- Email: [email protected]
MS methods for simultaneous determination of pitavastatin and its lactone metabolite in human plasma and urine involving a procedure for inhibiting the conversion of pitavastatin lactone to pitavastatin in plasma and its application to a pharmacokinetic study
Journal of Pharmaceutical and Biomedical Analysis 72 (2013) 8–15
Contents lists available at SciVerse ScienceDirect
Journal of Pharmaceutical and Biomedical Analysis journal homepage: www.elsevier.com/locate/jpba
Simple LC–MS/MS methods for simultaneous determination of pitavastatin and its lactone metabolite in human plasma and urine involving a procedure for inhibiting the conversion of pitavastatin lactone to pitavastatin in plasma and its application to a pharmacokinetic study Xiemin Qi a , Li Ding a,∗ , Aidong Wen b,∗∗ , Na Zhou a , Xiaolang Du a , Shailendra Shakya a a b
Department of Pharmaceutical Analysis, China Pharmaceutical University, 24 Tongjiaxiang, Nanjing 210009, China Department of Pharmacy, The First Affiliated Hospital of the Fourth Military Medical University of PLA, Xi’an 710032, China
a r t i c l e
i n f o
Article history: Received 15 July 2012 Received in revised form 14 September 2012 Accepted 22 September 2012 Available online 29 September 2012 Keywords: Pitavastatin Pitavastatin lactone LC-ESI-MS/MS Interconversion Stability Pharmacokinetics
a b s t r a c t Sometimes, drugs and their metabolites in plasma may convert to each other. This phenomenon is called interconversion, which may result in the instability problem of the plasma samples. The instability problem caused by interconversion of the co-existing metabolites may often be ignored, since there is no drug metabolite in the quality control samples prepared for method validation. Pitavastatin lactone (Pi-LAC), a main metabolite of pitavastatin (Pi), is very unstable and easily converted to Pi in plasma. In this paper, simple and rapid LC-ESI-MS/MS methods were developed for the simultaneous determination of Pi and Pi-LAC in human plasma and urine. The sample stability was examined under different conditions. The interconversion of Pi and Pi-LAC was prevented by adding a pH 4.2 buffer solution to the freshly collected plasma samples. Detection was performed using an electrospray ionization (ESI) operating in positive ion multiple reaction monitoring mode by monitoring the ion transitions from m/z 422.2→290.3 (Pi), 404.2→290.3 (Pi-LAC) and m/z 611.3→423.2 (candesartan cilextetil, the internal standard), respectively. The calibration curve of Pi and Pi-LAC in both human plasma and urine showed good linearity over the concentration range of 0.1–200 ng/mL. The established methods were successfully applied to a pharmacokinetic study of pitavastatin calcium tablets in healthy Chinese volunteers after oral administration of 1, 2 and 4 mg single and multiple doses of pitavastatin calcium. The pharmacokinetic parameters of Pi and Pi-LAC in Chinese volunteers were given respectively. The urinary excretion profiles of Pi and Pi-LAC in Chinese volunteers were also presented. After receiving a single 4 mg oral dose of pitavastatin calcium, the average cumulative urinary excretion percentages of Pi and Pi-LAC in Chinese volunteers were (0.41 ± 0.16)% and (6.1 ± 5.0)%, respectively. © 2012 Elsevier B.V. All rights reserved.
1. Introduction Pitavastatin (Pi) is a synthetic competitive inhibitor of HMG-CoA reductase, which can reduce plasma levels of low-density lipoprotein cholesterol by 40% in hypercholesterolaemic patients. It has the cholesterol-lowering effects and can reduce the risk of cardiovascular diseases in everyday medical practice [1,2]. Based on the preclinical findings, Pi has been widely used as a first-line agent to lipid-modifying therapies [3]. Lactonization is the major metabolic pathway of Pi in humans [4], and its main metabolite pitavastatin lactone (Pi-LAC) may be converted to Pi in plasma at room
∗ Corresponding author. Tel.: +86 25 83271289; fax: +86 25 83271289. ∗∗ Corresponding author. Tel.: +86 29 84773636; fax: +86 29 84773636. E-mail addresses: [email protected], [email protected] (L. [email protected] (A. Wen). 0731-7085/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jpba.2012.09.026
Ding),
temperature, see Fig. 1 [5,6]. This interconversion phenomenon of Pi-LAC and Pi in the samples may occur in the storage and sample preparation stages, and seriously affect the accuracy of sample determination results. So as to accurately characterize the clinical pharmacokinetic properties of Pi and Pi-LAC, it is critical to avoid their interconversion in plasma samples. The pH value of the plasma could significantly affect the interconversion of Pi and PiLAC. Pi-LAC would convert to Pi in the plasma samples without any treatment, and this conversion can be avoided by acidifying the samples. However, if the pH level of the plasma samples is too low, the conversion of Pi to Pi-LAC would occur. So it is important to find an optimal plasma pH value to avoid the interconversion of them. Interestingly, Pi is stable and would not convert to Pi-LAC in plasma samples without any treatment. This is the reason why the unstable problem of the incurred plasma samples containing both Pi and Pi-LAC has been ignored in the most reported methods for the determination of Pi only [7–15]. Different from the incurred
X. Qi et al. / Journal of Pharmaceutical and Biomedical Analysis 72 (2013) 8–15
9
2. Experimental 2.1. Reference standards and chemicals Pitavastatin calcium reference and pitavastatin calcium tablets were provided by Chongqing Institute of Pharmaceutical Industry. Pi-LAC was purchased from Nanjing Ou Xin Medical Technology Co., Ltd. Candesartan cilextetil, the internal standard (IS), was obtained from National Institute for Control of Pharmaceutical and Biological Products (Beijing, China). Methanol and acetonitrile were purchased from Merck KGaA (Darmstadt, Germany). Formic acid, acetic acid and ammonium acetate were of analytical grade and purchased from Nanjing Chemical Reagent Co., Ltd. (Nanjing, China). Distilled water was used throughout the study. 2.2. Instrumentation and operating conditions The liquid chromatography was performed on an Agilent 1200 Series liquid chromatography (Agilent Technologies, Palo Alto, CA, USA), which included an Agilent 1200 binary pump (model G1312B), vacuum degasser (model G1322A), Agilent 1200 autosampler (model G1367C), temperature controlled column compartment (model G1330B). The LC system was coupled with an Agilent 6410B triple quadrupole mass spectrometer (USA) equipped with an electrospray ionization source (model G1956B). The signal acquisition and peak integration were performed using the Masshunter Qualitative Analysis Software (B.03.01Build 346) supplied by Agilent Technologies. Separation of the analytes was achieved using Hedera ODS-2 C18 column (150 mm × 2.1 mm, 5 m) with a mobile phase of 0.2% formic acid–acetonitrile (10:90, v/v) at a flow rate of 0.28 mL/min. The column temperature was 43 ◦ C (for plasma sample assay) or 38 ◦ C (for urine sample assay). ESI was used in the positive mode with the drying gas temperature of 300 ◦ C, gas (N2 ) flow of 10 L/min, nebulizer of 40 psi and capillary voltage of 4000 V. The MRM transitions were chosen to be m/z 422.2→290.3 for Pi, m/z 404.2→290.3 for Pi-LAC and m/z 611.3→423.2 for the IS. The fragmentor voltage values set for Pi, Pi-LAC and the IS were 180 V, 195 V and 135 V, respectively. The collision energy values set for Pi, Pi-LAC and the IS were 30 eV, 28 eV and 10 eV, respectively. Fig. 1. Structures of Pi (A), Pi-LAC (B) and esterification product of Pi-LAC (C).
