Construction and validation of a cell death-related genes prognosis signature of hepatocellular carcinoma

Introduction

Primary liver cancer is the sixth most common tumor and the fourth most deadly tumor in the world (1). Hepatocellular carcinoma (HCC) accounts for about 70–85% of primary liver cancer cases (2), with the main risk factors being hepatitis B virus infection, alcohol, aflatoxin, obesity, and diabetes (3,4). However, the overall prognosis of HCC is unsatisfactory due to its hidden onset, high postoperative recurrence and metastasis rate, and multidrug resistance (5-7). Despite the advances in early diagnosis, surgical treatment, and comprehensive treatment of HCC, the prognosis remains very poor (8). An important factor contributing to the poor prognosis of HCC is the lack of an accurate prognostic grading system and effective early screening methods. Therefore, studying the molecular mechanisms of HCC development provides a theoretical basis for identifying effective predictive markers and potential therapeutic targets, and holds great clinical significance for early diagnosis and treatment.

Classic cell death pathways include apoptosis, autophagy, and necroptosis. Numerous studies have reported that cell death is closely associated with tumor development, including HCC. Apoptosis is a highly controlled cell death process that maintains homeostasis in normal tissues (9). The occurrence and development of multiple malignant tumors have been associated with dysfunction and impairment of apoptosis. Myeloid cell leukemia 1 (Mcl-1) is an anti-apoptotic member of the B-cell lymphoma 2 (Bcl-2) family and is significantly upregulated in HCC tissues compared to normal tissues, suggesting that the inhibition of intracellular apoptosis facilitates the development of HCC (10). Autophagy is a process in which cells transport damaged organelles and macromolecules to lysosomes for digestion and degradation under the regulation of autophagy-related genes (ATGs) (11). Normally, autophagy acts as a defensive process against cancer by eliminating damaged organelles and stabilizing the cell genome. However, mutations in ATGs may lead to autophagy disorders and promote the occurrence and progression of tumors. Beclin1, a specific mammalian gene involved in autophagy, was found to be significantly downregulated in HCC tissues compared to adjacent normal tissues, and the expression of Beclin1 was negatively correlated with the pathological grade of HCC (12). Necroptosis is a passive cell death process caused by nonspecific or pathological stress (13). Some solid tumors such as HCC produce an ischemic and hypoxic environment due to the rapid growth of tumor cells, which promotes tumor cell and tissue necroptosis (14). A growing number of studies have demonstrated that tumor necroptosis involves the release of damage-associated molecules such as high mobility group box 1 (HMGB1) and heat shock protein 90 (HSP90) via direct or indirect pathways, thereby promoting tumor progression and metastasis (15,16). These results suggest that cell death is not only a pathological feature but also has predictive implications for disease progression and prognosis.

In this study, the relationship between the expression of cell death-related genes (CDRGs) and HCC prognosis was systematically investigated based on distinct cell death modes. Besides, our classification system provides a map showing disease progression and immune status in the difference of patients with different heterogeneity, guiding the selection of therapeutic strategies for HCC patients. We present this article in accordance with the TRIPOD reporting checklist (available at https://tcr.amegroups.com/article/view/10.21037/tcr-24-1315/rc).

Methods

Data processing

The data of 371 HCC patients from The Cancer Genome Atlas (TCGA) database (https://portal.gdc.cancer.gov/) was downloaded from the University of California, Santa Cruz (UCSC) Xena browser (http://xena.ucsc.edu/) as a validation cohort, which provided RNA expression profiles and corresponding clinical information. Furthermore, fragments per kilobase million (FPKM) values were converted to transcripts per million (TPM) values for subsequent analyses.

The training cohort consisted of samples from two public database platforms, Gene Expression Omnibus (GEO) (https://www.ncbi.nlm.nih.gov/geo/) and International Cancer Genome Consortium (ICGC) (https://dcc.icgc.org/releases), which included GSE14520 (n=242), GSE76427 (n=115), and LIRI-JP (n=230). The ComBat function of the “sva” package in R was used to eliminate batch effects between different datasets, and the three datasets were combined. Moreover, the overlapping genes among the three datasets were reserved for further analysis (n=8,418). Additionally, a total of 264 cytokine-related genes involved in the cytokine-cytokine receptor interaction pathway were obtained from a previous study (17). The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013).

