049: Assessment of the Effect of HBV Viral Genotype on Bepirovirsen Efficacy Using a Population Pharmacokinetic/Pharmacodynamic Modeling Approach

Statement of Purpose, Innovation or Hypothesis: Bepirovirsen (BPV), an antisense oligonucleotide, reduces the level of HBV RNA in participants with chronic HBV infection who were not on nucleotide (NA) therapy and those who were on stable NA therapy. A mechanistic pharmacokinetic/pharmacodynamic (PK/PD) model was developed to simultaneously capture the time course of PK and HBsAg changes observed in BPV Phase 2 studies (Youssef A et al. Presented at EASL 2022 [abstract 3592]). Baseline HBsAg level is a significant predictor of response to BPV treatment, while the effect of HBV virus genotype could not be accurately assessed in the previous analysis due to a high percentage of missing data in virally suppressed participants. The aim of this post-hoc work was to assess the effect of different HBV genotypes as predictors of BPV efficacy in addition to baseline HBsAg level using the PKPD model and assess the correlation between HBV genotype and baseline HBsAg level.

Description of Methods & Materials: A serological assay for HBV genotyping was added to determine HBV genotype in virally suppressed participants from B-Clear (Study no. 209668) and B-Together (Study no. 209348) studies. The effect of genotypes A, B, C and D on the response to BPV was examined visually. First, response at end of BPV treatment and end of study, grouped by study and cohort, was plotted against baseline HBsAg for each genotype. Effect of genotype was assessed on the model parameters using ETA-covariate plots from the base model as well as the final model which includes baseline HBsAg as a covariate on IC50. Finally, Genotype B was tested as a covariate on IC50 in the base and final models.

Data & Results: HBV genotype B was correlated with lower baseline HBsAg compared to other genotypes. The ETA-genotype plots indicate a slightly lower IC50 in genotype B participants compared to other genotypes. HBV Genotype B was a significant covariate on BPV IC50 in the base model (ΔOFV = - 5.45), however the addition of baseline HBsAg as a covariate on IC50 resulted in more pronounced decrease in the objective function (ΔOFV = - 20.32). Genotype B was not a statistically significant covariate on BVP IC50 after the inclusion of baseline HBsAg as covariate on IC50. On the other hand, including genotype B as a covariate on the predicted baseline HBsAg parameter in the final model was shown to be significant (ΔOFV = - 38.16), and to reduce the correlation between genotype B and IC50.

Interpretation, Conclusion or Significance: HBV genotype appears to be correlated with baseline HBsAg level in HBV patients. However, it does not provide additional predictive power or explanation of the variability in response to BPV once the effects of baseline HBsAg are accounted for. BPV is currently being studied in Phase 3 studies in HBV patients with baseline HBsAg of 3000 IU/mL or less. BPV is not expected to have a different efficacy profile across HBV genotypes, nor is dose adjustment needed based on HBV genotype.

Funding: GSK (studies 209668 and 209348)

Disclosures: ASY, JB, RE, MP and AN are GSK employees and hold GSK shares. ACCC is a student worker at GSK.

Citations/References:

Additional Information/Authors: n/a