Journal of Pharmaceutical and Biomedical Analysis 96 (2014) 31–36
Contents lists available at ScienceDirect
Journal of Pharmaceutical and Biomedical Analysis journal homepage: www.elsevier.com/locate/jpba
Short communication
Validated HILIC–MS/MS assay for determination of vindesine in human plasma: Application to a population pharmacokinetic study Rong-Hua Zhu a,b,∗ , Huan-De Li a , Hua-Lin Cai a , Zhi-Ping Jiang c , Ping Xu a , Li-Bo Dai a,b , Wen-Xing Peng a a Clinical Pharmacy and Pharmacology Research Institute, the Second Xiang-Ya Hospital, Central-South University, Changsha, Hunan Province 410011, People’s Republic of China b School of Pharmacy, Central-South University, Changsha 410005, People’s Republic of China c Laboratory of Clinical Pharmacology, Department of Hematology, Xiang-Ya Hospital, Central-South University, Changsha 410008, People’s Republic of China
a r t i c l e
i n f o
Article history: Received 16 December 2013 Received in revised form 11 March 2014 Accepted 13 March 2014 Available online 23 March 2014 Keywords: Vindesine Vinorelbine HILIC MS/MS Pharmacokinetics
a b s t r a c t The first HILIC–tandem mass spectrometry (MS/MS) method for determination of vindesine (VDS) in human plasma using vinorelbine as an internal standard (IS) has been developed and validated. Plasma samples clean-up consisted of solid phase extraction with a strataTM -X column. The compounds were separated on a HILIC column with an isocratic mobile phase consisting of acetonitrile and 15 mM ammonium acetate buffer containing 0.15% formic acid (80:20, v/v). The detection was performed on a triple quadrupole tandem mass spectrometer via electrospray positive ionization (ESI+ ). The ion transitions recorded in multiple reaction monitoring mode were m/z 754.6 → 123.8 for VDS and 779.4 → 323.3 for IS, respectively. Linear calibration curves were obtained in the concentration range of 0.3–28 ng/ml and the lower limit of quantification for VDS was 0.3 ng/ml. The coefficient of variation of the assay precision was less than 13%, and the accuracy exceeded 96%. The developed assay method was successfully applied for the evaluation of population pharmacokinetics of VDS after intravenous infusion of Xi Ai Ke Vial® (3 mg of Vindesine Sulfate for Injection) to Chinese Han subjects with hematological malignant disorders. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Vindesine (VDS) is a semi-synthetic vinca alkaloids derivatived from vinblastine, which acts by causing the arrest of cells in metaphase mitosis through its inhibition tubulin mitotic functioning. This drug is cell-cycle specific for the S phase and is indicated for the treatment of acute leukemia, malignant lymphoma, Hodgkin’s disease, acute erythraemia and acute panmyelosis. VDS has demonstrated activity in patients who have relapsed while receiving multiple-agent treatment that included vincristine. It is currently being examined for its potential to synergize with the interferons and for its value as prolonged therapy in preventing metastasis [1–3].
∗ Corresponding author at: Clinical Pharmacy and Pharmacology Research Institute, the Second Xiang-Ya Hospital, Central-South University, Changsha, Hunan Province 410011, People’s Republic of China. Fax: +86 731 84896038. E-mail address: [email protected] (R.-H. Zhu). http://dx.doi.org/10.1016/j.jpba.2014.03.017 0731-7085/© 2014 Elsevier B.V. All rights reserved.
A variety of methods, including radioimmunoassay [4,5], enzyme-linked immunosorbent assay [6], nonaqueous capillary electrophoresis [7], high performance thin-layer chromatography (HPTLC) [8,9] and liquid chromatography (HPLC) coupled with ultraviolet (UV) detector [10,11], fluorescence detector (FD) [12] or electrochemical detector (ECD) [13] have been used to analysis VDS. However, of all reported methods, only radioimmunoassay and HPLC-ECD assay were applied for quantify concentration of VDS in clinical pharmacokinetic studies. Although a sensitive radioimmunoassay can provided valuable information on tissue distribution, elimination and disposition kinetics of VDS in an early stage of clinical and preclinical investigations, radiolabelled compounds had to be used for quantification purposes in most cases and this techniques lack selectivity as well. Electrochemical detector providing a highly sensitive detection of trace amounts of electrically active substances such as vinca alkaloids, but the detector system could not work continuously with such troubles as clogging and deterioration of electrode performance, making it impossible to perform analysis with good reproducibility over long period. Recently, Maeda and Miwa [14] reported a HPLC–tandem mass
32
R.-H. Zhu et al. / Journal of Pharmaceutical and Biomedical Analysis 96 (2014) 31–36
spectrometry method for analysis ten chemotherapeutic drugs including VDS in wipe samples with LLOQ of 5 ng/ml. However, if introducing this method for quantifying the concentration of VDS in biological sample, its sensitivity should be greatly improved to meet the very low concentration of VDS in human plasma. Although vindesine has been used as an antineoplastic drug for almost 40 years, data on vindesine pharmacokinetics and pharmacodynamics are scarce [15–18]. One of the reasons for this is the lack of a specific, sensitive and simple assay suitable for small plasma volumes. Therefore the authors aimed to improve an existing highperformance liquid chromatography (HPLC) assay by changing the solid phase and by using a more sensitive and controlled tandem mass spectrometry as detector. Based on HILIC separation mode, this study established a rapid, accurate, sensitive and reliable HILIC–MS/MS method for the determination of VDS concentrations in human plasma. The developed assay method was successfully applied for the evaluation of population pharmacokinetics of VDS after intravenous infusion of Xi Ai Ke Vial® (3 mg of Vindesine Sulfate for Injection) to Chinese Han subjects with hematological malignant disorders. 2. Materials and methods 2.1. Reagents and chemicals Vindesine (VDS, 97.9% of purity), and vinorelbine (IS, 98.6% of purity) were purchased from National Institute of Zhejiang institute for food and drug control (Hangzhou, China). Vindesine Sulfate for Injection (Xi Ai Ke Vial® , 1 mg/ramus) was the product of Hangzhou Minsheng Pharmaceutical Group Co., Ltd., China). Methanol and acetonitrile (HPLC grade) were purchased from Merck (Darmstadt, Germany). Formic acid and ammonium acetate (HPLC grade) were purchased from Tedia (Fairfield, OH, USA). Water was redistilled and filtered through 0.22 m membrane filters before it was used. 2.2. Apparatus and operation conditions 2.2.1. Liquid chromatography The chromatography was performed on AcquityTM UPLC system (Waters Corp., Milford, MA, USA) with an autosampler and an oven that ensure the temperature control of the analytical column. A Waters Acquity UPLC® BEH HILIC column (1.7 m, 50 mm × 2.1 mm) was employed. The column temperature was maintained at 30 ◦ C. The auto-sampler was conditioned at 4 ◦ C and the sample volume injected was 4 l. The mobile phase consisted of 80% acetonitrile and 20% 15 mM ammonium acetate buffer (containing 0.15% formic acid) (v/v). The flow rate was 0.3 ml/min. 2.2.2. Mass spectrometry Triple quadrupole tandem mass spectrometric detection was carried out on a Quattro Premier XE mass spectrometer (Waters Corp., Milford, MA, USA) with an electrospray ionization (ESI) interface. The ESI source was set in positive ionization mode. The operating conditions were optimized by flow injection of a mixture of all analytes and were determined as follows: capillary 1.00 kV; extractor 4.0 V; source temperature 120 ◦ C; desolvation temperature 400 ◦ C; cone gas flow 50 L/h; desolvation gas flow 750 L/h; gas cell Pirani pressure 3.39 × 10−3 mbar; multiplier 650; dwell time 0.10 s. Quantification was performed by multiple reaction monitoring (MRM) of the deprotonated precursor ion and the related product ion for VDS using internal standard method with peak area ratios and a weighting factor. The ion for monitor and optimized cone and collision energy were as shown in Table 1. All data collected in centroid mode were acquired and processed using MassLynxTM 4.1