Identification of gene set classifications based on biological functions

All the CDRGs were integrated from the Molecular Signatures Database (MSigDB) (http://www.gsea-msigdb.org/gsea/msigdb) using package ‘msigdbr’ in R. The C2 (curated gene sets), C5 (ontology gene sets), and H (hallmark gene sets) collections included genes whose functional descriptions were related to autophagy, apoptosis, and necroptosis, which were identified as CDRGs. Furthermore, all CDRGs were divided into four gene sets, namely the ATGs (n=393), apoptosis-related genes (n=3,493), necroptosis-related genes (n=430), and CDRGs (n=3,796, union of the first three gene sets).

Constructing prediction models based on different gene sets

First, univariate Cox regression analysis was conducted to screen candidate genes significantly related to the overall survival (OS) of patients with HCC (P<0.001). Then, Lasso Cox regression analysis was performed to determine prognosis-related genes using package ‘glmnet’ in R and enhance the adaptability of the prediction model. The risk score of each sample was calculated as follows: risk score = β₁ * Exp₁ + β₂ * Exp₂ + … + β_i * Exp_i, where β represents regression coefficients generated from the Lasso Cox regression analysis, and Exp represents the expression levels of CDRGs.

Evaluation and validation of the prediction model

Based on the median risk score, all samples in the training cohort were divided into the high- and low-risk groups. Subsequently, a Kaplan-Meier survival analysis was performed to compare the differences in survival time between the two groups. The receiver-operating characteristic (ROC) curves were plotted and the corresponding area under the curve (AUC) were calculated to estimate the accuracy of the prediction model. In addition, principal component analysis (PCA) was applied to assess the clustering ability of the prognostic model. Finally, the data of HCC patients from the TCGA database was used for validation.

The data of HCC patients were retrieved from the TCGA cohort to investigate the relationship between the prediction model and clinical characteristics. Then, the distribution of risk scores across different clinical characteristics was compared to understand the prognostic value of the predictive model.

Analysis of differences in the infiltration of 22 immune cell types

The cell-type identification by estimating relative subsets of RNA transcripts (CIBERSORT) online analytical platform (https://cibersort.stanford.edu/) was used to quantify the abundances of 22 immune cell types in a mixed cell population based on the leukocyte gene signature matrix (547 genes) (18). The differences in infiltration of the 22 immune cell types were compared between the high- and low-risk groups.

Identifying differentially expressed cytokine-related genes between high- and low-risk groups

Extensive evidence has demonstrated that chemokines play a crucial role in the occurrence and progression of HCC. Therefore, the differentially expressed genes (DEGs) between the high- and low-risk groups were explored, indicating the mechanism underlying the effects of cytokines on the development of HCC. The R package ‘DEseq2’ was applied to identify DEGs (|log2fold change| >1, P<0.01) between high- and low-risk groups.

Analysis of difference in the expression levels of immune checkpoint molecules between the high- and low-risk groups

Tumor cells can regulate the expression of immune checkpoint molecules to inhibit T cell antitumor immune response or enhance tumor immune escape, promoting tumor malignancy. In the present study, the Wilcoxon test was used to compare the differential expression of immune checkpoint molecules, including programmed cell death protein 1 (PD-1) and cytotoxic T-lymphocyte-associated protein 4 (CTLA-4), between the two groups.

Statistical analysis

All bioinformatics analyses were performed using R software (version 4.2.3). The Wilcoxon test and the Kruskal-Wallis test were used to compare the differences between two groups and three groups or more, respectively. A univariate Cox regression analysis was performed to identify prognostic genes, and the Lasso Cox regression analysis was employed to calculate the risk score. Subsequently, the Kaplan-Meier curve and Log-rank test were used to evaluate the difference in prognosis between the high- and low-risk groups. The ROC curve was plotted to estimate the prediction accuracy of the prognostic model. All reported P values were two-sided and P<0.05 was considered statistically significant.

Results

Identification of four gene set classifications

The study flow is shown in Figure 1. Based on respective biological processes, all CDRGs were classified into four gene sets: apoptosis-related genes (n=5,692), ATGs (n=598), necroptosis-related genes (n=708), and CDRGs (n=6,234, union of the first three gene sets).

Figure 1 Study flow chart. GEO, Gene Expression Omnibus; ICGC, International Cancer Genome Consortium; ROC, receiver operating characteristic; PCA, principal component analysis; TCGA, The Cancer Genome Atlas; DEG, differentially expressed gene.

