BACKGROUND Chemokines not only regulate immune cells but also play significant roles in development and treatment of tumors and patient prognoses. However, these effects have not been fully explained in hepatocellular carcinoma (HCC). MATERIALS AND METHODS We conducted a clustering analysis of chemokine-related genes. We then examined the differences in survival rates and analyzed immune levels using single sample gene set enrichment analysis (ssGSEA) for each subtype. Based on chemokine-related genes of different subtypes, we built a prognostic model in The Cancer Genome Atlas (TCGA) dataset using the survival package and glmnet package and validated it in the Gene Expression Omnibus (GEO) dataset. We used univariate and multivariate regression analyses to select independent prognostic factors and used R package rms to draw a nomogram reflecting patient survival rates at 1, 3, and 5 years. RESULTS We identified two chemokine subtypes, and after screening, found that Cluster1 had higher survival rates than Cluster2. In addition, in terms of immune evaluation, stromal evaluation, ESTIMATE evaluation, immune abundance, immune function, and expressions of various immune checkpoints, immune levels of Cluster1 were significantly better than those of Cluster2. The immunophenoscore (IPS) of HCC patients in Cluster1 was significantly higher than that in Cluster2. Furthermore, we established a prognostic model consisting of 9 genes, which correlated with chemokines. Through testing, RiskScore was revealed as an independent prognostic factor, and the model could effectively predict HCC patients' prognoses in both TCGA and GEO datasets. CONCLUSION This study resulted in the development of a novel prognostic model related to chemokine genes, providing new targets and theoretical support for HCC patients.

中文翻译：

---

基于趋化因子相关基因的肝细胞癌分型和预后模型构建。

背景趋化因子不仅调节免疫细胞，而且在肿瘤的发展和治疗以及患者预后中发挥重要作用。然而，这些作用在肝细胞癌（HCC）中尚未得到充分解释。材料和方法我们对趋化因子相关基因进行了聚类分析。然后，我们检查了存活率的差异，并使用每个亚型的单样本基因集富集分析 (ssGSEA) 分析了免疫水平。基于不同亚型的趋化因子相关基因，我们使用survival包和glmnet包在癌症基因组图谱（TCGA）数据集中建立了预后模型，并在基因表达综合（GEO）数据集中进行了验证。我们使用单变量和多变量回归分析来选择独立的预后因素，并使用 R 包 rms 绘制反映患者 1 年、3 年和 5 年生存率的列线图。结果我们鉴定了两种趋化因子亚型，经过筛选，发现Cluster1的存活率高于Cluster2。另外，在免疫评估、基质评估、ESTIMATE评估、免疫丰度、免疫功能、各免疫检查点表达等方面，Cluster1的免疫水平明显优于Cluster2。Cluster1中HCC患者的免疫表型评分（IPS）显着高于Cluster2。此外，我们建立了一个由 9 个基因组成的预后模型，这些基因与趋化因子相关。通过测试，RiskScore被揭示为独立的预后因素，该模型在TCGA和GEO数据集中都能有效预测HCC患者的预后。结论本研究开发了一种与趋化因子基因相关的新型预后模型，为HCC患者提供了新的靶点和理论支持。