Table 1 Instrument method in MRM mode for the quantitative determination of vindesine by HILIC–MS/MS using vinorelbine as IS. Compound
Precursor ion (m/z)
Product ion (m/z)
Cone (V)
Collision energy (eV)
Vindesine Vinorelbine
754.6 779.4
123.8 323.3
55 40
50 24
software with QuanLynxTM program (Waters Corp., Milford, MA, USA). 2.3. Preparation of standard and quality control samples Standard stock solutions of VDS, and IS were both prepared in methanol at the concentration of 100.0 g/ml. The internal standard working solution was diluted with methanol to 1.0 g/ml. All the solutions were stored at −20 ◦ C. Standard solutions of VDS in human plasma were prepared by spiking with an appropriate volume of the variously diluted stock solutions, giving final concentrations of 0.30, 0.65, 1.50, 3.30, 6.50, 13.0, 28.0 ng/ml. The calibration curves were prepared and assayed along with quality control (QC) samples and each batch of clinical plasma samples. The QC samples were prepared at three different concentration levels of 0.65, 3.30 and 23.0 ng/ml in blank plasma. All prepared plasma samples were stored at −80 ◦ C. 2.4. Sample preparation A 1.0 ml aliquot of human plasma sample was mixed with a 50 l of internal standard working solution (1.0 g/ml in methanol) in a 1.5 ml EP tube. The EP tube was shaken in a vortex for 30 s, and then centrifuged at 15,000 rpm for 5 min at 4 ◦ C. VDS and IS were extracted with strataTM -X solid phase extraction columns (30 mg, Phenomenex, Torrance, CA, USA). Briefly, the solid phase extraction (SPE) columns were conditioned with methanol and equilibrated with water prior to loading sample. After washing with 30% methanol, VDS and IS were eluted with 1 ml methanol (containing 0.1% formic acid). Following SPE, the samples were dried evaporated to dryness at ambient temperature in a SpeedVac (SPD121P-230, Thermo, USA). The residue was reconstituted with a 100 l aliquot of the mobile phase, and a 4 l aliquot was injected directly onto the UPLC-MS/MS system. 2.5. Method validation The method validation procedure was carried out according to the FDA guidance for bioanalytical method validation (2001) [19]. 2.5.1. Specificity Specificity of the method was established by comparing chromatograms of six different batches of blank plasma obtained from six subjects with those of corresponding standard plasma samples spiked with VDS and IS. 2.5.2. Linearity and lower limit of quantification (LLOQ) Linearity calibration curves were prepared by making serial dilution of the working stocks and assaying standard plasma samples at seven concentrations of VDS ranging 0.3–28 ng/ml. The linearity of calibration curve was determined by plotting the peak area ratio (y) of VDS to IS versus the nominal concentrations (x) of VDS. The calibration curves were constructed by being weighted (1/x) least square linear regression. The lower limit of quantification (LLOQ) for VDS was determined based on at least 10 times of signal-to-noise ratio and sufficient precision (within 20%) and accuracy (80–120%).
R.-H. Zhu et al. / Journal of Pharmaceutical and Biomedical Analysis 96 (2014) 31–36
33
Fig. 1. The chemical structures of vindesine and vinorelbine.
2.5.3. Precision and accuracy Intra- and inter-day accuracy and precision for this method were determined at three different concentration levels on three consecutive days, and on each day five replicates were analyzed with an independently prepared calibration curve. The accuracy was expressed by (mean observed concentration)/(nominal concentration) × 100% and the precision by relative standard deviation (RSD, %). 2.5.4. Extraction recovery and matrix effect The extraction recovery of VDS and IS from the sample preparation procedure was determined by a comparison of the peak area of VDS and IS in spiked plasma samples (three replicates each of QC levels) with that of VDS and IS in samples prepared by spiking extracted drug-free plasma samples with the same amounts of VDS and IS. To evaluate the matrix effect on the ionization, i.e. the potential ion suppression or enhancement, VDS at three QC levels (five replicates) were prepared with the supernatant extract of blank plasma. The corresponding peak areas (A) were compared with those of VDS standard solutions (B). The ratio (A/B × 100)% was used to evaluate the matrix effect. The matrix effect of internal standard was also evaluated using the same method. 2.5.5. Stability The stability of stock solutions of VDS and IS at a concentration of 100.0 g/ml was evaluated at −20 ◦ C for 30 days by comparison with freshly prepared solutions at corresponding concentration. The stability of VDS in plasma samples were assessed under four conditions: after short-term storage for 4 h at ambient temperature (25 ◦ C), after long-term storage for 30 days under frozen condition (−80 ◦ C), after three freeze–thaw cycles, and after sample preparation (for 12 h at 4 ◦ C, stability in autosampler). 2.6. Clinical application The method was applied to determine the plasma concentrations of VDS from a pharmacokinetic study in which unrelated Chinese Han subjects (62 males and 38 females, aged 36 ± 17
years) with hematological malignant disorders were enrolled. The pharmacokinetic study was approved by the Ethical Committee of Xiangya Hospital of Central South University and all volunteers gave their signed informed consent to participating in the study according to the principles of the Declaration of Helsinki. All the subjects received 3 mg of Xi Ai Ke Vial® (Vindesine Sulfate for Injection) as a 1 min rapid intravenous infusion. Three random sampling times derived from 0 min, 5 min, 20 min, 1 h, 3 h, 6 h, 12 h, 24 h, 48 h, 72 h after dosing were made by using a nonlinear mixed effects model (NONMEM) program and approximately 4 ml blood samples were collected into chilled EDTA tubes. Blood samples were centrifuged immediately at 2000 × g for 8 min at 4 ◦ C, the plasma was separated and stored at −80 ◦ C until analysis. The population pharmacokinetic parameters of VDS were estimated by NONMEM software (VLevel 1.1, GloboMaxLLC, USA). 3. Results and discussion 3.1. Optimization of chromatographic separation and MS/MS detection As other family members of vinca alkaloids, VDS and vinorelbine have a large dimeric asymmetric structure composed of an indole nucleus (catharanthine) and a dihydroindole nucleus (vindoline) linked by a carbon–carbon bond as shown in Fig. 1. In contrast with VDS, vinorelbine has structural modifications to the catharanthine nucleus. Apart from three tertiary amine groups and one secondary amine group in the molecule as vinorelbine, VDS also has one methanamide. VDS and vinorelbine are mild base with pKa of 11.34 and 10.87, respectively. In most of the reports about determination of VDS in plasma, reversed-phase chromatographic columns such as C18 column have been used for separation of the analytes and IS. During method development stage prior to HILIC column, we tried a revised-phase (RP) column for analytes separation. When using a C18 or C8 column, it was observed that these columns provided peak shapes with serious tailing, which did not meet the requirement of pharmacokinetic study for peak symmetry. By excluding other possible influencing factors, we considered that the poor peak shapes obtained on RP
34
R.-H. Zhu et al. / Journal of Pharmaceutical and Biomedical Analysis 96 (2014) 31–36
Fig. 2. Positive-ion electrospray mass spectra obtained in full-scan mode from authentic samples of vindesine and vinorelbine.