Construction and analysis of the apoptosis-based prediction model

A total of 17 apoptosis-related genes (Table S1) were identified by the univariate (Figure S1A) and Lasso Cox regression analyses (Figure S1B) and were integrated into the prediction model. To assess the predictive ability of this model, 587 HCC samples (training cohort) were categorized into the high- (n=293) and low-risk (n=294) groups according to the median risk score. Similarly, 363 HCC samples (validation cohort) were split into two the high-risk score group (n=181) and the low-risk score group (n=182).

The Kaplan-Meier survival analysis revealed poorer survival in the high-risk groups compared to the low-risk groups (Figure 2A,2B). To evaluate the model reliability, ROC curves were developed using the data obtained from the training and validation cohorts. The AUC values confirmed that the apoptosis-based model had strong predictive power in assessing the 1-, 2-, and 3-year patient survival (training cohort: 0.737, 1-year OS; 0.749, 2-year OS; 0.740, 3-year OS; validation cohort: 0.691, 1-year OS; 0.669, 2-year OS; 0.675, 3-year OS) (Figure 2C,2D). Besides, PCA showed that patients in different risk groups were distributed in two separate clusters (Figure 2E,2F). These results indicated the apoptosis-based prediction model had good predictive capability.

Figure 2 Estimation and validation of the prognostic model based on apoptosis-related genes. (A,B) Kaplan-Meier survival curve and log-rank test. (C,D) ROC curve and corresponding AUC value. (E,F) PCA analysis. (G) The relationship between risk score and clinicopathological characteristics in TCGA. AUC, area under the curve; ROC, receiver operating characteristic; PCA, principal component analysis; TCGA, The Cancer Genome Atlas.

Finally, the association between the risk score and clinical characteristics was analyzed. The risk score was positively correlated with the tumor, node, metastasis (TNM) stage, indicating that the prediction model can reveal the development of HCC (Figure 2G).

Construction and analysis of autophagy-based prediction model

Based on the univariate (Figure S1C) and Lasso (Figure S1D) regression analyses, 34 ATGs were identified (Table S2) and used to construct the prognostic model. Moreover, 587 HCC samples (training cohort) and 363 HCC samples (validation cohort) were classified into two groups based on the risk score. The survival analysis uncovered that patients with a high-risk score had a significantly shorter survival time than those with a low-risk score (P<0.001 and P=0.03) (Figure 3A,3B). The AUC values were 0.783 (training cohort, 1-year OS), 0.798 (training cohort, 2-year OS), 0.777 (training cohort, 3-year OS), 0.651 (validation cohort, 1-year OS), 0.630 (validation cohort, 2-year OS), and 0.626 (validation cohort, 3-year OS) (Figure 3C,3D). PCA illustrated that patients were grouped into two major clusters with few overlapping regions (Figure 3E,3F).

Figure 3 Estimation and validation of prognostic model based on autophagy-related genes. (A,B) Kaplan-Meier survival curve and log-rank test. (C,D) ROC curve and corresponding AUC value. (E,F) PCA analysis. (G) The relationship between risk score and clinicopathological characteristics in TCGA. AUC, area under the curve; ROC, receiver operating characteristic; PCA, principal component analysis; TCGA, The Cancer Genome Atlas.

Additionally, the autophagy-based risk score was confirmed to be significantly correlated with HCC progression (Figure 3G).

Construction and analysis of necroptosis-based prediction model

According to the same analysis strategy, 22 necroptosis-related genes (Table S3) were used to establish the predictive model (Figure S1E,S1F). Additionally, 587 HCC samples (training cohort) and 363 HCC samples (validation cohort) were divided into two groups based on the risk score. The survival analysis indicated that a high-risk score was significantly associated with an unfavorable prognosis (P<0.001 and P=0.008) (Figure 4A,4B). The AUC was calculated to represent the accuracy of the prediction, showing AUCs of 0.765, 0.782, and 0.775 for the 1-, 2-, and 3-year OS, respectively, in the training cohort. In the validation cohort, the AUC for the 1-, 2-, and 3-year OS were 0.687, 0.635, and 0.627, respectively (Figure 4C,4D). In the PCA results, patients from different groups were placed into different clusters with few overlapping regions (Figure 4E,4F).

