Purpose: Valproic acid (VPA), a widely used antiepileptic, also inhibits histone deacetylase (HDAC), and is undergoing evaluation as an anti-cancer agent. We studied the pharmacokinetics of VPA in the plasma and cerebrospinal fluid (CSF) in a non-human primate model that is highly predictive of human CSF penetration to determine if levels of VPA required to inhibit HDAC in in vivo models can be attained.
Methods: Oral VPA, 75 mg/kg, was administered to four non-human primates. Serial samples of blood (n = 4) and CSF (n = 3) were obtained for pharmacokinetic studies of total and free VPA. Plasma and CSF VPA concentrations were measured using the commercially available Abbott AxSYM VPA assay reagent system (Abbott Laboratories, Abbott Park, IL, USA). The resultant plasma and CSF data were evaluated using pharmacokinetic modeling methods.
Results: At a dose of 75 mg/kg, the maximum plasma concentration of VPA was 130.1 +/- 70.6 microg/ml (mean +/- standard deviation) for total drug and 53.3 +/- 44.4 microg/ml for free drug. The mean plasma area under the curve (AUC) for total drug was 680 +/- 233 microg/ml h and for free drug 146 +/- 89 microg/ml hr. The maximum CSF concentration occurred 2-3 h after administration and was 28.2 +/- 18.6 microg/ml. The CSF AUC for VPA was 108 +/- 52 microg/ml h. The CSF penetration of VPA was 12.9 +/- 5.1% for total drug and 57.0 +/- 8.7% for free drug. Disappearance from the plasma followed non-linear kinetics with a V (max) of 321.2 +/- 65.6 microg/kg/min and a K (m) of 17.2 +/- 13.7 mg/l.
Conclusion: Valproic acid deserves further study for the treatment of CNS tumors given its high CSF penetration after oral dosing coupled with the anti-tumor activity observed in preclinical studies.