免疫相关基因预后模型的构建对喉癌患者预后的影响

The effect of constructing a prognostic model of immune-related genes on the prognosis of patients with laryngeal cancer

目的探讨基于生物信息学方法构建免疫相关基因(immune-related genes,IRG)预后模型以准确预测喉癌患者的预后.方法从癌症基因组图谱(the Cancer Genome Atlas,TCGA)数据库获得111个喉癌组织和12个正常相邻组织之间的差异表达基因(differentially expressed genes,DEG).利用ImmPort数据库识别出差异表达的IRG.Cox单变量生存分析用于筛选与生存相关的IRG.差异表达的与生存相关的IRG被认为是预后相关的免疫基因.然后构建免疫基因预后模型计算患者风险值,受试者ROC曲线分析验证模型准确性.通过该模型行单因素、多因素独立预后分析证明其独立预测能力.最后分析关键免疫基因与临床病理参数的关联.结果鉴定喉癌的DEG并筛选出IRG.接着与预后生存时间结合,鉴定8个关键免疫基因(CXCL11、RBP1、AQP9、CYSLTR2、BTC、STC2、UCN和FCGR3B)作为免疫基因预后模型,这种预后模型可以准确的将患者分为高危和低危人群.总体生存分析表明,高危患者的生存时间比低危患者要短(P<0.0001).模型的ROC曲线下面积为0.810,提示预后模型具有较高的敏感性和准确性.单因素和多因素Cox回归表明其为喉癌患者预后的独立预测因素.此外,我们发现模型中的5个关键基因与临床病理特征显著相关.结论基于生物信息学方法构建喉癌的免疫相关基因预后模型,发现8个基因有助于预测喉癌患者的预后,其中5个与临床病理特征显著相关.

Objective To explore the construction of an immune-related gene(IRG)prognosis model based on bioinformatics methods to accurately predict the prognosis of patients with laryngeal cancer.Methods From the Cancer Genome Atlas(TCGA)database,1I1 differentially expressed genes(DEG)between laryngeal cancer tissues and 12 normal adjacent tissues were obtained.Use ImmPort database to identify differentially expressed IRG.Cox univariate survival analysis is used to screen IRGs related to survival.Differentially expressed survival-related IRGs are considered to be prognostic-related immune genes.Then construct an immune gene prognosis model to calculate the patient's risk value,and receiver operating characteristic(ROC)curve analysis to verify the accuracy of the model.The independent prognostic analysis of single factor and multiple factors of this model proves its independent predictive ability.Finally,the relationship between key immune genes and clinicopathological parameters is analyzed.Results The DEG of laryngeal carcinoma was identified and the IRG was screened out.Then combined with prognostic survival time,eight key immune genes CXCL11,RBP1,AQP9,CYSLTR2,BTC,STC2,UCN and FCGR3B were identified as immune gene prognosis models.This prognostic model can accurately divide patients into high-risk and low-risk groups.The overall survival analysis showed that the survival time of high-risk patients was shorter than that of low-risk patients(P<0.0001).The area under the ROC curve of the model is 0.810,indicating that the prognostic model has high sensitivity and accuracy.Univariate and multivariate Cox regression indicate that it is an independent predictor of the prognosis of patients with laryngeal cancer.In addition,we found that five key genes in the model are significantly related to clinicopathological characteristics.Conclusions Based on bioinformatics methods,a prognostic model of immune-related genes for laryngeal cancer was constructed.Eight genes were found to help predict the prognosis of patients with laryngeal cancer,and five of them were significantly related to clinicopathological characteristics.