Figure 4 Estimation and validation of prognostic model based on necroptosis-related genes. (A,B) Kaplan-Meier survival curve and log-rank test. (C,D) ROC curve and corresponding AUC value. (E,F) PCA analysis. (G) The relationship between risk score and clinicopathological characteristics in TCGA. AUC, area under the curve; ROC, receiver operating characteristic; PCA, principal component analysis; TCGA, The Cancer Genome Atlas.

Moreover, the necroptosis-based risk signature was significantly correlated with the development of HCC (Figure 4G).

Construction and analysis of cell death-based prediction model

Likewise, a total of 6,234 CDRGs were subjected to univariate and Lasso Cox regression analyses. Interestingly, the cell death-based prediction model was the same as the model using apoptosis-related genes. Therefore, the predictive power of this model was consistent with the prognostic performance of the apoptosis-based model.

Analysis of difference in the immune status between the high- and low-risk groups

Based on the above predictive models, a differential gene expression analysis was performed between the high- and low-risk groups. Furthermore, the DEGs included numerous chemokine-related genes, with C-X-C motif chemokine ligand 5 (CXCL5), colony-stimulating factor 3 receptor (CSF3R), and bone morphogenetic protein 7 (BMP7) being downregulated, and interleukin 13 (IL13) being upregulated (Figure 5A-5C).

Figure 5 Investigating the chemokines and immune cell infiltration in HCC microenvironment based on three prognostic models. (A-C) Identification of chemokine-related DEGs between high- and low-risk HCC patients. The red dots represent upregulated genes, while the blue dots represent downregulated genes. (D-F) The differential immune infiltration of 22 immune cell types between high- and low-risk HCC patients. The symbols have the following meanings: “*” indicates P<0.05, “**” indicates P<0.01, “***” indicates P<0.001, and “ns” indicates not significant. HCC, hepatocellular carcinoma; DEG, differentially expressed gene.

Additionally, a total of seven types of immune cells showed significantly different levels between the high- and low-risk groups, which included naive B cells, memory B cells, CD8 T cells, resting memory CD4 T cells, M0 macrophages, M1 macrophages, and M2 macrophages (Figure 5D-5F). Importantly, the three predictive models were consistent in assessing the infiltration levels of CD8 T and M2 macrophages; patients in the high-risk group exhibited low levels of CD8 T cell infiltration and high levels of M2 macrophage infiltration.

The difference in the expression levels of immune checkpoint molecules between the high- and low-risk groups

The expression levels of programmed death-ligand 1 (PD-L1) (Figure 6A) were significantly higher in the low-risk groups, while the expression levels of CTLA-4 (Figure 6B) were significantly higher in the high-risk groups, indicating that changes in immune checkpoints are associated with the prognosis of patients with HCC.

Figure 6 Exploring the differential expression of PD-L1 (A) and CTLA-4 (B) in high- and low-risk HCC patients based on the prognostic model. PD-L1, programmed death-ligand 1; CTLA-4, cytotoxic T-lymphocyte-associated protein 4; HCC, hepatocellular carcinoma.

Discussion

HCC is a malignant tumor with high heterogeneity, which arises from multiple genetic mutations and epigenetic changes (19). Despite major breakthroughs in the diagnosis and treatment of HCC, the prognosis of HCC patients remains unsatisfactory. This is mainly due to the low early diagnosis rate of HCC, with the majority of the detected HCC cases being too advanced for radical surgical resection. Therefore, the molecular mechanisms of HCC should be urgently studied to identify biomarkers for disease diagnosis and treatment.

The term “cell death” includes two types of death, namely regulatory cell death (RCD) and accidental cell death (ACD) (20). In contrast to ACD, which is an uncontrolled and passive process, regulated cell death is orchestrated through a range of molecular mechanisms and signaling pathways (21). An increasing body of evidence suggests that the dysregulation of regulated cell death is closely associated with the occurrence of various diseases, particularly tumors. Furthermore, targeting cell death pathways holds promise for improving the effectiveness of cancer treatments (22). Cell death evasion is a distinct characteristic of HCC that supports the growth, progression, and resistance to therapy of liver cancer (23). Therefore, elucidating the programmed cell death (PCD) pathways in HCC will provide new insights and strategies for anticancer treatments.

