Tian Siyuan, Liu Jingyi, Sun Keshuai, Liu Yansheng, Yu Jiahao, Ma Shuoyi, Zhang Miao, Jia Gui, Zhou Xia, Shang Yulong, Han Ying
State Key Laboratory of Cancer Biology and National Clinical Research Center for Digestive Diseases, Xijing Hospital of Digestive Diseases, The Fourth Military Medical University, Xi'an, China.
Department of Radiation Oncology, Xijing Hospital, The Fourth Military Medical University, Xi'an, China.
Front Oncol. 2021 Jan 26;10:597996. doi: 10.3389/fonc.2020.597996. eCollection 2020.
Evidence from prevailing studies show that hepatocellular carcinoma (HCC) is among the top cancers with high mortality globally. Gene regulation at post-transcriptional level orchestrated by RNA-binding proteins (RBPs) is an important mechanism that modifies various biological behaviors of HCC. Currently, it is not fully understood how RBPs affects the prognosis of HCC. In this study, we aimed to construct and validate an RBP-related model to predict the prognosis of HCC patients.
Differently expressed RBPs were identified in HCC patients based on the GSE54236 dataset from the Gene Expression Omnibus (GEO) database. Integrative bioinformatics analyses were performed to select hub genes. Gene expression patterns were validated in The Cancer Genome Atlas (TCGA) database, after which univariate and multivariate Cox regression analyses, as well as Kaplan-Meier analysis were performed to develop a prognostic model. Then, the performance of the prognostic model was assessed using receiver operating characteristic (ROC) curves and clinicopathological correlation analysis. Moreover, data from the International Cancer Genome Consortium (ICGC) database were used for external validation. Finally, a nomogram combining clinicopathological parameters and prognostic model was established for the individual prediction of survival probability.
The prognostic risk model was finally constructed based on two RBPs (BOP1 and EZH2), facilitating risk-stratification of HCC patients. Survival was markedly higher in the low-risk group relative to the high-risk group. Moreover, higher risk score was associated with advanced pathological grade and late clinical stage. Besides, the risk score was found to be an independent prognosis factor based on multivariate analysis. Nomogram including the risk score and clinical stage proved to perform better in predicting patient prognosis.
The RBP-related prognostic model established in this study may function as a prognostic indicator for HCC, which could provide evidence for clinical decision making.
现有研究证据表明，肝细胞癌（HCC）是全球死亡率最高的癌症之一。由RNA结合蛋白（RBPs）精心调控的转录后水平基因调控是改变HCC多种生物学行为的重要机制。目前，RBPs如何影响HCC的预后尚不完全清楚。在本研究中，我们旨在构建并验证一个与RBP相关的模型，以预测HCC患者的预后。
基于基因表达综合数据库（GEO）中的GSE54236数据集，在HCC患者中鉴定出差异表达的RBPs。进行综合生物信息学分析以选择枢纽基因。在癌症基因组图谱（TCGA）数据库中验证基因表达模式，然后进行单变量和多变量Cox回归分析以及Kaplan-Meier分析，以建立预后模型。然后，使用受试者工作特征（ROC）曲线和临床病理相关性分析评估预后模型的性能。此外，国际癌症基因组联盟（ICGC）数据库的数据用于外部验证。最后，建立了一个结合临床病理参数和预后模型的列线图，用于个体生存概率预测。
最终基于两种RBPs（BOP1和EZH2）构建了预后风险模型，有助于对HCC患者进行风险分层。低风险组的生存率明显高于高风险组。此外，较高的风险评分与晚期病理分级和临床分期相关。此外，基于多变量分析发现风险评分是一个独立的预后因素。包括风险评分和临床分期的列线图在预测患者预后方面表现更好。
本研究建立的与RBP相关的预后模型可能作为HCC的预后指标，可为临床决策提供依据。
