E. Tousset (1), C. Reners (1) & B. Vrijens (1-2)
(1) AARDEX Ltd., Visé, Belgium (2) University of Liège, Belgium
Objectives: Occasional sampling of the concentrations of drug in plasma is used to study the pharmacokinetics of the drug in question, and to determine whether or not dosing is optimal. Both purposes are compromised by variance in the concentration of plasma, the origins of which include variations in both the sources and sinks for drug in the body. The main variation in the source of drug in ambulatory patients is variable execution of the prescribed regimen of drug administration. The main variations in the sinks for drug arise from food- and/or drug-induced changes in pharmacokinetic parameters, changing the relation between dose and concentration. To minimize these variations, sampling is routinely done at a ‘trough’ point in an interval between scheduled doses, i.e., just prior to the next scheduled dose in the prescribed dosing sequence.
The main objective of this research is to determine the magnitude and sources of variability in trough concentrations collected during ambulatory care.
Methods: Therapeutic drug monitoring data issued from several HIV studies were combined in order to quantify the within- and between-patient variability in trough concentrations and to identify major sources of PK variability in HIV studies. Electronically compiled dosing histories data were used to identify the proportion of variability in PK studies that could be attributed to patient non adherence to prescribed therapy.
Results: When therapeutic drug monitoring is used, many samples turned out not to be taken at the trough, but at some other point in the dosing cycle, and, furthermore, many samples turned out to be taken during an inter-dose cycle when the assumption of a steady-state is not justifiable, because of prior irregularities in dosing times. Those deviations result in considerable within-patient variability inducing sometimes difficult clinical interpretation of the results. Surprisingly, in some circumstances, the within-patient variability exceeds the between-patient variability. Through simulations we were able to show that electronically compiled times of dose taken prior to blood sampling can explain more than 50% of the residual, within-patient variability in trough concentrations.
Conclusions: These results suggest that electronically-compiled dosing histories may greatly improve information derived from both population PK studies and therapeutic drug monitoring.
Reference: PAGE 13 (2004) Abstr 514 [www.page-meeting.org/?abstract=514]
Poster: poster