Apoptosis is the most common form of regulated cell death, mediated by the sequential activation of a series of caspases (24). Apoptosis is spontaneous, controlled by genes, and occurs in an orderly manner (25). Moreover, tumorigenesis is closely related to dysfunctional cellular apoptosis. A substantial number of studies revealed that the induction of apoptosis is an important target for cancer therapy. For example, secretory leukocyte protease inhibitor (SLPI) inhibits HCC progression via endoplasmic reticulum stress-induced apoptosis (26). Furthermore, Ji et al. found that elevated miR-486-3p levels can reverse sorafenib resistance in sorafenib-resistant HCC cells by inducing apoptosis (27). Recently, several studies have focused on the potential cytotoxic effects of ropivacaine on HCC cells (28,29). Wang et al. revealed that ropivacaine promotes the apoptosis of HCC cells in a dose- and time-dependent manner by activating caspase-3 activity (30). These results suggested that targeting apoptosis could effectively improve the therapeutic effect on HCC.

Autophagy is an important degradation process of cellular contents, allowing the recirculation of structural cellular components and improved survival. Autophagy-dependent cell death is a rare type of PCD. During autophagy, proteins and organelles are degraded through the lysosomal pathway (31). Numerous ATGs are involved in autophagy, including microtubule-associated protein 1A/1B-light chain 3 (LC3), Beclin1, and p62 (32). Tumorigenesis, tumor development, and tumor treatment are all greatly influenced by autophagy. On the one hand, autophagy plays a critical role in tumor suppression by preserving genomic integrity and preventing proliferation (33). On the other hand, autophagy promotes tumor growth by providing energy substrates that are needed for cellular proliferation (34). Growing evidence suggests that autophagy suppresses growth in the early stage of tumor progression, while autophagy promotes survival during the later stages (35). Takamura et al. reported that the impairment of autophagy leads to the accumulation of p62 and contributes to the development of HCC (36). Additionally, the autophagic protein Beclin1 was found to be associated with HCC tumors (37). One study has confirmed that inducing or suppressing autophagy through the mammalian target of rapamycin (mTOR) pathway significantly impacts the development of HCC (38). Therefore, autophagy modulation represents a potential option for anti-HCC therapy.

Necroptosis is a regulated necrotic cell death modality, which is mainly mediated by receptor-interacting protein 1 (RIP1), receptor-interacting protein 3 (RIP3), and mixed lineage kinase domain-like (MLKL) (39). In tumorigenesis and development, necroptosis plays a dual role as it both protects against tumor development (40) and promotes cancer metastasis and progression (41). HCC, which results from chronic liver disease, arises almost exclusively from chronic hepatic inflammation (42). A recent study revealed for the first time that necroptosis contributes to chronic inflammation in the liver, leading to liver fibrosis and liver cancer (43). Necroptosis, as a regulated cell death pathway, is strongly associated with increased inflammation through the release of damage-associated molecular patterns (DAMPs) from necroptotic cells (44). Furthermore, DAMPs are associated with acute liver injury and chronic liver disease (45). Additionally, Lin et al. found that Tan IIA induces necroptosis in HepG2 cells (46). However, tumor progression has been suggested to resist necroptosis Therefore, targeting necroptosis might provide novel therapeutic biomarkers for HCC.

In this study, HCC patients with high-risk scores were accompanied by higher M2-type macrophage infiltration and lower CD8 T cell infiltration levels, which was consistent with previous studies. Numerous studies have shown that HCC progression is significantly associated with the development of immunosuppressive tumor microenvironment (TME). In addition, M2-type macrophages produce immunosuppressive factors, such as transforming growth factor beta (TGF-β) and interleukin 10 (IL-10), to support other immunosuppressive cells (47). M2-type macrophages could facilitate the development of epithelial-mesenchymal transformation and migration of HCC (48). Furthermore, long-term suppressive TME drives the CD8 T cells into a functionally impaired state called T cell exhaustion (49). These exhausted CD8 T cells express high amounts of inhibitory receptors (50). Therefore, the importance of chemokines and chemokine receptors cannot be overlooked in tumor immunity. For example, Yang et al. reported that the upregulation of C-X-C chemokine receptor type 4 (CXCR4) expression contributes to tumorigenesis and metastasis of HCC (51). Hence, a deeper understanding of the molecular mechanism of immune cells and chemokines in the TME may advance the development of immunotherapy for HCC.

