A Bayesian K-PD model for synergy: A case study.

Area of Science:

• Pharmacology
• Pharmacokinetics
• Pharmacodynamics

Background:

• Drug-drug interactions (DDIs) significantly impact drug efficacy and safety.
• In vitro studies are common for pharmacodynamic DDIs, but in vivo models offer more relevant insights.
• Understanding how coadministered drugs alter individual pharmacokinetic and pharmacodynamic profiles is crucial.

Purpose of the Study:

• To develop and apply an in vivo model for assessing pharmacodynamic drug-drug interactions.
• To utilize an indirect response model where one drug's pharmacokinetics influence another's pharmacodynamics.
• To implement a Bayesian inference framework for analyzing sequential dose-response experiments, incorporating historical data.

Main Methods:

• An indirect response model was employed to simulate the time-dependent changes in a safety biomarker.
• Virtual pharmacokinetic profiles of one compound were used to drive the pharmacodynamic effects of a coadministered compound.
• Sequential experiments were conducted across various dose level combinations.
• A Bayesian inference framework was utilized, integrating results from a historical dose-response experiment.

Main Results:

• The indirect response model successfully captured the dynamic interplay between coadministered drugs in an in vivo setting.
• The Bayesian approach allowed for the incorporation of prior knowledge from historical experiments, enhancing parameter estimation.
• The model demonstrated the ability to predict changes in a safety biomarker driven by the pharmacokinetic profile of another compound.

Conclusions:

• This study presents a robust in vivo modeling approach for evaluating pharmacodynamic drug-drug interactions.
• The integration of Bayesian inference with historical data provides a powerful tool for analyzing complex DDI scenarios.
• The developed model can aid in optimizing combination therapies and predicting potential safety concerns.