(W0930-10-68) Physiologically Based Pharmacokinetic (PBPK) Modeling to Predict the Pharmacokinetics of Pantoprazole According to CYP2C19 Genetic Polymorphism

Purpose: Pantoprazole, a proton pump inhibitor (PPI), is widely prescribed to alleviate the symptoms related to peptic ulcer disease (PUD) and gastroesophageal reflux disease (GERD). Pantoprazole is primarily metabolized by CYP2C19, to a lesser extent CYP3A4. CYP2C19 is genetically polymorphic and the polymorphism affects the exposure of pantoprazole. The aim of the present study was to establish the pantoprazole PBPK model according to CYP2C19 genetic polymorphism.

Methods: Whole-body PBPK model for pantoprazole was established based on Simcyp^® v21 simulator. Total of 29 clinical pharmacokinetic data for pantoprazole were collected and used to establish the PBPK model. Intrinsic clearance (CL_int) and CYP2C19 abundance were fitted to match the plasma concentration-time profiles of pantoprazole in different CYP2C19 phenotypes. Other parameters were obtained from drug databases or predicted by Simcyp^® simulator. Simulations were based on the healthy virtual populations (n=100) that the demographic data are similar to the clinical trials. Model was numerically and visually evaluated based on the comparison of the predicted pharmacokinetic parameters and profiles to the observed results.

Results: Predicted plasma concentration-time profiles were visually similar to the observed profiles in different CYP2C19 phenotype groups. In the model development, predicted AUC values of pantoprazole in CYP2C19IM, RM, and UM phenotypes were altered by 1.37-, 0.93-, and 0.56-fold compared to CYP2C9EM phenotype, respectively. The range of ratio scores for AUC and C_max were 0.73-0.98 and 0.63-0.74 respectively, which were satisfied in the two-fold error range. In the model validation, the range of ratio scores for AUC and C_max were 0.51-2.24 and 0.61-1.18, respectively.

Conclusion: The PBPK model of pantoprazole in different CYP2C19 phenotypes was appropriately developed and validated. The present model may contribute to the decision-making for optimal dose administration strategy