In general, our conclusions are based on retrospective analysis of publicly available datasets, which, as acknowledged, can introduce inherent biases such as patient selection and variability in data quality. Although the results were validated using the TCGA cohort, these findings still require confirmation in independent prospective cohorts to ensure robustness and reduce potential retrospective biases. Additionally, while we identified key death-related genes in HCC patients, their functional roles and predictive abilities remain speculative in the absence of direct experimental validation. To fully elucidate their roles, further in vivo and in vitro experiments are essential. This will help confirm the biological relevance of these genes and their contribution to HCC progression and patient prognosis. Moreover, the expression levels of these genes in biological specimens from clinical patients need to be validated, as such confirmation would strengthen the translational potential of our findings. Finally, the molecular mechanisms underlying how our prognostic model influences HCC remain to be explored. Future studies should focus on identifying these mechanisms through experimental investigation, which could lead to the development of novel therapeutic targets and strategies. Such efforts may enhance the efficacy of systemic treatments and ultimately improve survival outcomes for HCC patients.

Conclusions

In this paper, the potential value of CDRGs in the early diagnosis and treatment of HCC was analyzed with bioinformatics. Various bioinformatics analysis methods were used to evaluate the correlation between the risk score and prognosis, immune cell infiltration, the expression level of immune checkpoint molecules, and chemokine-related genes. Moreover, we plan to conduct further experiments to determine the precise molecular function of CDRGs in HCC, improving the treatment efficacy.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the TRIPOD reporting checklist. Available at https://tcr.amegroups.com/article/view/10.21037/tcr-24-1315/rc

Peer Review File: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-24-1315/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tcr.amegroups.com/article/view/10.21037/tcr-24-1315/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013).

