Identification and Validation of a Necroptosis-related Gene based Prognostic Signature for Patients with Hepatocellular Carcinoma

doi:10.21203/rs.3.rs-1601564/v1

Download PDF

Research Article

Identification and Validation of a Necroptosis-related Gene based Prognostic Signature for Patients with Hepatocellular Carcinoma

https://doi.org/10.21203/rs.3.rs-1601564/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Hepatocellular carcinoma (HCC) is the second leading cause of cancer-related death worldwide, nevertheless, its prognostic prediction remains obscure. Necroptosis is a tightly regulated form of cell death, which has been recently reported to be involved in HCC. However, the prognostic value of necroptosis-related genes (NRGs) in HCC has not been reported. In the present study, 159 NRGs of HCC subjects from the Cancer Genome Atlas (TCGA) cohort were downloaded and investigated. Of which, a total of thirty-six NRGs were identified to be significantly differentially expressed. Through univariate and multivariate Cox proportional regression analysis, four NRGs (HSP90AA1, PPIA, SQSTM1 and USP21) were found to have the excellent prognostic capacity for HCC, which could clearly divide patients into high-risk and low-risk groups. Survival analysis showed that the overall survival of high-risk patients was significantly shorter than that of low-risk patients. The independent prognostic value of necroptosis-related features was further confirmed by Cox regression analysis, and the area under the receiver operating characteristic curve of the panel comprising HSP90AA1, PPIA, SQSTM1 and USP21 was 0.807. Furthermore, the prognostic value of these NRGs was verified in another cohort of HCC patients from the ICGC database. Herein, this study shows that a novel necroptosis-related gene signature could be developed for prognostic prediction of HCC in future, which may also provide insights into the prevention and treatment of this disease clinically.

Hepatocellular carcinoma

Necroptosis

Prognostic Signature

Bioinformatics

Hepatocellular carcinoma (HCC) is globally one of the leading causes of cancer related death (Yang and Heimbach 2020). 90% of HCC cases originate from cirrhosis, during which hepatocytes undergo a chronic cycle of necrosis and regeneration (Kim and Viatour 2020). The treatment of HCC has greatly developed in recent years, unfortunately, the prognosis prediction seems stagnated (Siegel et al. 2018). The current prediction criteria for HCC prognosis have several disadvantages, such as aggressiveness. Therefore, it is highly necessary to establish an effective prognostic model based on the molecular heterogeneity of HCC (Forner et al. 2018; Agarwal et al. 2017).

Necroptosis is a novel form of cell death, which is mainly mediated by the RIPK1-RIPK3-MLKL signal axis (Christofferson and Yuan 2010). Accumulating evidence has suggested that necroptosis plays a key role in regulating cancer biology, including tumorigenesis, cancer metastasis, cancer immunity, and cancer subtypes (Gong et al. 2019; Seehawer et al. 2018). According to previous reports, necroptosis could serve as an alternative mode of programmed cell death to overcome apoptosis resistance, whichmay trigger and enhance antitumor immunity in cancer therapy (Gong et al. 2019). Increasing studies have suggested that necroptosis regulators could be used as biomarkers for the prognosis prediction of cancer (Park et al. 2020; Zhang et al. 2018). In a recent study, low-level expression of necroptosis factors was demonstrated to indicate poor prognosis in the non-small cell lung squamous cell carcinoma subtype (Lim et al. 2021). Recent studies have confirmed that necroptosis related genes can play a potential prognostic role in HCC (Nicole L et al. 2022). However, whether the necroptosis-related genes (NRGs) can be developed as biomarkers for predicting the HCC prognosis remains to be investigated.

In this study, we systematically analyzed the differentially expressed NRGs in HCC in TCGA database, and the differentially expressed NRGs that were significantly associated with OS were selected to construct a prognosis related NRGs signature. According to the cutoff value of risk score, HCC patients can be divided into high-risk and low-risk groups. Finally, we explored the prognostic role of the risk score in the TCGA database and in a validation dataset from the ICGC database. Collectively, this study may suggest a novel NRGs based biomarkers for the prognosis prediction of HCC clinically. The study design workflow is presented in Fig. 1.

Human necroptosis-related genes (NRGs) set

Based on the Kyoto Encyclopedia of genes and genomes (KEGG) database (https://www.kegg.jp/), A total of 159 genes were involved in the necroptosis signal pathway (Supplementary Table1).

