Characterization of decomposition products and preclinical and low dose clinical pharmacokinetics of decitabine (5-aza-2'-deoxycytidine) by a new liquid chromatography/tandem mass spectrometry quantification method.

Aberrant DNA methylation patterns resulting in gene transcriptional repression are observed in numerous cancers. Decitabine, a DNA methyltransferase inhibitor, is being clinically evaluated in patients with hematologic malignancies and solid tumors. Decitabine is rather unstable and decomposes to 1-beta-D-2'-deoxyribofuranosyl-3-guanylurea under basic conditions and several additional unknown products under neutral conditions. This has greatly limited application of pharmacokinetic assays to clinical development of decitabine. In this paper, a high-performance liquid chromatography/ultraviolet multi-stage mass spectrometry (HPLC-UV-MSn) study of the decomposition of decitabine in water and human plasma revealed that these previously unknown products are isomers of the intermediates formyl-1-beta-D-2'-deoxyribofuranosyl-3-guanylurea and 1-beta-D-2'-deoxyribofuranosyl-3-guanylurea. A HPLC tandem mass spectrometry (MS/MS) method for the determination of decitabine concentrations in human and rat plasma has been developed. This method was based on a specific fragmentation pathway of the molecular ion of decitabine at m/z 229 to generate a unique fragment ion at m/z 113 under collision-induced dissociation. Separation of decitabine and the stable internal standard dihydro-5-aza-cytidine from the endogenous interfering substance in plasma extract was carried out on a C18 Aquasil column under an isocratic elution with a mobile phase consisting of 5% water/acetonitrile and 10 mM ammonium formate. The detection of decitabine was via selected reaction monitoring (SRM, 229 > 113), and its ionization was enhanced by post-column addition of acetonitrile. Effects of sample preparation and handling parameters on the stability of decitabine were also evaluated in human plasma at various temperatures. The accuracy and precision of this assay showed a coefficient of variation of <15% over the range of 0.5-25 ng for rat plasma and 0.1-25 ng for human plasma injected on-column. Pharmacokinetics of decitabine in rats following intravenous doses of 1.0 and 5.0 mg/kg were characterized. In the rat, plasma concentration-time profiles were found to follow a biexponential decline and the pharmacokinetics was dose-independent. Application of this decitabine pharmacokinetic assay to human studies is therefore justified and ongoing. Copyright 2006 John Wiley & Sons, Ltd.