Survey on the current use of optimal design approaches and the developments needed in adaptive optimal design for model based analysis performed amongst DDMoRe’s EFPIA members
F. Mentré (1), M. Chenel (2), E. Comets (1), J. Grevel (3), A. Hooker (4), M.O. Karlsson (4), M. Lavielle (5), J. Nyberg (4), I. Gueorguieva (6)
(1) UMR 738, INSERM and University Paris Diderot, Paris, France; (2) Clinical Pharmacokinetic Department, Institut de Recherches Internationales Servier, Suresnes, France; (3) BAST Inc Ltd. Nottingham, UK; (4) Uppsala University, Uppsala, Sweden; (5) INRIA, Paris, France; (6) Lilly UK, Global PK/PD department, Erl Wood Manor, Windlesham, Surrey, UK
Objectives: One objective of the DDMoRe project is to develop tools for adaptive optimal design using nonlinear mixed effect models (NLMEM). As a first step, a survey was conducted in the 10 EFPIA members of the DDMoRe project to identify current practices and perceived shortcomings.
Methods: TThe working group designed a survey that was sent to all EFPIA partners in October 2011. It was composed of two parts, part 1: state of the art on the use of optimal design methods in industry, part 2: requests for future developments in adaptive optimal design. Results were obtained in November 2011 from AstraZeneca, GSK, Lilly, Merck Serono, Novartis, Novo Nordisk, Pfizer, Roche, Servier, UCB Pharma.
Results: > Part 1 of the survey, investigating the current situation, showed that optimal design is being used by nearly all companies (9/10), mostly during phase 1 and 2 for PKPD. All currently available software (PFIM, POPED, PODES, POPT) are used, some companies using several. These approaches are used for a large variety of investigations, including design evaluation, design optimisation, power evaluation, dose/input optimisation, and for a wide variety of designs, including sampling windows or designs with several groups of elementary designs. The most common limitation was the need to change software when moving from estimation to design.
For Part 2, adaptive design is of high priority for most companies, with the following specific needs: (1) start from prior information, (ii) design optimisation after each new cohort, (iii) use stopping rules. New developments in design approaches should also have the following priorities: (i) handling data below quantification limit, (ii) design for discrete data and joint continuous/discrete models, (iii) handling continuous covariates, (iv) robustness across models.
Conclusions: This is the first survey of its kind performed in the pharmaceutical industry and it demonstrates the important role of optimal design in population PKPD. There is a clear need for further developments, especially for adaptive design and for discrete data in NLMEM.
Acknowledgements: The research leading to these results has received support from the Innovative Medicines InitiativeJoint Undertaking under grant agreement n° 115156, resources of which are composed of financial contributions from the European Union's Seventh Framework Programme (FP7/2007-2013) and EFPIA companies’ in kind contribution. The DDMoRe project is also financially supported by contributions from Academic and SME partners.
