High-Throughput
Screening and
ADMET-PBPK

High-throughput physiologically based pharmacokinetic (HT-PBPK) modeling enables rapid, scalable prediction of chemical and drug kinetics across large datasets. By integrating in silico tools such as QSAR, cheminformatics, and curated in vitro data, HT-PBPK provides a systematic approach to anticipate exposure and internal concentrations without requiring compound-specific in vivo data. These approaches are key to both Next-Generation Risk Assessment (NGRA) and Model-Informed Drug Development (MIDD), supporting early-stage decision-making, candidate de-risking, and prioritization based on exposure-driven mechanisms.

At ESQlabs, we leverage open-source modeling platforms and custom automation pipelines to deliver quantitative, mechanistically informed predictions of ADME-T behavior. These applications support regulatory alignment, screening, and data integration across sectors ranging from industrial chemicals to pharmaceuticals.

What we can offer

Our high-throughput PBPK and ADME-T services are designed to support both broad-scale chemical screening and project-specific needs. Whether you’re evaluating thousands of compounds for regulatory prioritization or selecting the right candidate during early drug development, we offer a flexible and scalable approach to simulate exposure, interpret in vitro data, and inform kinetic decision-making. Through automation, open-source modeling, and integration of predictive algorithms, we ensure efficient, reproducible, and mechanistically grounded outcomes.

  • Automated HT-PBPK pipelines for rapid simulation of large chemical libraries using batch processing tools integrated with PK-Sim®, MoBi® and R
  • QSAR-driven parameter estimation for absorption, distribution, metabolism, and elimination inputs (e.g., logP, fup, metabolic clearance)
  • Integration with ADME databases and high-throughput in vitro screening data to refine input parameters and improve prediction fidelity
  • Large-scale exposure predictions across species, life stages, and exposure routes to inform risk prioritization and regulatory thresholds
  • Candidate de-risking and prioritization strategies for MIDD through exposure-led simulations in early development
  • Flexible modeling frameworks for screening-level assessments or detailed compound-specific refinements
  • Modular model design enabling future integration with QST, effect modeling, or toxicodynamic platforms
  • Quantitative IVIVE applications to translate in vitro assay data into internal dose metrics across thousands of chemicals

Related publications and initatives

Systematic evaluation of high-throughput PBK modelling strategies for the prediction of intravenous and oral pharmacokinetics in humans
High-Throughput PBPK Framework in R using Open Systems Pharmacology Software for Anti-Tuberculosis Drug Development
See all publications 

Meet the Team

Diane Lefaudeux

Scientist
Consultant
Google Scholar

Diane Lefaudeux is an interdisciplinary scientist with a strong drive to understand complex mechanisms and in particular those arising from biological systems.

Before joining ESQlabs, she worked at Novadiscovery where she developed PBPK-QSP models on various therapeutic areas to predict outcomes using virtual populations.

Diane obtained her Master’s degree in General Engineering from École Centrale Paris, France, as well as in Control Systems Engineering and Systems Biology from University of Stuttgart, Germany.

 

Marco Siccardi

Principal Scientist
Lead Toxicology & PBPK
Google Scholar

Marco is a Clinical Biologist by training with a PhD in molecular pharmacology and PK/PD modelling. He spent over 15 years at the University of Liverpool working on the topic of pharmacogenetics and in developing PBPK approaches for the optimisation of drug delivery, including HIV therapy optimisation.

Marco has most recently been working with CROs in taking this approaches for modelling and simulation approaches and PKTK (Systems Toxicology) models across a number of disease areas.

Marco leads the Systems Toxicology team with the aim to promote collaborative innovation and to develop novel modeling approaches to streamline the toxicological assessment.

Pavel Balazki

Senior Scientist
Lead Software ToolChain
Google Scholar

Pavel Balazki is a bioinformatitian by training with over 10 years of experience in PB(P)K and QSP/T modeling and software development.

Pavel’s modeling expertices is focused on developing complex PB-QSP/T disease models, such as the Diabetes Platform or the Thyroid Hormones Toxicology Platform.

Pavel leads the Software ToolChain team at ESQlabs, developing new technologies to enable an integrated ecosystem of software, models, and solutions.