Development and validation of a HPLC-MS/MS method with electrospray ionization for quantitation of potassium oxonate in human plasma: Application to a pharmacokinetic study

A rapid, sensitive and specific HPLC-MS/MS method was developed and validated for the quantification of potassium oxonate (Oxo) in human plasma using [13C2,15N3]-Oxo as an internal standard (IS). The target substance was separated from human plasma using the solid-phase extraction method. Chromatography separation was performed on a Waters:Atlantis dC18 column (150×4.6 mm, 5.0 μm) with a mobile phase consisting of H2O with 0.1% formic acid in acetonitrile (90:10, v/v). The mass spectrometer works with electrospray ionization and multiple reaction monitoring in its negative ion mode, using target ions at [M–H]− m/z 111.9 for Oxo and [M–H]− m/z 117.0 for the IS. The mean standard curve was linear (r=0.9991) over the concentration range of 2.0–200.0 ng/ml and had good back-calculated accuracy and precision. The intra- and inter-day precision were <6.33% and the accuracy was >99.38%. The extraction recovery was >60.26%. The lower limit of quantification achieved with this method was 2.0 ng/ml. This assay method was demonstrated to be accurate, sensitive and simple and was successfully applied to a pharmacokinetic study following single oral administration of a 40-mg S-1 capsule in 12 tumor patients.

To date, there are several methods reported in the literature to determine Oxo, including enzyme immunoassay (11), GC-NICI-MS (gas chromatography-negative ion chemical ionization mass spectrometry) (12,13) and LC-MS/MS (liquid chromatography-tandem mass spectrometry) (14)(15)(16)(17). The enzyme immunoassay method is not suitable for analysis in large batches. The GC-NICI-MS method has been shown to be unstable when used clinically. The LC-MS/MS method is reported to require long and complex pre-processing for derivatization. Therefore, a simple and novel method is required. In this study, a novel HPLC-MS/MS method was developed and successfully applied to a pharmacokinetic study, after single oral administration of a combination of FT, CDHP and Oxo to 12 tumor patients. This method was determined to be simple, rapid and stable for use in the quantitation of Oxo in human plasma.

Results and Discussion
Selection of IS. It is necessary to use an IS to obtain high accuracy when a mass spectrometer is used as the HPLC detector.  of Oxo to the IS versus the concentrations (X) of Oxo, using weighted least squares linear regression (the weighting factor was 1/X 2 ). In our study, the mean standard curve for Oxo was Y=0.0057X+0.000353 (r=0.9991). The concentrations of Oxo in unknown samples were obtained from the regression line. The LOQ was defined as the lowest concentration on the calibration curve, where precision was within ±20% and accuracy was within ±20%. This was established using six independent samples of standards.   Extraction recovery (%) = [Mean peak area (extraction samples)/mean peak area (reference samples)] x 100. RSD, relative standard deviation.

C B
A Intra-day and inter-day precision and accuracy. Intra-assay precision and accuracy were assessed by measuring the concentration of Oxo in six aliquots of three different quality control samples, which were extracted and analyzed on the same day. Inter-assay precision and accuracy were determined from the results of three different quality control samples, which were extracted and analyzed six-fold on three consecutive days. The results are presented in Table I.
Matrix effect (ME). The ME represents the potential ion suppression or enhancement effects of co-eluting and undetected matrix components in plasma. This was obtained by comparing the peak area of Oxo and IS spiked into post-extracted blank plasma samples to that of Oxo and IS spiked into the mobile phase at an equivalent concentration. In this study, the ME was evaluated by three quality control concentrations (5.0, 25.0 and 160.0 ng/ml) of Oxo and an IS concentration level of 250.0 ng/ml. Six samples at each concentration level were analyzed. The blank plasma used in this study was obtained from six different batches. If the peak area ratios were <85 or >115%, an endogenous ME was implied. The ME of plasma at concentrations of 5.0, 25.0 and 160.0 ng/ml were 95.50, 100.45 and 97.61%, respectively. The ME of IS was 95.6%. The results obtained were within the acceptable limit, suggesting that there was no ME observed in this study.
Stability. The stability of Oxo in plasma under various conditions was evaluated. The quality control plasma samples (5.0, 25.0 and 160.0 ng/ml) were stable when placed at room temperature for 4 h, following three freeze/thaw (-40˚C) cycles and when stored at -40˚C for 3 months. The processed sample, placed in the autosampler at an ambient temperature (20˚C) for 2 h, was also stable. These results (Table II) demonstrated that no significant degradation occurred under different conditions.
Pharmacokinetic application. The method was applied for the analysis of plasma samples obtained from 12 tumor patients, following single oral administration of a 40-mg S-1 capsule in the pharmacokinetics study. The pharmacokinetic parameters were estimated by DAS version 2.1.1 software. Pharmacokinetic analysis of Oxo was performed using the noncompartmental method. The concentration maximum (C max ) and the time to reach it (T max ) were recorded directly. The elimination rate constant (K e ) was calculated using linear regression of the terminal points from the semi-log plot of plasma concentration against time. The elimination half-life (t 1/2 ) was calculated as 0.693/K e . The area under the plasma concentration-time curve of Oxo, from time zero to infinity (AUC 0-∞ ), was determined using the linear trapezoidal rule to the last measurable plasma concentration (C t ), plus the additional area from time t to infinity, calculated as C t /K e . Finally, the mean plasma concentration versus time profile and pharmacokinetic parameters of Oxo were obtained ( Fig. 4; Table III). The primary pharmacokinetic parameters of Oxo in our study were similar to those as previously published (18)(19)(20). The method developed in this study was demonstrated to be accurate and sensitive when compared with the pharmacokinetic parameters and concentration-time curves of previous studies.

Conclusions.
A HPLC-MS/MS method was developed for the determination of Oxo in human plasma. There were numerous advantages to this method, including low volumes of sample requirement, simple sample processing, absence of the ME and a short analysis time. The method was successfully applied to a pharmacokinetic study after single oral administration of a 40-mg S-1 capsule in 12 tumor patients.