Design optimisation of a pharmacokinetic study in the paediatric development of a drug
Cyrielle Dumont (1), 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, Paris, France
Objectives: In the context of the Paediatric Investigation Plan  in place since 2007, pharmacokinetic (PK) studies in children are strongly supervised. Indeed, the blood volume, which can be taken in a child, is limited and studies are analysed by NonLinear Mixed Effect Models (NLMEM) [2,3]. The choice of the PK design has an impact on the precision of parameters estimates. To that end, approaches based on the evaluation of the Fisher information matrix (MF)  are used and implemented in software packages, as PFIM [5,6] in R. Our aim was to optimise the PK sampling time design for the paediatric trial of a drug X in development, taking into account clinical constraints. To evaluate designs, a priori information is needed. For the present study, we used ‘simulated' plasma concentration in children, obtained via the SIMCYP software from knowledge of the drug in adults and its physico-chemistry properties [7,8].
Methods: The molecule breaks down into parent drug and its metabolite, which is active. A first work was performed to find the starting dose of the drug X in the paediatric trial . The PK model, obtained from the ‘simulated data' with the software NONMEM, using a joint model for the parent drug and its metabolite as in adults, was a model with four compartments. The PK design, for a future study in 82 patients receiving a single dose of the drug, was then optimised by PFIM, considering several clinical constraints, as the number of groups, the timing of the clinical examination and the duration of stay at the hospital. Limit of quantification (LOQ) was not taken into account for design optimisation, as the simulated observation had no LOQ. For final evaluation of the proposed design, we used an approach based on the simulated proportion of LOQ at each sampling time to predict data below LOQ for the metabolite.
Results: Reaching a compromise between PFIM results and clinical constraints, the optimal design is composed of four samples at 0.1, 1.8, 5 and 10 h after drug injection. Concerning LOQ, we showed its limited influence on the design.
Conclusions: PFIM was a useful tool to find an optimal design in children, considering clinical constraints. Even if it was not forecast in the initial design at the beginning, it was necessary to include a late time at 10 hours for all children. Finally, we proposed to carry out an adaptive design after the inclusion of 20 children.
 EMEA. Guideline on the role of pharmacokinetics in the development of medicinal products in the paediatric population. Scientific guideline, 2006.
 Tod M, Jullien V, Pons G. Facilitation of drug evaluation in children by population methods and modelling. Clinical Pharmacokinetics, 2008; 47(4): 231-243.
 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.
 Mentré F, Mallet A, Baccar D. Optimal design in random-effects regression models. Biometrika, 1997; 84(2): 429-442.
 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.
 Turner D. Simcyp Population-based ADME Simulator, 2008.
 Brendel K, Gaynor C, Dumont C, Blesius A, Chenel M. Using Modelling & Simulation techniques to optimise the design of a paediatric PK/PD study. Population Approach Group in Europe, 2010; Abstr 1695 [www.page-meeting.org/?abstract=1695].