In Silico Modelling

#

In silico modeling—including Physiologically Based Pharmacokinetic (PBPK) models and digital twins—simulates drug behavior in the human body, enabling virtual clinical trials and predictive toxicology. By integrating computational methods with biological data, these approaches reduce reliance on animal testing, optimize dosing regimens, and accelerate the development of safe and effective therapies.

Physiologically Based Pharmacokinetic Modeling

#

PBPK models integrate in vitro data on absorption, distribution, metabolism, and excretion with physiological parameters to predict internal human exposure. The models correctly estimated systemic exposure of caffeine and coumarin, demonstrating that model-informed approaches can replace in vivo toxicokinetics. This impact has enabled virtual clinical trials and optimized dosing regimens without animal testing.
Advancing drug development with “Fit-for-Purpose” modeling informed approaches
SCCS Notes of guidance for the testing of cosmetic ingredients and their safety evaluation
The margin of internal exposure (MOIE) concept for dermal risk assessment based on oral toxicity data - A case study with caffeine
AI-driven virtual cell models in preclinical research
Organs-on-Chips in Drug Development: Engineering Foundations, Artificial Intelligence, and Clinical Translation

Bioequivalence Bridging for Tofacitinib

#

Pharmacokinetic/pharmacodynamic modeling was used to bridge the immediate-release formulation of tofacitinib to a new extended-release version. The computational model successfully established bioequivalence, satisfying regulatory safety and efficacy requirements without further animal testing. This supported FDA approval while avoiding new Phase 3 clinical trials, accelerating patient access to the formulation.
Integrating Clinical Variability into PBPK Models for Virtual Bioequivalence of Single and Multiple Doses of Tofacitinib Modified-Release Dosage Form Virtual Bioequivalence Assessment of Tofacitinib Once Daily Modified Release Dosage Form in Pediatric Subjects

Digital Twins for Clinical Trial Simulation

#

Digital twins are virtual representations of individuals that integrate clinical, genetic, and environmental data to revolutionize clinical trial design. Simulation of treatment strategies before patient enrollment has been shown to reduce both risks and costs. This technology could eventually eliminate the need for many traditional clinical trials by predicting patient-specific responses.
Increasing acceptance of AI‐generated digital twins through clinical trial applications
The Use of Digital Healthcare Twins in Early-Phase Clinical Trials: Opportunities, Challenges, and Applications
Enhancing randomized clinical trials with digital twins What are NAMs?

Quantitative Systems Pharmacology Models

#

QSP models combine mechanical simulations of physiology with molecular signaling pathways to predict immunogenicity and pharmacokinetics of complex biologics. The FDA highlighted QSP as a vital tool to reduce reliance on animal testing for “what-if” development scenarios. The use of these models has accelerated the development of biologics and personalized medicine.
FDA animal testing phaseout urges AI-based trial alternatives, organoids, other “NAMs”
Beyond lab animals

AlphaFold Predicts Protein Structures

#

AI-based prediction of protein 3D structures from amino acid sequences transformed structural biology and drug target identification. The open-access database covers over 200 million structures with atomic accuracy, even for architectures not previously discovered in animal research. This provides data that previously required years of laboratory work, significantly supporting the efficiency of NAM workflows.
Highly accurate protein structure prediction with AlphaFold