Influence of biological variables upon pharmacokinetic parameters of intramuscular methotrexate in rheumatoid arthritis

Abstract

The pharmacokinetics of methotrexate were studied in 22 patients receiving 5-15 mg per week in a single i.m. administration for rheumatoid arthritis. The data consisted of 3 plasma levels per patient, taken at 2, 6, and 12 hours after the administration. The concentration of methotrexate was determined by fluorescence polarization immunoassay. The pharmacokinetic parameters of a 2-compartment model were determined by Bayesian estimation using the population values of Bressolle et al. [1996]. The fitted parameters were: total plasma clearance of methotrexate (CL), first-order absorption constant (ka), volume of central compartment (V1), and transfer constants between the 2 compartments (k12 and k21). Additional parameters were derived from the fitted ones: maximal concentration (Cmax), time to maximum (tmax), volume of distribution at steady-state (Vss), and terminal half-life (t1/2). Twenty-one biological covariates were considered to explain the interpatient variability. The relationships between these covariates and the pharmacokinetic parameters were investigated by principal component analysis and multiple regression analysis. About 90% of the variability of CL were explained by 4 variables (sex, age, height and serum creatinine). About 50%-70% of the variability of the other pharmacokinetic parameters were explained by a set of covariates including age, height, creatinine, creatinine clearance, and dose. The effect of dose was noticed mainly on k12, Vss, and t1/2, thus suggesting that the transfer of the drug from plasma to tissues may be nonlinear. The possibility of predicting CL with a good precision would facilitate the computation of dosage regimens in these patients.