E. Pigeolet(1), P. Jacqmin(2), L. Sargentini-Maier(1) and A. Stockis(1)
(1) UCB, Clinical Pharmacology, Braine l’Alleud, Belgium; (2) Exprimo Consulting LLP, Colchester, Essex, United Kingdom
Objective: The aim of the analysis was to identify demographic and/or physiologic determinants of levetiracetam disposition in healthy subjects and in subjects with partial epilepsy, using two matched sets of studies performed in
Methods: 5408 plasma concentration-time data were available together with demographic variables and treatment information, from 524 unique subjects participating in six phase-I studies, and in two phase-III and two long-term follow-up studies in add-on treatment of partial epilepsy. The structural model was a one compartment model with first order absorption and elimination rates. Parameters were assumed log-normally distributed. Residual variability was modeled by two proportional error models, one for healthy and one for epileptic subjects. An inter-occasion variability term was used for Ka. Modeling was performed using NONMEM with FO estimation. The full data set was used for analysis and validation.
Results: Patients were taking one to three concomitant anti-epileptic drugs (AED) consisting mainly of carbamazepine, phenytoin, phenobarbital and valproate. Body weight, gender, creatinine clearance (CLcr) and concomitant AEDs had a statistically significant effect on levetiracetam clearance. Weight, health status and valproate had a statistically significant impact on the volume of distribution. Food significantly decreased Ka. Ethnicity was not a statistically significant covariate.
Clearance, typically 4.02 L/h in a 70 kg male subject with CLcr=110 mL/min, varied by 20% at most when body weight was halved or doubled from the population mean (from 70 to 35 or 140 kg). Enzyme inducers increased levetiracetam clearance by 9%, while valproate decreased it by 19%. Clearance was 10% lower in females than in males, and 10% lower when CLcr was decreased from 110 to 50 mL/min.
Distribution volume, typically 52.7 L in a 70 kg epileptic subject, increased linearly with body weight, and decreased by 23% when levetiractam was co-administered with valproate, probably as a consequence of body fat gain caused by valproate. Simulations from the final model showed that only body weight or valproate had a >20% effect on Cmax and AUCτ. These effects do not require dose adjustment, considering the low toxicity of levetiracetam and the recommended individual titration approach (from 1 to 3g daily).
Conclusion: Analysis of this large data set allowed the identification of several explanatory covariates that contribute to the modest pharmacokinetic variability of levetiracetam. Ethnicity was not statistically significant.
Reference: PAGE 14 (2005) Abstr 747 [www.page-meeting.org/?abstract=747]
Poster: poster