S. Jönsson* and M.O. Karlsson
Div. of Biopharmaceutics and Pharmacokinetics, Dept. of Pharmacy, Uppsala University, Sweden
In clinical drug development, dosing based on covariates needs to be considered and if necessary, a dosing algorithm has to be defined. We present, using a real example, a strategy that estimates optimal conditions for such an algorithm with the minimum number of dose sizes that fulfil the overall therapy goal.
The example drug is proposed for usage during acute conditions requiring hospitalisation and will be administered as an intravenous infusion. The table below presents the steps involved in the strategy in general and the outcome in the specific case.
Consideration of In the specific case
Definition of therapeutic target Css, unbound 70 m M Css, unbound 60 m M if CLcr < 50 ml/min
Definition of individual penalty: Quadratic loss in log domain function
Definition of population therapy goal: >90% of patients above 50 m M and <5% of patients above 100 m M
Establishment of pharmacokinetic model: 2 compartment and/or pharmacodynamic model CL linearly related to CLcr and Vc to body weight
Collection of relevant population Empirical distributions of CLcr and characteristics body weight from earlier studies in target population
Estimation of optimal individualisation Dose size(s) and cut-off based on body algorithm under predefined constraints weight estimated for loading infusion Dose size(s) and cut-off(s) based on CLcr estimated for maintenance infusion
Selection of dose regimen that fulfils Loading infusion independent of the population therapy goal covariates Three maintenance infusion rates dependent on CLcr
Css, unbound = unbound steady state concentration CLcr = creatinine clearance CL clearance Vc volume of central compartment
Reference: PAGE 9 (2000) Abstr 118 [www.page-meeting.org/?abstract=118]
Poster: poster