基于淋巴结转移相关基因的头颈部鳞状细胞癌预后模型构建和肿瘤免疫微环境分析

Construction of prognostic model of head and neck squamous carcinoma with lymph node metastasis-related gene andanalysis of tumor immunity microenvironment

目的基于癌症基因组图谱(TCGA)数据库筛选与头颈部鳞状细胞癌(head and neck squamous cell carcinoma,HNSCC)淋巴结转移相关的关键基因并构建预后模型。方法利用R软件筛选TCGA数据库中HNSCC数据集中肿瘤组织和癌旁组织之间的差异表达基因(differentially expressed genes,DEGs),并通过加权基因共表达网络(weighted gene co-expression network analysis,WGCNA)筛选与淋巴结转移相关的基因模块。使用单变量Cox回归和Lasso回归分析构建预后风险模型,通过生存分析和受试者工作特征(ROC)曲线验证预后模型的可靠性。最后使用CIBERSORT算法分析不同风险组的肿瘤微环境(tumor micro environment,TME)的差异。结果筛选出2565个DEGs,通过WGCNA分析得到与疾病预后和淋巴结转移高度相关的一组基因模块,相关性分析验证该基因模块中的基因表达与淋巴结转移高度相关。通过单因素Cox回归和Lasso回归筛选出了6个关键预后基因:CDKN2A、CCNE2、KNSTRN、AURKA、KPNA2和ORC1。基于这6个基因构建预后模型,生存分析显示高风险组的预后显著差于低风险组(P<0.0001)。ROC曲线表明该预后模型具有良好的预测性能。CIBERSORT分析显示不同风险组的肿瘤免疫微环境存在差异。结论筛选出的6个关键基因有助于预测HNSCC患者的预后,并与HNSCC免疫微环境密切相关,提示可能作为潜在的治疗靶点。

OBJECTIVE To identify the key genes associated with lymph node metastasis in head and neck squamous carcinoma(HNSCC)and construct a prognostic model based on The Cancer Genome Atlas(TCGA)database.METHODS Differentially expressed genes(DEGs)between tumor tissues and normal tissues in the HNSCC dataset in the TCGA database were screened by R software,and gene modules related to lymph node metastasis were screened by weighted gene co-expression network(weighted gene co-expression network analysis,WGCNA).Prognostic risk models were constructed by univariate cox regression and Lasso regression analyses.Survival analyses and ROC curves were performed to verify the Reliability of prognostic models.CIBERSORT,TIMER and ESTIMATE algorithms analysed the differences in the tumor micro environment(TME)of different risk groups.RESULTS There were 2565 DEGs screened,and a set of gene modules highly correlated with disease prognosis and lymph node metastasis were obtained by WGCNA analysis,and correlation analysis verified that the expression of genes in this gene module was highly correlated with lymph node metastasis.Univariate cox regression and Lasso regression were used to identify 6 key prognostic genes:CDKN2A,CCNE2,KNSTRN,AURKA,KPNA2,and ORC1.A prognostic model was constructed based on the 6 genes,and survival analysis showed that the prognosis of the high-risk group was significantly worse than that of the low-risk group(P<0.0001).The ROC curves demonstrated the good predictive performance of this prognostic model.CIBERSORT analyses revealed differences in the immune microenvironment of tumors in different risk groups.CONCLUSION The 6 key prognostic genes screened were helpful in predicting the prognosis of HNSCC patients and were closely associated with the immune microenvironment of HNSCC,suggesting that they may serve as potential therapeutic targets.