Premium
A population pharmacokinetic model for Cremophor EL using nonlinear mixed‐effect modeling: model building and validation
Author(s) -
Van Den Bongard H. J. G. D.,
Mathôt R. A. A.,
Van Tellingen O.,
Schellens J. H. M.,
Beijnen J. H.
Publication year - 2002
Publication title -
british journal of clinical pharmacology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.216
H-Index - 146
eISSN - 1365-2125
pISSN - 0306-5251
DOI - 10.1046/j.1365-2125.2002.161316.x
Subject(s) - nonmem , pharmacokinetics , compartment (ship) , paclitaxel , population , chemistry , pharmacology , chromatography , bootstrapping (finance) , medicine , surgery , mathematics , chemotherapy , oceanography , environmental health , geology , econometrics
Cremophor EL is a polyethoxylated castor oil that is used for the formulation of a variety of hydrophobic drugs. Recently, it has been demonstrated that the pharmacokinetic (PK) behaviour of Cremophor EL influences the disposition of paclitaxel [1]. Consequently, assessment of the clinical PK of Cremophor EL during paclitaxel administration is of major importance. The purpose of the present study was to develop a population PK model of Cremophor EL as a solubilizing agent for paclitaxel. Plasma concentration‐time data (85 courses) of Cremophor EL concentrations were collected from 71 patients with solid tumours treated with paclitaxel dissolved in Cremophor EL. Cremophor EL was intravenously infused during 3, 24 and 96 h (doses ranging from 8.3–20.8 ml m −2 ). Population PK analysis of plasma concentration‐time data was performed using the nonlinear mixed effect modeling software program NONMEM. Thirteen covariates were investigated for their influence on the PK parameters of Cremophor EL. The stability of the final model was evaluated by bootstrapping. The median parameter estimates obtained from 1000 bootstrap replicates were compared with those obtained from the original data set. The data were fitted to a 3‐compartment model with Michaelis‐Menten elimination from the central compartment. The following PK parameters were estimated: volume of the central compartment V 1=2.59 l), volume of the first peripheral compartment ( V 2=1.81 l), volume of the second peripheral compartment ( V 3=1.61 l), intercompartmental clearance between the central compartment and the first peripheral compartment (Q12=1.44 l h −1 ) and the second peripheral compartment (Q13=0.16 l h −1 ), maximal elimination rate ( V max =0.19 ml h −1 ), and the plasma concentration at half of the V max ( K m =0.12 ml l −1 ). Interindividual variability (IIV) was estimated using an exponential error model, and was quantified for V 1 (25%), V 2 (36%), and V max (31%). Residual variability was small with a combined additional and proportional error of 0.095 ml l −1 and 7%, respectively. Significant relationships ( P <0.001) between the covariates and the PK parameters were determined between V 1 and sex, between V 2 and BSA, and between V max and performance status. The median PK parameter values from the bootstrap procedure were in accordance with the parameter estimates of the population PK model. The non‐linear PK of Cremophor EL could adequately be described by the developed population PK model. Sex, BSA and PS partially explained the IIV in V 1, V 2 and V max , respectively. The PK model may be used when studying the clinical PK of paclitaxel.