The Landscape of Iron metabolism-related Genes for Overall survival prediction in Patients with Hepatocellular carcinoma

Background Hepatocellular carcinoma (HCC) is the seventh most commonly occurring cancer and the second most common cause of cancer-related death worldwide. Despite improvements in early detection and treatment, the morbidity and mortality remain high because of complex molecular mechanisms and cellular heterogeneity in HCC. However, novel model is still needed to predict the survival and clinical immunotherapy response in HCC. Methods 13 iron metabolism-related gene sets were identied from the GSEA. DEGs associated with iron metabolism were calculated between patients who survived < 1 year and more than 3 years for subsequent analysis. Univariate cox proportional hazard regression and LASSO analysis were performed to construct a gene signature. The Kaplan–Meier analysis, time-dependent receiver operating characteristic (ROC), Univariate and Multivariate Cox regression analysis, stratication analysis, Principal Component Analysis (PCA) analysis were used to assess the prognostic value of the gene signature. Furthermore, the reliability and validity were validated in external testing cohort, internal testing cohort. Moreover, Weighted gene co-expression network analysis(WGCNA), Gene set enrichment analysis(GSEA), Gene Ontology (GO) and KEGG analysis were performed to reveal signaling pathways, and two independent prognostic factors were combined to build Nomogram for predicting HCC prognosis. Finally, expression level of genes of gene signature in clinical samples was performed using quantitative real-time RT-PCR (qRT-PCR) analyses. Results Here we established a gene signature using iron metabolism-related genes form the Cancer Genome Atlas (TCGA), and the survival outcomes were validated from International Cancer Genome Consortium(ICGC). Distinct subtypes (high- and low-risk subtypes) were characterized by different clinical outcomes. The high-risk subtype was featured by better survival outcomes, upregulated cell cycle relevant pathways and better response for immunotherapy. Conclusions Our nding suggested that the novel model may be useful as a biomarker for prognostic predication, immunotherapy and cell cycle inhibitors may be ecacious for high-risk subtype of HCC patients.


Introduction
Due to the complex molecular mechanisms and cellular heterogeneity in HCC, [1][2][3] it remains high morbidity and mortality in the world. Despite the improvement of treatment, the morbidity and mortality remain high and it is the most rapidly increasing cancer types leading cause of tumor-related death. [4] In the US, 5-year survival rate is only 18%. [5] Surgery is the main treatment in early and middle period. However, most patients have locally advanced or widely metastatic tumors at the time of diagnosis, extra therapeutic strategies were required for patients in advanced and metastatic status. Unfortunately, the treatment method for advanced and metastatic cases still has a feeble role because of molecular mechanisms and cellular heterogeneity, and metabolic pathway may be involved in the occurrence and progression of HCC, which mediated by iron, nutrient, oxygen, energy stimulation. It is urgently required to construct prediction model to predict the prognosis for patients with HCC. Although several previous studies have revealed different types of signatures for predicting the prognosis of HCC, effective biomarker for prognosis prediction is still lack. Thus, novel prognostic biomarkers are urgently needed to predict the survival of HCC patients and outline an individualized treatment plan for HCC patients.
Iron participates in multiple types of basic biological processes, including cell replication, metabolism and growth. Disorder of iron metabolism is associated with cell proliferation and unraveled to be involved in most tumors.[6-8] Ferroptosis, as a hot topic of disorder of iron metabolism, regulates cell death via membrane lipid peroxidation. [9,10] In recently, increased studies have proved the vital role of iron mechanism on the pathogenesis and inhibition of ferroptosis may increase the anticancer activity in HCC. [11][12][13] Previous studies revealed iron metabolism-related genes in HCC, including NRF2, [13] MT1G, [12]Rb, [11]CISD1, [14] Inspired by all these works, we constructed an iron metabolic gene signature by performing training cohort from TCGA dataset. Then, the prognostic value of our 13-gene signature was validated in internal validation cohort, external validation cohort. Finally, functional enrichment analysis was performed to explore the underlying mechanisms and provide e cacious therapeutic schedule for patients in high risk group.

Data Source and grouping
365 primary HCCs candidates and clinical information(age, gender, BMI, vascular tumor cell, grade, TNM stage) were acquired from The cancer genome atlas-liver hepatocellular carcinoma (TCGA-LIHC, the 13 gene sets and selected 482 iron metabolism-related genes from TCGA and ICGC gene expression pro le.