samples, there was only Pi without Pi-LAC in the spiked QC samples used for the method validation. So, the authors [7–15] had not found the problem that the co-existing Pi-LAC in the incurred samples might convert to Pi. Methylation [16–18] of Pi with diazomethane was applied to inhibit the conversion of Pi to Pi-LAC, but this did not stop the conversion of Pi-LAC to Pi. Preparing plasma samples under lower temperature conditions might inhibit the interconversion of the two analytes [19–21]. But this measure could not solve the instability problem completely. Our study showed that in the plasma samples, which spiked with only Pi-LAC and exposed to one frozen-thaw circle (frozen at −70 ◦ C, thaw at 4 ◦ C), approximately 17% of Pi-LAC converted to Pi. But this fact had not been mentioned in the previously published articles. Most of the reported methods emphasized on the determination of Pi only, and did not pay any attention to the conversion of Pi-LAC to Pi in plasma. In this paper a simultaneous determination method for Pi and Pi-LAC in human plasma and urine was established and validated. The test results indicated that the interconversion of Pi and Pi-LAC was minimized by this method. This approach was successfully applied to accurately determine the Pi and Pi-LAC in human plasma and characterize the clinical pharmacokinetics of Pi and Pi-LAC in humans.
2.3. Preparation of stock and working solutions The stock solutions were prepared with a mixture of methanol–water (7:3, v/v) for pitavastatin calcium and acetonitrile for Pi-LAC to reach a final concentration of 1 mg/mL, respectively. Working solutions were prepared by further diluting the stock solutions with methanol for pitavastatin calcium and acetonitrile for Pi-LAC to 2, 10, 100, 200, 1000, 2000 ng/mL. The standard stock solution of the IS was prepared in methanol to 0.5 mg/mL and diluted to 250 ng/mL with methanol. All the working solutions were kept at −20 ◦ C. 2.4. Calibration curves and quality control sample preparation A unique procedure of making the plasma calibration standards and quality control (QC) samples was started with the addition of the working solutions to the pH adjusted blank plasma. To an aliquot of the blank plasma, 20 L ammonium acetate buffer solution (0.6 M; pH 4.2) was added and vortex-mixed to obtain the pH adjusted blank plasma. The plasma calibration standards were prepared by spiking the pH adjusted blank plasma samples with appropriate volumes of the above-mentioned working solutions, to produce the plasma calibration standards at concentrations of 0.1, 0.5, 2.5, 10, 30, 60, 120, 200 ng/mL for Pi and Pi-LAC, respectively. By using the same procedure for preparing the calibration
10
X. Qi et al. / Journal of Pharmaceutical and Biomedical Analysis 72 (2013) 8–15
standards, the QC samples were prepared at the concentration levels of 0.25, 10, 180 ng/mL for both Pi and Pi-LAC. The same concentration levels of the two analytes were prepared for the urine calibration standards and QC samples in the blank urine without the pH adjustment. 2.5. Sample preparation 2.5.1. Plasma sample pretreatment After being drawn from volunteers, the blood samples were centrifuged immediately to separate the plasma, and an aliquot of 0.3 mL plasma sample was transferred to a tube containing 20 L of ammonium acetate buffer (0.6 M; pH 4.2), and vortex-mixed at once for 10 s, then stored at −20 ◦ C until analysis. 2.5.2. Plasma and urine sample preparation An aliquot of 0.3 mL plasma sample was mixed with 30 L of the IS solution (251.5 ng/mL) and deproteinized with 720 L of acetonitrile. The mixture was centrifuged at 15,600 rpm for 10 min and then the supernatant was transferred into an injection vial for LC–MS/MS analysis. After being vortex-mixed with 25 L of the IS solution (251.5 ng/mL), an aliquot of 700 L urine sample was diluted with 300 L acetonitrile and centrifuged at 15,600 rpm for 10 min. Then the supernatant was transferred into an injection vial for LC–MS/MS analysis. The whole analysis procedure was carried out in a dark room to avoid light-induced decomposition of the analytes. 2.6. Method validation The analytical methodology was validated according to the FDA’s guidance for industry on bioanalytical method validation. The selectivity, linearity, matrix effects (ME), recovery, accuracy, precision and stability were assessed as described by Zhu et al. [22]. 2.7. Interconversion characters 2.7.1. Interconversion of Pi-LAC to Pi The interconversion of Pi-LAC and Pi in plasma was evaluated at low, middle and high QC concentration levels of the two analytes. Samples that were used to evaluate the influencing factors on the interconversion were prepared according to the sample preparation described in Section 2.5.2. In this paper, only the test results obtained at the high QC concentration level were given as the representations. Three sets of plasma samples at the high QC concentration level without adding the buffer solution were used to investigate the interconversion of Pi-LAC and Pi: set 1 containing Pi-LAC only, set 2 containing the both analytes, set 3 containing Pi only. Several replicate samples were prepared in each set, and analyzed after being kept at room temperature for different time. The changing trends of Pi and Pi-LAC in the plasma samples caused by the interconversion were shown as the time-dependent variations of the measured mass responses (chromatographic peak areas) to the two analytes. 2.7.2. Influence of the different stabilizers on the interconversion Ethylene diamine tetraacetic acid (EDTA), potassium fluoride (KF) and sodium fluoride (NaF) are the common esterase inhibitors (called stabilizer in this section) used to prevent the hydrolysis of the esters in plasma samples [23]. In view of the instability of Pi-LAC in plasma induced by hydrolysis degradation to Pi, different amounts of the above mentioned stabilizers were added in the plasma samples, and their efficiency on the inhibiting the conversion of Pi-LAC to Pi was compared. The Pi-LAC high QC samples
containing varied amounts of the stabilizers were prepared in triplicates for each stabilizer. After being kept at room temperature for 1 h, these samples were analyzed according to Section 2.5.2. The residual amount of Pi-LAC in each sample was calculated.