Data retrieval

From the TCGA database (https://portal.gdc.cancer.gov/), RNA sequencing (RNA-seq) data and corresponding clinical information of 374 HCC patients were extracted. As a validation cohort, RNA-seq and clinical parameters of 231 HCC patients were downloaded from the ICGC database (https://dcc.icgc.org/projects/LIRI-JP). The correlation between the NRGs and the survival of patients was established by univariate cox regression analysis.

Bioinformatics analysis

A Consensus clustering analysis and a principal components analysis were carried out by the R language (version 4.1) to verify the regulatory role of necroptosis in HCC. The R package limma was used to screen the differentially expressed NRGs. Univariate Cox proportional hazard regression analysis was used to evaluate the association between total survival time (OS) and gene expression values. The relationship between independent factors and risk characteristics were determined by the multivariate Cox proportional hazards regression analyses. The regression coefficient was calculated by the Cox regression model.

Construction of prognostic model based on NRGs

Multivariate cox regression was used to construct prognostic genes. After combining the expression value of each specific gene, the risk score formula of each patient was constructed and weighted by its estimated regression coefficient in multivariate cox regression analysis. According to the risk scoring formula, taking the median risk score as the dividing point, the patients were divided into low-risk and high-risk groups. Survival differences between the two groups were assessed by Kaplan-Meier method and compared using log-rank statistical methods. Multivariate Cox regression analysis and stratified analysis were used to test the role of risk score in predicting the prognosis of patients. The area under the ROC curve was calculated for each dataset to measure the prognostic role of the model.

Statistical analysis

Survival curves were created by the Kaplan-Meier method and compared by log-rank test. Univariate and multivariate analysis using cox proportional hazard model. All statistical analyses were performed using the R language. All statistical tests with p<0.05 were considered significant.

Identification of differential NRGs

RNA-seq and clinical data from 374 HCC and 50 healthy controls were downloaded from TCGA database. Expression values of 159 NRGs from HCC patients were extracted and compared. Of which, two genes were downregulated in the subjects with HCC such as CAMK2G and IL1B, meanwhile, thirty-four ones were upregulated, including USP21, HSP90AA1,PPIA,SQSTM1 et al (Fig. 2a, c). Heatmap revealed the expression patterns of these differentially expressed genes (Fig. 2b).

Functional enrichment analysis

The functions of the thirty-six differential genes were investigated by pathway enrichment analysis at the GO and KEGG levels. At the biological process of GO, these genes were involved in I-κB kinase/NF-κB and cytokine-mediated signaling pathway. At the cellular components, secretory granule lumen, cytoplasmic vesicle lumen and vesicle lumen were enriched. At the molecular function of GO, ubiquitin protein ligase binding and ubiquitin-like protein ligase binding were changed (Fig. 3a). Subsequently, in KEGG pathway enrichment analysis, these genes were demonstrated to be significantly associated with Necroptosis and NOD − like receptor signaling pathway (Fig. 3b). Most enriched pathways had z-scores greater than zero, indicating that most pathways were likely enriched (Supplementary Figure S1).

Identification of prognosis-related NRGs and construction of prognosis prediction model

A univariate Cox regression approach was first employed to identify prognostically relevant NRGs. Prognostic NRGs that were statistically significant in univariate analysis were selected for inclusion in subsequent multivariate analysis. The results of multivariate analysis showed that a total of four genes were significantly associated with prognosis. A prognostic model was constructed based on the expression coefficients of each independent risk gene derived from the results of multivariate Cox analysis (Table 1. and Fig. 4a). The results showed the risk score of patients, in which below the median are low risk group (n = 172) while above the median are high risk group (n = 171). (Fig. 4b). The scatter plot represents alive in blue and dead in red, showing the mortality rate increased with the increase of patient risk (Fig. 4c). The K-M survival curve showed the survival status of low risk group in blue and high risk group in red, which indicated the predicted survival time of low-risk group was significantly longer than that of high-risk group (Fig. 4d, P < 0.001). We used the ROC curve for OS to assess the ability of the risk score to predict survival. The AUC value of the prognostic model was 0.807, which was significantly higher than that associated with age, gender, grade, stage, T stage, M stage, and N stage (Fig. 4e). These results suggest that the risk signature has a better ability to predict the survival of HCC patients than these clinical factors.

