Pharmacokinetic-pharmacodynamic modelling of recombinant human erythropoietin in athletes : a population approach.

objectiveTo develop a pharmacokinetic model able to take into account the negative feedback loop of endogenous erythropoietin production observed after repeated administration of recombinant human erythropoietin (rHuEPO), and to propose a pharmacokinetic-pharmacodynamic model capable of assessing and quantifying the relationship between changes in: (i) serum soluble transferrin receptor (sTfR) levels, (ii) reticulocyte haematocrit (RetHct), and (iii) percentage macrocytes (%Macro) secondary to repeated administration of rHuEPO. SUBJECTS AND

methodsEighteen trained athletes (three females and 15 males) participated in this study. They received subcutaneous injections of rHuEPO-α 50 U/kg bodyweight for 26 days (days 1, 3, 5, 9, 10, 12, 15, 17, 19, 22, 24 and 26) with iron supplementation. Venous blood samples were collected before, during and after rHuEPO treatment for determination of serum erythropoietin concentrations, haematological parameters (RetHct, %Macro) and sTfR levels. Population pharmacokinetic-pharmacodynamic calculations were performed using NONMEM® software.

resultsThe serum erythropoietin concentration-time profile was compatible with a one-compartment open model and first-order input rate. The mean half-lives calculated from the first and the terminal log-linear parts of the curves were 5.2 and 35.8 hours, respectively. After subcutaneous administration of rHuEPO, the terminal part of the curve should correspond to the absorption rather than the elimination phase ('flip-flop' phenomenon). The total clearance divided by bio-availability was 4.33 L/h. The pharmacodynamic relationship based on a sigmoid E(max) model can be reasonably used to relate changes observed in haematological and biochemical markers after rHuEPO administration to changes in serum erythropoietin concentrations. rHuEPO induces a delayed increase in sTfR levels, RetHct and %Macro. The half-life (t1/2) k(0) (equilibration delay) values were 10.2 days for sTfR, 2 days for RetHct and 10.2 days for %Macro. The pharmaco-kinetic-pharmacodynamic approach developed in this study allowed below-base-line decreases in RetHct levels (i.e. from days 10-26 after the end of rHuEPO treatment) to be taken into account. A negative-feedback loop of red blood cell production further to high haemoglobin and haematocrit values could explain this decrease.

conclusionsThe approach described here may provide an additional tool in the war against drug abuse by athletes; indeed, the model could be useful for simulating pharmacokinetic-pharmacodynamic relationships according to different rHuEPO dosage schedules.