Establishing a tool for individualised dosing of gabapentin - The use of a non-parametric population kinetics model in the MM-USCPACK software, a new tool for clinical dosage optimization.
Kristin Cecilie Carlsson(1), Heidi Hansen(1), Roger Jelliffe(2), Nils Ove Hoem(3).
(1)Dept. of Pharmacology, School of Pharmacy, University of Oslo, Norway.(2) Laboratory of Applied Pharmacokinetics, University of Southern California, USA. (3) Smerud Medical Research, Oslo Norway.
To be able to use pharmacokinetic tools in clinical practice, they must be easily available to the physician at the clinic, or even bedside. To be clinically useful, even in many everyday situations, such tools must be able to handle relatively complex pharmacokinetic models, and also be able to take into account continuous changes in the patient’s condition.
The MM-USCPACK software employs an interacting multiple-model (non-parametric) strategy for analyzing previous serum concentration-time data and predicting future dosage regimens. The MM-USCPACK software can be used on a portable computer and have a jump function that makes it possible to handle continuously changing parameter estimates.
To facilitate individualized dosing of gabapentin, a population pharmacokinetic model has been developed consisting of an absorption and a central compartment, and with speed of elimination described as a linear function of creatinine clearance.
We have dose monitored a small number of patients who were treated with gabapentin against neuropathic pain (8 patients with a total of 19 dosage interval profiles).
Serum concentration levels, analgesic effects and adverse effects measured by a visual analog scale, were registered as time variables. The covariates gender, age, height, weight, serum creatinine and total daily dose were registered as patient characteristics.
The parameters for the pharmacokinetic model were first estimated by a parametric standard two-stage Bayesian population tool (IT2B) and then with the Non-Parametric Adaptive Grid (NPAG) tool recently developed by The Laboratory of Applied Pharmacokinetics. The resulting population pharmacokinetic model can together with the MM-USCPACK software be used to optimize dosing of gabapentin in the chosen patient population.