S. Samant,1 X. Jiang,1 R. B. Horenstein,2 A. R. Shuldiner,2 L. M. Yerges-Armstrong,2 L. A. Peletier,3 X. Zhang,1 M. N. Trame,1 L. J. Lesko,1 S. Schmidt1; 1Center for Pharmacometrics & Systems Pharmacology, University of Florida, Orlando, FL, 2Division of Endocrinology, Diabetes and Nutrition, University of Maryland School of Medicine, Baltimore, MD, 3Mathematical Institute, Leiden University, Leiden, Netherlands
BACKGROUND: Clopidogrel (CLOP), a widely used antiplatelet agent, is associated with high between-subject variability in response to antiplatelet therapy. Multiple intrinsic and extrinsic factors, like genetic polymorphisms in cytochrome P450 (CYP) and carboxyl esterase 1 (CES1) enzymes, age, body mass index and co-morbidities have been associated with high heterogeneity in response to CLOP treatment. The objective of this study was to use a modeling and simulation approach to guide the identification of clinically important sources of variability in CLOP responsiveness.
METHODS: Previously developed physiology-based and genotype-directed model for CLOP and its active metabolite (CLOP-AM) (CPT, 2014, 95, Suppl.1, S102, OIII-3) was evaluated using both dimensional and Sobol analyses based on parameter estimates obtained from PGXB2B crossover study PK/PD data from 18 healthy adults (NCT01341600). The final model was mathematically reduced to capture the sources of variability and to make it applicable for fitting to clinical data.
RESULTS: CES1 enzyme, with approximately 25-fold higher intrinsic clearance compared to CYP enzymes, was identified to have highest influence on the variability in PK of CLOP and CLOP-AM and subsequent platelet aggregation. This information was carried forward during model reduction and the simplified model was able to successfully describe CLOP and CLOP-AM data from the PGXB2B and PAPI (NCT0079936) study as well as the associated variability.
CONCLUSION: This analysis provides an example for how a single unifying model integrating information from clinical, in vitro and in silico experiments can be used to set up quantitative decision support tools that can guide the identification of clinically important sources of variability in treatment response.