(S-056) Quantitative Assessment of Glecaprevir and Pibrentasvir Exposures in Adults with Chronic or Acute HCV Infection: A Comprehensive Analysis of Phase 2 and 3 Studies Utilizing Subject Matching and Population PK Simulations

Maria Vittoria Barbarossa – Abbvie; Mong-Jen Chen – Abbvie; Doerthe Eckert – Abbvie; Insa Winzenborg – Abbvie; Tien-Yi Lee – Abbvie; Nael Mostafa – Abbvie; Sven Mensing – Abbvie

Objectives: The global burden of Hepatitis C Virus (HCV) infection remains significant, with new (acute) infections driven primarily by injection drug use. While the fixed-dose combination regimen of Glecaprevir/Pibrentasvir (GLE/PIB) 300 mg/120 mg QD has been indicated in treating adults and adolescents with chronic HCV infection [1], its safety and efficacy in adults with acute HCV infection were only recently explored in Phase 3 study M20-350 (N=286) [2] without pharmacokinetics (PK) sample collection.
GLE/PIB are expected to follow similar ADME pathways in both acute and chronic HCV infections, as supported by observed efficacy data. Based on this assumption, our work presents an innovative approach that uses statistical matching [3] and leverages data collected from 2664 adults with chronic HCV infection who received GLE/PIB, along with in-house developed GLE/PIB population PK (popPK) models [4], to infer expected drug levels in adults with acute HCV infection.

Methods: Statistical nearest neighbor matching (SNNM) (R package “MatchIt” [5]) was applied to define for each subject in the acute HCV (test) population the "closest" eligible subject in the chronic HCV (reference) population. The distance computed by the SNNM algorithm accounted for baseline demographic, hepatic and disease-related parameters, particularly those included as covariates in the previously developed popPK models [4].

Once pairs of subjects (reference - test) were identified, GLE/PIB steady-state exposures (AUC24ss) were computed based on the popPK models, individual baseline characteristics, and individual variability parameters. While covariates of subjects in the test population were accounted for, their individual variabilities (unknown due to lacking exposure data) were assigned from the reference population, allowing each subject with acute HCV to inherit variability from the matched subject with chronic HCV. Robustness of results was investigated by systematically evaluating other matching models with reduced or different sets of covariates.

Results: Comparison of baseline characteristic distributions between the chronic HCV matched group and the total reference population showed no major differences, indicating that the matched group is a representative subset of the overall reference population. This enabled reliable extrapolation of GLE/PIB exposures to adults with acute HCV infection, with predicted AUC24ss being well centered within the overall steady-state exposure ranges observed in the reference population. Variations of model inputs demonstrated robustness of the results, with variations in geometric means of GLE and PIB AUC24ss less than 20.1% and 6.82%, respectively.

Conclusions: This novel approach employed SNNM and popPK models to predict GLE/PIB steady-state exposures in adults with acute HCV infection. The analysis-predicted drug levels in adults with acute HCV infection were comparable to those of adults with chronic HCV infection, supporting the use of the GLE/PIB 300 mg/120 mg QD regimen in the former population. This methodology, validated through robustness analysis, provides a promising strategy to understand exposure levels in a population where PK collection may be unavailable.

Citations: [1] Mavyret™ (glecaprevir and pibrentasvir) [package insert]. North Chicago, IL. USA. Abbvie Inc. Revised October 2023.
[2] Study to Evaluate Adverse Events and Change in Disease Activity in Adult and Adolescent Participants With Acute Hepatitis C Virus (HCV) Infection on Treatment With Oral Tablets of Glecaprevir (GLE)/Pibrentasvir (PIB). ClinicalTrials.gov identifier: NCT04903626. Updated Dec 4, 2024. Accessed Apr 2,2025. https://clinicaltrials.gov/study/NCT04903626
[3] Ho DE, Imai K, King G, et al. Stuart. Matching as nonparametric preprocessing for reducing model dependence in parametric causal inference. Political Analysis. 2007;15(3):199-236
[4] Lin, C.-W. et. al. (2017). Glecaprevir and Pibrentasvir Exposures in Hepatitis C Virus-Infected Subjects in Phase 2 and 3 Studies. [Poster Presentation] 2017 AASLD
[5] Greifer, Noah. MatchIt: Getting Started. Accessed Apr 2, 2025. https://cran.r-project.org/web/packages/MatchIt/vignettes/MatchIt.html

Keywords: HCV, statistical matching, population PK