Optimal two-stage design for a population pharmacokinetic study in children
Cyrielle Dumont (1,2), Marylore Chenel (2), France Mentré (1)
(1) UMR 738, INSERM, University Paris Diderot, Paris, France; (2) Department of Clinical Pharmacokinetics, Institut de Recherches Internationales Servier, Suresnes, France
Objectives: Pharmacokinetic (PK) studies in children are mainly analysed by nonlinear mixed effect-models [1,2] as recommended in guideline [3]. Approaches based on the Fisher information matrix (MF) [4] can be used to optimize their designs and are based on a priori information. However, PK data in children are often not available and methods as allometry or PBPK are used to predict ‘initial’ PK parameters. Therefore adaptive designs [5,6], among which two-stage designs, are useful to provide some flexibility. Our aims are: i) to analyse concentration-time data obtained from PBPK simulations in children after oral absorption of a drug X in development; ii) to develop and evaluate the impact of two-stage designs when children ‘true’ parameters are different from initial ones.
Methods: Concentration of drug X are generated by PBPK with SIMCYP [7] for 100 children from 6 months to 18 years. PK model and parameter estimates are obtained using NONMEM 7 [8]. Optimal one-stage and two-stage designs are derived assuming a total of N=60 children with identical 5 sampling times using PFIM [9,10]. The two-stage design is defined as follows. From initial parameters Ψ0, we optimize design ξ1 for the first cohort of N1 children. From the obtained data set Y1, population parameters Ψ1 are estimated. The design ξ2 of the second cohort of N2 children, is optimized using a combined information matrix. The study is then performed in N2 children with design ξ2 and, finally, data Y1 and Y2 obtained from each cohort are analysed together. We evaluated one and two-stage designs for drug X assuming that the true CL is moderately or strongly higher than the initial one. A simulation study is on-going to evaluate the impact of the size of each cohort on the precision of population parameters estimation.
Results: The PK model is a 2-compartment model with first-order absorption. One-stage design from initial parameters shows a loss of efficiency when true CL is different. The two-stage design, with N1=N2=30, allows to partly compensate this loss of information and we show that the second stage design ξ2 is different from ξ1. The respective size of each cohort influences the gain in efficiency of the two-stage versus the one stage design.
Conclusions: Two articles in other contexts [11,12] discussed that two-stage designs could be more efficient than fully adaptive designs. Two-stage designs, which are easier to conduct, is a good alternative for designing PK studies in children.
References:
[1] Mentré F, Dubruc C, Thénot J.P. Population pharmacokinetic analysis and optimization of the experimental design for Mizolastine solution in children. Journal of Pharmacokinetics and Pharmacodynamics, 2001; 28(3): 299-319.
[2] Tod M, Jullien V, Pons G. Facilitation of drug evaluation in children by population methods and modelling. Clinical Pharmacokinetics, 2008; 47(4): 231-243.
[3] EMEA. Guideline on the role of pharmacokinetics in the development of medicinal products in the paediatric population. Scientific guideline, 2006.
[4] Mentré F, Mallet A, Baccar D. Optimal design in random-effects regression models. Biometrika, 1997; 84(2): 429-442.
[5] Foo L K, Duffull S. Adaptive Optimal Design for Bridging Studies with an Application to Population Pharmacokinetic Studies. Pharmaceutical Research, 2012; in press.
[6] Zamuner S, Di Iorio V L, Nyberg J, Gunn R N , Cunningham V J, Gomeni R and Hooker A C. Adaptive-Optimal Design in PET Occupancy Studies. Clinical Pharmacology & Therapeutics, 2010; 87 (5): 563–571.
[7] Jamei M , Marciniak S, Feng K, Barnett A, Tucker G, Rostami-Hodjegan A. The Simcyp Population-based ADME Simulator. Expert Opinion of Drug Metabolism and Toxicology, 2009; 5(2): 211-223.
[8] Dumont C, Chenel M, Mentré F. Design optimisation of a pharmacokinetic study in the paediatric development of a drug. Population Approach Group in Europe, 2011; Abstr 2160 [www.page-meeting.org/?abstract=2160].
[9] Bazzoli C, Retout S, Mentré F. Design evaluation and optimisation in multiple response nonlinear mixed effect models: PFIM 3.0. Computer Methods and Programs in Biomedicine, 2010; 98(1): 55-65.
[10] www.pfim.biostat.fr.
[11] Federov V, Wu Y, Zhang R. Optimal dose-finding designs with correlated continuous and discrete responses. Statistics in Medicine, 2012; 31(3): 217-234.
[12] Chen T.T. Optimal three-stage designs for phase II cancer clinical trials. Statistics in Medicine, 1997; 16(23): 2701-2711.