Prognostic signature construction
In the training cohort, DEGs associated with iron metabolism were calculated between patients who survived < 1 year and more than 3 years by using the Limma version 3.36.2 R package, and |Log 2 FC|> 1.5 and adjusted P < 0.05 were considered as selection criterion. Univariate Cox proportional hazards regression analysis was performed to assess the relationship between DEGs and the overall survival (OS) to screen prognostic genes, with adjusted P-value < 0.05 as the signi cance cutoff. An interaction network for the prognostic DEGs was generated by the STRING database (version 11.0) The LASSO Cox regression was applied to minimize over tting by performing R package glmnet. A multi-gene markerbased prognostic signature was established to assess risk score for HCC patients, the formula is as follows: Risk score = βgene1 × Egene1 + βgene2 × Egene2 + βgene3 × Egene3 ... βgeneN × EgeneN. Taking the median risk score as a cutoff value, HCC patients in training cohort were divided into low or high risk group.
2.4 The performance of gene signature KM survival curves was performed to compare the survival rate between low or high risk group, and Time-ROC curve analyses the predictive capacity of the gene signature. Next, Univariate and Multivariate Cox regression analysis were used to assess whether the gene signature could be independent of other clinical parameters, including age, gender, BMI, vascular tumor cell, grade, TNM stage. Furthermore, we applied strati cation analysis to assess the predictive capacity of the gene signature in different subgroups strati ed clinical variables. Finally, the gene signature was validated in external testing cohort, internal testing cohort.
2.5 Identi cation of Key Co-expression Modules associated with Risk score Using WGCNA To furtherly nd the gene modules associated with Risk score, the samples from HCC patients were divided into high-and low-high risk group according to median risk score, high and low groups were considered as external sample traits. Gene co-expression network was constructed to explore the modules highly associated with sample traits, 9 was selected as the soft thresholding power to produce a weighted network. The enrolled genes were clustered into 9 modules except the gray module, the modules with high correlation coe cient were considered candidates relevant to traits, which were picked out to perform further pathway enrichment analysis using an online-based web tool "Metascape" (http://metascape.org/).

Gene set enrichment analysis
To identify potential mechanisms underlying our constructed prognostic gene signature to affect the HCC, GSEA was performed to nd enriched terms using TCGA-LIHC dataset, and GSEA was performed based on high risk group and low risk group. The FDR < 0.05 was used to sort the pathways enriched in each phenotype.

Prediction of immunotherapy response
CIBERSORT computational method was implemented to estimate TIC abundance pro le in all tumor samples, and samples with p < 0.05 were selected for the following analysis. The difference analysis of immune cells and the activity of immune-related pathways were calculated by performing single-sample gene set enrichment analysis (ssGSEA) in the "gsva" R package.23 ImmuneCellAI (http://bioinfo.life.hust.edu.cn/web/ImmuCellAI/) was used to forecast the immunotherapy response of high and low risk group.
2.8 Building and validating a predictive nomogram for HCC Survival Prediction Independent prognostic factors identi ed by multivariate Cox regression analysis were selected to build a nomogram, The calibration curve plotted to observe the prediction probabilities against the observed rates. Time-ROC analysis was performed to compare the prediction accuracy between the nomogram and single independent prognostic factor.\ 2.9 Human tissue samples Twelve HCC and adjacent tissue samples were collected from patients -during surgery at The First A liated Hospital of Xiamen University; including. This study was approved Institutional Review Board of The First A liated Hospital of Xiamen University.

Quantitative real-time RT-PCR (qRT-PCR) analyses
Total RNA was extracted from the human tissue specimens using TRIzol reagent (Invitrogen; Thermo Fisher Scienti c, Inc.). cDNA was synthesized using ReverTra AceH qPCR RT Kit (TOYOBO) and qRT-PCRs were performed using a LightCycler480II (Roche) with SYBR Green technology (Takara). The relative expression levels of the target mRNA were calculated using the 2-ΔΔCt method (ΔCt = Cttarget gene -Ctinternal control). The primer sequences are shown in TableS1. qRT-PCR was performed for 30 cycles consisting of denaturation (at 94°C for 30 sec), annealing (at 56°C for 30 sec) and elongation (at 72°C for 1 min).