2.7.3. Influence of pH of the buffer solution on the interconversion Several different pH values of the added buffers were assessed to prevent the sample-collection transformation of Pi-LAC into its prodrug. The solutions used for test were ammonium acetate buffer solution (0.6 M; pH 4.2), ammonium acetate buffer solution (0.6 M; pH 2.0) and hydrochloric acid (1 M; pH 0), respectively. After being treated with 20 L different solutions of different pH values, and maintaining at room temperature for 1.5 h, the high QC samples containing both Pi and Pi-LAC were assessed by LC–MS/MS, and the conversion ratios (CR) of Pi-LAC to Pi (or Pi to Pi-LAC) were calculated. It is defined as: CR (%) = (b − a)/a × 100, in which a and b represent the values of peak areas of the analytes obtained from the freshly prepared QC samples and the QC samples kept at room temperature for 1.5 h, respectively.
3. Results and discussion 3.1. Method development 3.1.1. LC–MS–MS optimization To maximize the MS responses to Pi and Pi-LAC, some instrument parameters were investigated and optimized. Positive ion monitoring mode was chosen because of its better sensitivity than negative ion monitoring mode. Additionally, reducing drying gas temperature from 350 ◦ C (the recommended value of the instrument manufacturer) to 300 ◦ C is a matter of choice which presented a significant benefit of reproducibility. Acetonitrile was selected as the organic portion of the mobile phase, because it showed lower background noise. The test results showed that higher column temperature could improve the peak shape and symmetry of Pi. The acceptable peak shape and symmetry of Pi was achieved at the column temperature of 43 ◦ C for the plasma sample analysis.
3.1.2. Sample pretreatment Characters of interconversion confirmed that Pi-LAC was very unstable in plasma, and the interconversion of Pi-LAC to Pi in plasma catalyzed by esterase. As the rate of hydrolysis of Pi-LAC to Pi might be dependent on pH, ionic strength and protein concentration, adding the ammonium acetate buffer was chosen to guarantee the stabilities of the analytes in plasma in the whole sample treatment process. However the instability problem of Pi-LAC did not occur in human urine, for there was no esterase in the urine.
3.2. Method validation 3.2.1. LC-ESI-MS/MS characteristics No endogenous interferences were observed at the retention times of the analytes, see Fig. 2. Pi, Pi-LAC and the IS exhibited sharp peaks at 1.7, 2.0 and 2.2 min, respectively, while the full run time was 3.2 min. The matrix effects (ME) at concentrations of 0.25, 10, 180 ng/mL were (98.9 ± 5.4)%, (102.5 ± 2.9)% and (96.6 ± 3.7)% for Pi; and (97.8 ± 6.8)%, (99.4 ± 4.2)% and (100.7 ± 1.4)% for Pi-LAC. The matrix effect for the IS (251.5 ng/mL) was (99.4 ± 3.9)%. The extraction recoveries were (96.2 ± 4.1)%, (98.8 ± 3.6)% and (96.3 ± 2.7)% for Pi, and (99.3 ± 6.5)%, (94.9 ± 2.4)% and (96.8 ± 3.1)% for Pi-LAC (n = 5) at concentration levels of 0.25, 10 and 180 ng/mL, respectively.
X. Qi et al. / Journal of Pharmaceutical and Biomedical Analysis 72 (2013) 8–15
11
3.2.3. Stability Stability tests were designed to assess the stability of the analytes under the conditions expected during handling of clinical samples. The stability results of the analytes in plasma and urine were summarized in Tables 1 and 2, respectively. Pi, Pi-LAC and the IS were found to be stable in their working solutions after being placed at room temperature for 4.5 h and stored at −20 ◦ C for a period of 50 d. 3.2.4. Carry-over effect The carry-over effect was tested by instantly analyzing blank samples following upper limit of quantification (ULOQ). No residual was observed at the retention times of Pi, Pi-LAC and the IS in the chromatograms of blank plasma and urine samples. 3.3. Interconversion characters of Pi-LAC and Pi The immediate transformation of the unstable metabolite into its parent drug after sample collection was a major concern. The main purpose was to explore the characters of the interconversion of the two analytes under several different conditions and to solve the instability problem.
Fig. 2. MRM chromatograms of (A) blank plasma, (B) plasma spiked with Pi and Pi-LAC at 0.1 ng/mL (LLOQ) and the IS, (C) a plasma sample collected after 0.5 h of single 2 mg oral dose, and (D) a urine sample collected in the interval 2–4 h after single 4 mg oral dose.
3.2.2. Linearity, lower limit of quantification, precision and accuracy The calibration curves were obtained by weighted linear regression, the peak area ratio (analyte/IS) was plotted vs the analyte concentrations. The calibration curve was linear over the range of 0.1–200 ng/mL in human plasma and urine with coefficient of correlation (r2 ) > 0.997 for both analytes. At the LLOQ (0.1 ng/mL), the RSD (%) < 8.3% (n = 5) and the RE (%) ranged from −0.1% to 7.3%. Validation samples of five replicates of the QC samples were prepared and analyzed in three separate analytical batches to evaluate the accuracy and intra-batch and inter-batch precisions of the methods. The intra-batch and inter-batch precisions were measured to be less than 8.3% and 9.8%, respectively, and the accuracy to be within ±15%.
