Coproporphyrin I PBPK model in the open systems pharmacology suite enables precise prediction of OATP 1B- mediated drug interactions through calibrated In vivo inhibition constants

PBPK & PBBM

Program strategies and efficiencies

Trial design and analysis

Regulatory interactions

Front. Pharmacol. 2026;17:1828033

Kanacher T, Pellowe M, Kovar C, Eriksson J, Wu J, Busse D, Gómez-Mantilla JD, Stopfer P, Ince I, Huang F

When new drugs inhibit liver transporters (OATP1B1/3), it can cause harmful drug-drug interactions (DDIs) leading to severe side effects like muscle damage (myopathy) or liver toxicity. Coproporphyrin-I (CP-I) is a key natural biomarker in the body used to evaluate how well these liver transporters are functioning.

However, computational models (PBPK models) used to simulate CP-I have historically been locked behind proprietary, paid software, creating a barrier to open access and collaboration across the pharmaceutical industry.

This is why researchers from Boehringer Ingelheim, with the support of Pharmetheus, established the first transparent, open-source CP-I PBPK model built within the freely available Open Systems Pharmacology (OSP) Suite.

Published in Frontiers in Pharmacology, the study contains:

First Open-Source CP-I PBPK Model: Implemented and qualified a whole-body PBPK model in PK-Sim®/MoBi® describing endogenous synthesis and OATP1B1/MRP2-mediated disposition of CP-I.
Calibrated In Vivo inhibition constants: Estimated precipitant-specific in vivo Ki,OATP1B parameters using clinical CP-I biomarker data, which yielded superior predictive accuracy relative to in vitro values alone.
Validated DDI network: Successfully predicted OATP1B-mediated interaction ratios for object drugs (atorvastatin, pitavastatin, and rosuvastatin) across precipitants (rifampicin, probenecid, and cyclosporine A), with over 90% of predictions falling within standard 2-fold acceptance limits.
Population baseline adaptation: Demonstrated that calibrating the baseline CP-I synthesis rate to study-specific populations accounts for baseline variations (e.g., pharmacogenomics/ethnicity) without requiring subject genotyping.

This integrated biomarker–PBPK workflow enables prospective DDI predictions for new chemical entities during early clinical development. This approach has the potential to de-risk OATP-mediated DDIs, reduce the necessity of dedicated clinical DDI studies, accelerate development timelines, and advance Model-Informed Drug Development (MIDD) strategies.