Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2021 Mar 12:11:626654.
doi: 10.3389/fonc.2021.626654. eCollection 2021.

Identification of a Novel Four-Gene Signature Correlated With the Prognosis of Patients With Hepatocellular Carcinoma: A Comprehensive Analysis

Weihua Zhu  1 , Lixin Ru  2 , Zhenchao Ma  2   3
Affiliations

Identification of a Novel Four-Gene Signature Correlated With the Prognosis of Patients With Hepatocellular Carcinoma: A Comprehensive Analysis

Weihua Zhu et al. Front Oncol. .

Abstract

Purpose: Hepatocellular carcinoma (HCC) is a common solid-tumor malignancy with high heterogeneity, and accurate prognostic prediction in HCC remains difficult. This analysis was performed to find a novel prognostic multigene signature.

Methods: The TCGA-LIHC dataset was analyzed for differentially coexpressed genes through weighted gene coexpression network analysis (WGCNA) and differential gene expression analysis. A protein-protein interaction (PPI) network and univariate Cox regression analysis of overall survival (OS) were utilized to identify their prognostic value. Next, we used least absolute shrinkage and selection operator (LASSO) Cox regression to establish a prognostic module. Subsequently, the ICGC-LIRI-JP dataset was applied for further validation. Based on this module, HCC cases were stratified into high-risk and low-risk groups, and differentially expressed genes (DEGs) were identified. Functional enrichment analyses of these DEGs were conducted. Finally, single-sample gene set enrichment analysis (ssGSEA) was performed to explore the correlation between the prognostic signature and immune status.

Results: A total of 393 differentially coexpressed genes were obtained. Forty differentially coexpressed hub genes were identified using the CytoHubba plugin, and 38 of them were closely correlated with OS. Afterward, we established the four-gene prognostic signature with an acceptable accuracy (area under the curve [AUC] of 1-year survival: 0.739). The ICGC-LIRI-JP dataset also supported the acceptable accuracy (AUC of 1-year survival:0.752). Compared with low-risk cohort, HCC cases in the high-risk cohort had shorter OS, higher tumor grades, and higher T stages. The risk scores of this signature still act as independent predictors of OS (P<0.001). Functional enrichment analyses suggest that it was mainly organelle fission and nuclear division that were enriched. Finally, ssGSEA revealed that this signature is strongly associated with the immune status of HCC patients.

Conclusions: The proposed prognostic signature of four differentially coexpressed hub genes has satisfactory prognostic ability, providing important insight into the prediction of HCC prognosis.

Keywords: hepatocellular carcinoma; immune status; least absolute shrinkage and selection operator Cox regression; prognostic signature; single-sample gene set enrichment analysis; weighted gene coexpression network analysis.

PubMed Disclaimer

Conflict of interest statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Figures