column may be due to undesirable secondary effects caused by interactions between basic analytes and residual silanol groups on the surface of silica stationary phase in RP column. In order to improve the peak, mobile phase had to contain high concentration of mineral acid such as phosphate buffer (PBS) and the pH of mobile phase must be adjusted to basic. However, PBS belongs to nonvolatile salt and is not suitable for mass spectrometric detection, and also, basic mobile phase is unfavorable for solvation of VDS and decrease the sensitivity of mass spectrometric detection for VDS. For these reasons, we employed HILIC column for VDS separation. In the HILIC model, polar stationary phase and high proportion of organic solvent in the mobile phase are applied, which adds to the retention of polar material and improves the separation. Combined with ESI, HILIC can increase the sensitivity of mass spectrometry. In the present study, the LLOQ is 0.3 ng/ml for VDS in plasma, which is remarkable lower than the data published in literatures based on HPLC method [11–13]. The selection of MRM transitions and associated acquisition parameters (collision energy and cone voltage) were evaluated for best response under positive mode ESI conditions by infusing a standard solution, via a syringe pump, into the mobile phase. The corresponding full-scan ESI+ -MS/MS spectra for VDS and the IS were shown in Fig. 2. The very narrow chromatographic peaks with a peak width at half height about 5 s that was produced by UPLC indicates efficient chromatography of fast separation. VDS and IS were rapidly eluted with retention time less than 3.0 min (Fig. 3). The short analysis time meets the application requirements for high sample throughput.
3.2. Selection of IS The best internal standard for LC-MS assay is a deuterated form of the compound analyzed. However, deuterated VDS is not available for general laboratory, and thus a compound structurally
or chemically similar to VDS was considered in this study. In LC–MS/MS the IS should also have chromatographic and mass spectrometric behaviors similar to those of the analyte, and mimic the analyte through preparation steps and detection steps. Vinorelbine was chosen as the internal standard for the assay because of its similarity in chemical structure and physical properties to VDS. The results indicate that it is an appropriate IS for the analysis of VDS.
3.3. Method validation 3.3.1. Selectivity No interference peaks were detected for the VDS and IS from six different sources of human plasma. The typical chromatograms of blank plasma, plasma spiked with 0.3 ng/ml of VDS and IS, and a volunteer plasma at 3.0 h after intravenous infusion of Xi Ai Ke Vial® (3 mg of Vindesine Sulfate for Injection) were shown in Fig. 3. The retention times of IS and VDS were approximately 1.4 min and 2.3 min.
3.3.2. Linearity and lower limits of quantification The plasma concentration ranges were 0.3–28 ng/ml for VDS. The calibration model was selected on the basis of the analysis of the data by linear regression with intercepts and weighting factors. The best linear fit and least-squares residuals for the calibration curve were achieved with 1/x weighing factor VDS. The mean correlation coefficient (r) of the respective weighted calibration curves generated during the validation were 0.998 (0.997–0.999) for VDS. The lower limits of quantification (LLOQ) for VDS were 0.3 ng/ml (shown in Fig. 3). The back calculated value of each point of the calibration curve was comprised in the range of ±15% from the nominal concentration. The statistical data computed from the five analytical curves was listed in Table 2.
R.-H. Zhu et al. / Journal of Pharmaceutical and Biomedical Analysis 96 (2014) 31–36
35
Table 2 Deviation of calibration curves obtained for determination of vindensine in human plasma (n = 5). Back-calculated concentration for standards
Mean SD %CV %Bias
STD 1 (0.30 ng/ml)
STD 2 (0.65 ng/ml)
STD 3 (1.50 ng/ml)
STD 4 (3.30 ng/ml)
STD 5 (6.50 ng/ml)
STD 6 (13.0 ng/ml)
STD 7 (28.0 ng/ml)
0.326 0.024 7.4 8.7
0.624 0.020 3.1 −4.0
1.45 0.056 3.9 −3.3
3.16 0.108 3.4 −4.1
6.60 0.253 3.8 1.6
13.44 0.391 2.9 3.4
28.68 0.780 2.7 2.4
3.3.3. Precision and accuracy The intra- and inter-day accuracy and precision for VDS were listed in Table 3. The mean intra-day RSD (%) for VDS were 3.5–6.6%, the corresponding values for inter-day were 6.1–12.9%. The
Table 3 Intra- and inter-day precision and accuracy data for assays of vindesine in human plasma. Precision
Nominal concentration (ng/ml)
Meana (ng/ml) Intra-day (n = 5)
0.65 3.3 23.0
Inter-day (n = 3)
0.65 3.3 23.0
a
RSD (%)
Accuracy (%)
0.655 ± 0.023
3.5
100.8
3.18 ± 0.166 23.8 ± 1.58
5.2 6.6
96.3 103.3
0.634 ± 0.082
12.9
97.5
3.29 ± 0.199 22.4 ± 2.15
6.1 9.6
99.5 97.5
Values are means ± standard deviation.
intra-day accuracies of VDS were 96.3–103.3%, the corresponding values for inter-day were 97.5–99.5%. 3.3.4. Extraction recovery and matrix effect The extraction (absolute) recoveries for VDS at the three concentrations of 0.65, 3.0, 23.0 ng/ml from human plasma were 76.1 ± 6.09%, 75.2 ± 4.41%, 78.2 ± 6.39%. The value for IS at 50.0 ng/ml was 81.6 ± 5.47%. As for the matrix effect, the ratios (A/B × 100%) for VDS at 0.65, 3.0 and 23.0 ng/ml were 93.4 ± 6.4%, 96.1 ± 3.9% and 95.8 ± 3.5%, for IS at 50.0 ng/ml was 94.7 ± 4.5%. As defined in Section 2, all the ratios were between 85% and 115%, which means no matrix effect for VDS and IS in this method. 3.3.5. Stability Stock solutions of VDS and IS in methanol were stable for 30 days at −20 ◦ C; more than 95.0% of the spiked amounts were recovered. The stability data of VDS in human plasma under three conditions were listed in Table 4. As shown in Table 4, no significant degradation of VDS in human plasma was observed under all conditions studied. 3.4. Pharmacokinetic application After intravenous infusion of Xi Ai Ke Vial® (3 mg of Vindesine Sulfate for Injection) to 100 Chinese Han subjects with Table 4 Stability of vindesine in human plasma. Conditions
Percentage of control valuea 0.65 ng/ml
Fig. 3. Typical mass chromatograms (MRM) obtained from vindesine (channel 2) and vinorelbine (channel 1). (A) Drug-free plasma. (B) Blank plasma spiked with vindesine at the LLOQ level 0.3 ng/ml and vinorelbine. (C) Plasma sample from a patient 3.0 h after intravenous infusion 3 mg of Xi Ai Ke Vial® (Vindesine Sulfate for Injection).
