Nastya Kassir (1), Mohamad-Samer Mouksassi (1), Nathalie Gosselin (1), and JF Marier (1)
(1) Pharsight - A Certara Company, Montreal, Quebec, Canada
Objectives: Paediatric drug development is challenging and unique in several aspects. FDA recently provided new recommendations for designing optimal paediatric studies [relative standard error (RSE) and 90% confidence intervals (CI) <20%]. A modelling and simulations framework was developed to simultaneously determine an optimal pharmacokinetic (PK) sampling strategy and sample size of paediatric subjects that would result in robust PK parameters and ultimately identify a dose level that would result in drug exposure within the targeted range of efficacy and safety of the product.
Methods: The framework consists in the following steps: 1) perform simulations using a population PK model (Trial Simulator®) to determine dose levels that would result in exposure within the targeted range in different age groups, 2) optimize sample size of patients and PK sampling (2 or 3 blood draws) using WinPOPT®, 3) confirm the best study design using a simulation-fitting approach, which accounts for the covariate distribution in the target population, 4) determine the optimal design that provides RSE and CI <20% with the smallest sample size and limited number of PK samplings across age groups.
Results: The framework was used to optimize a paediatric study for a small molecule currently under development. A population PK model including an allometric function was used to predict PK parameters in paediatric patients. Simulations were performed in all age groups (1-2, 2-4, 4-6, 6-8, and 8-12 years) to determine dose levels that would result in drug exposure within the targeted range of efficacy and safety. A total of 25, 30, 40 and 50 subjects between 1 and 12 years were randomly selected and concentrations for the optimal time points (1 and 6 h vs. 1, 4 and 6 h) were predicted with the population PK model. Sampling strategies involving 2 and 3 samples and 25 patients (5 per each age group) resulted in RSE of CL/F and Vc/F with their relative CI <20%.
Conclusions: A modelling and simulations framework was developed to rapidly identify optimal study designs in paediatrics patients and allowed a precise estimation of RSE to ensure robustness of PK parameters and meet FDA's requirements for clinical studies in paediatrics. Although EMA did not provide regulatory requirements to define the level of robustness required in paediatric studies, the above modelling and simulations framework may be considered for optimizing paediatric studies in Europe.
References:
[1] Gobburu J: How to double success rate of pediatric trials? http://www.slideshare.net/JogaGobburu/how-to-double-success-rate-of-pediatric-trials.
Reference: PAGE 21 (2012) Abstr 2478 [www.page-meeting.org/?abstract=2478]
Poster: Paediatrics