Lin Zhiying, Wang Rongsheng, Huang Cuilan, He Huiwei, Ouyang Chenghong, Li Hainan, Zhong Zhiru, Guo Jinghua, Chen Xiaohong, Yang Chunli, Yang Xiaogang
Jiangxi Provincial People's Hospital, The First Affiliated Hospital of Nanchang Medical College, Nanchang, China.
Front Genet. 2022 Jun 17;13:926122. doi: 10.3389/fgene.2022.926122. eCollection 2022.
Glioblastoma (GBM) is the most common and malignant type of brain tumor. A large number of studies have shown that the immunotherapy of tumors is effective, but the immunotherapy effect of GBM is not poor. Thus, further research on the immune-related hub genes of GBM is extremely important. The GBM highly correlated gene clusters were screened out by differential expression, mutation analysis, and weighted gene co-expression network analysis (WGCNA). Least absolute shrinkage and selection operator (LASSO) and proportional hazards model (COX) regressions were implemented to construct prognostic risk models. Survival, receiver operating characteristic (ROC) curve, and compound difference analyses of tumor mutation burden were used to further verify the prognostic risk model. Then, we predicted GBM patient responses to immunotherapy using the ESTIMATE algorithm, GSEA, and Tumor Immune Dysfunction and Exclusion (TIDE) algorithm. A total of 834 immune-related differentially expressed genes (DEGs) were identified. The five hub genes (STAT3, SEMA4F, GREM2, MDK, and SREBF1) were identified as the prognostic risk model (PRM) screened out by WGCNA and LASSO analysis of DEGs. In addition, the PRM has a significant positive correlation with immune cell infiltration of the tumor microenvironment (TME) and expression of critical immune checkpoints, indicating that the poor prognosis of patients is due to TIDE. We constructed the PRM composed of five hub genes, which provided a new strategy for developing tumor immunotherapy.
胶质母细胞瘤（GBM）是最常见且恶性程度最高的脑肿瘤类型。大量研究表明肿瘤免疫疗法是有效的，但GBM的免疫治疗效果不佳。因此，对GBM免疫相关枢纽基因进行进一步研究极为重要。通过差异表达、突变分析和加权基因共表达网络分析（WGCNA）筛选出GBM高度相关的基因簇。采用最小绝对收缩和选择算子（LASSO）及比例风险模型（COX）回归构建预后风险模型。利用生存分析、受试者工作特征（ROC）曲线以及肿瘤突变负荷的复合差异分析进一步验证预后风险模型。然后，我们使用ESTIMATE算法、基因集富集分析（GSEA）和肿瘤免疫功能障碍与排除（TIDE）算法预测GBM患者对免疫治疗的反应。共鉴定出834个免疫相关差异表达基因（DEG）。通过对DEG进行WGCNA和LASSO分析筛选出五个枢纽基因（信号转导和转录激活因子3（STAT3）、Ⅳ型膜联蛋白超家族成员4F（SEMA4F）、生长调节致癌基因2（GREM2）、中脑多巴胺神经营养因子（MDK）和固醇调节元件结合转录因子1（SREBF1））作为预后风险模型（PRM）。此外，PRM与肿瘤微环境（TME）的免疫细胞浸润以及关键免疫检查点的表达呈显著正相关，表明患者预后不良是由于TIDE所致。我们构建了由五个枢纽基因组成的PRM，为开发肿瘤免疫疗法提供了新策略。
