# Assessment of peripheral gene expression signatures as predictive biomarkers for hepatocellular carcinoma following DAA treatment

## TL;DR

This study identifies blood-based gene expression patterns that could predict liver cancer risk in patients who cleared hepatitis C with drug treatment.

## Contribution

Introduces a non-invasive blood-based gene panel for HCC risk assessment in DAA-treated HCV patients.

## Key findings

- WNT10A, SPHK1, JUNB, and EDN1 were significantly upregulated in PBMCs post-SVR.
- KLF4 was downregulated in PBMCs, particularly after SVR.
- PBMC gene expression showed high diagnostic accuracy (AUROC > 0.92) for several genes.

## Abstract

Hepatocellular carcinoma (HCC) is a major cause of cancer mortality worldwide, with viral hepatitis accounting for about 80 % of incidences. In Egypt, HCV contributes to 63 % of HCC cases. Although DAAs have achieved high SVR rates, they do not eliminate the risk of HCV-related HCC. Persistent epigenetic changes induced by HCV-infection may establish an “oncogenic memory” that promotes HCC even after viral clearance. Peripheral blood mononuclear cells (PBMCs) offer a non-invasive platform for detecting systemic immune and oncogenic signatures, aiding HCC risk assessment. This study aimed to characterize the expression of an epigenetically induced gene panel, comprising JUNB, WNT10A, SPHK1, EDN1, and KLF4 in hepatic tissues and PBMCs from Egyptian HCC patients with HCV genotype 4 who achieved SVR. In silico analyses revealed strong epigenetic associations of these genes, including links to histone-modifying enzymes, protein–protein interaction networks, and enrichment in cancer-related pathways. Gene expression was analyzed using qRT-PCR in SVR individuals, chronic HCV patients, and healthy controls, with diagnostic performance evaluated using multivariate regression and ROC curve analyses. Our results showed significant upregulation of WNT10A, SPHK1, JUNB, and EDN1 and downregulation of KLF4 in PBMCs, particularly post-SVR. PBMC expression showed high diagnostic accuracy (AUROC > 0.92 for SPHK1, WNT10A, JUNB). In conclusion, combining PBMC gene expression profiling with in-silico analyses highlights JUNB, WNT10A, SPHK1, EDN1, and KLF4 as promising non-invasive biomarker panel for HCC risk in DAA-SVR patients, reflecting their integration into epigenetic and oncogenic networks and supporting their potential for risk stratification and therapeutic targeting.