A Population Pharmacokinetic Model for Low-Dose Methotrexate and its Polyglutamated Metabolites in Red Blood Cells

Abstract

Measurement of intracellular concentrations of methotrexate (MTX) and its polyglutamated metabolites (MTXGlu(2-5)) in red blood cells (RBCs) has been suggested as a potential means of monitoring low-dose MTX treatment of rheumatoid arthritis (RA). However, a possible correlation between RBC MTX and MTXGlu2-5 concentrations and clinical outcomes of MTX treatment in RA is debated. A better understanding of the dose-concentration-time relationship of MTX and MTXGlu(2-5) in RBCs by population pharmacokinetic modelling is desirable and will facilitate assessing a potential RBC concentration-effect relationship in the future. The purpose of this analysis was to describe the pharmacokinetics of MTX and MTXGlu(2-5) in RBCs. Secondary objectives included investigation of deglutamation reactions and the loss of MTX and MTXGlu(2-5) from the RBC. A model was developed using NONMEM(®) version 7.2 based on RBC data obtained from 48 patients with RA receiving once-weekly low-dose MTX treatment. This model was linked to a fixed two-compartment model that was used to describe the pharmacokinetics of MTX in the plasma. A series of five compartments were used to describe the intracellular pharmacokinetics of MTX and MTXGlu(2-5) in RBCs. Biologically plausible covariates were tested for a significant effect on MTX plasma clearance and the intracellular volume of distribution of all MTX species in RBCs ([Formula: see text]). The developed model was used to test hypotheses related to the enzymatic deglutamation of MTXGlu(2-5) and potential loss of MTXGlu(2-5) from RBCs. The final RBC pharmacokinetic model required the intracellular volumes of distribution for the parent and metabolites to be set to the value estimated for the parent drug MTX alone, and the rate constants describing the polyglutamation steps were fixed at literature values. Significant covariates included effect of body surface area-adjusted estimated glomerular filtration rate on renal plasma clearance and effect of allometrically scaled total body weight with a fixed exponent of 0.75 on non-renal plasma clearance of MTX. The only significant covariate with an effect on [Formula: see text] was mean corpuscular volume (MCV). The model supported single deglutamation steps and a single mechanism of MTX and MTXGlu(2-5) loss from RBCs. The developed model enabled acceptable description of the intracellular kinetics of MTX and MTXGlu(2-5) in RBCs. In the future it can form the basis of a full pharmacokinetic-pharmacodynamic model to assess the time-RBC concentration-effect relationship of low-dose MTX treatment in RA.