Table 1

Multivariate Cox regression analysis of necroptosis-related genes that can significantly affect OS.
id	coef	HR	HR.95L	HR.95H	P value
HSP90AA1	0.0029	1.0029	1.0006	1.0053	0.0121
PPIA	0.0086	1.0087	1.0028	1.0146	0.0038
SQSTM1	0.0025	1.0025	1.0013	1.0037	0.0001
USP21	0.0975	1.1024	1.0318	1.1779	0.00398

Abbreviations: HR, Hazardous Ratio.

The relationships between clinicopathological parameters and prognosis-related NRGs

Clinicopathological analysis to explore the association between clinical parameters and risk signature. The results showed that the signature was correlated with tumor stage (P = 0.002), T stage (P = 0.002), M stage (P = 0.014) and survival outcome (P = 0.007) (Fig. 5a-d). Furthermore, student's t-test analysis revealed that some signature related genes were similarly differentially expressed in different clinicopathological parameters (Supplementary Figure S2). To validate the independent predictive value of necroptosis-related features for OS, we performed univariate Cox regression analysis and multivariate Cox regression analysis. Univariate Cox analysis found that necroptosis-related features, tumor stage, T stages and M stages were all associated with survival of HCC patients (Fig. 5e). Further performed multivariate Cox analysis, the results demonstrated the predictive function of necroptosis-related features in prognosis (Fig. 5f). Therefore, our results confirm that necroptosis-related features can serve as independent prognostic factors in clinical practice.

Validation of the necroptosis-related features by an independent cohort

We calculated the risk score for each patient in the ICGA dataset using the same formula as an independent external validation cohort. The patients were divided into high-risk group and low-risk group according to the median risk score. OS in high-risk patients was shorter than that in low-risk patients as expected (Fig. 6a). The ROCs, with the AUC at 1 year value of 0.718; the AUC at 2 years value of 0.783; the AUC at 3 years value of 0.799, also proved the excellent survival prediction ability of the factor (Fig. 6b).

Necroptosis plays a crucial role in the migration and invasion of a variety of cancers. Although the correlation between necroptosis and liver cancer has been reported, no systematic study has focused on necroptosis-related features as prognostic indicators for the prognosis prediction of patients with liver cancer including HCC.

To address the question, we reanalyzed the necroptosis-related genes of HCC patients derived from public databases via introducing an integrated bioinformatics strategy. At the discovery stage, a total of thirty-six differential NRGs were identified in the TCGA database, which were mainly involved in the signaling pathways of necroptosis and NF kappaB. Studies have shown that necroptosis may be associated with the inhibition of the NF-kB pathway (Zhu et al. 2018; Liu et al. 2016). Tumor necrosis factor (TNF) binds and activates TNF receptor 1, which in turn activates the NF-kB pathway triggering the expression of pro-survival genes to promote cell survival. However, when protein synthesis is inhibited, activation of TNFR1 turns into a death signal, inducing necroptosis of the cell. In parallel, studies found that NF-kB plays an important role in hepatocellular injury, liver fibrosis, and HCC development, and the effects of NF-kB on HCC largely depend on the degree of its activation or inhibition (Luedde and Schwabe 2011).

By integrating univariate and multivariate Cox regression analysis, four NRGs (HSP90AA1, PPIA, SQSTM1 and USP21) were found to have the strong association with the prognosis parameters of HCC patients. Furthermore, we observed that the subjects with HCC could be clearly divided into the high-risk and low-risk groups according to the median risk score using the four NRGs. Of note, the patients in the high-risk group had a shorter OS time compared with that of the low-risk group, which indicated that the expression levels of HSP90AA1, PPIA, SQSTM1 and USP21 were strongly correlated with the survival outcomes and tumor stage of HCC patients.

