Search

BACKGROUND AND AIMS: We conducted haplotype analysis of complete hepatitis B virus (HBV) genomes following deep sequencing from 368 patients across multiple phases of chronic hepatitis B (CHB) infection from four major genotypes (A-D), analyzing 4,110 haplotypes to identify viral variants associated with treatment outcome and disease progression. APPROACH AND RESULTS: Between 18.2% and 41.8% of nucleotides and between 5.9% and 34.3% of amino acids were 100% conserved in all genotypes and phases examined, depending on the region analyzed. Hepatitis B e antigen (HBeAg) loss by week 192 was associated with different haplotype populations at baseline. Haplotype populations differed across the HBV genome and CHB history, this being most pronounced in the precore/core gene. Mean number of haplotypes (frequency) per patient was higher in immune-active, HBeAg-positive chronic hepatitis phase 2 (11.8) and HBeAg-negative chronic hepatitis phase 4 (16.2) compared to subjects in the "immune-tolerant," HBeAg-positive chronic infection phase 1 (4.3, P< 0.0001). Haplotype frequency was lowest in genotype B (6.2, P< 0.0001) compared to the other genotypes (A = 11.8, C = 11.8, D = 13.6). Haplotype genetic diversity increased over the course of CHB history, being lowest in phase 1, increasing in phase 2, and highest in phase 4 in all genotypes except genotype C. HBeAg loss by week 192 of tenofovir therapy was associated with different haplotype populations at baseline. CONCLUSIONS: Despite a degree of HBV haplotype diversity and heterogeneity across the phases of CHB natural history, highly conserved sequences in key genes and regulatory regions were identified in multiple HBV genotypes that should be further investigated as targets for antiviral therapies and predictors of treatment response.

Background Alanine aminotransferase (ALT) flares during NA therapy are uncommon but occur. Evaluation of ALT flares during nucleos(t)ide analogue (NA) therapy is important as new immunomodulatory therapies for hepatitis B virus (HBV) are developed. We evaluated the association between ALT flares and HBsAg loss during long-term therapy for genotype A CHB. Methods This analysis included genotype A subjects from a phase III study of tenofovir vs adefovir in HBeAg-positive HBV. ALT flare was defined as (i) a rise in ALT >2x ULN from normal ALT levels; or (ii) a rise in ALT >2x baseline ALT level. HBsAg response at week 384 was recorded as one of HBsAg loss vs HBsAg decline (≥1 log10 IU/mL decline) vs non-response. The primary analysis evaluated the association between ALT flare and HBsAg response. Results 54 subjects were included. 23/54 (43%) subjects experienced an on-treatment ALT flare. 45% achieved an HBsAg reduction ≥1 log10 IU/mL, and of these 67% achieved HBsAg loss at a median of 102 weeks [IQR: 64-156]. Flare was associated with HBsAg decline vs non-response (67% vs 23%, P = .002), and were more common in subjects who achieved HBsAg loss vs non-response (56% vs 23%), P = .049). There was a median delay of 56 weeks [IQR: 40-80] between a flare and HBsAg loss. Conclusion In genotype A subjects undergoing long-term NA therapy, ALT flares predict for HBsAg response. The delay between ALT flare and HBsAg loss has implications for clinical trial design for early phase development of immunomodulatory strategies aiming for HBsAg loss.