Figure 1
Figure 1
Study design and workflow of this study.
Figure 2
Figure 2
Identification of modules related to the clinical traits in the TCGA-LIHC dataset. (A) Sample dendrogram and trait heatmap. (B) Scale independence and Mean connectivity. (C) The cluster dendrogram of co-expression network modules is ordered by a hierarchical clustering of genes based on the 1-TOM matrix. Different colors represent different modules. (D) Module-trait relationships. Each row represents a color module and every column represents a clinical trait (normal and tumor). Each cell contains the corresponding correlation and P-value.
Figure 3
Figure 3
Identification of differentially expressed genes (DEGs) in TCGA-LIHC dataset with the cut-off criteria of |logFC|>1 and adj.P <0.05. (A) Heatmap of top 50 upregulated and 50 downregulated DEGs of TCGA-LIHC dataset. (B) Volcano plot of DEGs in the TCGA-LIHC dataset. (C) The Venn diagram of genes between DEGs and co-expression genes. A total of 393 overlapping differential co-expression genes are detected.
Figure 4
Figure 4
Visualization of the protein-protein interaction (PPI) network and hub genes. (A) PPI network of differential co-expression genes. (B) The identification of 40 differential co-expressed hub genes using the degree algorithm.
Figure 5
Figure 5
Identification differential co-expressed hub genes with prognostic values. (A) Univariate Cox analysis for overall survival (OS) of 38 differential co-expressed hub genes with prognostic values. (B) 38 differential co-expressed hub genes with prognostic values are significantly upregulated in HCC tissues. (C) The correlation network of candidate genes. The correlation coefficients are represented by different colors.
Figure 6
Figure 6
|Construction of the gene signature and nomogram in TCGA-LIHC dataset. (A, B) The construction of the four-gene signature module. (C) The construction of the nomogram of this module. (D–F) The calibration curves of 1-, 2-, and 3-year overall survival probability.
Figure 7
Figure 7
Prognostic analysis of the four-gene signature model in TCGA-LIHC dataset. (A) AUC of time-dependent ROC curves verifies the prognostic performance of the risk score in TCGA-LIHC dataset. (B) The distribution and the median value of the risk scores in TCGA-LIHC dataset. (C) The distributions of OS status, OS and risk score in the TCGA-LIHC dataset. (D) Kaplan-Meier curves for the OS of patients in the high-risk group and low-risk group in TCGA-LIHC dataset. (E) PCA plot of TCGA-LIHC dataset. (F) t-SNE analysis of TCGA-LIHC dataset.
Figure 8
Figure 8
Validation of the 10-gene signature in ICGC-LIRI-JP dataset. (A) AUC of time-dependent ROC curves verifies the prognostic performance of the risk score in ICGC-LIRI-JP dataset. (B) The distribution and the median value of the risk scores in ICGC-LIRI-JP dataset. (C) The distributions of OS status, OS and risk scores in ICGC-LIRI-JP dataset. (D) Kaplan-Meier curves for the OS of patients in the high-risk group and low-risk group in ICGC-LIRI-JP dataset. (E) PCA plot of ICGC-LIRI-JP dataset. (F) t-SNE analysis of ICGC-LIRI-JP dataset.
Figure 9
Figure 9
Independent prognostic role of the four-gene signature. (A)The univariate Cox regression analysis in TCGA-LIHC dataset. (B)The univariate Cox regression analysis in ICGC-RI-JP dataset. (C) The multivariate Cox regression analysis in TCGA-LIHC dataset. (D) The multivariate Cox regression analysis in ICGC-LIRI-JP dataset.
Figure 10
Figure 10
Functional enrichment analysis of differentially expressed genes (DEGs) between high-risk and low-risk groups. (A) Gene ontology (GO) enrichment analysis of DEGs of TCGA-LIHC dataset. (B) Gene ontology (GO) enrichment analysis of DEGs of ICGC-RI-JP dataset. (C) Kyoto encyclopedia of genes and genomes pathway analysis of DEGs of TCGA-LIHC dataset. (D) KEGG pathway analysis of DEGs of ICGC-RI-JP dataset.
Figure 11
Figure 11
Comparison of single-sample gene set enrichment (ssGSEA) scores between high-risk and low-risk groups in TCGA-LIHC and ICGC-LIRI-JP datasets. (A, B) The scores of 16 immune cells and 13 immune-related functions are displayed in boxplots in TCGA-LIHC dataset. (C, D) The scores of 16 immune cells and 13 immune-related functions are displayed in boxplots in ICGC-RI-JP dataset. Adjusted P values are showed as: ns, not significant; *P < 0.05; **P < 0.01; ***P < 0.001.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2020. CA Cancer J Clin (2020) 70(1):7–30. 10.3322/caac.21590 - DOI - PubMed
    1. Vogel A, Saborowski A. Current strategies for the treatment of intermediate and advanced hepatocellular carcinoma. Cancer Treat (2020) 82:101946. 10.1016/j.ctrv.2019.101946 - DOI - PubMed
    1. Kole C, Charalampakis N, Tsakatikas S, Vailas M, Moris D, Gkotsis E, et al. . Immunotherapy for Hepatocellular Carcinoma: A 2021 Update. Cancers (Basel) (2020) 12(10):E2859. 10.3390/cancers12102859 - DOI - PMC - PubMed
    1. Caines A, Selim R, Salgia R. The Changing Global Epidemiology of Hepatocellular Carcinoma. Clin Liver Dis (2020) 24(4):535–47. 10.1016/j.cld.2020.06.001 - DOI - PubMed
    1. Huo J, Wu L, Zang Y. A robust nine-gene prognostic signature associated with tumour doubling time for hepatocellular carcinoma. Life Sci (2020) 260:118396. 10.1016/j.lfs.2020.118396 - DOI - PubMed

LinkOut - more resources