Chongqing Key Laboratory of Molecular Oncology and Epigenetics, The First Affiliated Hospital of Chongqing Medical University, Chongqing, PR China.
Department of Intensive Care Unit, Affiliated Hangzhou First People's Hospital, Zhejiang University School of Medicine, PR China.
Gene. 2019 May 20;697:86-93. doi: 10.1016/j.gene.2019.01.046. Epub 2019 Feb 16.
The microRNAs (miRNAs) have been validated as prognostic markers in many cancers. The aim of this study was to identify new miRNA prognostic biomarkers in endometrial cancer (EC) and to develop an expression-based miRNA signature to provide survival risk prediction for EC patients.
From TCGA database, the miRNA datasets of EC and clinical information were downloaded in April 2018. Using univariate and multivariate Cox regression analyses identify prognostic factors. Using area under the curve (AUC) of receiver operating characteristic (ROC) curve assess the sensitivity and specificity of prognostic model.
530 patients were randomly divided into training set and testing set. Among 561 differentially expressed miRNAs, 4 miRNAs (miR-4758, miR-876, miR-142, miR-190b) were demonstrated to be predictive biomarkers of overall survival (OS) for EC patients in training set. Based on the risk score of 4-miRNA model, patients in the training set were divided into high-risk and low-risk groups with significantly different OS. This 4-miRNA model was validated in testing and entire set. The AUC for the ROC curves in the entire set was 0.704. Meanwhile, multivariate Cox regression combined with other traditional clinical parameters indicated that the 4-miRNA model can be used as an independent OS prognostic factor. Functional enrichment analysis revealed that these miRNAs are involved in biological processes and pathways that are closely related to cancer.
A robust 4-miRNA signature as an independent prognostic factor for OS in EC patients was established.
微小 RNA（miRNAs）已被验证为许多癌症的预后标志物。本研究旨在鉴定子宫内膜癌（EC）中的新型 miRNA 预后生物标志物，并开发基于表达的 miRNA 特征，为 EC 患者提供生存风险预测。
从 TCGA 数据库中，于 2018 年 4 月下载了 EC 的 miRNA 数据集和临床信息。使用单变量和多变量 Cox 回归分析确定预后因素。使用接受者操作特征（ROC）曲线的曲线下面积（AUC）评估预后模型的灵敏度和特异性。
530 名患者被随机分为训练集和测试集。在 561 个差异表达的 miRNAs 中，miR-4758、miR-876、miR-142 和 miR-190b 这 4 个 miRNA 被证明是训练集中 EC 患者总生存期（OS）的预测生物标志物。基于 4- miRNA 模型的风险评分，训练集中的患者被分为高风险和低风险组，两组的 OS 有显著差异。该 4- miRNA 模型在测试集和整个数据集均得到验证。整个数据集的 ROC 曲线的 AUC 为 0.704。同时，多变量 Cox 回归结合其他传统临床参数表明，4- miRNA 模型可作为 OS 的独立预后因素。功能富集分析表明，这些 miRNA 参与了与癌症密切相关的生物学过程和途径。
建立了一种稳健的 4- miRNA 特征，作为 EC 患者 OS 的独立预后因素。
