Sophie Peigné, Sylvain Fouliard, Adrien Tessier and Marylore Chenel
Institut de Recherches Internationales Servier,
Introduction: A safety dose ranging study that included a 3 days treatment period with 7 PK samples collected each day was conducted in 51 paediatric patients with drug S (marketed in adult) with the objectives to build a population PK model in the paediatric population, to quantify the variability and identify the sources of variability and finally to compute derived secondary PK parameters in order to perform dose recommendation for the following phase III studies.
Methodology: During the paediatric development, a first step consisted in performing an identifiability analysis to evaluate whether the 2 compartmental model in adult population could be applied to the available paediatric data [1]. Then a population modelling approach was used to characterize the PK of drug S in the paediatric population and a covariate analysis was performed testing the covariates found in adult as a starting point and some well-known features of the paediatric population i.e. age and weight effect [2]. More specifically, an optimal-design based approach was used in order to assess whether the proposed study design with the number of subjects would allow the characterization of a gender effect, a covariate that was found in adult.
Finally, using the PK model developed, simulations were performed based on the assumption that the relationship exposure/efficacy was similar between adult and children population in order to perform dose recommendation.
Results: The data did not support any influential covariates, as opposed to the adult model. It was shown that a poor power was achieved if the magnitude of this effect was 2-fold for gender effect, and satisfactory for a 4-fold effect. However, the limitations of data that were available for the paediatric population analysis, questions the broader validity of covariate analysis.
The identifiability analysis based on simulations and design evaluation showed that the PK sampling times were unfortunately not optimized because of the high amount of BLQ and the data only supports a 1 compartment model.
Individual PK parameters derived from the final PK model were calculated at each dose. A direct comparison was performed with adult values and showed no difference between the two populations.
Conclusion: This model allowed to perform dosing recommendation for the next phase III study and will be further refined with the new data coming, especially regarding covariate effect assessment.
References:
[1] Retout S, Duffull S, Mentré F. Development and implementation of the population Fisher information matrix for the evaluation of population pharmacokinetic designs. Computer Methods and Programs in Biomedicine 2001; 65: 141-51.
[2] Anderson BJ, Holford NH. Mechanistic basis of using body size and maturation to predict clearance in humans. Drug Metab Pharmacokinet. 2009;24(1):25-36.
Reference: PAGE 25 (2016) Abstr 6038 [www.page-meeting.org/?abstract=6038]
Poster: Drug/Disease modeling - Paediatrics