Intrahepatic quantification of HBV antigens in chronic hepatitis B reveals heterogeneity and treatment-mediated reductions in HBV core-positive cells

JHEP Rep. 2022 Dec 26;5(4):100664. doi: 10.1016/j.jhepr.2022.100664. eCollection 2023 Apr.

Abstract

Background & aims: Patterns of liver HBV antigen expression have been described but not quantified at single-cell resolution. We applied quantitative techniques to liver biopsies from individuals with chronic hepatitis B and evaluated sampling heterogeneity, effects of disease stage, and nucleos(t)ide (NUC) treatment, and correlations between liver and peripheral viral biomarkers.

Methods: Hepatocytes positive for HBV core and HBsAg were quantified using a novel four-plex immunofluorescence assay and image analysis. Biopsies were analysed from HBeAg-positive (n = 39) and HBeAg-negative (n = 75) participants before and after NUC treatment. To evaluate sampling effects, duplicate biopsies collected at the same time point were compared. Serum or plasma samples were evaluated for levels of HBV DNA, HBsAg, hepatitis B core-related antigen (HBcrAg), and HBV RNA.

Results: Diffusely distributed individual HBV core+ cells and foci of HBsAg+ cells were the most common staining patterns. Hepatocytes positive for both HBV core and HBsAg were rare. Paired biopsies revealed large local variation in HBV staining within participants, which was confirmed in a large liver resection. NUC treatment was associated with a >100-fold lower median frequency of HBV core+ cells in HBeAg-positive and HBeAg-negative participants, whereas reductions in HBsAg+ cells were not statistically significant. The frequency of HBV core+ hepatocytes was lower in HBeAg-negative participants than in HBeAg-positive participants at all time points evaluated. Total HBV+ hepatocyte burden correlated with HBcrAg, HBV DNA, and HBV RNA only in baseline HBeAg-positive samples.

Conclusions: Reductions in HBV core+ hepatocytes were associated with HBeAg-negative status and NUC treatment. Variation in HBV positivity within individual livers was extensive. Correlations between the liver and the periphery were found only between biomarkers likely indicative of cccDNA (HBV core+ and HBcrAg, HBV DNA, and RNA).

Impact and implications: HBV infects liver hepatocyte cells, and its genome can exist in two forms that express different sets of viral proteins: a circular genome called cccDNA that can express all viral proteins, including the HBV core and HBsAg proteins, or a linear fragment that inserts into the host genome typically to express HBsAg, but not HBV core. We used new techniques to determine the percentage of hepatocytes expressing the HBV core and HBsAg proteins in a large set of liver biopsies. We find that abundance and patterns of expression differ across patient groups and even within a single liver and that NUC treatment greatly reduces the number of core-expressing hepatocytes.

Keywords: ADV, adefovir; ALT, alanine aminotransferase; Biomarkers; CHB, chronic hepatitis B; CNN, convolutional neural network; HBV; HBV core; HBV core, hepatitis B core antigen; HBV, Hepatitis B Virus; HBcrAg, hepatitis B core-related antigen; HBeAg; HBeAg, Hepatitis B e antigen; HBsAg; HBsAg, Hepatitis B surface antigen; HCC, hepatocellular carcinoma; IF, immunofluorescence; NUC; NUC, nucleo(t)side; Na+K+-ATPase, sodium–potassium ATPase; QC, quality control; TDF, tenofovir disoproxil fumarate; cccDNA, covalently closed circular DNA; dslDNA, double-stranded linear DNA.