Short-term stability (4 h) Three freeze–thaw cycles Autosampler stability (12 h) Long-term stability (30 days) a
110.7 108.4 93.3 104.8
± ± ± ±
Values are means ± standard deviation.
5.0 8.2 9.4 7.3
3.3 ng/ml 101.6 103.0 99.0 98.4
± ± ± ±
7.9 10.0 6.2 6.1
23.0 ng/ml 100.9 101.1 96.1 101.9
± ± ± ±
3.6 2.6 9.5 8.2
36
R.-H. Zhu et al. / Journal of Pharmaceutical and Biomedical Analysis 96 (2014) 31–36
after intravenous infusion Xi Ai Ke Vial® (3 mg of Vindesine Sulfate for Injection) to Chinese Han subjects with hematological malignant disorders. References
Fig. 4. Mean plasma concentration–time profiles of intravenous injection of Xi Ai Ke Vial® (3 mg of Vindesine Sulfate for Injection) to 100 unrelated Chinese Han subjects with hematological malignant disorders. Vertical bars represent standard deviation.
hematological malignant disorders, the mean plasma concentration–time profiles of VDS were shown in Fig. 4. Pharmacokinetic parameters of VDS in individual 100 subjects were obtained from population estimates according to Bayesian’ theorem. The results showed that the pharmacokinetics of VDS conformed with three compartment open model and the first order kinetics in this population. The mean value of systemic clearance (CL), terminal half-life (t1/2 ), distribution volume (V), and area under the curve (AUC) was estimated to be 0.381 ± 0.108 L/h/kg, 22.8 ± 10.4 h, 170.12 ± 69.29 L, 149 ± 67 ng h/ml, respectively. The terminal half-life (t1/2 ) was similar to early reports [16–18]. The systemic clearance is a little larger than the values (0.252 L/h/kg and 0.132 L/h/kg) reported by Nelson [16] and Ohnuma et al. [17], but is slightly less than the value (0.53 L/h/kg) reported by Rahmani et al. [18]. As compared with those classical pharmacokinetic studies which had limited sample size, our study recruited more subjects with wide age ranges and different extent of diseases. On statistical grounds, the smaller the sample size, the greater chance to obtain smaller or larger observed values. Therefore, we speculate that this difference among the studies is more likely to reflect the sampling error, and do not represent the actual difference between populations. 4. Conclusion This is the first reported study of plasma concentration of VDS using ultra-performance hydrophilic interaction liquid chromatography–tandem mass spectrometry (HILIC–MS/MS). It allows a suitable sensitivity of VDS (LLOQ 0.3 ng/ml) and can be carried out in a short time within 3.5 min. This method has been successfully applied for the evaluation of pharmacokinetics of VDS
[1] M. Bayssas, J. Gouveia, F. de Vassal, J.L. Misset, L. Schwarzenberg, P. Ribaud, M. Musset, C. Jasmin, M. Hayat, G. Mathé, Vindesine: a new vinca alkaloid, Recent Results Cancer Res. 74 (1980) 91–97. [2] J. Dancey, W.P. Steward, The role of vindesine in oncology-recommendations after 10 years’ experience, Anticancer Drugs 6 (1995) 625–636. [3] E. Cvitkovic, E. Wasserman, Role of vindesine as neoadjuvant chemotherapy for non-small cell lung and head and neck cancers, Anticancer Drugs 8 (1997) 734–745. [4] V.S. Sethi, S.S. Burton, D.V. Jackson, A sensitive radioimmunoassay for vincristine and vinblastine, Cancer Chemother. Pharmacol. 4 (1980) 183–187. [5] R. Rahmani, J. Barbet, J.P. Cano, A 125 I-radiolabelled probe for vinblastine and vindesine radioimmunoassays: applications to measurements of vindesine plasma levels in man after intravenous injections and long-term infusions, Clin. Chim. Acta 129 (1983) 57–69. [6] Y. Nakano, T. Saita, H. Fujito, Development of a specific and sensitive enzymelinked immunosorbent assay for vindesine, Yakugaku Zasshi 132 (2012) 727–732. [7] L. Barthe, J.P. Ribet, M. Pélissou, M.J. Degude, J. Fahy, A. Duflos, Optimization of the separation of vinca alkaloids by nonaqueous capillary electrophoresis, J. Chromatogr. A 968 (2002) 241–250. [8] R.L. Hussey, W.M. Newlon, High-performance liquid chromatographic and TLC determinations of desacetylvinblastine amide (vindesine) and its monosulfate salt, J. Pharm. Sci. 67 (1978) 1319–1320. [9] A. Paci, L. Mercier, P. Bourget, Identification and quantitation of antineoplastic compounds in chemotherapeutic infusion bags by use of HPTLC: application to the vinca-alkaloids, J. Pharm. Biomed. Anal. 30 (2003) 1603–1610. [10] J.H. Beijnen, D.E. Vendrig, W.J. Underberg, Stability of vinca alkaloid anticancer drugs in three commonly used infusion fluids, J. Parenter. Sci. Technol. 43 (1989) 84–87. [11] M. De Smet, S.J. Van Belle, G.A. Storme, D.L. Massart, High-performance liquid chromatographic determination of vinca-alkaloids in plasma and urine, J. Chromatogr. 345 (1985) 309–321. [12] D.E. Vendrig, J. Teeuwsen, J.J. Holthuis, Determination of vinca alkaloids in plasma and urine using ion-exchange chromatography on silica gel and fluorescence detection, J. Chromatogr. 434 (1988) 145–155. [13] D.E. Vendrig, J. Teeuwsen, J.J. Holthuis, Analysis of vinca alkaloids in plasma and urine using high-performance liquid chromatography with electrochemical detection, J. Chromatogr. 424 (1988) 83–94. [14] S. Maeda, Y. Miwa, Multicomponent high-performance liquid chromatography/tandem mass spectrometry analysis of ten chemotherapeutic drugs in wipe samples, J. Chromatogr. B 921–922 (2013) 43–48. [15] R. Rahmani, X.J. Zhou, M. Placidi, M. Martin, J.P. Cano, In vivo and in vitro pharmacokinetics and metabolism of vinca alkaloids in rat. I. Vindesine (4deacetyl-vinblastine 3-carboxyamide), Eur. J. Drug Metab. Pharmacokinet. 15 (1990) 49–55. [16] R.L. Nelson, The comparative clinical pharmacology and pharmacokinetics of vindesine, vincristine, and vinblastine in human patients with cancer, Med. Pediatr. Oncol. 10 (1982) 115–127. [17] T. Ohnuma, L. Norton, A. Andrejczuk, J.F. Holland, Pharmacokinetics of vindesine given as an intravenous bolus and 24-hour infusion in humans, Cancer Res. 45 (1985) 464–469. [18] R. Rahmani, R. Samak, P. Bore, J.P. Cano, Clinical pharmacokinetics of vinca alkaloids, Bull. Cancer 75 (1988) 195–200. [19] Guidance for Industry Bioanalytical Method Validation from FDA, US Department of Health and Human Services, Food and Drug Administration, CDER, Rockville, USA, 2001.
