Table 1_Evaluating the toxicity profile of combination immune checkpoint inhibitors: a disproportionality analysis of real-world adverse events from the FDA Adverse Event Reporting System for tremelimumab, durvalumab, ipilimumab, and nivolumab.xls

BackgroundAs one of the therapeutic modalities for treating tumors, immune checkpoint inhibitors (ICIs) have gained widespread application in clinical practice, including non-small cell lung cancer, melanoma, head and neck squamous cell carcinoma, hepatocellular carcinoma, and other types of cancers. However, the safety profile of combining ICIs remains inadequately understood, which poses limitations on the clinical utilization of this novel class of medications. To investigate the toxicity spectrum associated with combination immunotherapy, we conducted an extensive data mining and analysis of the US Food and Drug Administration Adverse Event Reporting System (FAERS) database.
MethodsBy mining adverse event (AE) reports from the FAERS database covering the period from the first quarter of 2011 through the second quarter of 2024, baseline data were analyzed using Cramer’s V coefficient and p-value. Subsequently, two methods, the reporting odds ratio (ROR) and the Bayesian confidence propagation neural network, were employed to detect AE signals for single immune checkpoint inhibitors (sICIs) and dual immunotherapy group (tremelimumab plus durvalumab and ipilimumab plus nivolumab, DIG).
ResultsA total of 55,052 patients and 118,001 AEs were selected. The DIG exhibited a higher incidence of AE signals across 14 distinct system organ class level. Moreover, DIG exhibited higher positive signal intensity compared to sICIs in the following preferred terms: myocarditis [ROR 2.221, 95% confidence interval lower limit of information component (IC025) 0.486], immune-mediated myocarditis (ROR 2.922, IC025 0.610), adrenal insufficiency (ROR 2.503, IC025 0.602), hyperthyroidism (ROR 1.872, IC025 0.305), thyroiditis (ROR 2.669, IC025 0.546), immune-mediated enterocolitis (ROR 3.948, IC025 0.937), pyrexia (ROR 1.570, IC025 0.290), hepatic function abnormality (ROR 2.582, IC025 0.591), hepatitis (ROR 2.705, IC025 0.637), liver disorder (ROR 2.718, IC025 0.646), immune-mediated hepatitis (ROR 5.504, IC025 0.994), immune-mediated liver disorder (ROR 5.322, IC025 0.966), cytokine release syndrome (ROR 7.650, IC025 1.103), autoimmune diseases (ROR 1.754, IC025 0.275), sepsis (ROR 1.414, IC025 0.062), diabetic ketoacidosis (ROR 2.294, IC025 0.472), type 1 diabetes mellitus (ROR 2.421, IC025 0.508), arthritis (ROR 1.562, IC025 0.113), myositis (ROR 2.204, IC025 0.412), and acute kidney injury (ROR 1.708, IC025 0.264).
ConclusionsOur findings indicate that the AEs associated with dual ICI predominantly originate from immune-related AEs, including myotoxicity, endocrine toxicity, and hepatotoxicity. Notably, cytokine release syndrome, a rarely reported AE with a strongly positive signal, warrants particular attention in clinical decision-making.

背景
作为肿瘤治疗手段之一，免疫检查点抑制剂（immune checkpoint inhibitors, ICIs）已在非小细胞肺癌、黑色素瘤、头颈部鳞状细胞癌、肝细胞癌等多种癌症的临床实践中得到广泛应用。然而，联合使用ICIs的安全性特征仍未得到充分阐明，这限制了这类新型药物的临床应用。为探究联合免疫治疗的毒性谱，本研究对美国食品药品监督管理局不良事件报告系统（US Food and Drug Administration Adverse Event Reporting System, FAERS）数据库开展了大规模数据挖掘与分析。
方法
本研究挖掘了2011年第一季度至2024年第二季度FAERS数据库中的不良事件（adverse event, AE）报告，采用克莱姆V系数与p值对基线数据进行分析。随后采用报告比值比（reporting odds ratio, ROR）与贝叶斯置信传播神经网络两种方法，分别检测单一免疫检查点抑制剂（single immune checkpoint inhibitors, sICIs）与双免疫治疗组（曲美木单抗联合度伐利尤单抗、伊匹木单抗联合纳武利尤单抗，即DIG）的不良事件信号。
结果
本研究共纳入55052例患者与118001例不良事件。双免疫治疗组在14个不同的系统器官分类层级的不良事件信号发生率更高。此外，在以下不良事件首选术语中，双免疫治疗组的阳性信号强度均高于单一免疫检查点抑制剂组：心肌炎[ROR=2.221，信息组分95%置信区间下限（IC025）=0.486]、免疫介导性心肌炎（ROR=2.922，IC025=0.610）、肾上腺功能不全（ROR=2.503，IC025=0.602）、甲状腺功能亢进症（ROR=1.872，IC025=0.305）、甲状腺炎（ROR=2.669，IC025=0.546）、免疫介导性小肠结肠炎（ROR=3.948，IC025=0.937）、发热（ROR=1.570，IC025=0.290）、肝功能异常（ROR=2.582，IC025=0.591）、肝炎（ROR=2.705，IC025=0.637）、肝脏疾病（ROR=2.718，IC025=0.646）、免疫介导性肝炎（ROR=5.504，IC025=0.994）、免疫介导性肝脏疾病（ROR=5.322，IC025=0.966）、细胞因子释放综合征（ROR=7.650，IC025=1.103）、自身免疫性疾病（ROR=1.754，IC025=0.275）、败血症（ROR=1.414，IC025=0.062）、糖尿病酮症酸中毒（ROR=2.294，IC025=0.472）、1型糖尿病（ROR=2.421，IC025=0.508）、关节炎（ROR=1.562，IC025=0.113）、肌炎（ROR=2.204，IC025=0.412）与急性肾损伤（ROR=1.708，IC025=0.264）。
结论
本研究结果显示，双免疫检查点抑制剂相关不良事件主要源于免疫相关不良事件，包括肌毒性、内分泌毒性与肝毒性。值得注意的是，细胞因子释放综合征作为一种报告率较低但阳性信号极强的不良事件，在临床决策中需予以特别关注。