In HCC, the tumor microenvironment comprises a diverse array of components, including cancer-associated fibroblasts, immune cells, blood vessels, and extracellular matrix proteins[7]. This intricate network regulates crucial processes such as angiogenesis, invasion, and immune evasion, impacting the overall behavior of the tumor. Immune system plays central roles in cancer development and progress. In this study, our objective was to unveil the immune-related signature in HCC in silico. This approach relies on extensive sequencing data and diverse bioinformatics algorithms, which are increasingly used to identify tumor subtypes and provide a valuable reference for the discovery of potential biomarkers.
Although a few articles have analyzed the roles of immune-related genes in predicting the prognosis and drug response of HCC, there is still a lack of relevant studies using single-cell sequencing results [19–21]. Previous studies were based on RNA sequencing, and did not take into consideration of the cell types in the microenvironment of HCC, which had significant limitations. In current study, single-cell sequencing results from the TISCH database were utilized to perform cluster analysis of cell types in HCC tissue and calculate the DEGs between CD8 + T cells and other cell types.
We identified 92 immune-related DEGs by intersecting them with 2533 known immune-related genes. Using univariate Cox regression analysis and the LASSO algorithm, we identified 11 key immune-related genes (FCER1G, RGS2, GRN, CTSB, CTSL, S100A9, IL7R, S100A11, CCR7, SPON2, PTGER4) including in our prognostic model. FCER1G had been identified as a key gene involved in cancer immune infiltration and tumor microenvironment [22]. RGS2, which was highly expressed in gastric cancer cells, was significantly associated with TMB, TIDE, and CD8 + T-cell infiltration [23], but its role in HCC remained unclear, which we were currently investigating. GRN was upregulated in HCC and interacted with NUPR1, playing a key role in HCC progression [24]. CTSB had been identified as a negative prognostic biomarker and therapeutic target associated with immune cell infiltration and immunosuppression in gliomas [25]. While few studies had focused on CTSB in HCC, the results suggested that high CTSB expression predicted poor prognosis and more frequent metastasis in HCC patients [26]. CTSL high expression was associated with higher ESTIMATE score, immune checkpoints expression, and lower TIDE score in lung adenocarcinoma [27]. Tumor-infiltrating monocytes/macrophages promoted tumor invasion and migration by elevating S100A9 expression in cancer cell [28]. Higher S100A9 level in HCC tissue or serum predicted a worse outcome for HCC patients [29]. IL7R was associated with tumor microenvironmental status and immune cell infiltration in lung adenocarcinoma [30]. S100A11 correlates with the immunosuppressive microenvironment in pan-cancer [31, 32]. CCR7 was highly expressed in human breast cancer cells, which induces chemotactic and invasive responses by mediating actin polymerization and pseudopodia formation, and subsequently [33]. SPON2 plays a crucial role in facilitating the immune response against the growth and migration of tumor cells in HCC[34]. PTGER4, interaction with its ligand SPP1, mainly contributed by the communication of malignant cells and TAMs, and elevated expressions were associated with poor patient outcome in HCC[35]. Overall those genes play a crucial role in tumor progression and immune cell function which may account for the better performance of the combined score of these 11 genes than the assessment of one immune checkpoint gene alone.
In our model, high-risk patients had significantly worse survival rates compared to the low-risk group, and risk scores can independently predict the prognosis of HCC patients. Based on DEGs between different risk groups, we conducted GSEA, GVSA, and GO functional analyses, finding enrichment in tumor-associated biological processes and pathways. Additionally, we found that TMB, immune score, stromal score, and ESTIMATE score were higher in the high-risk group compared to the low-risk group. Most immune checkpoints, including CTLA4, PD1, and PD-L1, were expressed at significantly higher levels in the high-risk group. Therefore, this model might offer a new pathway for immunotherapy in HCC.