Peripheral immune signatures associated with the risk of hepatocarcinogenesis in cirrhotic Egyptian HCV patients before and after treatment with direct-acting antivirals

Abstract

Background Although direct-acting antivirals (DAAs) have revolutionized the management of chronic HCV, the debatable association with hepatocellular carcinoma (HCC) occurrence/recurrence has raised major concerns about their long-term use, especially in cirrhotic cases. The role of epithelial tight junction proteins (TJPs) in hepatocarcinogenesis has been highlighted; however, the association of their expression in peripheral blood mononuclear cells (PBMCs) with HCC has rarely been reported. This study aimed to explore the role of peripheral claudin (Cldn)1 in liver pathogenesis and its crosstalk with soluble immune mediators in HCC prognosis. Methods The study population included six independent subgroups: healthy controls, cirrhotic/non-cirrhotic treatment-naïve HCV patients, DAA-SVR patients, and anticancer treatment-naïve de novo HCC patients. The laboratory tests included serum levels of alpha-fetoprotein (AFP), albumin, liver transaminases, total bilirubin, and CBC profiling. The serum levels of soluble cluster of differentiation (sCD)163, IL-10, and IL-12 were estimated by corresponding ELISA kits, whereas the levels of Cldn1 and transforming growth factor (TGF)-β in PBMCs were quantified using quantitative PCR (qPCR). Results Serum sCD163, IL-10, and IL-12 levels were significantly higher in the HCC patient group than in the control and non-malignant patient groups (P < 0.0001). No significant difference was detected in the serum levels of the three markers between cirrhotic and non-cirrhotic patients, whereas their levels were significantly different between cirrhotic and non-cirrhotic patients (P < 0.0001). Similarly, the transcriptional levels of peripheral Cldn1 and TGF-β were significantly higher in patients with HCC and non-malignant cirrhosis than in patients without cirrhosis (P = 0.0185–<0.0001 and 0.0089–<0.0001, respectively). Logistic regression analysis revealed a significant association between all the abovementioned markers and HCC (P = 0.0303 to < 0.0001), which was further confirmed by the results of receiver operating characteristic (ROC) analysis, which revealed an area under the curve (AUC) value ranging from 0.883 to 0.996. The calculated cutoff values demonstrated remarkable prognostic capacity, with ranges of 88–99.41% and 82.14–97.92% and positive/negative predictive values ranging from 84.62 to 98.3% and 92–98%, respectively. Conclusion The proposed HCC predictors are novel non-invasive HCC biomarkers that maintain their predictive power under different pathological conditions and circumvent the drawbacks of conventional prognostic markers in patients with mild cirrhosis and/or normal AFP, albumin, and/or platelet counts.