Contents lists available at ScienceDirect
Journal of Pharmaceutical and Biomedical Analysis journal homepage: www.elsevier.com/locate/jpba
Short communication
Validated HILIC–MS/MS assay for determination of vindesine in human plasma: Application to a population pharmacokinetic study Rong-Hua Zhu a,b,∗ , Huan-De Li a , Hua-Lin Cai a , Zhi-Ping Jiang c , Ping Xu a , Li-Bo Dai a,b , Wen-Xing Peng a a Clinical Pharmacy and Pharmacology Research Institute, the Second Xiang-Ya Hospital, Central-South University, Changsha, Hunan Province 410011, People’s Republic of China b School of Pharmacy, Central-South University, Changsha 410005, People’s Republic of China c Laboratory of Clinical Pharmacology, Department of Hematology, Xiang-Ya Hospital, Central-South University, Changsha 410008, People’s Republic of China
a r t i c l e
i n f o
Article history: Received 16 December 2013 Received in revised form 11 March 2014 Accepted 13 March 2014 Available online 23 March 2014 Keywords: Vindesine Vinorelbine HILIC MS/MS Pharmacokinetics
a b s t r a c t The first HILIC–tandem mass spectrometry (MS/MS) method for determination of vindesine (VDS) in human plasma using vinorelbine as an internal standard (IS) has been developed and validated. Plasma samples clean-up consisted of solid phase extraction with a strataTM -X column. The compounds were separated on a HILIC column with an isocratic mobile phase consisting of acetonitrile and 15 mM ammonium acetate buffer containing 0.15% formic acid (80:20, v/v). The detection was performed on a triple quadrupole tandem mass spectrometer via electrospray positive ionization (ESI+ ). The ion transitions recorded in multiple reaction monitoring mode were m/z 754.6 → 123.8 for VDS and 779.4 → 323.3 for IS, respectively. Linear calibration curves were obtained in the concentration range of 0.3–28 ng/ml and the lower limit of quantification for VDS was 0.3 ng/ml. The coefficient of variation of the assay precision was less than 13%, and the accuracy exceeded 96%. The developed assay method was successfully applied for the evaluation of population pharmacokinetics of VDS after intravenous infusion of Xi Ai Ke Vial® (3 mg of Vindesine Sulfate for Injection) to Chinese Han subjects with hematological malignant disorders. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Vindesine (VDS) is a semi-synthetic vinca alkaloids derivatived from vinblastine, which acts by causing the arrest of cells in metaphase mitosis through its inhibition tubulin mitotic functioning. This drug is cell-cycle specific for the S phase and is indicated for the treatment of acute leukemia, malignant lymphoma, Hodgkin’s disease, acute erythraemia and acute panmyelosis. VDS has demonstrated activity in patients who have relapsed while receiving multiple-agent treatment that included vincristine. It is currently being examined for its potential to synergize with the interferons and for its value as prolonged therapy in preventing metastasis [1–3].
∗ Corresponding author at: Clinical Pharmacy and Pharmacology Research Institute, the Second Xiang-Ya Hospital, Central-South University, Changsha, Hunan Province 410011, People’s Republic of China. Fax: +86 731 84896038. E-mail address: [email protected] (R.-H. Zhu). http://dx.doi.org/10.1016/j.jpba.2014.03.017 0731-7085/© 2014 Elsevier B.V. All rights reserved.
A variety of methods, including radioimmunoassay [4,5], enzyme-linked immunosorbent assay [6], nonaqueous capillary electrophoresis [7], high performance thin-layer chromatography (HPTLC) [8,9] and liquid chromatography (HPLC) coupled with ultraviolet (UV) detector [10,11], fluorescence detector (FD) [12] or electrochemical detector (ECD) [13] have been used to analysis VDS. However, of all reported methods, only radioimmunoassay and HPLC-ECD assay were applied for quantify concentration of VDS in clinical pharmacokinetic studies. Although a sensitive radioimmunoassay can provided valuable information on tissue distribution, elimination and disposition kinetics of VDS in an early stage of clinical and preclinical investigations, radiolabelled compounds had to be used for quantification purposes in most cases and this techniques lack selectivity as well. Electrochemical detector providing a highly sensitive detection of trace amounts of electrically active substances such as vinca alkaloids, but the detector system could not work continuously with such troubles as clogging and deterioration of electrode performance, making it impossible to perform analysis with good reproducibility over long period. Recently, Maeda and Miwa [14] reported a HPLC–tandem mass
32
R.-H. Zhu et al. / Journal of Pharmaceutical and Biomedical Analysis 96 (2014) 31–36
spectrometry method for analysis ten chemotherapeutic drugs including VDS in wipe samples with LLOQ of 5 ng/ml. However, if introducing this method for quantifying the concentration of VDS in biological sample, its sensitivity should be greatly improved to meet the very low concentration of VDS in human plasma. Although vindesine has been used as an antineoplastic drug for almost 40 years, data on vindesine pharmacokinetics and pharmacodynamics are scarce [15–18]. One of the reasons for this is the lack of a specific, sensitive and simple assay suitable for small plasma volumes. Therefore the authors aimed to improve an existing highperformance liquid chromatography (HPLC) assay by changing the solid phase and by using a more sensitive and controlled tandem mass spectrometry as detector. Based on HILIC separation mode, this study established a rapid, accurate, sensitive and reliable HILIC–MS/MS method for the determination of VDS concentrations in human plasma. The developed assay method was successfully applied for the evaluation of population pharmacokinetics of VDS after intravenous infusion of Xi Ai Ke Vial® (3 mg of Vindesine Sulfate for Injection) to Chinese Han subjects with hematological malignant disorders. 2. Materials and methods 2.1. Reagents and chemicals Vindesine (VDS, 97.9% of purity), and vinorelbine (IS, 98.6% of purity) were purchased from National Institute of Zhejiang institute for food and drug control (Hangzhou, China). Vindesine Sulfate for Injection (Xi Ai Ke Vial® , 1 mg/ramus) was the product of Hangzhou Minsheng Pharmaceutical Group Co., Ltd., China). Methanol and acetonitrile (HPLC grade) were purchased from Merck (Darmstadt, Germany). Formic acid and ammonium acetate (HPLC grade) were purchased from Tedia (Fairfield, OH, USA). Water was redistilled and filtered through 0.22 m membrane filters before it was used. 2.2. Apparatus and operation conditions 2.2.1. Liquid chromatography The chromatography was performed on AcquityTM UPLC system (Waters Corp., Milford, MA, USA) with an autosampler and an oven that ensure the temperature control of the analytical column. A Waters Acquity UPLC® BEH HILIC column (1.7 m, 50 mm × 2.1 mm) was employed. The column temperature was maintained at 30 ◦ C. The auto-sampler was conditioned at 4 ◦ C and the sample volume injected was 4 l. The mobile phase consisted of 80% acetonitrile and 20% 15 mM ammonium acetate buffer (containing 0.15% formic acid) (v/v). The flow rate was 0.3 ml/min. 2.2.2. Mass spectrometry Triple quadrupole tandem mass spectrometric detection was carried out on a Quattro Premier XE mass spectrometer (Waters Corp., Milford, MA, USA) with an electrospray ionization (ESI) interface. The ESI source was set in positive ionization mode. The operating conditions were optimized by flow injection of a mixture of all analytes and were determined as follows: capillary 1.00 kV; extractor 4.0 V; source temperature 120 ◦ C; desolvation temperature 400 ◦ C; cone gas flow 50 L/h; desolvation gas flow 750 L/h; gas cell Pirani pressure 3.39 × 10−3 mbar; multiplier 650; dwell time 0.10 s. Quantification was performed by multiple reaction monitoring (MRM) of the deprotonated precursor ion and the related product ion for VDS using internal standard method with peak area ratios and a weighting factor. The ion for monitor and optimized cone and collision energy were as shown in Table 1. All data collected in centroid mode were acquired and processed using MassLynxTM 4.1