3.3.1. Interconversion of Pi-LAC to Pi Pi-LAC is unstable in methanol because of methyl esterification. A stability test showed that 17% of Pi-LAC was converted to its methyl ester (Fig. 1) in its methanol solution after keeping at −20 ◦ C for 20 d. Therefore, acetonitrile was selected as the solvent to prevent the methyl esterification of Pi-LAC. In this study, an acetic acid–ammonium acetate buffer solution (0.6 M; pH 4.2) was used as stabilizing agent, which was promptly added to the freshly collected plasma samples. The interconversion profiles of the two analytes were presented by plotting the MS response intensity of them vs time curves (see Fig. 3). In Fig. 3, the tendency of the interconversion of the two analytes is observed clearly. The test results of the sets 1 and 2 indicated a prominent degradation of Pi-LAC, and over 55% of Pi-LAC degraded and mainly converted to Pi in the first 1.5 h. At the same time, a significant increase of Pi in the plasma samples was observed. The presence of Pi in plasma may inhibit the conversion of Pi-LAC to Pi to some extent. For the initial presence of a large amount of Pi, the decreasing process of Pi-LAC in set 2 was slower than that in set 1. Interestingly, Pi was stable in human plasma and no Pi-LAC was found in the samples of the set 3 in the whole test period, see Fig. 3C. It demonstrated that Pi was stable and would not convert to Pi-LAC in plasma under normal conditions, but Pi-LAC was susceptible to hydrolysis. 3.3.2. Influence of the different stabilizers on the interconversion The test results of the influence of the different stabilizers on the interconversion are summarized in Fig. 4. The ordinate of Fig. 4 represents the residual amounts of Pi-LAC in the samples treated with different stabilizers. It was observed that the esterase inhibitors EDTA (2 mol/L) and KF (3.4 mol/L) did not play a key role in the inhibition of the conversion of Pi-LAC, but the addition of the pH 4.2 buffer solution can inhibit the conversion efficiently. 3.3.3. Influence of the pH values of the buffer solutions on the interconversion The influence of the different pH values of the added buffer solutions on the interconversion of the analytes was investigated. Fig. 5 shows that the different solutions added to the plasma could significantly affect the interconversion of Pi and Pi-LAC. The conversion of Pi to Pi-LAC would occur when the pH value of the plasma was adjusted too low. But, when the pH value of the plasma was not controlled, much amount of Pi-LAC would easily convert to Pi. So it
12
X. Qi et al. / Journal of Pharmaceutical and Biomedical Analysis 72 (2013) 8–15
Table 1 Stability of Pi and Pi-LAC in human plasma at three QC levels (n = 3). Storage conditions
Analytes
Short-term stability (6 h, room temperature)
Pi
Pi-LAC
Pre-preparative stability at 10 ◦ C for 7 h (Autosampler)
Pi
Pi-LAC
Freeze/thaw stability (3 cycles)
Pi
Pi-LAC
35 d long-term stability
Pi
Pi-LAC
Concentration levels (ng/mL)
RSD %
RE %
0.2415 9.908 183.6 0.2616 9.931 180.9
5.5 3.7 3.1 6.7 2.4 3.5
−3.6 −1.1 1.8 3.6 −1.7 −0.5
0.2505 10.02 180.4 0.2525 10.10 181.8
0.2571 10.81 186.9 0.2618 10.89 189.2
4.9 2.2 2.6 6.4 3.5 3.1
2.6 7.9 3.6 3.7 7.8 4.1
0.2505 10.02 180.4 0.2525 10.10 181.8
0.2477 10.07 180.9 0.2431 10.33 171.3
4.2 3.3 3.0 4.1 2.0 2.8
−1.1 0.5 0.3 −3.7 2.3 −5.8
0.2505 10.02 180.4 0.2525 10.10 181.8
0.2439 9.789 184.1 0.2437 9.597 181.7
5.3 2.8 3.1 4.3 4.5 3.1
−2.6 −2.3 2.1 −3.5 −5.0 −0.1
Added
Found
0.2505 10.02 180.4 0.2525 10.10 181.8
RSD, relative standard deviation; RE %, [(found − added)/added] × 100.
is important to find an optimal plasma pH value to avoid the interconversion of the analytes. Interestingly, Pi is stable and would not convert to its lactone in plasma samples under normal conditions. That is why the instability problem of the Pi plasma samples, which also containing Pi-LAC in them, was ignored by those bioanalysis methods reported in the literatures [7–15]. Commonly, we develop a method for the quantification of a parent drug using the spiked blank plasma (in vitro) but not the incurred plasma (in vivo). Care must be taken, however, as validations in vitro are not absolutely equal to those in vivo for some drugs. Sometimes, the metabolites
in plasma samples may convert to their parent drugs. However, the instability problem of the plasma samples caused by the coexisting metabolites has been largely ignored, for QC samples used for the validation of an analytical method, which are prepared in the same blank plasma as the calibration samples, will not reveal metabolite-related instability observed in the incurred samples. For unstable analytes, the greatest challenge is how to keep them stable before analyzing them. Tian et al. [19] and Ashwini et al. [20] reported that Pi-LAC in human plasma was stable in the freezethaw cycles. In fact, Pi-LAC degrades rapidly in plasma at room
Table 2 Stability of Pi and Pi-LAC in human urine at three QC levels (n = 3). Storage conditions
Short-term stability (3.5 h, room temperature)
Analytes
Pi
Pi-LAC
Pre-preparative stability at 10 ◦ C for 4 h (Autosampler)
Pi
Pi-LAC
Freeze/thaw stability (3 cycles)
Pi
Pi-LAC
15 d long-term stability
Pi
Pi-LAC
Concentration levels (ng/mL)
RSD %
RE %
0.2538 9.654 185.3 0. 2702 10.32 186.7
3.2 4.8 2.7 6.4 3.5 3.2
1.3 −3.7 2.7 7.0 2.2 2.7
0.2505 10.02 180.4 0.2525 10.10 181.8
0.2489 9.973 165.8 0.2449 9.915 174.3
5.1 3.7 2.9 3.3 4.5 3.6
−0.6 −0.5 −8.1 −3.0 −1.8 4.1
0.2505 10.02 180.4 0.2525 10.10 181.8
0.2491 10.22 170.5 0.2456 9.406 188.9
4.6 6.7 3.1 5.7 4.1 3.4
−0.6 2.0 −5.5 −2.7 −6.9 3.9
0.2505 10.02 180.4 0.2525 10.10 181.8
0.2441 9.398 185.4 0.2433 10.47 177.6
4.8 4.0 2.6 4.3 2.7 2.6
−2.6 −6.2 2.8 −3.6 3.7 −2.3
Added
Found
0.2505 10.02 180.4 0.2525 10.10 181.8
X. Qi et al. / Journal of Pharmaceutical and Biomedical Analysis 72 (2013) 8–15
13
Fig. 4. The residual amounts of Pi-LAC in the QC samples treated with different stabilizers after keeping at room temperature for 1 h.
Fig. 3. The interconversion profiles (the MS response intensity vs time curves) of Pi and Pi-LAC in the plasma samples of set 1 (A), set 2 (B) and set 3 (C).
temperature, and even during the procedure of freezing-thawing. Enzymes have a catalytic activity and some chemical reactions cannot happen in the absence of enzymes. But according to our test results, the interconversion of Pi-LAC to Pi was still going on with a lower speed even after the plasma samples were treated with deproteinization (data were not given). It is supposed that apart from enzymes, there may be some other influencing factors. Hence it is vital to prevent the degradation of Pi-LAC in plasma by using the method described in this article. Both analytes can be well stabilized by immediate addition of the buffer solution, and there is no need to use an ice bath [19,21] or derivatization [16–18] for the sample preparation. 3.4. Pharmacokinetic study The established methods were applied in a pharmacokinetic study. Thirty healthy Chinese volunteers, 15 males and 15 females,
Fig. 5. The CR (%) of Pi (A) and Pi-LAC (B) in the plasma samples added with the solutions of different pH values.