Study Process and Patients' Information
The entire work of construction process of prediction model for HCC patients was presented in Fig. 1A, and the clinical information from training cohort, internal testing cohort and external testing cohort was summarized in Table 1. 3.2 Establishment of the four-gene-based prognostic gene signature 510 iron metabolism-related genes were identi ed from the 13 gene sets, and 481 overlapping iron metabolism-related genes were obtained between training cohort and external cohort. A total of 37 DEGs were obtained when compared patients who survived < 1 year and more than 3 years using training set in the TCGA database (Fig. 1B). 370 patients with complete survival time from training cohort were included for subsequent univariate Cox proportional hazards regression analysis, 25 genes were signi cantly associated with OS( Fig. 2C,D). Then the LASSO analysis was performed to narrow the gene range. 13 genes were identi ed and subsequently used to construct a gene signature. Risk-formula: Risk score = expression of CYP3A5*-0.00144471228429986 + expression of CCNB1*0.00932548817179557 + expression of ABCB6*0.115364054791514 + expression of FLVCR1*0.00981797872412028 + expression of OSBP2*0.0960318647350786 + expression of G6PD*0.00325860881257371 + expression of RAP1GAP*0.00568586514806877 + expression of SLC7A11*0.0217165375535805 + expression of PPAT*0.196619205112227 + expression of CYP2C9*-0.000726941457117346 + expression of PLOD2*0.010247507631253 + expression of IGSF3*0.000979749106599136 + expression of RRM2*0.00176828844166581. The interaction network among these genes was presented in Fig. 2E and the correlation between these genes is presented in Fig. 1F. We then calculated the risk score for each patient in training cohort and median risk score was set as cut-off for risk score, 370 patients in training cohort were divided into high risk (185)and low risk (185)group, 184 patients from internal testing cohort were divided into high risk (86)and low risk (98)group, 243 patients from external testing cohort were divided into high risk (185) and low risk (58)group.

Diagnostic Values of iron metabolism-related signature
In training cohort, Kaplan-Meier curve, time-dependent ROC and PCA analysis were performed to assess the integrated effects of iron metabolism-related signature on the prognosis (Fig. 2). Patients in high-risk group shown signi cantly poorer OS than patients in the low-risk group (

Further Validation of the iron metabolism-related signature using Independent internal testing cohort and external testing cohort
In order to verify the reliability of the 13-Gene Signature, similar procedures were performed in internal testing cohort (Fig. 3), external testing cohort (Fig. 4). Patients in the high-risk group indicated signi cantly poorer OS than patients in the low-risk group in internal testing cohort (Fig. 3B)and external testing cohort ( Iron metabolism-related signature could be considered as an independent factor in predicting OS. Univariate and Multivariate Cox regression analysis were conducted to assess independent predictive values. In training cohort, Univariate Cox regression analysis indicated that BMI, TNM stage and signature had a prognostic value (Fig. 2F), Multivariate Cox regression analysis revealed that BMI, TNM stage and signature were independent prognostic factor correlation with worse OS (Fig. 2G). In internal testing cohort, Univariate Cox analysis suggested that BMI, TNM stage and signature were associated with OS( Fig. 3F), Multivariate Cox regression analysis showed that BMI, TNM stage and signature were independent prognostic factor associated with OS (Fig. 3G). In external testing cohort, Univariate Cox analysis suggested that gender, TNM stage and signature had a prognostic value (Fig. 4F), Multivariate Cox regression analysis indicated that gender, TNM stage and signature were independent prognostic factor associated with OS (Fig. 4G). These results con rmed that gene signature is an independent predictor of prognosis in HCC patients. 3.6 Validation of expression level of the four genes and risk score in different subgroups As showed in Fig. 5A-M, the expression level of ABCB6, CCNB1, FLVCR1, G6PD, IGSF3, OSBP2, PLOD2, PPAT, RAP1GAP, RRM2 and SLC7A11 was signi cantly higher in high risk group compared with low risk group in training cohort, The expression level of CYP3A5 and CYP2C9 was signi cantly lower in high risk group compared with high risk group in training cohort. Moreover, the risk score in tumor was obviously higher in than normal liver (Fig. 5N), the AUC was 0.932 (95%CI: 0.919-0.945) (Fig. 5O). In addition, the heatmap showed the expression of the nine genes in high-and low-risk patients in the TCGA dataset. It showed that high-and low-risk group associated with age, sex, survival status, grade and stage (Fig. 6A).