Table 1 Instrument method in MRM mode for the quantitative determination of vindesine by HILIC–MS/MS using vinorelbine as IS. Compound
Precursor ion (m/z)
Product ion (m/z)
Cone (V)
Collision energy (eV)
Vindesine Vinorelbine
754.6 779.4
123.8 323.3
55 40
50 24
software with QuanLynxTM program (Waters Corp., Milford, MA, USA). 2.3. Preparation of standard and quality control samples Standard stock solutions of VDS, and IS were both prepared in methanol at the concentration of 100.0 g/ml. The internal standard working solution was diluted with methanol to 1.0 g/ml. All the solutions were stored at −20 ◦ C. Standard solutions of VDS in human plasma were prepared by spiking with an appropriate volume of the variously diluted stock solutions, giving final concentrations of 0.30, 0.65, 1.50, 3.30, 6.50, 13.0, 28.0 ng/ml. The calibration curves were prepared and assayed along with quality control (QC) samples and each batch of clinical plasma samples. The QC samples were prepared at three different concentration levels of 0.65, 3.30 and 23.0 ng/ml in blank plasma. All prepared plasma samples were stored at −80 ◦ C. 2.4. Sample preparation A 1.0 ml aliquot of human plasma sample was mixed with a 50 l of internal standard working solution (1.0 g/ml in methanol) in a 1.5 ml EP tube. The EP tube was shaken in a vortex for 30 s, and then centrifuged at 15,000 rpm for 5 min at 4 ◦ C. VDS and IS were extracted with strataTM -X solid phase extraction columns (30 mg, Phenomenex, Torrance, CA, USA). Briefly, the solid phase extraction (SPE) columns were conditioned with methanol and equilibrated with water prior to loading sample. After washing with 30% methanol, VDS and IS were eluted with 1 ml methanol (containing 0.1% formic acid). Following SPE, the samples were dried evaporated to dryness at ambient temperature in a SpeedVac (SPD121P-230, Thermo, USA). The residue was reconstituted with a 100 l aliquot of the mobile phase, and a 4 l aliquot was injected directly onto the UPLC-MS/MS system. 2.5. Method validation The method validation procedure was carried out according to the FDA guidance for bioanalytical method validation (2001) [19]. 2.5.1. Specificity Specificity of the method was established by comparing chromatograms of six different batches of blank plasma obtained from six subjects with those of corresponding standard plasma samples spiked with VDS and IS. 2.5.2. Linearity and lower limit of quantification (LLOQ) Linearity calibration curves were prepared by making serial dilution of the working stocks and assaying standard plasma samples at seven concentrations of VDS ranging 0.3–28 ng/ml. The linearity of calibration curve was determined by plotting the peak area ratio (y) of VDS to IS versus the nominal concentrations (x) of VDS. The calibration curves were constructed by being weighted (1/x) least square linear regression. The lower limit of quantification (LLOQ) for VDS was determined based on at least 10 times of signal-to-noise ratio and sufficient precision (within 20%) and accuracy (80–120%).
R.-H. Zhu et al. / Journal of Pharmaceutical and Biomedical Analysis 96 (2014) 31–36
33
Fig. 1. The chemical structures of vindesine and vinorelbine.
2.5.3. Precision and accuracy Intra- and inter-day accuracy and precision for this method were determined at three different concentration levels on three consecutive days, and on each day five replicates were analyzed with an independently prepared calibration curve. The accuracy was expressed by (mean observed concentration)/(nominal concentration) × 100% and the precision by relative standard deviation (RSD, %). 2.5.4. Extraction recovery and matrix effect The extraction recovery of VDS and IS from the sample preparation procedure was determined by a comparison of the peak area of VDS and IS in spiked plasma samples (three replicates each of QC levels) with that of VDS and IS in samples prepared by spiking extracted drug-free plasma samples with the same amounts of VDS and IS. To evaluate the matrix effect on the ionization, i.e. the potential ion suppression or enhancement, VDS at three QC levels (five replicates) were prepared with the supernatant extract of blank plasma. The corresponding peak areas (A) were compared with those of VDS standard solutions (B). The ratio (A/B × 100)% was used to evaluate the matrix effect. The matrix effect of internal standard was also evaluated using the same method. 2.5.5. Stability The stability of stock solutions of VDS and IS at a concentration of 100.0 g/ml was evaluated at −20 ◦ C for 30 days by comparison with freshly prepared solutions at corresponding concentration. The stability of VDS in plasma samples were assessed under four conditions: after short-term storage for 4 h at ambient temperature (25 ◦ C), after long-term storage for 30 days under frozen condition (−80 ◦ C), after three freeze–thaw cycles, and after sample preparation (for 12 h at 4 ◦ C, stability in autosampler). 2.6. Clinical application The method was applied to determine the plasma concentrations of VDS from a pharmacokinetic study in which unrelated Chinese Han subjects (62 males and 38 females, aged 36 ± 17
years) with hematological malignant disorders were enrolled. The pharmacokinetic study was approved by the Ethical Committee of Xiangya Hospital of Central South University and all volunteers gave their signed informed consent to participating in the study according to the principles of the Declaration of Helsinki. All the subjects received 3 mg of Xi Ai Ke Vial® (Vindesine Sulfate for Injection) as a 1 min rapid intravenous infusion. Three random sampling times derived from 0 min, 5 min, 20 min, 1 h, 3 h, 6 h, 12 h, 24 h, 48 h, 72 h after dosing were made by using a nonlinear mixed effects model (NONMEM) program and approximately 4 ml blood samples were collected into chilled EDTA tubes. Blood samples were centrifuged immediately at 2000 × g for 8 min at 4 ◦ C, the plasma was separated and stored at −80 ◦ C until analysis. The population pharmacokinetic parameters of VDS were estimated by NONMEM software (VLevel 1.1, GloboMaxLLC, USA). 3. Results and discussion 3.1. Optimization of chromatographic separation and MS/MS detection As other family members of vinca alkaloids, VDS and vinorelbine have a large dimeric asymmetric structure composed of an indole nucleus (catharanthine) and a dihydroindole nucleus (vindoline) linked by a carbon–carbon bond as shown in Fig. 1. In contrast with VDS, vinorelbine has structural modifications to the catharanthine nucleus. Apart from three tertiary amine groups and one secondary amine group in the molecule as vinorelbine, VDS also has one methanamide. VDS and vinorelbine are mild base with pKa of 11.34 and 10.87, respectively. In most of the reports about determination of VDS in plasma, reversed-phase chromatographic columns such as C18 column have been used for separation of the analytes and IS. During method development stage prior to HILIC column, we tried a revised-phase (RP) column for analytes separation. When using a C18 or C8 column, it was observed that these columns provided peak shapes with serious tailing, which did not meet the requirement of pharmacokinetic study for peak symmetry. By excluding other possible influencing factors, we considered that the poor peak shapes obtained on RP
34
R.-H. Zhu et al. / Journal of Pharmaceutical and Biomedical Analysis 96 (2014) 31–36
Fig. 2. Positive-ion electrospray mass spectra obtained in full-scan mode from authentic samples of vindesine and vinorelbine.