14
X. Qi et al. / Journal of Pharmaceutical and Biomedical Analysis 72 (2013) 8–15
Table 3 Pharmacokinetic parameters of Pi and Pi-LAC in Chinese volunteers (n = 30). Dose (mg)
Single dose 1
Multi-dose 2
4
2
Pi
AUC (ng h/mL) Cmax (ng/mL) Tmax (h) t1/2 (h) CL (L/h) Vd (L) R
59.60 ± 23.27 ± 0.6 ± 11.6 ± 16.1 ± 264.9 ± –
12.30 2.799 0.1 1.6 3.1 54.12
116.7 ± 57.65 ± 0.6 ± 10.1 ± 26.0 ± 369.3 ± –
89.8 40.58 0.2 1.8 17.1 252.4
213.0 ± 88.78 ± 0.6 ± 12.8 ± 18.5 ± 339.0 ± –
29.17 27.81 0.1 1.2 2.5 65.00
142.0 53.67 0.8 11.6 20.7 312.4 1.2
± ± ± ± ± ± ±
126.2 45.33 0.1 3.0 11.2 159.2 0.3
Pi-LAC
AUC (ng h/mL) Cmax (ng/mL) Tmax (h) t1/2 (h) CL (L/h) Vd (L) R
137.4 ± 20.21 ± 1.1 ± 10.8 ± 7.1 ± 108.3 ± –
18.69 4.348 0.4 1.8 1.0 19.27
234.7 ± 47.78 ± 1.0 ± 11.7 ± 9.5 ± 158.2 ± –
143.2 32.60 0.3 1.3 3.7 74.55
438.6 ± 84.09 ± 1.1 ± 12.1 ± 9.3 ± 159.9 ± –
115.8 28.45 0.3 1.4 2.4 31.16
270.9 47.38 1.2 12.3 8.9 153.6 1.1
± ± ± ± ± ± ±
201.2 33.87 0.3 1.2 4.8 87.98 0.3
AUC, area under the drug concentration–time curve; Cmax , maximum concentration; Tmax , time to reach peak concentration; t1/2 , elimination half life; R, the accumulation ratio was calculated from AUC /AUC0− where AUC is the AUC in a dosing interval ( = 12) during multi-dose administration and AUC0− is the AUC within time span 0–12 h after single dose of 2 mg. The volume of distribution (Vd) and clearance (CL) were calculated from the formulas: Vd = CL × t1/2 /0.693 and CL = Dose/AUC0−∞ , respectively.
were screened to enroll in the study after signing the informed consent form. The study was approved by the Ethics Committee of the First Affiliated Hospital of the Fourth Military Medical University of PLA (Xi’an, China) and conducted at this study center, in accordance with the GCP requirements and the Declaration of Helsinki. Written informed consent was obtained from each subject. A randomized, open-label and parallel-group design was used to evaluate the pharmacokinetic profiles of Pi and Pi-LAC after single oral dosing. Twenty healthy volunteers were randomized into 1 and 4 mg dose groups (10 volunteers in each, half male and half female) for the determination of the Pi and Pi-LAC pharmacokinetic profiles of the single-dose of pitavastatin calcium tablets. Another 10 volunteers, half male and half female, were administrated 2 mg dose of pitavastatin calcium tablets on day 1; received no treatment on day 2; and continued to receive the study drug once daily from days 3 through 7 to assess multiple-dose pharmacokinetics. The volunteers were required to fast overnight (12 h) before drug administration, while standard meals and water intake were provided 4 h post-dose. Blood samples were collected at 0 h (pre-dose), 0.25, 0.5, 0.75, 1, 1.5, 2, 3, 4, 6, 8, 12, 24, 36 and 48 h post-dose for the single-dose study. For the multiple-dose study, blood samples were collected prior to dosing on days 1, 5, 6 and 7 (0 h prior to dosing) and at 0.25, 0.5, 0.75, 1, 1.5, 2, 3, 4, 6, 8, 12, 24, 36 and 48 h post-dose on day 1 and day 7. Plasma was separated and stored at −20 ◦ C until analysis. The urine samples were collected pre-dose and over the intervals 0–2, 2–4, 4–6, 6–8, 8–12, 12–24, 24–36 and 36–48 h post-dose on day 1 of the 4 mg dose group. The collected plasma samples and the urine samples were assessed by the established methods. The concentration levels of Pi and Pi-LAC in each sample were calculated and the pharmacokinetics of Pi and PiLAC were evaluated. The mean plasma concentration–time profiles and the urinary excretion characteristics of Pi and Pi-LAC are illustrated in Figs. 6 and 7, respectively. The urinary excretion profile of Pi-LAC in Chinese volunteers was reported for the first time in this paper. In the interval of 0–2 h post-dose, the urinary excretion rates for both analytes reached the peak values of (5.04 ± 2.53) g/h and (85.3 ± 59.8) g/h, respectively. The average cumulative urinary excretion percentages of Pi and Pi-LAC in 48 h post-dose were (0.41 ± 0.16)% and (6.1 ± 5.0)%, respectively, suggesting that Pi and Pi-LAC may be subject to non-renal elimination in healthy subjects. Pi is a hydrophilic compound. And it was reported for pitavastatin that the biliary excretion of the intact form was the main elimination pathway in some experimental animals including rats, rabbits and dogs [24]. The pharmacokinetic parameters for Pi and Pi-LAC
are presented in Table 3. Both the AUC and Cmax of two analytes were dose-proportional over the dose range of 1–4 mg. No accumulation and gender differences were found. The pharmacokinetic parameters showed that the AUC and Cmax of Pi and Pi-LAC in extensive metabolizers were about four times more than that of poor metabolizers in the 2 mg dose group. This suggested a significant individual difference in the pharmacokinetic characteristics of Pi and its metabolite. Several single-nucleotide polymorphisms (SNPs) of OATP1B1 are known which associate with the transport capacity and significantly alter the disposition of Pi [25]. This may explain the wide inter-individual variation in Pi pharmacokinetics.
Fig. 6. Mean plasma concentration–time profiles of Pi (A) and Pi-LAC (B) in Chinese volunteers after the single dose administration (n = 30).
X. Qi et al. / Journal of Pharmaceutical and Biomedical Analysis 72 (2013) 8–15
Fig. 7. Mean urinary cumulative excretion percentage–time profiles of Pi and Pi-LAC in Chinese volunteers after a single oral administration of 4 mg dose pitavastatin calcium tablets (n = 10).
