Pharmacokinetics of doripenem in cerebrospinal fluid
G. Charkoftaki (1), R. Nalda-Molina (2), H. Dimaraki (1), K. Margetis (3), N. Boutos (3), D. E. Sakas (3), E. Vryonis (4), A. Skoutelis (4), S. Markantonis (1), H. Archontaki (5), A. Dokoumetzidis (1), G. Valsami (1)
(1) School of Pharmacy, University of Athens, Greece; (2) School of Pharmacy, University of Alicante, Spain; (3) School of Medicine, Department of Neurosurgery, University of Athens, Greece; (4) Fifth Department of Internal Medicine, Evangelismos General Hospital, Athens, Greece; (5) Department of Chemistry, University of Athens, Greece
Objectives: To characterise the pharmacokinetics of the antibiotic doripenem (DRP) in cerebrospinal fluid (CSF).
Methods: 44 neurological patients received a single 500 mg prophylactic dose of DRP at various times before surgery for implantation of a pump for intrathecal administration of baclofen or after lumbar puncture for the trial intrathecal infusion of baclofen. Patients had neither active neurological disease nor infection of the CNS. One and in some cases two samples of CSF per patient were collected, as well as one blood sample. Samples were analysed using an HPLC method  and DRP concentrations were quantified. A NONMEM pharmacokinetic analysis was carried out in two stages. The first stage used the plasma samples and literature population priors for a two-compartment model  to estimate the Empirical Bayesian Estimates (EBE) of the PK parameters of each patient for DRP in plasma. The EBE of the PK parameters where used as covariates to estimate the PK parameters of a third distribution compartment corresponding to CSF . The structural parameters included in the model were the rate constant, kCSF, and the partition coefficient PC.
Results: The VPC of the literature PK model for the plasma together with the plasma data showed that the model describes the data reasonably. In the second stage a model was developed for the CSF data, using EBE estimates for the PK parameters from the plasma model as covariates. The final model estimated the mean values of kCSF and PC and the interindividual variability for PC, while the residual variability was fixed to the corresponding plasma value. The model was validated internally using VPC and bootstrap. The estimates for the mean of the parameters were kCSF=0.11 h-1, PC=0.05 while IIV for PC was 56%. These values correspond to a mean steady state CSF concentration of 0.22 mg/L, for 1500 mg daily dose and a mean half-life time to equilibrium of 6.3 hours. Comparing MIC concentrations for different bacteria to this CSF value, is promising for the potential use of DRP for CNS infections, taking into account that in case of meningitis drug penetration through BBB will be significantly more extensive. More clinical trials in patients are needed with richer samples, to characterise the kinetics of DRP in the CSF more accurately.
Conclusion: The present NONMEM analysis of DRP CSF data shows that DRP crosses BBB significantly even in healthy non-inflammating meninges and therefore may be appropriate to treat certain CNS infections.
 Ikeda K, Ikawa K, Morikawa N, Kameda K, Urakawa N, Ohge H, Sueda T. Quantification of doripenem in human plasma and peritoneal fluid by high-performance liquid chromatography with ultraviolet detection. J Chromatogr B Analyt Technol Biomed Life Sci. 867:20-5 (2008).
 Nandy P, Samtani MN, Lin R. Population pharmacokinetics of doripenem based on data from phase 1 studies with healthy volunteers and phase 2 and 3 studies with critically ill patients. Antimicrob Agents Chemother. 54:2354-9 (2010).
 Anderson BJ, Holford NH, Woollard GA, Chan PL. Paracetamol plasma and cerebrospinal fluid pharmacokinetics in children. Br J Clin Pharmacol. 46:237-43 (1998).