S. Jönsson* and M.O. Karlsson
Division of Pharmacokinetics and Drug Therapy,
Introduction
Individualised dosing may be based on a covariate, on plasma concentration measurements, on the therapeutic effect or on the weighted sum of desired and unwanted effects i.e. the utility of the drug. Factors affecting covariate-based dosing strategies under the restriction of a discrete number of dose sizes have been identified previously [1]. The aim of this simulation study was to distinguish factors influencing the choice of a dosing strategy based on plasma concentrations, effect or utility of the drug under the same restriction, with the purpose of estimating optimal dose sizes for the population and conditions for an individualisation strategy.
Methods
Aspects of dosing strategies were estimated by minimising a therapeutically defined loss function given known target and known characteristics of the drug in the population.
Results and conclusions
The results substantiate that it is possible to optimise individualised dosing strategies using estimation.
The choice of optimal dosing strategies for individualisation was found to be dependent of several characteristics. Influencing factors were related to the characteristics of the drug, e.g. distribution of pharmacokinetic and pharmacodynamic parameters (shape and variability) or the definition of the target, e.g. extent of desired effect, shape of loss function.
Although the relations found were expected on a theoretic basis, it is unlikely that optimal dosing strategies will be identified without estimation.
1. Jönsson S., Karlsson M.O, Abstract submitted to Eighth Meeting Population Approach Group in Europe, Saintes, France, June 17-18 1999.
Reference: PAGE 10 () Abstr 188 [www.page-meeting.org/?abstract=188]
Poster: poster