The pharmacokinetics of erythropoietin in the cerebrospinal fluid after intravenous administration of recombinant human erythropoietin

Objectives: Erythropoietin (EPO) was originally described as a regulator of erythropoiesis. Recently, synthesis of EPO and expression of the EPO receptor (EPO-R) have been reported for the central nervous system (CNS). The potential use of EPO to prevent or reduce CNS injury and the paucity of information regarding its entry into the human CNS led us to examine the pharmacokinetics (PK) of recombinant human EPO (r-HuEPO) in the serum and cerebrospinal fluid (CSF). Methods: Four patients with Ommaya reservoirs were enrolled to facilitate serial CSF sampling. R-HuEPO was given intravenously (IV) in single doses of 40,000 IU or 1,500 IU/kg and in multiple doses of 40,000 IU daily for 3 days. Results: The EPO concentrations in the CSF increased after a period of slow equilibration. Linear first-order distribution kinetics were observed for serum and CSF. The concentration of EPO in the CSF was proportional to the serum concentration of EPO and the permeability of the blood-brain barrier (BBB), as determined by the albumin quotient (Q(A)=[ albumin] CSF/[albumin] serum). A rise in the CSF concentration was seen as early as 3 h after IV administration. Peak levels (C-max) were reached between 9 h and 24 h. After a single dose of 1,500 IU/kg, the C-max in the CSF ranged from 11 mIU/ ml to 40 mIU/ ml, and the ratios of CSF/serum C-max ranged from 3.6 x 10(-4) to 10.2 x 10(-4). The terminal half-life (t(1/2)) values of EPO in serum and CSF were similar. The t(1/2) of r-HuEPO in the CSF ranged from 25.6 h to 35.5 h after a single dose of 1,500 IU/l. Using these parameters a PK model was generated that predicts the concentration-time profile of EPO in the CSF. Conclusions: We report that r-HuEPO can cross the human BBB and describe for the first time the PK of EPO in the CSF after IV administration. Our data suggest that the concentration-time profile of EPO in the CSF can be predicted for individual patients if the serum concentration of EPO and the Q(A) are known. This information may be useful in the design of clinical trials to explore the potential therapeutic effects of EPO during CNS injury.