Quyen Tran (1), Woo Jin Jung (1), Seon Jong Park (1), Chung Hee Lee (1), Sung Yoon Yang (1), Lien Ngo (1), Jung-Woo Chae (1), Hwi-yeol Yun (1)
(1) Chungnam National University, Daejeon, Korea
Objectives: The prediction of human pharmacokinetics is an extremely important work for further performing human clinical trial. Among various available methodologies, interspecies allometric scaling provides a simple and fast option to extrapolate pharmacokinetic parameters or drug dose in human. However, allometric scaling method is not a user-friendly approach for unprofessional people. Therefore, our aim in this study is to develop an easy-to-use R shiny web-based allometric scaling for prediction pharmacokinetic parameters in human.
Methods: Web-based allometric scaling is developed using R shiny platform, which provide a user-friendly interface. In our platform, allometric scaling method was classified into clearance (CL) and volume of distribution (Vd) and the number of species (n). In case of CL, if n equal to 1, (method 1) single species scaling method (CLhuman/kg = a*CLrat/kg) was used, in which a is coefficient, with value of 0.152, 0.41 and 0.407 for rat, dog and monkey, respectively [1]. If n is greater than 1, drugs was subsequently divided based on extraction ratio. With high extraction ratio (greater than 0.7), (method 2) simple allometry method [2] (CL = a*BWb) was used while with the other values of extraction ratio, (method 3) allometric scaling after normalization by CLint, in vitro [2] [Clhuman = CLanimal * (Human CLint, in vitro/Animal CLint, in vitro)] was applied. In case of volume of distribution, single species scaling method (Vd,u,human = Vd,u,animal*(fu,human/fu,animal) and simple allometry method (equation same as in CL) was applied following the number of species [3]. Accuracy of the prediction method using R shiny web-based was assessed by fold-errors (predicted value/observed value) and percentage of outliers falling out of the preselected fold-error ranges. Datasets for evaluation were extracted from study of Larry et al., (2005) (103 drugs) [4] and Lave et al., (1997) (10 drugs) [5].
Results: The values of fold-error and percentage of outliers of three methods using to predict CL and Vd showed that our platform works well comparing to parameter values in reference. For prediction CL using simple allometry method, 103 drugs were tested and result provided with 53% of fold-errors within two-fold. While prediction CL using allometric scaling after normalization by CLint, in vitro with 10 drugs extracted from Lave et al., study, 80% of fold-error fell into two-fold error range. A difference between predicted values and observed values was shown, however it has no significant statistic.
Conclusion: R shiny web-based allometric scaling provides an easy tool to predict precisely and reliably human pharmacokinetic parameters across species. It can be widely applied to quickly predict and reduce time and unnecessary effort on prediction pharmacokinetic parameters and first-in-human dose.
Acknowledgement: This research was supported by a grant (18182MFDS405 and 19182MFDS427) from Ministry of Food and Drug Safety in 2018 and 2019, respectively.
References:
[1] H. Tang, A. Hussain, M. Leal, M. Mayersohn, and E. Fluhler, “Interspecies prediction of human drug clearance based on scaling data from one or two animal species,” Drug Metab. Dispos., vol. 35, no. 10, pp. 1886–1893, 2007, doi: 10.1124/dmd.107.016188.
[2] P. Zou et al., “Applications of human pharmacokinetic prediction in first-in-human dose estimation,” AAPS J., vol. 14, no. 2, pp. 262–281, 2012, doi: 10.1208/s12248-012-9332-y.
[3] N. A. Hosea et al., “Prediction of human pharmacokinetics from preclinical information: Comparative accuracy of quantitative prediction approaches,” J. Clin. Pharmacol., vol. 49, no. 5, pp. 513–533, 2009, doi: 10.1177/0091270009333209.
[4] L. J. Jolivette and K. W. Ward, “Extrapolation of human pharmacokinetic parameters from rat, dog, and monkey data: Molecular properties associated with extrapolative success or failure,” J. Pharm. Sci., vol. 94, no. 7, pp. 1467–1483, 2005, doi: 10.1002/jps.20373.
[5] T. Lavé, P. Coassolo, and B. Reigner, “Prediction of hepatic metabolic clearance based on interspecies allometric scaling techniques and in vitro in vivo correlations,” Clin. Pharmacokinet., vol. 36, no. 3, pp. 211–231, 1999, doi: 10.2165/00003088-199936030-00003.
Reference: PAGE () Abstr 9421 [www.page-meeting.org/?abstract=9421]
Poster: Methodology - Estimation Methods