Objective: The aim of this analysis was to develop a population pharmacokinetic model to describe the pharmacokinetics of recombinant human erythropoietin (rHuEPO) in healthy subjects, after intravenous and subcutaneous administration over a wide dose range, and to examine the influence of demographic characteristics and other covariates on the pharmacokinetic parameters of rHuEPO.
Methods: Erythropoietin serum concentration data were available from 16 studies comprising 49 healthy subjects who received rHuEPO intravenous doses from 10 to 300 IU/kg, 427 healthy subjects who received rHuEPO subcutaneous doses from 1 to 2400 IU/kg, and 57 healthy subjects who received placebo and where endogenous erythropoietin concentrations were measured. Different pharmacokinetic models were fitted to the dataset using nonlinear mixed-effects modeling software (NONMEM, Version V, Level 1). Several patient covariates were tested in order to quantify the effect on rHuEPO pharmacokinetic parameters. Model evaluation was examined using a posterior predictive check.
Results: Erythropoietin showed a diurnal baseline variation of +/-20%, described with a dual cosine model. Disposition was described with a two-compartment model with a small volume of distribution (6L) and parallel linear and nonlinear clearance. Total clearance varied between 0.3 and 0.9 L/h over the concentration range studied. A dual absorption model was used to characterise the rHuEPO absorption from the subcutaneous formulation and consisted of a faster pathway described as a sequential zero- and first-order absorption process and a parallel slower pathway characterised as a zero-order process. The bioavailability of subcutaneous rHuEPO increased from 30% at low doses to 71% at the highest dose of 160 kIU and was described using a hyperbolic model. The most important covariate effects were a decrease in the first-order absorption rate constant (k(a)) with increasing age, an increase in subcutaneous bioavailability with increasing baseline haemoglobin, and a decrease in bioavailability with increasing bodyweight. A posterior predictive check showed no systematic deviation of the simulated data from the observed values.
Conclusion: The population pharmacokinetic model developed is suitable to describe the pharmacokinetic behaviour of rHuEPO after intravenous and subcutaneous administration in healthy subjects, over a wide dose range.