4. Conclusion The interconversion of Pi-LAC to Pi in plasma and its influencing factors were evaluated. The studies indicate that Pi-LAC undergoes fast degradation by hydrolysis in plasma. Addition of the buffer solution (pH 4.2) to the freshly collected plasma samples can prevent the interconversion of Pi-LAC to Pi. The simple and rapid LC-ESI-MS/MS methods were developed for the simultaneous determination of Pi and Pi-LAC in human plasma and urine. The methods were used to measure the concentration of Pi and Pi-LAC in bio-samples from healthy volunteers after oral administration of pitavastatin calcium tablets. The simple method totally solved the instability problems of Pi and Pi-LAC in plasma, and was successfully used in the clinical pharmacokinetic studies. The urinary excretion profile of Pi-LAC in human was reported for the first time in this paper. References [1] Y. Saito, N. Yamada, T. Teramoto, H. Itakura, Y. Hata, N. Nakaya, H. Mabuchi, M. Tushima, J. Sasaki, N. Ogawa, Y. Goto, A randomized, double-blind trial comparing the efficacy and safety of pitavastatin versus pravastatin in patients with primary hypercholesterolaemia, Atherosclerosis 162 (2002) 373–379. [2] A. Tonkin, P. Aylward, D. Colquhoun, P. Glasziou, P. Harris, S. MacMahon, P. Magnus, D. Newel, P. Nestel, N. Sharpe, D. Hunt, J. Shaw, R.J. Simes, P. Thompson, A. Thomson, M. West, H. White, S. Simes, W. Hague, S. Caleo, J. Hall, A. Martin, S. Mulray, P. Barter, L. Beilin, R. Collins, J. McNeil, P. Meier, H. Willimott, D. Smithers, P. Wallace, D. Sullivan, A. Keech, Prevention of cardiovascular events and death with pravastatin in patients with coronary heart disease and a broad range of initial cholesterol levels, N. Engl. J. Med. 339 (1998) 1349–1357. [3] R.J. Havel, E. Rapaport, Management of primary hyperlipidemia, N. Engl. J. Med. 332 (1995) 1491–1498. [4] I. Yamada, H. Fujino, S. Shimada, J. Kojima, Metabolic fate of pitavastatin, a new inhibitor of HMG-CoA reductase: similarities and difference in the metabolism of pitavastatin in monkeys and humans, Xenobiotica 33 (2003) 789–803. [5] H. Fujino, I. Yamada, S. Shimada, M. Yoneda, J. Kojima, Metabolic fate of pitavastatin, a new inhibitor of HMG-CoA reductase: human UDPglucuronosyl-transferase enzymes involved in lactonization, Xenobiotica 33 (2003) 27–41.
15
[6] H. Fujino, I. Yamada, S. Shimada, J. Kojima, Metabolic fate of pitavastatin, a new inhibitor of HMG-CoA reductase-effect of Cmoat deficiency on hepatobiliary excretion in rats and of mdr 1 a/b gene disruption on tissue distribution in mice, Drug. Metab. Pharmacokinet. 17 (2002) 449–456. [7] H. Lv, J.G. Sun, G.J. Wang, X.Y. Zhu, Y. Zhang, S.H. Gu, Y. Liang, J. Sun, Determination of pitavastatin in human plasma via HPLC-ESI-MS/MS and subsequent application to a clinical study in healthy Chinese volunteers, Clin. Chim. Acta 386 (2007) 25–30. [8] B. Di, M.X. Su, F. Yu, L.J. Qu, L.P. Zhao, M.C. Cheng, L.P. He, Solid-phase extraction and liquid chromatography/tandem mass spectrometry assay for the determination of pitavastatin in human plasma and urine for application to phase I clinical pharmacokinetic studies, J. Chromatogr. B 868 (2008) 95–101. [9] J.H. Wen, Y.Q. Xiong, Determination of the concentration of pitavastatin in human plasma by LC–MS and subsequent application to a clinical study, Chin. JMAP 27 (2010) 62–65. [10] C. Qiu, M. Zhang, X. Liu, Y.G. Wen, The research of pharmacokinetics of pitavastatin in healthy volunteers, Pharm. Care Res. 9 (2009) 225–227. [11] C. Deng, T. Yi, L. Tang, Y.J. Li, A simple and sensitive LC–MS/MS method for determining the concentration of pitavastatin in human plasma, Chin. J. New Drug 18 (2009) 1113–1115. [12] C.X. Chen, G.W. Li, Y.G. Wen, Rapid determination of pitavastatin in plasma by LC-ESI-MS/MS, Chin. JMAP 27 (2010) 642–645. [13] J.Z. Shen-Tu, X. Xu, J. Liu, X.J. Hu, J.C. Chen, L.H. Wu, M.Z. Huang, H.L. Zhou, Determination of pitavastatin in human plasma by LC–MS–MS, J. Chromatogr. 69 (2009) 1041–1047. [14] Y. Chen, Z.R. Tan, G. Zhou, D.S. Ou-Yang, J. Zhang, H.H. Zhou, LC–MS determination of pitavastatin in human plasma and its application, J. Clin. Pharmacol. 46 (2011) 134–137. [15] J.W. Deng, K.B. Kim, I.S. Song, J.H. Shon, H.H. Zhou, K.H. Liu, J.G. Shin, Determination of two HMG-CoA reductase inhibitors, pravastatin and pitavastatin, in plasma samples using liquid chromatography–tandem mass spectrometry for pharmaceutical study, Biomed. Chromatogr. 22 (2008) 131–135. [16] H. Fujino, J. Kojima, Y. Yamada, H. Kanda, H. Kimata, Studies on the metabolic fate of NK-104, a new inhibitor of HMG-CoA Reductase (4): interspecies variation in laboratory animals and humans, Xenobio. Metabol. Dispos. 14 (1999) 79–91. [17] J. Kojima, H. Fujino, M. Yosimura, H. Morikawa, H. Kimata, Simultaneous determination of NK-104 and its lactone in biological samples by column-switching high-performance liquid chromatography with ultraviolet detection, J. Chromatogr. B 724 (1999) 173–180. [18] C.K. Hui, B.M.Y. Cheung, G.K.K. Lau, Pharmacokinetics of pitavastatin in subjects with child – Pugh A and B cirrhosis, Br. J. Clin. Pharmacol. 59 (2004) 291–297. [19] L. Tian, Y.L. Huang, Y.H. Jia, L. Hua, Y.S. Li, Development and validation of a liquid chromatography-tandem mass spectrometric assay for pitavastatin and its lactone in human plasma and urine, J. Chromatogr. B 865 (2008) 127–132. [20] O. Ashwini, G. Swati, V. Chintan, P. Harish, Determination of pitavastatin from human plasma using high performance liquid chromatography with fluorescence detection, Chin. J. Chromatogr. 25 (2007) 715–718. [21] L.L. Wang, X.Y. Chen, B. Song, Y.F. Zhang, D.F. Zhong, Determination of pitavastatin in human plasma by liquid chromatography–tandem mass spectrometry, Chin. J. New Drugs Clin. Rem. 26 (2007) 805–808. [22] H. Zhu, L. Ding, S. Shakya, X.M. Qi, L.L. Hu, X.L. Yang, Z.L. Yang, Simultaneous determination of asperosaponin VI and its active metabolite hederagenin in rat plasma by liquid chromatography–tandem mass spectrometry with positive/negative ion-switching electrospray ionization and its application in pharmacokinetic study, J. Chromatogr. B 30 (2011) 3407–3414. [23] Y.R. Zhao, L. Ding, H.W. Fan, Y. Yu, X.M. Qi, Y. Leng, Y.K. Rao, Determination of the unstable drug otilonium bromide in human plasma by LC-ESI-MS and its application to a pharmacokinetic study, J. Chromatogr. B 878 (2010) 2896–2900. [24] J. Kojima, H. Fujino, H. Abe, M. Yosimura, H. Kanda, H. Kimata, Identification of metabolites of NK-104, an HMG-CoA reductase inhibitor, in rat, rabbit and dog bile, Biol. Pharm. Bull. 22 (1999) 142–150. [25] J. Wen, Y. Xiong, OATP1B1 388A>G polymorphism and pharmacokinetics of pitavastatin in Chinese healthy volunteers, J. Clin. Pharm. Ther. 35 (2010) 99–104.
