Establishment of a prognostic signature based on fatty acid metabolism genes in HCC associated with hepatitis B

Abstract

Background: Hepatitis B virus (HBV)-associated hepatocellular carcinoma (HCC) is one of the most common and deadly cancer and often accompanied by varying degrees of liver damage, leading to the dysfunction of fatty acid metabolism (FAM). This study aimed to investigate the relationship between FAM and HBV-associated HCC and identify FAM biomarkers for predicting the prognosis of HBV-associated HCC.

Methods: Gene Set Enrichment Analysis (GSEA) was used to analyze the difference of FAM pathway between paired tumor and adjacent normal tissue samples in 58 HBV-associated HCC patients from the Gene Expression Omnibus (GEO) database. Next, 117 HBV-associated HCC patients from The Cancer Genome Atlas (TCGA) database were analyzed to establish a prognostic signature based on 42 FAM genes. Then, the prognostic signature was validated in an external cohort consisting of 30 HBV-associated HCC patients. Finally, immune infiltration analysis was performed to evaluate the FAM-related immune cells in HBV-associated HCC.

Results: As a result, FAM pathway was clearly downregulated in tumor tissue of HBV-associated HCC, and survival analysis demonstrated that 12 FAM genes were associated with the prognosis of HBV-associated HCC. Lasso-penalized Cox regression analysis identified and established a five-gene signature (ACADVL, ACAT1, ACSL3, ADH4 and ECI1), which showed effective discrimination and prediction for the prognosis of HBV-associated HCC both in the TCGA cohort and the validation cohort. Immune infiltration analysis showed that the high-risk group, identified by FAM signature, of HBV-associated HCC had a higher ratio of Tregs, which was associated with the prognosis.

Conclusions: Collectively, these findings suggest that there is a strong connection between FAM and HBV-associated HCC, indicating a potential therapeutic strategy targeting FAM to block the accumulation of Tregs into the tumor microenvironment of HBV-associated HCC.

Keywords: Fatty acid metabolism; HBV; Hepatocellular carcinoma; Prognosis; Tregs.

Fig. 1

Flowchart of data collection and method implementation

Fig. 2

GSEA analysis of FAM pathway in HBV-associated HCC and paired normal tissue. A GSEA results in GSE94660; B Specific FAM genes in GSE94660; C GSEA results in GSE121248; D Specific FAM genes in GSE121248

Fig. 3

Establishment of the prognostic signature for HBV-associated HCC based on 42 FAM genes in the TCGA database. A LASSO regression coefficient profile of the 42 FAM genes; B LASSO deviance profile of the 42 FAM genes; C From top to bottom are the risk score distribution, survival and death status distribution, and heat map of genes in the prognostic signature between the low-risk and high-risk groups; D Kaplan-Meier curves for the OS; E Time-dependent ROC curves for predicting OS; F Kaplan-Meier curves for the RFS; G Time-dependent ROC curves for predicting RFS

Fig. 4

Nomogram of the prognostic signature for predicting HBV-associated HCC OS at 1-, 2-, 3-, and 5-year in the TCGA database (n = 117)

Fig. 5

Nomogram of the prognostic signature for predicting HBV-associated HCC RFS at 1-, 2-, 3-, and 5-year in the TCGA database (n = 117)

Fig. 6

Performance of the prognostic signature for HBV-associated HCC in the validation cohort. A Relative mRNA expression of ACADVL, ACAT1, ACSL3, ADH4 and ECI1 in 30 HBV-associated HCC with paired normal tissue; B Kaplan-Meier curves for the OS; C Time-dependent ROC curves for predicting OS; D Kaplan-Meier curves for the RFS; E Time-dependent ROC curves for predicting RFS

Fig. 7

Differential expression of the prognostic signature genes between HBV-positive and -negative HCC patients and cell lines. A Relative mRNA expression between 30 HBV-positive HCC and 15 HBV-negative HCC patients; B Relative mRNA expression between HBV-positive and -negative HCC cell lines

Fig. 8

Differential analysis of immune infiltration. A Immune infiltration in HBV-positive (n = 117) and -negative (n = 254) HCC; B Immune infiltration between high-risk and low-risk group identified by the prognostic signature in HBV-positive HCC

Fig. 9

A Correlation analysis of 5 prognostic signature genes and 22 immune cells; Kaplan-Meier curves of four immune cells associated with overall survival in HBV-associated HCC: B T cells CD4 memory resting; C T cells regulatory (Tregs); D Macrophages M2; E Neutrophils