基于生物信息学构建头颈部鳞癌微小RNAs预后预测模型

Construction of a micro RNAs prognostic model for head and neck squamous cell carcinoma based on bioinformatics

目的探究头颈部鳞癌(HNSCC)中微小RNAs(miRNAs)表达情况与无复发生存期之间的相关性,并构建预后预测模型。方法利用TCGA数据库获取头颈部鳞癌的miRNAs表达谱与临床信息,通过单因素和多因素COX分析筛选出与无复发生存期相关的miRNAs,构建预后预测模型,并用受试者工作特征(ROC)曲线验证模型的准确度。通过京都基因与基因组百科全书(KEGG)和基因本体(GO)富集分析进一步探究miRNAs在细胞通路和分子功能方面的生物学意义。结果筛选出7个与预后相关的miRNAs(miR-502-5p、miR-499-5p、miR-216a-5p、miR-548e-3p、miR-203-3p、miR-6837-5p和miR-320b)用以构建模型,Kaplan-Meier结果显示高风险组患者的无复发生存期低于低风险组患者(P<0.001)。ROC曲线显示,训练集、测试集和整集的5 a生存期曲线下面积分别为0.888、0.599、0.701。基因富集分析结果提示,miRNAs与局部黏附通路、调节干细胞多能性信号通路、转化生长因子-β信号通路、Wnt信号通路、FoxO信号通路等密切相关。结论本研究构建的HNSCC miRNAs预后预测模型可有效预测患者的无复发生存期,并作为判断预后的重要指标。

Objective To explore the correlation between micro RNAs(miRNAs)expression profiles in head and neck squamous cell carcinoma(HNSCC)and the recurrence-free survival period,and to build a prognostic prediction model.Methods miRNAs expression profiles and clinical information of HNSCC were obtained from the The Cancer Genome Atlas(TCGA)database.Univariate and multivariate COX regression analysis were performed to screen for miRNAs related to disease-free survival,and a prognostic model was constructed.The accuracy of the model was validated using receiver operating characteristic(ROC)curve analysis.Kyoto encyclopedia of genes and genomes(KEGG)and Gene Ontology(GO)enrichment analysis were used to further investigate the biological significance of miRNAs in cellular pathways and molecular functions.Results Seven miRNAs(miR-502-5p,miR-499-5p,miR-216a-5p,miR-548e-3p,miR-203-3p,miR-6837-5p and miR-320b)related to prognosis were screened and used to construct the model.Kaplan-Meier analysis showed that the disease-free survival of high-risk group patients was lower than that of low-risk group patients(P<0.001).The ROC curve analysis showed that the 5-year survival area under the curve of the training set,test set and whole set were 0.888,0.599 and 0.701,respectively.Gene enrichment analysis suggested that miRNAs were closely related to local adhesion pathways,regulation of stem cell pluripotency signaling pathways,transforming growth factor-βsignaling pathways,Wnt signaling pathways,FoxO signaling pathways,etc.Conclusion The miRNAs prognostic model of HNSCC constructed in this study can effectively predict the disease-free survival of patients and serve as an important indicator for assessing prognosis.