Subgroup Analysis of OS by clinical variables
We performed survival analysis strati ed by TNM stage in training cohort (Fig. 6C,D), internal testing cohort (Fig. 6F,G) and external testing cohort (Fig. 6I,J). Risk score grouped based on 13-gene signature, the log-rank test indicated that HCC patients in high risk group still had obviously worser OS than patients in low risk group in different subgroups in stage I-II and stage III-IV. Furthermore, as the TNM stage and tumor grade increased, the risk score increased in training cohort( Fig. 6B), internal testing cohort (Fig. 6E) and external testing cohort (Fig. 6H) HCC patients were divided into high-and low-risk groups according to median risk score. We identi ed 936 up-upregulated DEGs and 162 down-regulated DEGs(corrected P-value < 0.05 and absolute log fold change (FC) > 1), Go enrichment analysis revealed that DEGs were mainly involved in mitotic nuclear division, nuclear division, chromosome segregation, nuclear chromosome segregation (Fig. 7A). KEGG enrichment analysis revealed that DEGs were mainly involved in Cell cycle, ECM − receptor interaction, Glycolysis / Gluconeogenesis and so on (Fig. 7B).
Moreover, we conducted GSEA between high and low risk group compared with the median level of risk score to identify the signi cant pathways (NOM P-value < 0.05), the genes in high risk group were mainly enriched in cell cycle related pathways, including KEGG_CELL_CYCLE, KEGG_OOCYTE_MEIOSIS, KEGG_PURINE_METABOLISM, KEGG_PYRIMIDINE_METABOLISM, KEGG_RNA_DEGRADATION. As to low risk group, the genes were enriched in metabolic pathways, such as KEGG_DRUG_METABOLISM_CYTOCHROME_P450,KEGG_FATTY_ACID_METABOLISM, KEGG_GLYCINE_SERINE_AND_THREONINE_METABOLISM, KEGG_PRIMARY_BILE_ACID_BIOSYNTHESIS,KEGG_TRYPTOPHAN_METABOLISM (Fig. 7C).
To furtherly investigate the potential mechanisms of Risk score in HCC, WGCNA method was performed to cluster genes that highly correlated with the high risk and low risk group. 9 was selected as the soft thresholding power to produce a weighted network (Fig. 8A). We identi ed 9 modules (Fig. 8B,C) in the study(excluding a gray module). As showed in Fig. 8D, we found genes in blue module was highly positively associated with Risk score. Go enrichment analysis was performed to explore the potential mechanisms of genes in blue module using online-based web tool "Metascape" (http://metascape.org/), as showed in Fig. 8E and Fig. 8F, genes in the blue module were signi cantly enriched in metabolic process and cell cycle-related process.

Predictive immunotherapy response of identi ed HCC subtypes
We employed CIBERSORT algorithm to identify tumor in ltrating immune subsets proportions in HCC, and 21 kinds of immune cell pro les in HCC samples were constructed and showed in Fig. 9A,B. The difference analysis revealed that almost all types of immune cells (excluding B. cells. naive )were signi cantly different between high-and low-risk group (Fig. 9C). To further explore the correlation between the risk score and immune status, the enrichment scores of ssGSEA associated with functions or pathways were quanti ed, Cytolytic_activity, MHC_class_I, Type_I_IFN_Reponse and Type_II_IFN_Reponse were signi cantly different between the low risk and high risk group (Fig. 9D). Finally, ImmuneCellAI ( http://bioinfo.life.hust.edu.cn/web/ImmuCellAI/) was performed to predict immunotherapy response, Fig. 9E and Fig. 9F suggested that patients in the high-risk group might present with a better response for immunotherapy.