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Villanueva A. Hepatocellular Carcinoma. N Engl J Med 2019;380:1450-62. [Crossref] [PubMed]
2. Jemal A, Bray F, Center MM, et al. Global cancer statistics. CA Cancer J Clin 2011;61:69-90. [Crossref] [PubMed]
3. Yang JD, Hainaut P, Gores GJ, et al. A global view of hepatocellular carcinoma: trends, risk, prevention and management. Nat Rev Gastroenterol Hepatol 2019;16:589-604. [Crossref] [PubMed]
4. Fan JY, Huang TJ, Jane SW, et al. Prevention of Liver Cancer Through the Early Detection of Risk-related Behavior Among Hepatitis B or C Carriers. Cancer Nurs 2015;38:169-76. [Crossref] [PubMed]
5. Zhou Y, Wang Y, Zhou W, et al. YAP promotes multi-drug resistance and inhibits autophagy-related cell death in hepatocellular carcinoma via the RAC1-ROS-mTOR pathway. Cancer Cell Int 2019;19:179. [Crossref] [PubMed]
6. Wu H, Zhang G, Li Z, et al. Thrombospondin-4 expression as a prognostic marker in hepatocellular carcinoma. Gene 2019;696:219-24. [Crossref] [PubMed]
7. Sia D, Villanueva A, Friedman SL, et al. Liver Cancer Cell of Origin, Molecular Class, and Effects on Patient Prognosis. Gastroenterology 2017;152:745-61. [Crossref] [PubMed]
8. Wada H, Yamamoto H, Kim C, et al. Association between ephrin-A1 mRNA expression and poor prognosis after hepatectomy to treat hepatocellular carcinoma. Int J Oncol 2014;45:1051-8. [Crossref] [PubMed]
9. Chen Q, Kang J, Fu C. The independence of and associations among apoptosis, autophagy, and necrosis. Signal Transduct Target Ther 2018;3:18. [Crossref] [PubMed]
10. Fleischer B, Schulze-Bergkamen H, Schuchmann M, et al. Mcl-1 is an anti-apoptotic factor for human hepatocellular carcinoma. Int J Oncol 2006;28:25-32.
11. Levine B, Kroemer G. Biological Functions of Autophagy Genes: A Disease Perspective. Cell 2019;176:11-42. [Crossref] [PubMed]
12. Sun H, Yu J, Wen Z, et al. Decreased expression of Beclin-1 in patients with hepatocellular carcinoma. J BUON 2019;24:634-41.
13. Wang G, Qu FZ, Li L, et al. Necroptosis: a potential, promising target and switch in acute pancreatitis. Apoptosis 2016;21:121-9. [Crossref] [PubMed]
14. Chandler KB, Costello CE, Rahimi N. Glycosylation in the Tumor Microenvironment: Implications for Tumor Angiogenesis and Metastasis. Cells 2019;8:544. [Crossref] [PubMed]
15. Jacobsen AV, Lowes KN, Tanzer MC, et al. HSP90 activity is required for MLKL oligomerisation and membrane translocation and the induction of necroptotic cell death. Cell Death Dis 2016;7:e2051. [Crossref] [PubMed]
16. Vakkila J, Lotze MT. Inflammation and necrosis promote tumour growth. Nat Rev Immunol 2004;4:641-8. [Crossref] [PubMed]
17. Ahluwalia P, Ahluwalia M, Mondal AK, et al. Immunogenomic Gene Signature of Cell-Death Associated Genes with Prognostic Implications in Lung Cancer. Cancers (Basel) 2021;13:155. [Crossref] [PubMed]
18. Newman AM, Liu CL, Green MR, et al. Robust enumeration of cell subsets from tissue expression profiles. Nat Methods 2015;12:453-7. [Crossref] [PubMed]
19. Park IY, Sohn BH, Yu E, et al. Aberrant epigenetic modifications in hepatocarcinogenesis induced by hepatitis B virus X protein. Gastroenterology 2007;132:1476-94. [Crossref] [PubMed]
20. Jia Y, Wang X, Deng Y, et al. Pyroptosis Provides New Strategies for the Treatment of Cancer. J Cancer 2023;14:140-51. [Crossref] [PubMed]
21. Tang D, Kang R, Berghe TV, et al. The molecular machinery of regulated cell death. Cell Res 2019;29:347-64. [Crossref] [PubMed]
22. Sun Y, Peng ZL. Programmed cell death and cancer. Postgrad Med J 2009;85:134-40. [Crossref] [PubMed]
23. Ajoolabady A, Tang D, Kroemer G, et al. Ferroptosis in hepatocellular carcinoma: mechanisms and targeted therapy. Br J Cancer 2023;128:190-205. [Crossref] [PubMed]
24. Kouroumalis E, Tsomidis I, Voumvouraki A. Pathogenesis of Hepatocellular Carcinoma: The Interplay of Apoptosis and Autophagy. Biomedicines 2023;11:1166. [Crossref] [PubMed]
25. Wang B, Han D, Li F, et al. Elevated IL-22 in psoriasis plays an anti-apoptotic role in keratinocytes through mediating Bcl-xL/Bax. Apoptosis 2020;25:663-73. [Crossref] [PubMed]
26. Sun J, Li J, Wu Z, et al. SLPI suppresses hepatocellular carcinoma progression via endoplasmic reticulum stress induced apoptosis. Int J Biol Sci 2022;18:140-53. [Crossref] [PubMed]
27. Ji L, Lin Z, Wan Z, et al. miR-486-3p mediates hepatocellular carcinoma sorafenib resistance by targeting FGFR4 and EGFR. Cell Death Dis 2020;11:250. [Crossref] [PubMed]
