Gao Hong, Liu Yanhong, Hu Yue, Ge Meiling, Ding Jie, Ye Qing
Biobank of Nanjing Drum Tower Hospital, The Affiliated Hospital of Nanjing University Medical School, Nanjing, China.
Department of Pathology, The First Affiliated Hospital of USTC, Division of Life Sciences and Medicine, University of Science and Technology of China, Hefei, China.
Front Genet. 2022 Apr 25;13:850101. doi: 10.3389/fgene.2022.850101. eCollection 2022.
Lung adenocarcinoma (LUAD) is a highly heterogeneous tumor. Tumor mutations and the immune microenvironment play important roles in LUAD development and progression. This study was aimed at elucidating the characteristics of patients with different tumor immune microenvironment and establishing a prediction model of prognoses and immunotherapy benefits for patients with LUAD. We conducted a bioinformatics analysis on data from The Cancer Genome Atlas and Gene Expression Omnibus (training and test sets, respectively). Patients in the training set were clustered into different immunophenotypes based on tumor-infiltrating immune cells (TIICs). The immunophenotypic differentially expressed genes (IDEGs) were used to develop a prognostic risk score (PRS) model. Then, the model was validated in the test set and applied to evaluate 42 surgery patients with early LUAD. Patients in the training set were clustered into high (Immunity_H), medium (Immunity_M), and low (Immunity_L) immunophenotype groups. Immunity_H patients had the best survival and more TIICs than Immunity_L patients. Immunity_M patients had the worst survival, characterized by most CD8 T and Treg cells and highest expression of PD-1 and PD-L1. The PRS model, which consisted of 14 IDEGs, showed good potential for predicting the prognoses of patients in both training and test sets. In the training set, the low-risk patients had more TIICs, higher immunophenoscores (IPSs) and lower mutation rates of driver genes. The high-risk patients had more mutations of DNA mismatch repair deficiency and APOBEC (apolipoprotein B mRNA editing enzyme catalytic polypeptide-like). The model was also a good indicator of the curative effect for immunotherapy-treated patients. Furthermore, the low-risk group out of 42 patients, which was evaluated by the PRS model, had more TIICs, higher IPSs and better progression-free survival. Additionally, IPSs and PRSs of these patients were correlated with EGFR mutations. The PRS model has good potential for predicting the prognoses and immunotherapy benefits of LUAD patients. It may facilitate the diagnosis, risk stratification, and treatment decision-making for LUAD patients.
肺腺癌（LUAD）是一种高度异质性肿瘤。肿瘤突变和免疫微环境在LUAD的发生发展中起着重要作用。本研究旨在阐明不同肿瘤免疫微环境患者的特征，并建立LUAD患者预后和免疫治疗获益的预测模型。我们对来自癌症基因组图谱（The Cancer Genome Atlas）和基因表达综合数据库（Gene Expression Omnibus，分别为训练集和测试集）的数据进行了生物信息学分析。基于肿瘤浸润免疫细胞（TIICs），将训练集中的患者聚类为不同的免疫表型。利用免疫表型差异表达基因（IDEGs）建立预后风险评分（PRS）模型。然后，在测试集中对该模型进行验证，并应用于评估42例早期LUAD手术患者。训练集中的患者被聚类为高（免疫_H）、中（免疫_M）和低（免疫_L）免疫表型组。免疫_H组患者的生存期最佳，且TIICs比免疫_L组患者更多。免疫_M组患者的生存期最差，其特征是CD8 T细胞和调节性T细胞最多，PD-1和PD-L1表达最高。由14个IDEGs组成的PRS模型在预测训练集和测试集患者的预后方面显示出良好的潜力。在训练集中，低风险患者的TIICs更多、免疫表型评分（IPSs）更高且驱动基因的突变率更低。高风险患者的DNA错配修复缺陷和载脂蛋白B mRNA编辑酶催化多肽样蛋白（APOBEC）的突变更多。该模型也是免疫治疗患者疗效的良好指标。此外，通过PRS模型评估的42例患者中的低风险组具有更多的TIICs、更高的IPSs和更好的无进展生存期。此外，这些患者的IPSs和PRSs与表皮生长因子受体（EGFR）突变相关。PRS模型在预测LUAD患者的预后和免疫治疗获益方面具有良好的潜力。它可能有助于LUAD患者的诊断、风险分层和治疗决策。
