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.

中文翻译：

---

RNA结合蛋白相关的肝细胞癌预后模型的开发和验证

肝细胞癌（HCC）是最致命的癌症之一。尽管已经证明RNA结合蛋白（RBPs）是肿瘤发生和肿瘤进展的关键调节剂，但在HCC的背景下它们的失调仍有待充分表征。下载并分析了来自癌症基因组图谱-肝HCC（TCGA-LIHC）数据库的数据，以鉴定相对于健康正常组织在HCC肿瘤中差异表达的RBP。然后使用GO和KEGG数据库进行这些RBP的功能富集分析，以了解其机理作用。然后通过Cox回归分析检测与HCC患者预后相关的中枢枢纽RBP，并将其纳入预后模型。然后通过使用Kaplan-Meier曲线评估该模型的预后价值，时间相关的ROC分析，单变量和多变量Cox回归分析以及列线图。最后，使用Kaplan-Meier曲线评估了各个中心枢纽RBP与HCC患者总体生存（OS）之间的关系。最后，与轮毂RBP相关的发现蛋白编码基因（PCG）被用于构建轮毂RBP-PCG共表达网络。总共，我们鉴定出相对于健康组织在HCC肿瘤中差异表达的81个RBP（54个上调，27个下调）。然后使用七个与预后相关的中心RBP（SMG5，BOP1，LIN28B，RNF17，ANG，LARP1B和NR0B1）生成预后模型，然后根据结果风险评分将HCC患者分为高风险和低风险组价值观。在训练和测试数据集中，我们发现，相对于低风险患者，高风险肝癌患者的OS降低，ROC曲线值下的时间依赖性面积分别为0.801和0.676。因此，该模型表现出良好的预后性能。我们还基于这七个枢纽RBP生成了预后列线图，并发现其他四个基因与OS显着相关。我们在本文中确定了7个RBP签名，这些签名可以可靠地用于预测HCC患者的OS，强调了这些基因的预后相关性。