Overview of Korean Pharmacometrics Modeling Library and web-based pharmacometrics platform
Quyen Thi Tran (1), Chung Hee Lee (2), Min-Gul Kim (3), Minji Kim (3), Hansung Kim (4), Jung-Woo Chae (1), Hwi-yeol Yun (1)
(1) College of Pharmacy, Chungnam National University, Republic of Korea, (2) Clinical Research Division, National Institute of Food and Drug Safety Evaluation, Ministry of Food and Drug Safety, Cheongju, Republic of Korea, (3) Center for Clinical Pharmacology and Biomedical Research Institute, Jeonbuk National University Hospital, Jeonju, Republic of Korea, (4) Carpediem, Dongtan-si, Kyunggi-do, Republic of Korea
Objectives: Since, an increasing number of pharmacometrics (PM) models have been developed based on data in Korea, which leads to the needs of archiving and further using developed models. In addition, a growing demand for prediction pharmacokinetic (PK) parameters in an easy way leads to motivation of developing an easy-to-use platform where one can easily estimate PK parameters in diverse methods. As those of trends, Korean Pharmacometrics Model Library (KPML) has been developed by collaboration of Ministry of Food and Drug Safety (MFDS) and Korean Society for Clinical Pharmacology and Therapeutics (KSCPT) as a web repository and platform to store the developed PM models in Korea and provides PM platform for estimate PK parameters using various methods such as non-compartment (NCA), compartment model (CA), allometric scaling (AS), or in vitro in vivo extrapolation (IVVIE).
Methods: KPML was targeted to contain four sections with KPML, PM platform, PM training and Guidelines. KPML section is to store the model code and dataset, for example of developed PK/pharmacodynamic (PD) model and physiologically based PK/PD (PBPK/PD) model. PM training section is for repository of hands-on materials related to PM works where one can learn and practice it by themselves. Guidelines section contains updated PM guidelines from Food and Drug Administration, European Medicines Agency, Pharmaceuticals and Medical Devices Agency and MFDS. PM platform section provides an easy access to the PM approaches without barrier, where NCA, CA, AS and IVIVE were constructed to generates PM results in various steps of drug discovery. NCA was developed based on some R packages, including bear (v2.8.7) [1], NonCompart (v0.4.5), and ncar (v0.4.2) [2]. CA was developed using nlmixr package in R, which package was validated in the study of Schoemaker et al [3]. AS platform was established to extrapolate PK parameters from animal to human using single species scaling method or simple allometric scaling. IVIVE platform was developed to estimate in vivo hepatic clearance (CLh) from in vitro CLh using three different methods (e.g., well-stirred, parallel and dispersion method).
Results: We compile and upload all the available developed models and platform into webpage (http://repository.kscpt.org/) and provide this repository to the users with an open access online. Currently, 15 PK models, 4 PK/PD models and 5 PBPK/PD model were uploaded, and many pharmacometrics-related documents were available for reference in webpage. All four platforms are currently available for using as well. NCA and CA were developed using validated R packages and the requirement for those platforms is the dataset prepared in .csv file. NCA will provide results as same as Winnonlin, while results from CA will be same as from NONMEM. Validation result of AS and IVIVE showed a high success rate for human t1/2 prediction from 59% to 97% within 4-fold of observation values, while result of IVIVE showed prediction of in vivo CLh ~79% within 4-fold of observed values.
Conclusions: KPML with updated PM models is a precious repository which can be referred for developing new models or used in clinical practice. In addition, KPML with potential platforms would be an easy tool to predict precisely and reliably PK parameters. It can be widely applied to quickly predict and reduce time and unnecessary effort on prediction PK parameters.
References:
[1] “The data analysis tool for average bioequivalence (ABE) and bioavailability (BA).” [Online]. Available: http://pkpd.kmu.edu.tw/bear/. [Accessed: 11-May-2021].
[2] K. ISHIDA, S. WAKIMOTO, T. UEDA, and T. KANDA, “Development of R packages: ‘NonCompart’ and ‘ncar’ for noncompartmental analysis (NCA),” Proc. JSME Annu. Conf. Robot. Mechatronics, vol. 2018, no. 0, pp. 2A2-H13, 2018.
[3] R. Schoemaker et al., “Performance of the SAEM and FOCEI Algorithms in the Open-Source, Nonlinear Mixed Effect Modeling Tool nlmixr,” CPT Pharmacometrics Syst. Pharmacol., vol. 8, no. 12, pp. 923–930, 2019, doi: 10.1002/psp4.12471.
