Marc Lavielle1, Benjamin Ribba2. 1. Inria Saclay & CMAP, Ecole Polytechnique, University Paris-Saclay, Saint-Aubin, France. 2. Roche Pharma Research and Early Development, Roche Innovation Center Basel, Grenzacherstrasse 124, 4070, Basel, Switzerland. benjamin.ribba@roche.com.
Abstract
PURPOSE: For nonlinear mixed-effects pharmacometric models, diagnostic approaches often rely on individual parameters, also called empirical Bayes estimates (EBEs), estimated through maximizing conditional distributions. When individual data are sparse, the distribution of EBEs can "shrink" towards the same population value, and as a direct consequence, resulting diagnostics can be misleading. METHODS: Instead of maximizing each individual conditional distribution of individual parameters, we propose to randomly sample them in order to obtain values better spread out over the marginal distribution of individual parameters. RESULTS: We evaluated, through diagnostic plots and statistical tests, hypothesis related to the distribution of the individual parameters and show that the proposed method leads to more reliable results than using the EBEs. In particular, diagnostic plots are more meaningful, the rate of type I error is correctly controlled and its power increases when the degree of misspecification increases. An application to the warfarin pharmacokinetic data confirms the interest of the approach for practical applications. CONCLUSIONS: The proposed method should be implemented to complement EBEs-based approach for increasing the performance of model diagnosis.
PURPOSE: For nonlinear mixed-effects pharmacometric models, diagnostic approaches often rely on individual parameters, also called empirical Bayes estimates (EBEs), estimated through maximizing conditional distributions. When individual data are sparse, the distribution of EBEs can "shrink" towards the same population value, and as a direct consequence, resulting diagnostics can be misleading. METHODS: Instead of maximizing each individual conditional distribution of individual parameters, we propose to randomly sample them in order to obtain values better spread out over the marginal distribution of individual parameters. RESULTS: We evaluated, through diagnostic plots and statistical tests, hypothesis related to the distribution of the individual parameters and show that the proposed method leads to more reliable results than using the EBEs. In particular, diagnostic plots are more meaningful, the rate of type I error is correctly controlled and its power increases when the degree of misspecification increases. An application to the warfarin pharmacokinetic data confirms the interest of the approach for practical applications. CONCLUSIONS: The proposed method should be implemented to complement EBEs-based approach for increasing the performance of model diagnosis.
Entities:
Keywords:
model diagnostics; modeling and simulation; pharmacokinetics and pharmacodynamics
Authors: John C Panetta; Yiwei Liu; Hope D Swanson; Seth E Karol; Ching-Hon Pui; Hiroto Inaba; Sima Jeha; Mary V Relling Journal: Pediatr Blood Cancer Date: 2020-04-23 Impact factor: 3.167
Authors: Paolo Denti; Anneke C Hesseling; Louvina E van der Laan; Anthony J Garcia-Prats; H Simon Schaaf; Tjokosela Tikiso; Lubbe Wiesner; Mine de Kock; Jana Winckler; Jennifer Norman; Helen McIlleron Journal: Antimicrob Agents Chemother Date: 2018-01-25 Impact factor: 5.191
Authors: Darren Wethington; Olivia Harder; Karthik Uppulury; William C L Stewart; Phylip Chen; Tiffany King; Susan D Reynolds; Alan S Perelson; Mark E Peeples; Stefan Niewiesk; Jayajit Das Journal: J R Soc Interface Date: 2019-11-27 Impact factor: 4.118