In previous studies, the genes (HSP90AA1, PPIA, SQSTM1 and USP21) have been reported to be strongly related to cancer. HSP90AA1 belongs to the category of heat shock proteins and is a member of the HSP90 family, which involved in maintaining proper folding of client proteins, playing an important role in regulating protein synthesis / degradation as well as localization balance (Xiao et al. 2021; Jego et al. 2013). Many client proteins regulated by HSP90AA1 are proto oncogene products or important signal transducers during tumor pathogenesis, which are closely related to tumor development and progression (Xiang et al. 2018; Shi et al. 2020). Besides, PPIA was found to be significantly elevated in HCC, which was in accordance with previous reports (Lee and Kim 2010; Ye et al. 2013). A study conducted by Wang et al. (2019) demonstrated that PPIA was highly correlated with survival in patients with liver cancer. According to previous studies, overexpression of PPIA played a central role in a wide of pathological processes, such as aging, cancer metastasis, and the progression of inflammatory diseases (Nigro et al. 2013). SQSTM1 is a macroautophagy/autophagy receptor protein that can be degraded by selective autophagy. Studies have found that increased accumulation of SQSTM1 activation is frequently observed in various cancers including HCC. This may be related to the involvement of SQSTM1 in biological processes such as autophagy, defense against oxidative stress, and protein aggregation as well as apoptosis (Chao et al. 2022). USP21 belongs to the ubiquitin specific peptidases (USPS) subfamily and plays an important role in tumorigenesis (Chen et al. 2017). Yang et al. (2020) found that USP21 was elevated in HCC and promoted the disease progression by targeting mir-637.

Based on ROC analysis, the panel comprising HSP90AA1, PPIA, SQSTM1 and USP21 showed a high AUC of 0.807, suggesting that it had the capacity of distinguishing the prognostic progression of HCC. Furthermore, the diagnostic value of this panel was verified using another cohort of patients, in which its AUCs were over 0.7.

In conclusion, we depicted a necroptosis-related gene signature that had a high predictive value for the prognosis of HCC patients which could be further developed as a novel potential diagnostic biomarker for assessing the clinical outcomes and precise progression of this disease in future.

Acknowledgements Thank the TCGA and ICGC databases.

Author Contributions All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Junyan Lin and Zhiwei Wang. Data validation and visualization were undertaken by Yonghong Jiang and Huihui Su. The first draft of the manuscript was written by Zhandong Yang. All authors carefully revised the manuscript and approved the final version.

Funding This work was supported by Guangdong Basic and Applied Basic Research Foundation (No. 2020A1515110406), and the GDAS′ Projection of Science and Technology Development (No. 2021GDASYL-20210103037).

Competing Interests The authors have no relevant financial or non-financial interests to disclose.

Data Availability All RNA-seq transcriptome data are available in the TCGA database (https://portal.gdc.cancer.gov/) and ICGC database (https://dcc.icgc.org/projects/LIRI-JP).

