and as potential diagnostic biomarkers for HBV-induced liver fibrosis in chronic hepatitis B.

BACKGROUND

The main cause of the liver fibrosis (LF) remains hepatitis B virus (HBV) infection, especially in China. Histologically, liver fibrosis still occurs progressively in chronic hepatitis B (CHB) patients, even if HBV-DNA is negative or undetectable. The diagnosis of LF is beneficial to control the development of it, also it may promote the reversal of LF. Although liver biopsy is the gold standard of diagnosis in LF at present, it isa traumatic diagnosis. There are no diagnostic biomarkers as yet for the condition. It is badly in need of biomarkers clinically, which is simple to test, minimally invasive, highly specific, and sensitive. Early detection of HBV-LF development is crucial in the prevention, treatment, and prognosis prediction of HBV-LF. Cytokines are closely associated with both immune regulation and inflammation in the progression of hepatitis B virus associated-liver fibrosis (HBV-LF). In this bioinformatic study, we not only analyzed the relationship between HBV-LF and immune infiltration, but also identified key genes to uncover new therapeutic targets.

OBJECTIVES

To find potential biomarkers for liver fibrosis in the development of chronic hepatic B patients.

MATERIALS AND METHODS

We obtained two sets of data including CHB/healthy control and CHB/HBV-LF from the Integrated Gene Expression (GEO) database to select for differential expression analysis. Protein-protein interaction (PPI) network was also generated, while key genes and important gene modules involved in the occurrence and development of HBV-LF were identified. These key genes were analyzed by functional enrichment analysis, module analysis, and survival analysis. Furthermore, the relationship between these two diseases and immune infiltration was explored.

RESULTS

Among the identified genes, 150 were individually associated with CHB and healthy control in the differential gene expression (DGE) analysis. While 14 with CHB and HBV-LF. It was also analyzed in the Robust rank aggregation (RRA) analysis, 34 differential genes were further identified by Cytohubba. Among 34 differential genes, two core genes were determined: and . was able to predict CHB positivity (area under the receiver operating characteristic curve [AUC-ROC] = 0.883, 95% confidence interval [CI] 0.786-0.963), while HBV-LF positivity ([AUC-ROC] = 0.687, 95% confidence interval [CI] 0.592-0.779). And as able to predict CHB positivity ([AUC-ROC] = 0.960, 95% confidence interval [CI] 0.915-0.992), while HBV-LF positivity ([AUC-ROC] = 0.773, 95% confidence interval [CI] 0.680-0.856). Relationship between gene expression and LF grades was  < 0.05, as well as .

CONCLUSION

and were found to be potential biomarkers and therapeutic targets for HBV-LF. It is instructive for research on the progression of LF in HBV patients, suppression of chronic inflammation, and development of molecularly targeted-therapy for HBV-LF.