Building and validating a Nomogram to Predict OS
Univariate and Multivariate Cox regression analysis have revealed that TNM stage and gene signature were independent prognostic factor in training cohort and testing cohorts. Thus TNM stage and signature were used to build a nomogram to predict 1-year, 3-year, and 5-year OS in the TCGA cohort (Fig. 11A). The calibration curve analysis con rmed the reliability of the nomogram, which demonstrated that 1-, 3-, and 5-year survival prediction probabilities were closely related to the observed survival probability (Fig. 11B,C,D), and the C-index for the nomogram is 0.746342(0.708523-0.784161), which was more large than Signature(0.726819(0.680894-0.772744)) and Stage(0.664469(0.589597-0.73934))( Table 2)  3.13 Expression level of genes of risk score in clinical samples Finally, we compared expression level of genes in risk score between normal and cancer tissues using gene expression pro le in TCGA and ICGC, Six genes were signi cantly different between normal and cancer tissues in TCGA (Fig. 12A-F) and ICGC (Fig. 12G-L). Mrna expression level of G6PD, CCNB1, FLVCR1, ABCB6, RRM2 were signi cant higher in cancer tissues than normal tissues. However, Mrna expression level of CYP2C9 was signi cant lower in cancer tissue than normal tissues. In addition, The six genes were signi cant associated with overall survival (OS), disease-free survival (DFS), disease speci c survival (DSS) and progression free survival (PFS) by performing Kaplan-Meier(https://kmplot.com/analysis/). Furthermore, we performed qRT-PCR to detect the expression level in HCC tissues and adjacent samples (Fig. 12M-R). It revealed the same result, Mrna expression level of G6PD, CCNB1, FLVCR1, ABCB6, RRM2 were signi cant higher in cancer tissue than normal tissue, and Mrna expression level of CYP2C9 was signi cant lower in cancer tissue than normal tissues.