Agarwal R, Narayan J, Bhattacharyya A et al (2017) Gene expression profiling, pathway analysis and subtype classification reveal molecular heterogeneity in hepatocellular carcinoma and suggest subtype specific therapeutic targets. Cancer Genet 216–217:37–51. https://doi.org/10.1016/j.cancergen.2017.06.002
Christofferson DE, Yuan J (2010) Necroptosis as an alternative form of programmed cell death. Curr Opin Cell Biol 22:263–268. https://doi.org/10.1016/j.ceb.2009.12.003
Chen Y, Zhou B, Chen D (2017) USP21 promotes cell proliferation and metastasis through suppressing EZH2 ubiquitination in bladder carcinoma. Onco Targets Ther 10:681–689. https://doi.org/10.2147/OTT.S124795
Chao X, Ni HM, Ding WX (2022) An unexpected tumor suppressor role of SQSTM1/p62 in liver tumorigenesis. Autophagy 18:459–461. https://doi.org/10.1080/15548627.2021.2008693
Forner A, Reig M, Bruix J (2018) Hepatocellular carcinoma. Lancet 391:1301–1314. https://doi.org/10.1016/S0140-6736(18)30010-2
Gong Y, Fan Z, Luo G et al (2019) The role of necroptosis in cancer biology and therapy. Mol Cancer 18:100. https://doi.org/10.1186/s12943-019-1029-8
Jego G, Hazoume A, Seigneuric R et al (2013) Targeting heat shock proteins in cancer. Cancer Lett 332:275–285. https://doi.org/10.1016/j.canlet.2010.10.014
Kim E, Viatour P (2020) Hepatocellular carcinoma: old friends and new tricks. Exp Mol Med 52:1898–1907. https://doi.org/10.1038/s12276-020-00527-1
Lim JH, Oh S, Kim L et al (2021) Low-level expression of necroptosis factors indicates a poor prognosis of the squamous cell carcinoma subtype of non-small-cell lung cancer. Transl Lung Cancer Res 10:1221–1230. https://doi.org/10.21037/tlcr-20-1027
Lee J, Kim SS (2010) Current implications of cyclophilins in human cancers. J Exp Clin Cancer Res 29:97. https://doi.org/10.1186/1756-9966-29-97
Luedde T, Schwabe RF (2011) NF-kappaB in the liver–linking injury, fibrosis and hepatocellular carcinoma. Nat Rev Gastroenterol Hepatol 8:108–118. https://doi.org/10.1038/nrgastro.2010.213
Liu XJ, Zhou M, Mei L et al (2016) Key roles of necroptotic factors in promoting tumor growth. Oncotarget 7:22219–22233. https://doi.org/10.18632/oncotarget.7924
Nicole L, Sanavia T, Cappellesso R et al (2022) Necroptosis-driving genes RIPK1, RIPK3 and MLKL-p are associated with intratumoral CD3(+) and CD8(+) T cell density and predict prognosis in hepatocellular carcinoma. J Immunother Cancer 10(3). https://doi.org/10.1136/jitc-2021-004031
Nigro P, Pompilio G, Capogrossi MC (2013) Cyclophilin A: a key player for human disease. Cell Death Dis 4:e888. https://doi.org/10.1038/cddis.2013.410
Park JE, Lee JH, Lee SY et al (2020) Expression of key regulatory genes in necroptosis and its effect on the prognosis in non-small cell lung cancer. J Cancer 11:5503–5510. https://doi.org/10.7150/jca.46172
Shi W, Feng L, Dong S et al (2020) FBXL6 governs c-MYC to promote hepatocellular carcinoma through ubiquitination and stabilization of HSP90AA1. Cell Commun Signal 18(1):100. https://doi.org/10.1186/s12964-020-00604-y
Seehawer M, Heinzmann F, D'Artista L et al (2018) Necroptosis microenvironment directs lineage commitment in liver cancer. Nature 562:69–75. https://doi.org/10.1038/s41586-018-0519-y
Siegel RL, Miller KD, Jemal A (2018) Cancer statistics. CA Cancer J Clin 68:7–30. https://doi.org/10.3322/caac.21442
Wang S, Li M, Xing L et al (2019) High expression level of peptidylprolyl isomerase A is correlated with poor prognosis of liver hepatocellular carcinoma. Oncol Lett 18:4691–4702. https://doi.org/10.3892/ol.2019.10846
Xiang X, You XM, Li LQ (2018) Expression of HSP90AA1/HSPA8 in hepatocellular carcinoma patients with depression. Onco Targets Ther 11:3013–3023. https://doi.org/10.2147/OTT.S159432
Xiao H, Wang B, Xiong HX et al (2021) A novel prognostic index of hepatocellular carcinoma based on immunogenomic landscape analysis. J Cell Physiol 236:2572–2591. https://doi.org/10.1002/jcp.30015
Yang TB, Yi F, Liu WF et al (2020) Identification of hsa_circ_0039053 as an up-regulated and oncogenic circRNA in hepatocellular carcinoma via the miR-637-mediated USP21 activation. J Cancer 11:6950–6959. https://doi.org/10.7150/jca.48998
Ye Y, Huang A, Huang C et al (2013) Comparative mitochondrial proteomic analysis of hepatocellular carcinoma from patients. Proteomics Clin Appl 7:403–415. https://doi.org/10.1002/prca.201100103
Yang JD, Heimbach JK (2020) New advances in the diagnosis and management of hepatocellular carcinoma. BMJ 371:m3544. https://doi.org/10.1136/bmj.m3544
Zhang Z, Xie G, Liang L et al (2018) RIPK3-Mediated Necroptosis and Neutrophil Infiltration Are Associated with Poor Prognosis in Patients with Alcoholic Cirrhosis. J Immunol Res 2018:1509851. https://doi.org/10.1155/2018/1509851
Zhu K, Liang W, Ma Z et al (2018) Necroptosis promotes cell-autonomous activation of proinflammatory cytokine gene expression. Cell Death Dis 9:500. https://doi.org/10.1038/s41419-018-0524-y

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Identification and Validation of a Necroptosis-related Gene based Prognostic Signature for Patients with Hepatocellular Carcinoma

Status:

Version 1

Abstract

Figures

Introduction

Materials And Methods

Results

Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1