Using Fitness-for-Use to Define Design Space for Analytical Methods
View Presentation View Presentation
Bruno Boulanger, Arlenda S.A.  *Pierre Lebrun, University of Liège 

Keywords: Design Space, Fitness-for-use, Optimization, Prediction, Bayesian model, Specifications

A framework to identify Design Space (DS) will be presented based on Bayesian modeling. The ways to conceive and apply this methodology to bioassays will be shown. Through a Ligand-Binding Assay (LBA) example, the use of Bayesian modeling for the development of a robust optimal assay will be illustrated, with the relationship to the specifications applying on the precision of measurements and the dosing range. The ways to derive the predictive precision profile and to identify the set of conditions that guarantee the future results within the specifications will be examined. From these perspectives, predictions and specifications are the keys to define the DS for an analytical procedure. The similarities of the DS derived from assay optimization and the DS used in the validation phase will be highlighted. Finally, the ways to define acceptance limits for future runs will also be shown.