column may be due to undesirable secondary effects caused by interactions between basic analytes and residual silanol groups on the surface of silica stationary phase in RP column. In order to improve the peak, mobile phase had to contain high concentration of mineral acid such as phosphate buffer (PBS) and the pH of mobile phase must be adjusted to basic. However, PBS belongs to nonvolatile salt and is not suitable for mass spectrometric detection, and also, basic mobile phase is unfavorable for solvation of VDS and decrease the sensitivity of mass spectrometric detection for VDS. For these reasons, we employed HILIC column for VDS separation. In the HILIC model, polar stationary phase and high proportion of organic solvent in the mobile phase are applied, which adds to the retention of polar material and improves the separation. Combined with ESI, HILIC can increase the sensitivity of mass spectrometry. In the present study, the LLOQ is 0.3 ng/ml for VDS in plasma, which is remarkable lower than the data published in literatures based on HPLC method [11–13]. The selection of MRM transitions and associated acquisition parameters (collision energy and cone voltage) were evaluated for best response under positive mode ESI conditions by infusing a standard solution, via a syringe pump, into the mobile phase. The corresponding full-scan ESI+ -MS/MS spectra for VDS and the IS were shown in Fig. 2. The very narrow chromatographic peaks with a peak width at half height about 5 s that was produced by UPLC indicates efficient chromatography of fast separation. VDS and IS were rapidly eluted with retention time less than 3.0 min (Fig. 3). The short analysis time meets the application requirements for high sample throughput.
3.2. Selection of IS The best internal standard for LC-MS assay is a deuterated form of the compound analyzed. However, deuterated VDS is not available for general laboratory, and thus a compound structurally
or chemically similar to VDS was considered in this study. In LC–MS/MS the IS should also have chromatographic and mass spectrometric behaviors similar to those of the analyte, and mimic the analyte through preparation steps and detection steps. Vinorelbine was chosen as the internal standard for the assay because of its similarity in chemical structure and physical properties to VDS. The results indicate that it is an appropriate IS for the analysis of VDS.
3.3. Method validation 3.3.1. Selectivity No interference peaks were detected for the VDS and IS from six different sources of human plasma. The typical chromatograms of blank plasma, plasma spiked with 0.3 ng/ml of VDS and IS, and a volunteer plasma at 3.0 h after intravenous infusion of Xi Ai Ke Vial® (3 mg of Vindesine Sulfate for Injection) were shown in Fig. 3. The retention times of IS and VDS were approximately 1.4 min and 2.3 min.
3.3.2. Linearity and lower limits of quantification The plasma concentration ranges were 0.3–28 ng/ml for VDS. The calibration model was selected on the basis of the analysis of the data by linear regression with intercepts and weighting factors. The best linear fit and least-squares residuals for the calibration curve were achieved with 1/x weighing factor VDS. The mean correlation coefficient (r) of the respective weighted calibration curves generated during the validation were 0.998 (0.997–0.999) for VDS. The lower limits of quantification (LLOQ) for VDS were 0.3 ng/ml (shown in Fig. 3). The back calculated value of each point of the calibration curve was comprised in the range of ±15% from the nominal concentration. The statistical data computed from the five analytical curves was listed in Table 2.
R.-H. Zhu et al. / Journal of Pharmaceutical and Biomedical Analysis 96 (2014) 31–36
35
Table 2 Deviation of calibration curves obtained for determination of vindensine in human plasma (n = 5). Back-calculated concentration for standards
Mean SD %CV %Bias
STD 1 (0.30 ng/ml)
STD 2 (0.65 ng/ml)
STD 3 (1.50 ng/ml)
STD 4 (3.30 ng/ml)
STD 5 (6.50 ng/ml)
STD 6 (13.0 ng/ml)
STD 7 (28.0 ng/ml)
0.326 0.024 7.4 8.7
0.624 0.020 3.1 −4.0
1.45 0.056 3.9 −3.3
3.16 0.108 3.4 −4.1
6.60 0.253 3.8 1.6
13.44 0.391 2.9 3.4
28.68 0.780 2.7 2.4
3.3.3. Precision and accuracy The intra- and inter-day accuracy and precision for VDS were listed in Table 3. The mean intra-day RSD (%) for VDS were 3.5–6.6%, the corresponding values for inter-day were 6.1–12.9%. The
Table 3 Intra- and inter-day precision and accuracy data for assays of vindesine in human plasma. Precision
Nominal concentration (ng/ml)
Meana (ng/ml) Intra-day (n = 5)
0.65 3.3 23.0
Inter-day (n = 3)
0.65 3.3 23.0
a
RSD (%)
Accuracy (%)
0.655 ± 0.023
3.5
100.8
3.18 ± 0.166 23.8 ± 1.58
5.2 6.6
96.3 103.3
0.634 ± 0.082
12.9
97.5
3.29 ± 0.199 22.4 ± 2.15
6.1 9.6
99.5 97.5
Values are means ± standard deviation.
intra-day accuracies of VDS were 96.3–103.3%, the corresponding values for inter-day were 97.5–99.5%. 3.3.4. Extraction recovery and matrix effect The extraction (absolute) recoveries for VDS at the three concentrations of 0.65, 3.0, 23.0 ng/ml from human plasma were 76.1 ± 6.09%, 75.2 ± 4.41%, 78.2 ± 6.39%. The value for IS at 50.0 ng/ml was 81.6 ± 5.47%. As for the matrix effect, the ratios (A/B × 100%) for VDS at 0.65, 3.0 and 23.0 ng/ml were 93.4 ± 6.4%, 96.1 ± 3.9% and 95.8 ± 3.5%, for IS at 50.0 ng/ml was 94.7 ± 4.5%. As defined in Section 2, all the ratios were between 85% and 115%, which means no matrix effect for VDS and IS in this method. 3.3.5. Stability Stock solutions of VDS and IS in methanol were stable for 30 days at −20 ◦ C; more than 95.0% of the spiked amounts were recovered. The stability data of VDS in human plasma under three conditions were listed in Table 4. As shown in Table 4, no significant degradation of VDS in human plasma was observed under all conditions studied. 3.4. Pharmacokinetic application After intravenous infusion of Xi Ai Ke Vial® (3 mg of Vindesine Sulfate for Injection) to 100 Chinese Han subjects with Table 4 Stability of vindesine in human plasma. Conditions
Percentage of control valuea 0.65 ng/ml
Fig. 3. Typical mass chromatograms (MRM) obtained from vindesine (channel 2) and vinorelbine (channel 1). (A) Drug-free plasma. (B) Blank plasma spiked with vindesine at the LLOQ level 0.3 ng/ml and vinorelbine. (C) Plasma sample from a patient 3.0 h after intravenous infusion 3 mg of Xi Ai Ke Vial® (Vindesine Sulfate for Injection).