28. Jurj A, Tomuleasa C, Tat TT, et al. Antiproliferative and Apoptotic Effects of Lidocaine on Human Hepatocarcinoma Cells. A preliminary study. J Gastrointestin Liver Dis 2017;26:45-50. [Crossref] [PubMed]
29. Xing W, Chen DT, Pan JH, et al. Lidocaine Induces Apoptosis and Suppresses Tumor Growth in Human Hepatocellular Carcinoma Cells In Vitro and in a Xenograft Model In Vivo. Anesthesiology 2017;126:868-81. [Crossref] [PubMed]
30. Wang W, Zhu M, Xu Z, et al. Ropivacaine promotes apoptosis of hepatocellular carcinoma cells through damaging mitochondria and activating caspase-3 activity. Biol Res 2019;52:36. [Crossref] [PubMed]
31. Khalil H, Abd ElHady A, Elawdan KA, et al. The Mechanical Autophagy as a Part of Cellular Immunity; Facts and Features in Treating the Medical Disorders. Immunol Invest 2022;51:266-89. [Crossref] [PubMed]
32. Nishimura T, Tooze SA. Emerging roles of ATG proteins and membrane lipids in autophagosome formation. Cell Discov 2020;6:32. [Crossref] [PubMed]
33. Cheng Y, Ren X, Hait WN, et al. Therapeutic targeting of autophagy in disease: biology and pharmacology. Pharmacol Rev 2013;65:1162-97. [Crossref] [PubMed]
34. Casimiro MC, Di Sante G, Di Rocco A, et al. Cyclin D1 Restrains Oncogene-Induced Autophagy by Regulating the AMPK-LKB1 Signaling Axis. Cancer Res 2017;77:3391-405. [Crossref] [PubMed]
35. Lee YJ, Jang BK. The Role of Autophagy in Hepatocellular Carcinoma. Int J Mol Sci 2015;16:26629-43. [Crossref] [PubMed]
36. Takamura A, Komatsu M, Hara T, et al. Autophagy-deficient mice develop multiple liver tumors. Genes Dev 2011;25:795-800. [Crossref] [PubMed]
37. Yazdani HO, Huang H, Tsung A. Autophagy: Dual Response in the Development of Hepatocellular Carcinoma. Cells 2019;8:91. [Crossref] [PubMed]
38. Yang J, Pi C, Wang G. Inhibition of PI3K/Akt/mTOR pathway by apigenin induces apoptosis and autophagy in hepatocellular carcinoma cells. Biomed Pharmacother 2018;103:699-707. [Crossref] [PubMed]
39. Gong Y, Fan Z, Luo G, et al. The role of necroptosis in cancer biology and therapy. Mol Cancer 2019;18:100. [Crossref] [PubMed]
40. Höckendorf U, Yabal M, Herold T, et al. RIPK3 Restricts Myeloid Leukemogenesis by Promoting Cell Death and Differentiation of Leukemia Initiating Cells. Cancer Cell 2016;30:75-91. [Crossref] [PubMed]
41. Strilic B, Yang L, Albarrán-Juárez J, et al. Tumour-cell-induced endothelial cell necroptosis via death receptor 6 promotes metastasis. Nature 2016;536:215-8. [Crossref] [PubMed]
42. Leone V, Ali A, Weber A, et al. Liver Inflammation and Hepatobiliary Cancers. Trends Cancer 2021;7:606-23. [Crossref] [PubMed]
43. Mohammed S, Thadathil N, Selvarani R, et al. Necroptosis contributes to chronic inflammation and fibrosis in aging liver. Aging Cell 2021;20:e13512. [Crossref] [PubMed]
44. Newton K, Manning G. Necroptosis and Inflammation. Annu Rev Biochem 2016;85:743-63. [Crossref] [PubMed]
45. Wen S, Li X, Ling Y, et al. HMGB1-associated necroptosis and Kupffer cells M1 polarization underlies remote liver injury induced by intestinal ischemia/reperfusion in rats. FASEB J 2020;34:4384-402. [Crossref] [PubMed]
46. Lin CY, Chang TW, Hsieh WH, et al. Simultaneous induction of apoptosis and necroptosis by Tanshinone IIA in human hepatocellular carcinoma HepG2 cells. Cell Death Discov 2016;2:16065. [Crossref] [PubMed]
47. Han S, Bao X, Zou Y, et al. d-lactate modulates M2 tumor-associated macrophages and remodels immunosuppressive tumor microenvironment for hepatocellular carcinoma. Sci Adv 2023;9:eadg2697. [Crossref] [PubMed]
48. Jiang Y, Han Q, Zhao H, et al. Promotion of epithelial-mesenchymal transformation by hepatocellular carcinoma-educated macrophages through Wnt2b/β-catenin/c-Myc signaling and reprogramming glycolysis. J Exp Clin Cancer Res 2021;40:13. [Crossref] [PubMed]
49. Wherry EJ, Kurachi M. Molecular and cellular insights into T cell exhaustion. Nat Rev Immunol 2015;15:486-99. [Crossref] [PubMed]
50. Wang X, He Q, Shen H, et al. TOX promotes the exhaustion of antitumor CD8(+) T cells by preventing PD1 degradation in hepatocellular carcinoma. J Hepatol 2019;71:731-41. [Crossref] [PubMed]
51. Yang J, Zhang L, Jiang Z, et al. TCF12 promotes the tumorigenesis and metastasis of hepatocellular carcinoma via upregulation of CXCR4 expression. Theranostics 2019;9:5810-27. [Crossref] [PubMed]