基于生物信息学的结直肠腺癌miRNA预后模型的建立

目的通过生物信息学分析构建结直肠腺癌的miRNA预后模型。方法通过癌症基因组图谱(The Cancer Genome Atlas,TCGA)获取结直肠腺癌患者的miRNA表达谱与相应的临床数据,通过差异分析明确肿瘤组与正常对照组之间差异表达的miRNA。随后,将具有生存状态和生存时间的患者(n=484)随机分为训练集(n=244)和测试集(n=240),对训练集的差异miRNAs和患者的总生存期进行单因素和多因素Cox分析来构建预后模型,并用Kaplan-Meier法与受试者工作特征曲线(receiver operating characteristic,ROC)在训练集与合并数据集中验证模型的效能。此外,用单因素和多因素Cox分析验证miRNA模型是否为结直肠腺癌患者的独立预后因素。为了明确miRNAs的生物学功能,对miRNAs的靶基因进行了KEGG、GO富集分析。结果我们从TCGA数据库中获得了肿瘤组织与正常组织之间520个差异表达的miRNAs,在训练集中通过单基因与多因素Cox分析筛选出7个miRNA(hsa-miR-891a-5p、hsa-miR-664b-3p、hsa-miR-485-5p、hsa-miR-486-5p、hsa-miR-3615、hsa-miR-21-3p和hsa-miR-3677-3p)用以构建模型,并将患者分为高风险组与低风险组,Kaplan-Meier分析结果提示高风险组患者的总生存期明显低于低风险组患者。训练集、测试集与合并数据集的ROC表明该模型预测5年生存期的曲线下面积(area under the curve,AUC)分别是0.722、0.747和0.735,提示该模型有较好的预测能力。此外,多因素Cox分析提示该模型是患者的独立风险因素。基因富集分析结果提示miRNAs相关的靶基因与Hippo信号通路、cGMP-PKG信号通路、细胞生长等通路密切相关。结论本研究建立了一个miRNA预后模型能够有效地预测结直肠腺癌患者的总生存时间,并为研究miRNA在结直肠腺癌发生发展的作用提供了新的方向。

Identifying a new miRNA signature as a prognostic biomarker in colorectal adenocarcinoma

Objective To identify a new miRNA signature as a prognostic biomarker in colorectal adenocarcinoma.Methods miRNA expression profiles and clinical information of colorectal adenocarcinoma patients were collected from The Cancer Genome Atlas(TCGA)database.Differential gene expression analysis methods were first applied to identify differentially expressed miRNAs between normal tissues and patients with colorectal adenocarcinoma.Subsequently,all the patients(n=484)with complete survival time and status were divided into training dataset(n=244)and test dataset(n=240).Univariate and multivariate Cox regression analysis were performed to construct the miRNA expression signature in the training dataset.The prognostic power of the signature was evaluated through receiver operating characteristic(ROC)analysis and Kaplan-Meier analysis in the testing dataset and combined dataset.Univariate and multivariate analysis were performed to determine whether the prognostic value of the prognostic risk model was independent of other clinicopathological parameters.In order to predict the potential biological functions of the miRNA signature,Kyoto Encyclopedia of Genes and Genomes(KEGG)signaling pathway and Gene Ontology(GO)enrichment analysis were performed to explore the functions of target genes of miRNAs.Results We obtained 520 differentially expressed miRNAs between normal tissues and patients with colorectal adenocarcinoma from the TCGA dataset.After univariate and multivariate Cox regression analysis were performed for these differentially expressed miRNAs and overall survival of colorectal adenocarcinoma patients in the training dataset,a 7-miRNA signature(hsa-miR-891a-5p,hsa-miR-664b-3p,hsa-miR-485-5p,hsa-miR-486-5p,hsa-miR-3615,hsa-miR-21-3p,hsa-miR-3677-3p)was constructed to divide colorectal adenocarcinoma patients into high-and low-risk subgroups with significantly different overall survival.Area under the curve(AUC)of ROC curve for predicting 5 years’survival in training dataset,testing dataset,and combined dataset were 0.722,0.747,and 0.735,respectively,which demonstrated better predictive power of prognostic model.Multivariate Cox regression further showed that the 7-miRNA signature could serve as an independent prognostic factor.In addition,functional analysis of the target genes suggested that cell growth,the Hippo and cGMP-PKG signaling pathway were enriched.Conclusion Our study establishes a novel signature from 7 miRNAs to predict the overall survival of colorectal adenocarcinoma,which may provide new insights into the role of miRNAs in the tumorigenesis and progression of colorectal adenocarcinoma.