Irina Bondareva
The Research Institute of Physical – Chemical Medicine, Russia
Objectives: Antiepileptic drug (AED) monotherapy is preferable to polytherapy, but some patients require more than one AED for successful seizure control. Among others, potential PK drug-drug interactions of so called old AEDs are indications for TDM of AEDs. The objective of the study is to evaluate the predictability of individualized dosage regimens for CBZ and VPA given as combination therapy with other AEDs based on TDM data of epileptic patients and the earlier developed nonlinear population models for PK drug-drug interactions.
Methods: TDM data were routinely collected in the Laboratory of Pharmacokinetics of Moscow Medical University. Levels of AEDs were measured by high performance liquid chromatography. The PK analysis was performed using the USC*PACK software based on 226 TDM (peak-trough strategy) data files of adult epileptic patients who received chronic CBZ- or VAL-monotherapy or AED duotherapy. This study included epileptic patients with long and rich TDM stories: repeated measurements during 1 – 2 years related to modifications in dosages and/or addition of AED to monotherapy. These data were not included in the population models. In the nonlinear models of heteroinduction, the metabolic rate asymptotically changes from a monotherapy value (D) during time-lag (Lambda) to a value (D+A) after heteroinduction. The prediction errors were estimated as the difference between observed levels after changes in AED dosage regimen and those predicted based on the patient-specific Bayesian posterior PK model and the TDM data before changes.
Results: Absolute value of prediction error was less or equal to 25% and considered as “acceptable” in 42 (75%) cases for CBZ-monotherapy group, 40 (76.9%) for VPA-monotherapy, in 35 (72.9%) for CBZ+VPA, in 24 (66.7%) for VPA+CBZ, and in 22 (64.7%) for CBZ+AED. Intraindividual variability of predictions varied from 23 – 25% for monotherapy to 26 – 28% for polytherapy. No tendency to overestimate or underestimate the concentration levels was observed.
Conclusions: This study has demonstrated that, in most cases, predictions of future AED concentrations based on the population PK models, supplemented by TDM data and patient-specific Bayesian posterior parameter modelling provided clinically acceptable estimates. The individual prediction errors were slightly higher for changes in polytherapy compared to monotherapy, which highlights the value of TDM and individualizing AED dosage regimen in the setting of polypharmacy.
Reference: PAGE 25 () Abstr 5834 [www.page-meeting.org/?abstract=5834]
Poster: Drug/Disease modeling - CNS