Contents lists available at SciVerse ScienceDirect
Journal of Pharmaceutical and Biomedical Analysis journal homepage: www.elsevier.com/locate/jpba
Simple LC–MS/MS methods for simultaneous determination of pitavastatin and its lactone metabolite in human plasma and urine involving a procedure for inhibiting the conversion of pitavastatin lactone to pitavastatin in plasma and its application to a pharmacokinetic study Xiemin Qi a , Li Ding a,∗ , Aidong Wen b,∗∗ , Na Zhou a , Xiaolang Du a , Shailendra Shakya a a b
Department of Pharmaceutical Analysis, China Pharmaceutical University, 24 Tongjiaxiang, Nanjing 210009, China Department of Pharmacy, The First Affiliated Hospital of the Fourth Military Medical University of PLA, Xi’an 710032, China
a r t i c l e
i n f o
Article history: Received 15 July 2012 Received in revised form 14 September 2012 Accepted 22 September 2012 Available online 29 September 2012 Keywords: Pitavastatin Pitavastatin lactone LC-ESI-MS/MS Interconversion Stability Pharmacokinetics
a b s t r a c t Sometimes, drugs and their metabolites in plasma may convert to each other. This phenomenon is called interconversion, which may result in the instability problem of the plasma samples. The instability problem caused by interconversion of the co-existing metabolites may often be ignored, since there is no drug metabolite in the quality control samples prepared for method validation. Pitavastatin lactone (Pi-LAC), a main metabolite of pitavastatin (Pi), is very unstable and easily converted to Pi in plasma. In this paper, simple and rapid LC-ESI-MS/MS methods were developed for the simultaneous determination of Pi and Pi-LAC in human plasma and urine. The sample stability was examined under different conditions. The interconversion of Pi and Pi-LAC was prevented by adding a pH 4.2 buffer solution to the freshly collected plasma samples. Detection was performed using an electrospray ionization (ESI) operating in positive ion multiple reaction monitoring mode by monitoring the ion transitions from m/z 422.2→290.3 (Pi), 404.2→290.3 (Pi-LAC) and m/z 611.3→423.2 (candesartan cilextetil, the internal standard), respectively. The calibration curve of Pi and Pi-LAC in both human plasma and urine showed good linearity over the concentration range of 0.1–200 ng/mL. The established methods were successfully applied to a pharmacokinetic study of pitavastatin calcium tablets in healthy Chinese volunteers after oral administration of 1, 2 and 4 mg single and multiple doses of pitavastatin calcium. The pharmacokinetic parameters of Pi and Pi-LAC in Chinese volunteers were given respectively. The urinary excretion profiles of Pi and Pi-LAC in Chinese volunteers were also presented. After receiving a single 4 mg oral dose of pitavastatin calcium, the average cumulative urinary excretion percentages of Pi and Pi-LAC in Chinese volunteers were (0.41 ± 0.16)% and (6.1 ± 5.0)%, respectively. © 2012 Elsevier B.V. All rights reserved.
1. Introduction Pitavastatin (Pi) is a synthetic competitive inhibitor of HMG-CoA reductase, which can reduce plasma levels of low-density lipoprotein cholesterol by 40% in hypercholesterolaemic patients. It has the cholesterol-lowering effects and can reduce the risk of cardiovascular diseases in everyday medical practice [1,2]. Based on the preclinical findings, Pi has been widely used as a first-line agent to lipid-modifying therapies [3]. Lactonization is the major metabolic pathway of Pi in humans [4], and its main metabolite pitavastatin lactone (Pi-LAC) may be converted to Pi in plasma at room
∗ Corresponding author. Tel.: +86 25 83271289; fax: +86 25 83271289. ∗∗ Corresponding author. Tel.: +86 29 84773636; fax: +86 29 84773636. E-mail addresses: [email protected], [email protected] (L. [email protected] (A. Wen). 0731-7085/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jpba.2012.09.026
Ding),
temperature, see Fig. 1 [5,6]. This interconversion phenomenon of Pi-LAC and Pi in the samples may occur in the storage and sample preparation stages, and seriously affect the accuracy of sample determination results. So as to accurately characterize the clinical pharmacokinetic properties of Pi and Pi-LAC, it is critical to avoid their interconversion in plasma samples. The pH value of the plasma could significantly affect the interconversion of Pi and PiLAC. Pi-LAC would convert to Pi in the plasma samples without any treatment, and this conversion can be avoided by acidifying the samples. However, if the pH level of the plasma samples is too low, the conversion of Pi to Pi-LAC would occur. So it is important to find an optimal plasma pH value to avoid the interconversion of them. Interestingly, Pi is stable and would not convert to Pi-LAC in plasma samples without any treatment. This is the reason why the unstable problem of the incurred plasma samples containing both Pi and Pi-LAC has been ignored in the most reported methods for the determination of Pi only [7–15]. Different from the incurred
X. Qi et al. / Journal of Pharmaceutical and Biomedical Analysis 72 (2013) 8–15
9
2. Experimental 2.1. Reference standards and chemicals Pitavastatin calcium reference and pitavastatin calcium tablets were provided by Chongqing Institute of Pharmaceutical Industry. Pi-LAC was purchased from Nanjing Ou Xin Medical Technology Co., Ltd. Candesartan cilextetil, the internal standard (IS), was obtained from National Institute for Control of Pharmaceutical and Biological Products (Beijing, China). Methanol and acetonitrile were purchased from Merck KGaA (Darmstadt, Germany). Formic acid, acetic acid and ammonium acetate were of analytical grade and purchased from Nanjing Chemical Reagent Co., Ltd. (Nanjing, China). Distilled water was used throughout the study. 2.2. Instrumentation and operating conditions The liquid chromatography was performed on an Agilent 1200 Series liquid chromatography (Agilent Technologies, Palo Alto, CA, USA), which included an Agilent 1200 binary pump (model G1312B), vacuum degasser (model G1322A), Agilent 1200 autosampler (model G1367C), temperature controlled column compartment (model G1330B). The LC system was coupled with an Agilent 6410B triple quadrupole mass spectrometer (USA) equipped with an electrospray ionization source (model G1956B). The signal acquisition and peak integration were performed using the Masshunter Qualitative Analysis Software (B.03.01Build 346) supplied by Agilent Technologies. Separation of the analytes was achieved using Hedera ODS-2 C18 column (150 mm × 2.1 mm, 5 m) with a mobile phase of 0.2% formic acid–acetonitrile (10:90, v/v) at a flow rate of 0.28 mL/min. The column temperature was 43 ◦ C (for plasma sample assay) or 38 ◦ C (for urine sample assay). ESI was used in the positive mode with the drying gas temperature of 300 ◦ C, gas (N2 ) flow of 10 L/min, nebulizer of 40 psi and capillary voltage of 4000 V. The MRM transitions were chosen to be m/z 422.2→290.3 for Pi, m/z 404.2→290.3 for Pi-LAC and m/z 611.3→423.2 for the IS. The fragmentor voltage values set for Pi, Pi-LAC and the IS were 180 V, 195 V and 135 V, respectively. The collision energy values set for Pi, Pi-LAC and the IS were 30 eV, 28 eV and 10 eV, respectively. Fig. 1. Structures of Pi (A), Pi-LAC (B) and esterification product of Pi-LAC (C).
