基于WGCNA分析及风险评分构建头颈部鳞状细胞癌的预后模型

A prognosis model for head and neck squamous cell carcinoma based on WGCNA and risk score

目的:构建1个基因预后模型来预测头颈部鳞状细胞癌(HNSCC)患者的预后。方法:利用R软件的加权基因共表达网络分析(WGCNA)和limma包分析来自GEO及ArrayExpress的HNSCC数据集,识别差异基因及共表达网络模块基因。用TCGA数据库识别与预后相关基因,用LASSO-Cox回归分析构建预后风险模型,进行预后模型的生存分析、单因素和多因素Cox分析来验证模型的可靠性。结果:在TCGA数据库中得到11个基因与HNSCC总体生存相关,LASSO-Cox回归分析建立11基因的预后模型,低风险组的预后显著优于高风险组(P=8.002e-07),多因素Cox分析显示风险评分是独立的预后因子(HR:1.784,95%CI:1.461-2.180,P<0.001)。最后建立了可以预测HNSCC 1、3、5年生存率的列线图。结论:该研究构建并验证了一个新的HNSCC基因预后模型,可能作为HNSCC预后的生物标志物。

Objective:To identify a gene signature for predicting head and neck squamous cell carcinoma(HNSCC)prognosis.Methods:The data sets from gene expression omnibu(GEO)and ArrayExpress were analyzed by weighted gene co-expression network analysis(WGCNA)and limma,overlapping differentially expressed genes(DEGs)and co-expression module genes in HNSCC were identified.Next,hub genes related to HNSCC survival were identified based on the cancer genome atlas(TCGA)dataset.A prognostic risk model was then established using LASSO-Cox regression analysis and validated via survival analysis,univariate and multivariate Cox analysis.Results:In the TCGA dataset,11 genes were found to be significantly correlated with overall survival(OS)of HNSCC patients and were employed to develop a 11-gene prognostic signature for dividing patients into high-risk group and low-risk group.Survival was remarkably different between the 2 groups(P=8.002e-07).Multivariate Cox regression data indicated that the 11-gene risk model's risk score was independent of other clinical features(HR:1.784,95%CI:1.461-2.180,P<0.001).A nomogram was established to estimate the one-,three-and five-year HNSCC survival.Conclusion:A new 11-gene clinical signature for estimating HNSCC prognosis is developed and verified.The 11-gene model may act as an independent biomarker for HNSCC patients.