Kerepesi Csaba, Abushukair Hassan Mohammed, Ricciuti Biagio, Nassar Amin H, Adib Elio, Alessi Joao V, Pecci Federica, Rakaee Mehrdad, Fadlullah Muhammad Zaki Hidayatullah, Tőkés Anna-Mária, Rodig Scott J, Awad Mark M, Tan Aik Choon, Bakacs Tibor, Naqash Abdul Rafeh
Institute for Computer Science and Control (SZTAKI), Hungarian Research Network (HUN-REN), Budapest, Hungary.
Jordan University of Science and Technology, Irbid, Jordan.
JCO Precis Oncol. 2024 Feb;8:e2300439. doi: 10.1200/PO.23.00439.
Recent evidence has shown that higher tumor mutational burden strongly correlates with an increased risk of immune-related adverse events (irAEs). By using an integrated multiomics approach, we further studied the association between relevant tumor immune microenvironment (TIME) features and irAEs.
Leveraging the US Food and Drug Administration Adverse Event Reporting System, we extracted cases of suspected irAEs to calculate the reporting odds ratios (RORs) of irAEs for cancers treated with immune checkpoint inhibitors (ICIs). TIME features for 32 cancer types were calculated on the basis of the cancer genomic atlas cohorts and indirectly correlated with each cancer's ROR for irAEs. A separate ICI-treated cohort of non-small-cell lung cancer (NSCLC) was used to evaluate the correlation between tissue-based immune markers (CD8, PD-1/L1+, FOXP3+, tumor-infiltrating lymphocytes [TILs]) and irAE occurrence.
The analysis of 32 cancers and 33 TIME features demonstrated a significant association between irAE RORs and the median number of base insertions and deletions (INDEL), neoantigens (r = 0.72), single-nucleotide variant neoantigens (r = 0.67), and CD8 T-cell fraction (r = 0.51). A bivariate model using the median number of INDEL neoantigens and CD8 T-cell fraction had the highest accuracy in predicting RORs (adjusted r = 0.52, = .002). Immunoprofile assessment of 156 patients with NSCLC revealed a strong trend for higher baseline median CD8 T cells within patients' tumors who experienced any grade irAEs. Using machine learning, an expanded ICI-treated NSCLC cohort (n = 378) further showed a treatment duration-independent association of an increased proportion of high TIL (>median) in patients with irAEs (59.7% 44%, = .005). This was confirmed by using the Fine-Gray competing risk approach, demonstrating higher baseline TIL density (>median) associated with a higher cumulative incidence of irAEs ( = .028).
Our findings highlight a potential role for TIME features, specifically INDEL neoantigens and baseline-immune infiltration, in enabling optimal irAE risk stratification of patients.
最近的证据表明，较高的肿瘤突变负荷与免疫相关不良事件（irAE）风险增加密切相关。通过使用综合多组学方法，我们进一步研究了相关肿瘤免疫微环境（TIME）特征与irAE之间的关联。
利用美国食品药品监督管理局不良事件报告系统，我们提取了疑似irAE病例，以计算免疫检查点抑制剂（ICI）治疗癌症的irAE报告比值比（ROR）。基于癌症基因组图谱队列计算了32种癌症类型的TIME特征，并与每种癌症的irAE ROR间接相关。使用一个单独的接受ICI治疗的非小细胞肺癌（NSCLC）队列来评估基于组织的免疫标志物（CD8、PD-1/L1+、FOXP3+、肿瘤浸润淋巴细胞[TIL]）与irAE发生之间的相关性。
对32种癌症和33种TIME特征的分析表明，irAE ROR与碱基插入和缺失（INDEL）中位数、新抗原（r = 0.72）、单核苷酸变异新抗原（r = 0.67）以及CD8 T细胞分数（r = 0.51）之间存在显著关联。使用INDEL新抗原中位数和CD8 T细胞分数的双变量模型在预测ROR方面具有最高的准确性（调整后r = 0.52，P = 0.002）。对156例NSCLC患者的免疫特征评估显示，在发生任何级别irAE的患者肿瘤内，基线CD8 T细胞中位数较高有明显趋势。使用机器学习，一个扩大的接受ICI治疗的NSCLC队列（n = 378）进一步显示，irAE患者中高TIL（>中位数）比例增加与治疗持续时间无关（59.7%对44%，P = 0.005）。使用Fine-Gray竞争风险方法证实了这一点，表明较高的基线TIL密度（>中位数）与irAE的累积发生率较高相关（P = 0.028）。
我们的研究结果突出了TIME特征，特别是INDEL新抗原和基线免疫浸润，在实现患者最佳irAE风险分层方面的潜在作用。
