PK/PD sampling design optimization following a sustained release formulation of triptorelin using optimal experimental design
Elba Romero (1), Sebastian Ueckert (2), Joakim Nyberg(2), Josep-María Cendrós (3), Concepción Peraire (3), Rosendo Obach (3), Iñaki F. Trocóniz (1), and Andrew C. Hooker (2)
(1), Department of Pharmacy and Pharmaceutical Technology, School of Pharmacy,University of Navarra, Pamplona 31080, Spain; (2), Pharmacometrics Research Group, Department of Pharmaceutical Biosciences, Uppsala University;(3) ,Pharmacokinetics and Metabolism Department, Ipsen Pharma S.A., Sant Feliu de Llobregat, Barcelona, Spain.
Objectives: Sustained release administrations of drugs have improved long-term treatments for the patient. However, the design of clinical trials in such situations is complex due to the high number of samples required to obtain a precise prediction of the drug response. In the case of triptorelin (TPT) administered to supress testosterone (TST) levels in prostate cancer patients the study duration was 4 month and involved 32 samples per patient. Therefore, the aim of this work was to use optimal design theory to reduce the number of samples per patient based on a previously developed receptor-based pharmacokinetic/pharmacodynamic (PK/PD) model for the TST effects of TPT.
Methods: Pharmacometric Model: Data (CTPT and CTST) from a population of advanced several prostate cancer patients from four clinical trials were used to develop a receptor-based pharmacokinetic/pharmacodynamic model [1]. Data from one of these studies (n=24), where a slow release formulation was tested, were used to test the optimal design theory. Optimal Design: The PK/PD model was implemented in PopED [2] and optimization was performed using the D and Ds optimality criteria. For the later, only the PD parameters were considered interesting. Modified Fedorov Exchange algorithm with a grid of one sample per day and no replicates was used for the optimization.
Results: Comparable coefficients of variations as for the original design were obtained with 62.5 % optimal samples. Similarly, to achieve 100% efficiency only 10 samples with optimal time were needed. Focusing on the PD parameters using Ds optimality permitted a reduction to 87.5 % of the initial number of samples while maintaining 100% efficiency.
Conclusions: Using optimal design theory the number of samples in a long term sustained release trial could be substantially reduced, lowering both costs and patient burden.
References:
[1] PAGE 19 (2010) Abstr 1921 [www.page-meeting.org/?abstract=1921]
[2] Foracchia M, Hooker A, Vicini P, Ruggeri A. POPED, a software for optimal experiment design in population kinetics. Computer Methods and Programs in Biomedicine. 2004 Apr;74(1):29-46.
