Department of Hepatobiliary Surgery, Guangxi Medical University Cancer Hospital, Nanning, China.
Department of Chinese Medicine, Guangxi Medical University Cancer Hospital, Nanning, China.
BMC Cancer. 2020 Nov 23;20(1):1136. doi: 10.1186/s12885-020-07625-3.
Hepatocellular carcinoma (HCC) is among the deadliest forms of cancer. While RNA-binding proteins (RBPs) have been shown to be key regulators of oncogenesis and tumor progression, their dysregulation in the context of HCC remains to be fully characterized.
Data from the Cancer Genome Atlas - liver HCC (TCGA-LIHC) database were downloaded and analyzed in order to identify RBPs that were differentially expressed in HCC tumors relative to healthy normal tissues. Functional enrichment analyses of these RBPs were then conducted using the GO and KEGG databases to understand their mechanistic roles. Central hub RBPs associated with HCC patient prognosis were then detected through Cox regression analyses, and were incorporated into a prognostic model. The prognostic value of this model was then assessed through the use of Kaplan-Meier curves, time-related ROC analyses, univariate and multivariate Cox regression analyses, and nomograms. Lastly, the relationship between individual hub RBPs and HCC patient overall survival (OS) was evaluated using Kaplan-Meier curves. Finally, find protein-coding genes (PCGs) related to hub RBPs were used to construct a hub RBP-PCG co-expression network.
In total, we identified 81 RBPs that were differentially expressed in HCC tumors relative to healthy tissues (54 upregulated, 27 downregulated). Seven prognostically-relevant hub RBPs (SMG5, BOP1, LIN28B, RNF17, ANG, LARP1B, and NR0B1) were then used to generate a prognostic model, after which HCC patients were separated into high- and low-risk groups based upon resultant risk score values. In both the training and test datasets, we found that high-risk HCC patients exhibited decreased OS relative to low-risk patients, with time-dependent area under the ROC curve values of 0.801 and 0.676, respectively. This model thus exhibited good prognostic performance. We additionally generated a prognostic nomogram based upon these seven hub RBPs and found that four other genes were significantly correlated with OS.
We herein identified a seven RBP signature that can reliably be used to predict HCC patient OS, underscoring the prognostic relevance of these genes.
肝细胞癌（HCC）是最致命的癌症之一。虽然已经证明 RNA 结合蛋白（RBPs）是癌发生和肿瘤进展的关键调节剂，但它们在 HCC 中的失调仍有待充分表征。
从癌症基因组图谱 - 肝 HCC（TCGA-LIHC）数据库中下载并分析数据，以鉴定在 HCC 肿瘤中相对于健康正常组织表达差异的 RBPs。然后使用 GO 和 KEGG 数据库对这些 RBPs 进行功能富集分析，以了解它们的机制作用。通过 Cox 回归分析检测与 HCC 患者预后相关的核心 hub RBPs，并将其纳入预后模型。然后使用 Kaplan-Meier 曲线、时间相关 ROC 分析、单因素和多因素 Cox 回归分析以及列线图评估该模型的预后价值。最后，使用 Kaplan-Meier 曲线评估单个 hub RBPs 与 HCC 患者总生存期（OS）的关系。最后，使用 Kaplan-Meier 曲线评估与 hub RBPs 相关的蛋白编码基因（PCGs），构建 hub RBP-PCG 共表达网络。
总共鉴定出 81 个在 HCC 肿瘤中相对于健康组织表达差异的 RBPs（54 个上调，27 个下调）。然后使用七个预后相关的 hub RBPs（SMG5、BOP1、LIN28B、RNF17、ANG、LARP1B 和 NR0B1）生成预后模型，之后根据所得风险评分值将 HCC 患者分为高风险和低风险组。在训练和测试数据集，我们发现高风险 HCC 患者的 OS 低于低风险患者，时间依赖性 ROC 曲线下面积分别为 0.801 和 0.676。因此，该模型具有良好的预后性能。我们还根据这七个 hub RBPs 生成了一个预后列线图，并发现另外四个基因与 OS 显著相关。
我们在此鉴定了一个由七个 RBP 组成的特征，可以可靠地用于预测 HCC 患者的 OS，突出了这些基因的预后相关性。