Short-term stability (4 h) Three freeze–thaw cycles Autosampler stability (12 h) Long-term stability (30 days) a
110.7 108.4 93.3 104.8
± ± ± ±
Values are means ± standard deviation.
5.0 8.2 9.4 7.3
3.3 ng/ml 101.6 103.0 99.0 98.4
± ± ± ±
7.9 10.0 6.2 6.1
23.0 ng/ml 100.9 101.1 96.1 101.9
± ± ± ±
3.6 2.6 9.5 8.2
36
R.-H. Zhu et al. / Journal of Pharmaceutical and Biomedical Analysis 96 (2014) 31–36
after intravenous infusion Xi Ai Ke Vial® (3 mg of Vindesine Sulfate for Injection) to Chinese Han subjects with hematological malignant disorders. References
Fig. 4. Mean plasma concentration–time profiles of intravenous injection of Xi Ai Ke Vial® (3 mg of Vindesine Sulfate for Injection) to 100 unrelated Chinese Han subjects with hematological malignant disorders. Vertical bars represent standard deviation.
hematological malignant disorders, the mean plasma concentration–time profiles of VDS were shown in Fig. 4. Pharmacokinetic parameters of VDS in individual 100 subjects were obtained from population estimates according to Bayesian’ theorem. The results showed that the pharmacokinetics of VDS conformed with three compartment open model and the first order kinetics in this population. The mean value of systemic clearance (CL), terminal half-life (t1/2 ), distribution volume (V), and area under the curve (AUC) was estimated to be 0.381 ± 0.108 L/h/kg, 22.8 ± 10.4 h, 170.12 ± 69.29 L, 149 ± 67 ng h/ml, respectively. The terminal half-life (t1/2 ) was similar to early reports [16–18]. The systemic clearance is a little larger than the values (0.252 L/h/kg and 0.132 L/h/kg) reported by Nelson [16] and Ohnuma et al. [17], but is slightly less than the value (0.53 L/h/kg) reported by Rahmani et al. [18]. As compared with those classical pharmacokinetic studies which had limited sample size, our study recruited more subjects with wide age ranges and different extent of diseases. On statistical grounds, the smaller the sample size, the greater chance to obtain smaller or larger observed values. Therefore, we speculate that this difference among the studies is more likely to reflect the sampling error, and do not represent the actual difference between populations. 4. Conclusion This is the first reported study of plasma concentration of VDS using ultra-performance hydrophilic interaction liquid chromatography–tandem mass spectrometry (HILIC–MS/MS). It allows a suitable sensitivity of VDS (LLOQ 0.3 ng/ml) and can be carried out in a short time within 3.5 min. This method has been successfully applied for the evaluation of pharmacokinetics of VDS
[1] M. Bayssas, J. Gouveia, F. de Vassal, J.L. Misset, L. Schwarzenberg, P. Ribaud, M. Musset, C. Jasmin, M. Hayat, G. Mathé, Vindesine: a new vinca alkaloid, Recent Results Cancer Res. 74 (1980) 91–97. [2] J. Dancey, W.P. Steward, The role of vindesine in oncology-recommendations after 10 years’ experience, Anticancer Drugs 6 (1995) 625–636. [3] E. Cvitkovic, E. Wasserman, Role of vindesine as neoadjuvant chemotherapy for non-small cell lung and head and neck cancers, Anticancer Drugs 8 (1997) 734–745. [4] V.S. Sethi, S.S. Burton, D.V. Jackson, A sensitive radioimmunoassay for vincristine and vinblastine, Cancer Chemother. Pharmacol. 4 (1980) 183–187. [5] R. Rahmani, J. Barbet, J.P. Cano, A 125 I-radiolabelled probe for vinblastine and vindesine radioimmunoassays: applications to measurements of vindesine plasma levels in man after intravenous injections and long-term infusions, Clin. Chim. Acta 129 (1983) 57–69. [6] Y. Nakano, T. Saita, H. Fujito, Development of a specific and sensitive enzymelinked immunosorbent assay for vindesine, Yakugaku Zasshi 132 (2012) 727–732. [7] L. Barthe, J.P. Ribet, M. Pélissou, M.J. Degude, J. Fahy, A. Duflos, Optimization of the separation of vinca alkaloids by nonaqueous capillary electrophoresis, J. Chromatogr. A 968 (2002) 241–250. [8] R.L. Hussey, W.M. Newlon, High-performance liquid chromatographic and TLC determinations of desacetylvinblastine amide (vindesine) and its monosulfate salt, J. Pharm. Sci. 67 (1978) 1319–1320. [9] A. Paci, L. Mercier, P. Bourget, Identification and quantitation of antineoplastic compounds in chemotherapeutic infusion bags by use of HPTLC: application to the vinca-alkaloids, J. Pharm. Biomed. Anal. 30 (2003) 1603–1610. [10] J.H. Beijnen, D.E. Vendrig, W.J. Underberg, Stability of vinca alkaloid anticancer drugs in three commonly used infusion fluids, J. Parenter. Sci. Technol. 43 (1989) 84–87. [11] M. De Smet, S.J. Van Belle, G.A. Storme, D.L. Massart, High-performance liquid chromatographic determination of vinca-alkaloids in plasma and urine, J. Chromatogr. 345 (1985) 309–321. [12] D.E. Vendrig, J. Teeuwsen, J.J. Holthuis, Determination of vinca alkaloids in plasma and urine using ion-exchange chromatography on silica gel and fluorescence detection, J. Chromatogr. 434 (1988) 145–155. [13] D.E. Vendrig, J. Teeuwsen, J.J. Holthuis, Analysis of vinca alkaloids in plasma and urine using high-performance liquid chromatography with electrochemical detection, J. Chromatogr. 424 (1988) 83–94. [14] S. Maeda, Y. Miwa, Multicomponent high-performance liquid chromatography/tandem mass spectrometry analysis of ten chemotherapeutic drugs in wipe samples, J. Chromatogr. B 921–922 (2013) 43–48. [15] R. Rahmani, X.J. Zhou, M. Placidi, M. Martin, J.P. Cano, In vivo and in vitro pharmacokinetics and metabolism of vinca alkaloids in rat. I. Vindesine (4deacetyl-vinblastine 3-carboxyamide), Eur. J. Drug Metab. Pharmacokinet. 15 (1990) 49–55. [16] R.L. Nelson, The comparative clinical pharmacology and pharmacokinetics of vindesine, vincristine, and vinblastine in human patients with cancer, Med. Pediatr. Oncol. 10 (1982) 115–127. [17] T. Ohnuma, L. Norton, A. Andrejczuk, J.F. Holland, Pharmacokinetics of vindesine given as an intravenous bolus and 24-hour infusion in humans, Cancer Res. 45 (1985) 464–469. [18] R. Rahmani, R. Samak, P. Bore, J.P. Cano, Clinical pharmacokinetics of vinca alkaloids, Bull. Cancer 75 (1988) 195–200. [19] Guidance for Industry Bioanalytical Method Validation from FDA, US Department of Health and Human Services, Food and Drug Administration, CDER, Rockville, USA, 2001.
