Comment for best oral presentation award of JSSX meeting 2017
Sugiyama Laboratory, RIKEN, Japan / Daewoong Pharmaceutical, Korea
Above all, I would like to express my gratitude to Dr. Yuichi Sugiyama for providing this great opportunity to attend the 32nd JSSX annual meeting and present about “Physiologically Based Pharmacokinetic (PBPK) Analyses of Nonlinear Pharmacokinetics of Taxanes”. I would also like to give thanks to Kota Toshimoto who provided hands on help in establishing the PBPK model and performing analyses. Since last March in 2017, I have worked in the Sugiyama Laboratory at RIKEN as a visiting scientist from Daewoong Pharmaceuticals in Korea. Not only was I grateful to get a chance to work with a world leading pharmaceutical scientist, Dr. Sugiyama and his colleagues, it was also a great opportunity to gain experiences in PBPK modeling and simulation.
I had acquired a doctor’s degree in Pharmaceutics at University of Washington in 2013. During my Ph.D, I focused on characterization of drug transporters, particularly on organic cation transporters using in vitro systems as well as animal models. Although I had analyzed the data from animals using the modeling and simulation programs such as WinNonlin (or Phoenix), I had not have chance to develop a PBPK model. In the last decade, PBPK modeling has advanced rapidly and become increasingly accepted by regulatory agencies as an integral tool to quantitatively predict the pharmacokinetics of drugs. Therefore, I was very excited to obtain knowledge and training in PBPK modeling that can be used in drug discovery and development. The study I have been working on for the last 10 months is about PBPK modeling of paclitaxel.
Paclitaxel is one of the most widely prescribed taxanes for cancer treatment. Paclitaxel exhibits nonlinear pharmacokinetics, particularly when administered as an intravenous (IV) infusion over a short period of time. During the short IV infusion, nonlinear kinetics occurs possibly due to saturation of plasma protein binding, hepatic uptake, and metabolism or other mechanism like binding to excipient. The aim of this study was to elucidate the mechanisms of nonlinear pharmacokinetics of paclitaxel by establishing PBPK model.
Paclitaxel is formulated in polyoxyethylated castor oil (Cremophor EL) due to poor aqueous solubility and CrEL is known to form micelles and entrap drugs inside them. Previously, it was reported that unbound fraction of paclitaxel in plasma decreases as the CrEL level increases. Consistent with previously reported data, our PBPK analysis demonstrated that unbound fraction of paclitaxel decreases with increasing CrEL amounts. In addition, when the effect of CrEL was not considered in plasma, paclitaxel could not exhibit nonlinear kinetics, suggesting that decreased unbound fraction by CrEL entrapment results in nonlinearity of paclitaxel.
Our PBPK analyses showed that the dose-dependent decrease in unbound fraction could not be reproduced when only plasma protein binding was considered (no CrEL effect in plasma). Similarly, unbound fraction of blood cell did not exhibit the dose-dependent decrease. These results suggest that saturation of plasma protein and blood cell binding is unlikely the mechanism of nonlinear kinetics of paclitaxel.
In vitro Km value for hepatic transporter estimated in human hepatocyte was 5 μM whereas best-fitted value in our PBPK model was 0.02μM, which was approximately 10 fold lower than the reported maximum unbound concentration (0.3μM). These results indicate that in vivo Km for hepatic uptake of paclitaxel could be different from the in vitro Km. In our PBPK analyses, nonlinear kinetics could be reproduced regardless of Km values. Nevertheless, the PBPK simulations with Km value of 0.02μM described better the clinical observations with higher degree of nonlinearity compared with Km of 5 μM. Together, these results suggest that saturation of hepatic uptake may contribute to some extent to nonlinear kinetics of paclitaxel.
Our established PBPK model of paclitaxel could describe the observed data and decreased unbound concentrations of paclitaxel by CrEL micellar entrapment can be the major mechanism of nonlinear pharmacokinetics of paclitaxel. Using the PBPK model established in this study, further studies including PBPK-pharmacodynamics (PD) analysis and virtual clinical study are warranted to investigate interindividual variability in pharmacokinetics and drug response, and determine optimal dosage and dosing regimen for cancer treatment.
The journey has not been easy, but I believe all the hard work and guidance provided by Dr. Sugiyama and Toshimoto-san has paid off well in terms of our study outcomes. The PBPK model of paclitaxel required a lot of effort to establish because multiple binding parameters needed to be incorporated and optimized in order to best describe the pharmacokinetics of paclitaxel. I hope that our established paclitaxel PBPK model can be useful to predict drug disposition and response and eventually to inform individual cancer patients of their optimal dose/regimen for effective treatment.