Optimization of a Physiology Based Pharmacokinetic Model for Drug Interaction Predictions

Thursday, March 14, 2019

10:00 AM-12:00 PM

BIOMED Master's Thesis Defense

Title:
Optimization of a Physiology Based Pharmacokinetic Model for Drug Interaction Predictions with Application to Esketamine

Speaker:
Shayne Watson, Master's Candidate
School of Biomedical Engineering, Science and Health Systems
Drexel University

Advisor:
Andres Kriete, PhD
Teaching Professor
School of Biomedical Engineering, Science and Health Systems
Drexel University

Abstract:
Introduction: Drug-drug interactions (DDI) are commonly evaluated during development, as is recommended by health authorities. These interactions may alter the pharmacokinetics (PK), the toxicity, and efficacy of the drug. Cytochrome P450 (CYP) enzymes are commonly considered for DDIs as they play a major role in the metabolism of small molecules. However, the perpetrator drugs may interact with other proteins related to the ADME of the victim drug. A PBPK platform is ideal for exploring these interactions as it can take into account many aspects of ADME processes.

Method: A PBPK model will be built utilizing an iterative approach. The model was first built for intravenous (IV) esketamine. The oral model expanded on the IV model with only oral processes being optimized. Lastly, DDI was tested for perpetrator drugs, clarithromycin and itraconazole.

Results: Using the iterative process allowed for a successful development of a PBPK model describing both IV and oral esketamine. Total error for the final model reduced from 7.33 to 1.22 and from 3.52 to 2.18 for the IV and oral training datasets, respectively. DDI simulations showed an increase in AUC for clarithromycin and itraconazole, showing a lesser effect to itraconazole. AUC ratios were over predicted compared to observed values, but the simulations upheld the difference in effect of the two inhibitors.

Conclusion: The modeling process described in the paper, although timely, has many benefits. It decouples fitting of IV and oral data together, which aids identification of ADME processes and allows use of limited oral data.

Contact Information

Ken Barbee
215-895-1335
barbee@drexel.edu