumab, avelumab, durvalumab) [1]. By removing the inhibitory effect and releasing the restrained anti-tumor immune response [2], ICIs had shown significant efficacy and changed the therapeutic landscape of a variety of solid tumors including non-small cell lung cancer (NSCLC) [3], hepatocellular carcinoma [4], renal cell cancer [5], urothelial carcinoma [6], and others. Beside impressive clinical effects, a unique spectrum of side effects referred to as immune-related adverse events (irAEs) was reported [7]. It had been found that anti-PD-1/PD-L1 and anti-CTLA-4 had differences in terms of types, rates, time to onset and severity of irAEs [8]. For example, ICI-associated colitis occurred earliest in ipilimumab among all monotherapies [9]. With the rapid expansion of oncology indications of ICIs, clinicians were increasingly facing more complex and rarer irAEs. In line with the mechanism of action mediated by the immune system, irAEs resembled classic autoimmune disease and involved any organ system [10]. Frequently reported irAEs involved lung, liver, skin and gut [11]. Common irAEs could be found in pre-market clinical trials of ICIs, but it was difficult to evaluate profiles of irAEs and obtain accurate data on incidence or prevalence, due to strictly diagnosis standards, selection criteria, relatively small sample sizes and limited duration of follow-ups [12]. Hence, post-marketing monitoring was necessary to obtain uncommon AE reporting data. Real-world data played an important role in assessing irAEs associated with ICIs. Generally speaking, life-threatening irAEs were rare and may be considered other well-known conditions. Through investigating pharmacovigilance databases, especially large-samples of real-world data, it could possible to better obtain clinical characterization of irAEs, such as timing, outcomes and prognosis [13]. People could better monitor the safety profile of medicines, timely detection of rare and life-threatening irAEs.

Food and Drug Adverse Event Reporting System (FAERS) was one of the largest pharmacovigilance databases with a large number of adverse events and personal information of patients related to immune checkpoint inhibitors. Data mining techniques were increasingly used to explore and analyze massive data to identify potential associations between drugs and adverse events. This study aimed to conduct data mining to comprehensively evaluate and characterize ICI-associated irAEs by FAERS databases, serving as a reference for further prevention and management.