Discussion
In the study, we downloaded RNA sequencing, clinical information from TCGA and ICGC dataset. Next, we identi ed DEGs between patients with > 3 year OS and < 1 year OS by integrated bioinformatics analysis of iron metabolic genes in TCGA dataset. Then, we constructed and validated an iron metabolic signature in TCGA and ICGC dataset, which showed a strong prognostic performance for predicting the survival of HCC patients. In addition, the gene signature was an independent prognostic factor by conducting Multivariate Cox regression analysis. Meantime, WGCNA analysis, GSEA analysis, Go and KEGG enrichment analysis were performed to reveal molecular mechanisms, and cell cycle therapy which may be bene t to high-risk patients. And immunotherapy response prediction was analyzed and revealed immunotherapy may be signi cant for high risk group. Furthermore, two independent prognostic factors were combined to build Nomogram for predicting HCC prognosis. Finally, we analyzed the expression of the six genes using qRT-PCR in twelve HCC samples.
Iron plays a crucial role in biological processes in mammalian cells by functioning as iron-and haemcontaining enzymes, including the cellular stress response, mitochondrial metabolism, chromatin remodeling, DNA repair, mitochondrial metabolism, DNA replication. thus iron is an essential nutrient that facilitates cell proliferation and growth.3 Besides, iron is reported to be associated with tumor biological process, such as tumorigenesis, angiogenesis, invasion, and metastasis.4 Iron is unraveled to participate the formation of enzymes involved in DNA synthesis and the cell cycle, and it can regulate proteins that control cell cycle, Turner J et al 1 reported that a kind of iron chelator, tachpyridine, can induce G2 arrest and promote cancer cells to be sensitive to ionizing radiation, indicating iron chelators can be considered as radioenhancing agents for cancer treatment. Meantime, iron may promote tumour initiation via free radicals. Recently, proteins involved in iron metabolism are reported to contribute to cancer development, including iron e ux pumps, oxidases and reductases, iron regulators as well as siderophore-binding proteins.2 Ferroptosis is iron-dependent form of regulating cell death, which present different characteristics with apoptosis, necrosis. [9] Reactive oxygen species (ROS) may appear in mitochondria during iron-involving oxidative phosphorylation, and excessive ROS could induce oxidative stress response, which could damage large molecular substances such as nucleic acids, lipids and proteins. [21] Recently, an increasing number of studies has indicated that various kinds of diseases are associated with ferroptosis, including periventricular leukomalacia (PVL), Huntington's disease (HD) and renal functional damage. [22][23][24] In addition, many kinds of cancers have revealed the close relationships with ferroptosis, including liver cancer, osteosarcoma, prostate adenocarcinoma, cervical carcinoma and diffuse large B-cell lymphoma. [25,26]Ferroptosis is also con rmed to be associated with anti-tumor e cacy of chemotherapeutic drugs and immunotherapy. For advanced liver cancer, Sorafenib, a multikinase inhibitor, is the rst choice for treatment for 40% of newly diagnosed HCC patients. [27] which could induce and activate ferroptosis to play an anti-cancer role in HCC patients. The prognostic model contained 13 genes (ABCB6, CCNB1, FLVCR1, G6PD, IGSF3, OSBP2, PLOD2, PPAT,  RAP1GAP, RRM2, SLC7A11, CYP3A5 and CYP2C9), Six genes were signi cantly different between normal and cancer tissues by analyzing TCGA and ICGC dataset, qRT-PCR demonstrated the same trend, including G6PD, CCNB1, FLVCR1, ABCB6, RRM2 and CYP2C9. And survival analysis revealed the six genes were signi cantly associated OS,DFS, DSS, PFS using Kaplan-Meier. Thus we focused on the six genes. G6PD is a housekeeping enzyme expressed in most cells. Various studies suggested that G6PD de ciency may be protective against HCC in vitro and in vivo and patients with G6PD de ciency showed a 55% risk reduction for HCC compared with patients with normal G6PD activity.
[28] Aberrant activation of G6PD is linked to tumorigenesis and malignancy in rapidly growing cancer cells. Previous studies reported that silencing of PPP enzyme G6PD prevents erastin-induced ferroptosis, which reveals G6PD act a protective role by inducing ferroptosis. [9] Now it's known, RRM2 is one of the genes with high expression and poor prognosis in malignant tumors. During cell division, RR is essential for DNA synthesis. It encodes enzymes that catalyze the conversion of ribonucleotides to deoxyribonucleic acids. The activity of RRM2 subunit reductase requires two Fe3 + ions to form tyrosil radical on tyr122, and tyrosil radical plays a key role in the enzyme activity. [29]Animal experiments revealed that iron de ciency decreased the expression of RRM2 and reduced the protein abundance of RRM2. [30]And RRM2 may inhibit ferroptosis. [31] The expression level of CCNB1 was abnormally elevated in HCC and high expression is associated with better prognosis, as an oncogene, CCNB could inhibit the apoptosis of HCC cells and promote cell invasion. [32] FLVCR1 is a cell surface heme exporter, essential for erythropoiesis and systemic iron homeostasis. [33] The defect of FLVCR1 gene may lead to porphyrin gene overload, excess porphyrins could cause oxidative damage to the cell membrane components and eventually produce cell death. [34]Francesca Vinchi et al investigated the role of Flvcr1a in liver function in mice, the result revealed that FLVCR1 regulates heme synthesis and degradation as well as cytochromes P450 expression and activity. [35] Further study demonstrated that FLVCR1 mediates heme export from macrophages that ingest senescent red cells and regulates hepatic iron. Iron metabolism is disordered in tumors compared with normal tissues and the iron intake gene, FLVCR1, were the most relevant iron metabolism genes in Hepatocellular Carcinoma.
[36] ABCB6 encode a mitochondrial transporter. It localizes to both the outer mitochondrial membrane and the plasma membrane. Expression of ABCB6 has been associated with iron homeostasis, [37] and ABCB6 may play a role in heme metabolism.
Previous study has found ABCB6 can rescue the rapid growth phenotype of ATM1 deleted yeast cells, which means that ABCB6 acts as a backup system. [38] In HCC, prognostic analysis has indicated that high expression of ABCB6 was associated with worse prognosis. Because ABCB6 may inhibit the ferroptosis of liver cancer cell, and ferroptosis is considered to be a potential anticancer therapy. [39] Increased expression of cytochrome P450 CYP2C9, together with elevated level of its products epoxyeicosatrienoic acids(EET), is associated with aggressiveness in cancer, previous study revealed that high activity CYP2C9 genotype is associated with increased risk of colorectal cancer, [40] while a low activity CYP2C9 genotype (CYP2C9*2) is associated with increased risk of colorectal adenoma [41] and lung cancer [42] and low expression is associated with poor prognosis in HCC. [43] which is consistent with our study.
As a notable immunotherapy, immune checkpoint blockade have emerged as powerful tools in the management of multiple cancer types, and immune checkpoint drugs have already been FDA approved in renal cell carcinoma, melanoma, non-small cell lung cancer, Hodgkin lymphoma and urothelial bladder cancer, including pembrolizumab, ipilimumab and nivolumab. [44] Several trials investigating the e cacy of immune checkpoint blockades in HCC are in progress (NCT02837029, NCT02947165, NCT02572687, NCT02740985, NCT01853618, NCT02668770, NCT02239900, NCT02423343, NCT01658878). The treatment of anti-PD-1 antibody nivolumab in HCC demonstrated a signi cant and stable response. [45] Thus it is promising for HCC by applying immunotherapy. In our study, we found risk score has a strongly positive correlation with immune checkpoints, and the high-risk subtype had signi cantly higher immune