引用文本：孙学增, 苏日顺, 李文超, 等. CD8+ T细胞免疫相关基因标记有效预测胃癌预后和治疗反应性[J/CD]. 消化肿瘤杂志（电子版）, 2025, 17(1):41-55.

作者：孙学增1，苏日顺1，李文超2，王小群1，尹松成1，陈景耀1

单位：1. 中山大学附属第七医院 消化医学中心，广东 深圳 518107；2. 山东第一医科大学附属皮肤病医院 中心实验室，山东 济南 250000

Authors：Sun Xuezeng1, Su Rishun1, Li Wenchao2, Wang Xiaoqun1, Yin Songcheng1, Chen Jingyao1

Unit：1. Digestive Diseases Center, the Seventh Affiliated Hospital of Sun Yat-sen University, Shenzhen 518107, Guangdong, China；2. Central Laboratory of Hospital for Skin Diseases, Shandong First Medical University，Jinan 250000, Shandong, China

摘要：

目的 探索CD8⁺ T细胞免疫相关基因与胃癌免疫治疗及患者预后的关系。 方法 从癌症基因组图谱(the cancer genome atlas, TCGA)数据库下载胃癌相关转录组测序数据和对应的患者临床资料, 并将TCGA数据集的数据（n=407）和基因表达综合 (gene expression omnibus, GEO）数据集的数据（n=433）分别作为训练集和验证集，2个数据集合并作为总体样本（n=840）。基于单因素Cox回归、Lasoo回归分析构建CD8⁺ T细胞免疫相关基因的风险评分模型并进行各数据集的验证。根据训练集的中位风险评分，将训练集、验证集和总体样本分别划分为高风险和低风险。Kaplan-Meier生存分析和受试者操作特征曲线(receiver operating characteristic curve, ROC曲线) 被用于验证模型的预测性能。通过单因素和多因素Cox回归分析评估风险评分等临床指标的预后价值。结合风险评分、年龄和肿瘤TNM分期构建预测胃癌患者总生存率的列线图，ROC曲线与校准曲线被用于评估列线图的预测准确性。借助风险评分进行免疫细胞浸润分析、基因突变分析和免疫治疗疗效预测。使用IMvigor210数据集验证风险评分模型用于预测其他肿瘤患者对免疫治疗反应的效能。收集2023年8月至2024年4月中山大学附属第七医院收治的15例胃癌患者的胃癌组织样本和匹配的癌旁组织样本，采用荧光定量聚合酶链反应检测CD8⁺ T细胞免疫相关基因在胃癌组织中的差异表达。 结果 共鉴定出10个具有预后价值的CD8⁺ T细胞免疫相关基因。风险评分是胃癌患者预后的独立影响因素（HR=3.495，95%CI 2.072~5.894，P＜0.001）。Kaplan-Meier生存曲线显示高风险患者的总生存率低于低风险患者(P＜0.05)。构建的列线图预测胃癌患者的1年、3年和5年总生存率与实际总生存率相接近。高风险患者具有较差的预后，且伴有免疫抑制、低频基因突变和较低的免疫检查点分子表达。风险评分模型在IMvigor210数据集中具有较好的预测效能。CD8⁺ T细胞免疫相关基因在胃癌组织样本中的表达与癌旁组织样本存在差异(P＜0.05)。结论 本研究所构建的风险评分模型对于胃癌肿瘤微环境、预后、微卫星不稳定性和肿瘤突变负荷具有重要意义，为改善胃癌患者预后和治疗反应性奠定基础。

关键词：CD8⁺ T细胞； 胃癌； 免疫特征； 免疫浸润； 免疫治疗

Abstract：

Objective To explore the relationship between CD8⁺ T cell immune-related genes and immunotherapy and prognosis of gastric cancer. Method The transcriptome sequencing data of gastric cancer and the corresponding clinical data of patients were downloaded from the cancer genome atlas (TCGA) database, and the data of TCGA dataset (n=407) and gene expression omnibus (GEO) dataset (n=433) were used as the training set and validation set, respectively, and the two datasets were combined as population samples (n=840). Based on univariate Cox regression and Lasoo regression analysis, the risk scoring model of CD8⁺ T cell immune-related genes was constructed, and each dataset was validated. According to the median risk score of the training set, the training set, validation set, and population samples were classified as high-risk and low-risk, respectively. Kaplan-Meier survival analysis and receiver operating characteristic curve (ROC curve) were used to validate the predictive performance of the model. Univariate and multivariate Cox regression analyses were used to evaluate the prognostic value of clinical indicators such as risk scores. A nomogram for predicting the overall survival rate of gastric cancer patients was constructed based on risk score, age and tumor TNM stage, and the ROC curve and calibration curve were used to evaluate the prediction accuracy of the nomogram. Immune cell infiltration analysis, gene mutation analysis and immunotherapy efficacy prediction were conducted with the help of risk scores. IMvigor210 dataset was used to validate the performance of risk scoring model on prediction for the response to immunotherapy of other tumor patients. Gastric cancer tissue samples and paired adjacent tissue samples of 15 gastric cancer patients from the Seventh Affiliated Hospital of Sun Yat-sen University between August 2023 and April 2024 were collected, and the differential expression of CD8⁺ T cell immune-related genes in gastric cancer tissues were detected by real-time quantitative polymerase chain reaction. Result Totally, 10 CD8⁺ T cell immune-related genes with prognostic value were identified. Risk score was an independent prognostic factor for gastric cancer patients (HR=3.495，95%CI 2.072-5.894，P＜0.001). The Kaplan-Meier survival curve showed that the overall survival rate of high-risk patients was lower than that of low-risk patients (P＜0.05). The 1-year, 3-year, and 5-year overall survival rates of gastric cancer patients predicted by nomogram were close to the actual overall survival rates. The high-risk patients had a poor prognosis with immunosuppression, low-frequency gene mutations, and low expression of immune checkpoint molecules. The risk scoring model demonstrated good predictive performance in the IMvigor210 dataset. The expression levels of CD8⁺ T cell immune-related genes were different between gastric cancer tissue samples and adjacent tissue samples (P＜0.05). Conclusion The risk scoring model constructed in this study is of great significance for tumor microenvironment, prognosis, microsatellite instability and tumor mutational burden of gastric cancer, and lays a foundation for improving the prognosis and treatment responsiveness of gastric cancer patients.

Key words：CD8⁺ T cell; Gastric cancer; Immune signature; Immune infiltration; Immunotherapy

注：网络优先发布