A novel model based on ubiquitination-related gene to predict prognosis and immunotherapy response in hepatocellular carcinoma

Abstract

Background: Hepatocellular carcinoma (HCC) is a common cancer that is increasingly becoming a global health problem and a major public health concern. In order to improve patient outcomes, additional biomarkers and targets must be explored. Ubiquitination-related genes (URGs), as tumor regulators, exhibit multiple functions in tumor development. Our objective was to examine the influence of URGs on the prognosis of patients with HCC.

Methods: By utilizing unsupervised cluster analysis, we were able to identify URGs in the database and create a risk score profile for predicting the prognosis of patients with HCC. The model's clinical application was explored using subject operating characteristic curves, survival analysis, and correlation analysis. We additionally examined the variances in clinical traits, immune infiltration, somatic genetic alterations, and responsiveness to treatment among high- and low-risk populations identified by the prognostic model. Scores for immune cell infiltration and immune-related pathway activity were determined by performing ssGSEA enrichment analysis. Additionally, to investigate potential mechanisms, we utilized GO, KEGG and GSVA analyses.

Results: We developed a risk scoring model that relies on genes associated with ubiquitination. As the risk score increased, the malignancy and prognosis of the tumor worsened. The high-risk and low-risk groups exhibited notable disparities in relation to the immune microenvironment, genes associated with immune checkpoints, sensitivity to drugs, and response to immunotherapy.

Conclusion: The utilization of a risk model that relies on genes associated with ubiquitination can serve as a biomarker to assess the prognosis of patients with HCC, and aid in the selection of suitable therapeutic agents.

Keywords: Gene expression; Hepatocellular carcinoma; Immune cells; Immunotherapy; Ubiquitination-related genes.

Fig. 1

Enrichment analysis of differentially expressed URGs and consistency clustering analysis. (A) GO enrichment analysis of differentially expressed URGs. (B) KEGG enrichment analysis of differentially expressed URGs. (C) Consensus clustering CDF for k = 2 to 9. (D) Relative change in area under the cumulative CDF curve for k = 2 to 9. (E) The overall survival (OS) probability of the patients in the two clusters.

Fig. 2

Construction of prognostic model. (A) LASSO coefficient profiles of the 155 URGs. (B) LASSO cross‐validation for selecting optimal tuning parameter (λ). (C) Relevance of URGs involved in model construction. (D) Distribution of risk scores and survival status between low- and high-risk groups in the TCGA Cohort. (E) Survival rate of high- and low-risk group in TCGA. (F) PCA analysis in TCGA.

Fig. 3

Evaluation of URGs signatures as independent prognostic factors for HCC. (A) Univariate Cox regression analysis. (B) Multivariate Cox regression analysis. (C) ROC curve for 1-year survival in HCC patients. (D) The correlation between risk score and pathological grade. (E) The correlation between risk score and tumor stage.

Fig. 4

(A) The TME component analysis. (B) Immune checkpoint differences between high and low risk groups. (C) association between mDNAsi and risk score. (D) association between mRNAsi and risk score. (E) The analysis of differences in immune cell infiltration between the two groups with ssGSEA. (F) The analysis of differences in immune functions between the two groups with ssGSEA.

Fig. 5

(A) The somatic gene mutations in the high-risk group. (B) The somatic gene mutations in the low-risk group. (C) Correlation of risk scores with TMB. (D) The CNV gain and loss of URGs.

Fig. 6

Drug sensitivity analysis in the high-risk group. (A) 5-Fluorouracil. (B) Cediranib. (C) Crizotinib. (D) Lapatinib.

Fig. 7

Prediction of immunotherapy. (A) TIDE analysis revealing the difference of tumor immune dysfunction and exclusion in two groups. (B) Prediction immunotherapy response in IMVigor210.