PBPK modeling of drug-drug interaction in a regulatory submission

In oncology drug development, where patient conditions limit traditional clinical pharmacology studies, physiologically-based pharmacokinetics (PBPK) modeling emerges as a powerful solution to unravel complex drug-drug interactions (DDIs).

27.08.2026

Case

In this case study, a Pharmetheus embedded scientist applied PBPK modeling to address questions of interest in the clinical development of brigimadlin, a candidate drug targeting dedifferentiated liposarcoma. Brigimadlin presented unique challenges in predicting its behavior both as a DDI object and precipitant. The chosen middle-out PBPK modeling strategy involved integrating physicochemical properties and in vitro data, model refinement by leveraging clinical data, model qualification by reproducing published DDI outcomes, and simulations to predict brigimadlin’s potential DDI risks.

The PBPK model accurately predicted brigimadlin’s pharmacokinetics across dosing regimens and DDI scenarios. The prospective simulations successfully addressed the uncertainty of the DDI risks of complex enzyme–transporter interactions, crucial for regulatory submission. In summary, Pharmetheus’s modeling effort delivered the mechanistic clarity needed for regulatory confidence.

Background and challenge

The study of drug-drug interactions in oncology clinical trials is uniquely challenging as patients may require concomitant medications as part of their standard of care, making it difficult to isolate the effects of specific drug combinations. There are also ethical and safety concerns with prospective DDI studies, which might involve intentionally underdosing or overdosing a candidate anticancer drug in seriously ill patients.

Brigimadlin is an oral MDM2 p53 antagonist developed by Boehringer Ingelheim for the treatment of dedifferentiated liposarcoma. The current chemotherapy options for these patients have limited effectiveness with short duration. The pharmacological properties of brigimadlin introduced complexities in the documentation of potential drug-drug interactions for the regulatory submission package.

Brigimadlin has multiple enzyme- and transporter-mediated pathways, presenting challenges in predicting its behavior both as a DDI object and precipitant. Brigimadlin is a substrate of hepatic uptake transporters OATP1B1 and OATP1B3 and primarily metabolized by UGT1A3 (85%) with minor CYP3A4 (15%) contribution.

Additional complexity was highlighted in the clinical DDI data with rifampicin that showed short-term inhibition and long-term induction effects on CYP3A4 and OATP1B.

The questions of interest for the regulatory submission were:

Should the brigimadlin dose regimen be adapted when co-administered with OATP1B1, OAT1B3 and CYP3A4 precipitants?
What is the potential DDI risk for CYP3A4 substrates when co-administered with brigimadlin?

Client background

Boehringer Ingelheim is a research-driven biopharmaceutical company with over 53,500 employees worldwide.

Our solution

The goal for the embedded Pharmetheus scientist was to create a brigimadlin PBPK model that accurately described clinical data while maintaining mechanistic alignment. A middle-out modeling strategy was applied.

Category	Details
Strategy	Model and simulate drug-drug interactions using a middle-out PBPK workflow
Tools used	Open Systems Pharmacology Suite, PK-Sim® and MoBi®
Key service	Pharmetheus scientist with expertise in PBPK
Timeline	July 2023–April 2025

Model development and refinement

An overview of the modeling workflow for development, qualification and application oft he brigimadlin PBPK model is described and shown in the figure below:

brigimadlin-pbpk-drug-model-development-and-qualification-pharmetheus-2048x1280.png (1)

The brigimadlin structural base PBPK model was developed (top panel) using Physicochemical data, in vitro data, and pre-clinical observations to construct the disposition (i.e., distribution and elimination) in the PBPK model of brigimadlin after both intravenous (IV) and oral (PO) administrations. The elimination of the compound was structurally implemented via CYP3A4 and UGT1A3 metabolism, organic anion transporting polypeptide (OATP)1B uptake into the liver, and passive glomerular filtration. Clinical data was leveraged to optimize the key parameters tissue distribution, intestinal permeability, glomerular filtration rate, OATP1B1/B3 kinetics, as well as CYP3A4 and UGT1A3 metabolic fractions.

Model qualification

Published PBPK models for rifampicin (a CYP3A4 inducer) and midazolam (a CYP3A4 substrate) were used together with the model to reproduce clinical DDI outcomes, and to qualify the model for the questions of interest and context of use.

Model application and prospective simulations

The qualified model was used to predict DDIs with itraconazole (a strong CYP3A4 precipitant), clarithromycin (a CYP3A4-OATP1B1/-B3 inhibitor), and ethinylestradiol (a CYP3A4 substrate).

Overview of the implemented DDIs in the brigimadlin and the compounds used for development, qualification and application.

Results and outcomes

The developed PBPK model accurately predicted brigimadlin’s pharmacokinetics across dosing regimens and DDI scenarios. The potential brigimadlin DDI risks were successfully assessed by prospective simulations.

DDI Scenario	Prediction
CYP3A4 inhibition/induction	Brigimadlin exposure not impacted
OATP1B1/B3 inhibition	Brigimadlin exposure not impacted
Combined CYP3A4 and OATP1B1/B3 inhibition	Brigimadlin exposure weakly impacted
Brigimadlin as CYP3A4 precipitant	Unlikely to significantly affect substrates

Conclusions

PBPK modeling successfully supported brigimadlin’s regulatory submission preparation by assessing potential DDI risk through complex enzyme–transporter interactions.
This approach provides mechanistic insight beyond traditional PK modeling, particularly valuable in oncology where clinical DDI data are limited.
PBPK modeling bridges the gap between limited clinical data and regulatory requirements.

Related publications

Kanacher T, Lindauer A, Mezzalana E, et al. A Physiologically-Based Pharmacokinetic (PBPK) Model Network for the Prediction of CYP1A2 and CYP2C19 Drug-Drug-Gene Interactions with Fluvoxamine, Omeprazole, S-mephenytoin, Moclobemide, Tizanidine, Mexiletine, Ethinylestradiol, and Caffeine. Pharmaceutics. 2020;12(12):1191.

Sofia-Friberg-Hietala-and-Giovanni-Smania-Pharmetheus-2048x1366.jpg

Whether evaluating early disposition mechanisms or building robust DDI models for regulatory filing, our experts are here to design PBPK strategies tailored to your clinical questions.

Ready to optimize your PBPK modeling and regulatory filing strategy?

Get in touch with us now

19/06/23