Summary: Belumosudil is a selective ROCK2 inhibitor approved for treating chronic graft-versus-host disease (cGVHD) in adult and pediatric patients. Its metabolism involves multiple cytochrome P450 (CYP) enzymes, with CYP3A4 contributing nearly 42% of its clearance.
At the 2024 American Association of Pharmaceutical Scientists (AAPS) PharmSci 360 conference, Quotient Sciences’ poster showcased the development of a Physiologically Based Pharmacokinetic (PBPK) model to simulate drug-drug interactions (DDIs) of belumosudil with itraconazole, rifampicin, and efavirenz.
Read highlights from the poster “Predicting Belumosudil Drug-Drug Interactions Using a Physiologically Based Pharmacokinetic Model” below. Download the poster to see additional findings.
Study design and methodology
This study employed a physiologically based pharmacokinetic and biopharmaceutics (PBPK/PBBM) modeling approach using the GastroPlus™ software platform (Simulations Plus, Inc.). The primary objective was to predict the impact of various cytochrome P450 3A4 (CYP3A4) inhibitors and inducers, including itraconazole (ITZ) and rifampicin (RIF), on the pharmacokinetics of belumosudil. The model's utility was further extended to simulate DDIs involving the moderate CYP3A4 inducer efavirenz (EFV), for which clinical data were not available.
A whole-body PBPK model for belumosudil was constructed using published in vitro and clinical pharmacokinetic data. This involved:
- Integrating key physicochemical properties and ADME (absorption, distribution, metabolism, and excretion) parameters into the model, using either available in vitro data or predictions from the ADMET Predictor™ tool.
- Refining enzyme kinetics by incorporating clinical data on absolute bioavailability and food effects.
- Validating the model against observed plasma concentration-time profiles derived from IV and oral dosing studies.
The model was used to simulate drug-drug interactions involving the CYP3A4 inhibitors ITZ and RIF. These predictions were then validated against existing clinical data to confirm the model's reliability and predictive performance. The model was subsequently applied to predict interactions with EFV, in the absence of available clinical trial data. This application showcases the model's utility in offering insights for potential drug combinations when clinical trial data are not available.
Key findings from the program
The PBPK model for belumosudil demonstrated robust predictive capabilities, accurately forecasting DDIs in healthy male subjects, demonstrating its value in various scenarios:
- ITZ Interaction: Successfully predicted an approximate 25% increase in belumosudil exposure when co-administered with ITZ (a strong CYP3A4 inhibitor), consistent with clinical data (17%-33% increase).
- RIF Interaction: Accurately predicted a substantial 72% decrease in belumosudil exposure with RIF (a strong CYP3A4 inducer), closely matching clinical observations (70%-74% reduction).
- EFV Prediction: Expanded its utility by predicting a moderate 40% reduction in belumosudil exposure (AUC0-t) with EFV (a moderate CYP3A4 inducer), potentially reducing the need for dedicated clinical studies.
The model also provided reliable estimates of bioavailability and metabolic clearance, demonstrating its utility in simulating complex pharmacokinetic scenarios. This offers particularly valuable insights for guiding dosing strategies, identifying potential risks, and informing regulatory decisions—particularly when future clinical DDI studies are not feasible or to confirm retrospective findings.
By enabling quantitative predictions of drug interactions, this model supports smarter, faster optimization and refinement of belumosudil and similar compounds, helping to ensure safe and effective use across diverse patient populations.
Download the poster to see additional findings and learn more about modeling & simulation services provided by Quotient Sciences.
To get alerted about new blogs, news, and updates from our company, make sure you're on our mailing list. Subscribe for email updates.