HCC is one of the most common malignant tumors and is estimated as the fourth leading cause of death (18). The liver is a huge immune organ and complex microenvironment under physiologic conditions that could suppress abnormal inflammatory responses and maintain immunotolerance (19). The development of immuno-oncology has transformed cancer treatment strategies over the past decade. Immune checkpoints are important in the maintenance of self-tolerance under physiologic conditions, including PD-1 and its ligand PD-L1, as well as CTLA-4 (3, 20). The therapeutic strategies for HCC included surgery and liver transplantation, but it’s not suitable for most patients with advanced liver cancer. Accordingly, immunotherapy is a potential method, such as immune checkpoint inhibition (ICI) and chimeric antigen receptor redirected T (CAR-T) cell therapy (2, 3, 20). To better understand the correlation between the immune system and HCC, we carried out the present work about the IRGs how does it affect the prognosis of HCC patients.
First, we identified eight optimal PIRDEGs from HCC patients which were downloaded in the TCGA database to conduct a precise prognostic model for predicting overall survival (OS) of HCC patients. These eight genes are all tumor protective factors, which prevented regression in HCC patients, including HSPA4, PSMD14, RBP2, MAPT, TRAF3, NDRG1, NRAS, IL17D. But IL17D only showed little significant difference in the survival models. The GO analysis and KEGG pathway analysis were performed in R to discover the important biological functions and cellular molecular pathways of the PIRDEGs. And the correlation between PIRDEGs and TFs was completed with a network diagram. Thereafter, we have sufficient preconditions to accurately predict the prognosis of HCC patients in our prognostic model.
A survival analysis produced via the prognostic model indicated that patients with a high-risk score showed a poorer prognosis than the low-risk score group (P < 0.05). Additionally, univariate Cox prognostic analysis indicated that the risk score, pathological stage, T, and M status were independent prognostic factors and the AUC value of ROC was 0.813 which indicated that the risk score was an accurate prognostic factor for predicting the prognosis of HCC patients. And the other clinical signatures were dependent on predicting factors, such as age, gender, histological grade, pathological stage, TNM status. Furthermore, it has been reported that AUC > 0.60 was supposed to be a receivable foretell, and AUC > 0.75 was deemed to have excellent predictive value (21). As evidence, the risk curve showed that survival time and the expression of eight selected genes were positively proportional to the risk score. Hence, the prognostic model had a precise assessment of HCC patients in many aspects, such as age, gender, pathological stage, and TNM status, for better improving the poor prognosis of patients.
We also analyzed the relationship between the sifted genes and other clinical factors including age, fustat, histological grade, pathological stage, TNM status, and OS. The fustat showed that these genes were significantly expressed higher in dead patients than alive ones. The expression level of HSPA4 was positively correlated with histological grade, and its expression was higher in stage 2 of HCC patients than stages 1 and 3, and it was similar to the changing pattern of primary tumor size. A previous study also demonstrated that HSPA4 was associated with poor prognosis of liver cancer (22) and breast cancer (23). Studies have shown that NDRG1 can promote the proliferation, progression, and poor prognosis of liver cancer (24, 25). Our studies further proved that the expression of NDRG1 was positively correlated with the death, histological grade, pathological stage, and primary tumor size of HCC-TCGA patients. Furthermore, in the stage and grade 2–3, as well as primary tumor size, the expression of PSMD14 was higher than the stage and grade 1. However, NRAS’s influence on staging and tumor size is very weak. The expression of IL17D and TRAF3 was higher in grade 3 and stage 3 respectively. A small number of researchers have used bioinformatics methods to demonstrate the differential expression of IL17D in HCC (26). And Lv, J., et al. showed that PSMD14 can promote the proliferation and metastasis of liver cancer cells (27). The survival analysis revealed that high expression of these genes was associated with poor prognosis in HCC patients.
Immune infiltration plays an important role in the genesis and development of the tumor. As reported, tumor-infiltrating cells were preferentially enriched in HCC, such as exhausted CD8+ T cells and Tregs (28). Increased tumor infiltration of Tregs is associated with CD8+ T cell dysfunction, HCC invasiveness as well as progression, and poor patient outcomes (3, 29). Macrophages and dendritic cells (DCs) which can have anti-tumor functions may be transformed into tumor-associated macrophages (TAMs) and myeloid-derived suppressor cells (MDSCs) phenotype which have an immunosuppressive and pro-tumor feature. And the high level of TAMs and DCs in HCC was correlated with poor prognosis of HCC patients (3, 30). In the present study, the results showed that the risk score was positively correlated with M2 macrophages, activated CD4+ T memory cells, follicular helper T cells, and eosinophils. In the previous studies, Junyu Long .et.al had found that HCC patients in the high-risk group exhibited higher fractions of follicular helper T cells than the low-risk group(31). And Oscar W H Yeung .et.al discovered that M2 macrophages contribute to invasiveness and poor prognosis of HCC(32).
Finally, we conducted GO and KEGG pathway analysis to deeply explored the role of these genes and DETFs selected from our prognostic model, further comprehending the molecular mechanisms of HCC development, deterioration, and poor prognosis. The KEGG pathway analysis showed that the expression of these genes and DETFs was highly associated with cell cycle, cell senescence, Epstein-Barr virus infection, hepatitis B, and hepatocellular carcinoma. And the GO analysis showed that these genes and DETFs were associated with chromatin modification, transcription regulator complex, and DNA-binding transcription activator activity. We selected partial immune-related biological processes associated with the selected TFs and genes, and further analysis revealed that IRF5, POLR3A, PRKDC, CEBPA, SRC, PSMD14, NRAS, IL17D, JMJD6, MAPT, SOX4, and TRAF3 were intricately associated with the immune functions as follow: positive regulation of type I interferon production, positive regulation of defense response, macrophage activation, pro-B cell differentiation, regulation of interferon-beta production, lymphoid progenitor cell differentiation and activation of the innate immune response.
In general, the results of our study exhibited a potential direction of factors related to liver cancer prognosis and immunity. Nothing but some deficiencies existed in the present work. On the one hand, our sample size was not large enough to take into account the differences between individual patients and the heterogeneity of the tumor environment. On the other hand, we have not verified the pathways and mechanisms of these genes by in vitro and in vivo experiments.