samples, there was only Pi without Pi-LAC in the spiked QC samples used for the method validation. So, the authors [7–15] had not found the problem that the co-existing Pi-LAC in the incurred samples might convert to Pi. Methylation [16–18] of Pi with diazomethane was applied to inhibit the conversion of Pi to Pi-LAC, but this did not stop the conversion of Pi-LAC to Pi. Preparing plasma samples under lower temperature conditions might inhibit the interconversion of the two analytes [19–21]. But this measure could not solve the instability problem completely. Our study showed that in the plasma samples, which spiked with only Pi-LAC and exposed to one frozen-thaw circle (frozen at −70 ◦ C, thaw at 4 ◦ C), approximately 17% of Pi-LAC converted to Pi. But this fact had not been mentioned in the previously published articles. Most of the reported methods emphasized on the determination of Pi only, and did not pay any attention to the conversion of Pi-LAC to Pi in plasma. In this paper a simultaneous determination method for Pi and Pi-LAC in human plasma and urine was established and validated. The test results indicated that the interconversion of Pi and Pi-LAC was minimized by this method. This approach was successfully applied to accurately determine the Pi and Pi-LAC in human plasma and characterize the clinical pharmacokinetics of Pi and Pi-LAC in humans.
2.3. Preparation of stock and working solutions The stock solutions were prepared with a mixture of methanol–water (7:3, v/v) for pitavastatin calcium and acetonitrile for Pi-LAC to reach a final concentration of 1 mg/mL, respectively. Working solutions were prepared by further diluting the stock solutions with methanol for pitavastatin calcium and acetonitrile for Pi-LAC to 2, 10, 100, 200, 1000, 2000 ng/mL. The standard stock solution of the IS was prepared in methanol to 0.5 mg/mL and diluted to 250 ng/mL with methanol. All the working solutions were kept at −20 ◦ C. 2.4. Calibration curves and quality control sample preparation A unique procedure of making the plasma calibration standards and quality control (QC) samples was started with the addition of the working solutions to the pH adjusted blank plasma. To an aliquot of the blank plasma, 20 L ammonium acetate buffer solution (0.6 M; pH 4.2) was added and vortex-mixed to obtain the pH adjusted blank plasma. The plasma calibration standards were prepared by spiking the pH adjusted blank plasma samples with appropriate volumes of the above-mentioned working solutions, to produce the plasma calibration standards at concentrations of 0.1, 0.5, 2.5, 10, 30, 60, 120, 200 ng/mL for Pi and Pi-LAC, respectively. By using the same procedure for preparing the calibration
10
X. Qi et al. / Journal of Pharmaceutical and Biomedical Analysis 72 (2013) 8–15
standards, the QC samples were prepared at the concentration levels of 0.25, 10, 180 ng/mL for both Pi and Pi-LAC. The same concentration levels of the two analytes were prepared for the urine calibration standards and QC samples in the blank urine without the pH adjustment. 2.5. Sample preparation 2.5.1. Plasma sample pretreatment After being drawn from volunteers, the blood samples were centrifuged immediately to separate the plasma, and an aliquot of 0.3 mL plasma sample was transferred to a tube containing 20 L of ammonium acetate buffer (0.6 M; pH 4.2), and vortex-mixed at once for 10 s, then stored at −20 ◦ C until analysis. 2.5.2. Plasma and urine sample preparation An aliquot of 0.3 mL plasma sample was mixed with 30 L of the IS solution (251.5 ng/mL) and deproteinized with 720 L of acetonitrile. The mixture was centrifuged at 15,600 rpm for 10 min and then the supernatant was transferred into an injection vial for LC–MS/MS analysis. After being vortex-mixed with 25 L of the IS solution (251.5 ng/mL), an aliquot of 700 L urine sample was diluted with 300 L acetonitrile and centrifuged at 15,600 rpm for 10 min. Then the supernatant was transferred into an injection vial for LC–MS/MS analysis. The whole analysis procedure was carried out in a dark room to avoid light-induced decomposition of the analytes. 2.6. Method validation The analytical methodology was validated according to the FDA’s guidance for industry on bioanalytical method validation. The selectivity, linearity, matrix effects (ME), recovery, accuracy, precision and stability were assessed as described by Zhu et al. [22]. 2.7. Interconversion characters 2.7.1. Interconversion of Pi-LAC to Pi The interconversion of Pi-LAC and Pi in plasma was evaluated at low, middle and high QC concentration levels of the two analytes. Samples that were used to evaluate the influencing factors on the interconversion were prepared according to the sample preparation described in Section 2.5.2. In this paper, only the test results obtained at the high QC concentration level were given as the representations. Three sets of plasma samples at the high QC concentration level without adding the buffer solution were used to investigate the interconversion of Pi-LAC and Pi: set 1 containing Pi-LAC only, set 2 containing the both analytes, set 3 containing Pi only. Several replicate samples were prepared in each set, and analyzed after being kept at room temperature for different time. The changing trends of Pi and Pi-LAC in the plasma samples caused by the interconversion were shown as the time-dependent variations of the measured mass responses (chromatographic peak areas) to the two analytes. 2.7.2. Influence of the different stabilizers on the interconversion Ethylene diamine tetraacetic acid (EDTA), potassium fluoride (KF) and sodium fluoride (NaF) are the common esterase inhibitors (called stabilizer in this section) used to prevent the hydrolysis of the esters in plasma samples [23]. In view of the instability of Pi-LAC in plasma induced by hydrolysis degradation to Pi, different amounts of the above mentioned stabilizers were added in the plasma samples, and their efficiency on the inhibiting the conversion of Pi-LAC to Pi was compared. The Pi-LAC high QC samples
containing varied amounts of the stabilizers were prepared in triplicates for each stabilizer. After being kept at room temperature for 1 h, these samples were analyzed according to Section 2.5.2. The residual amount of Pi-LAC in each sample was calculated.
2.7.3. Influence of pH of the buffer solution on the interconversion Several different pH values of the added buffers were assessed to prevent the sample-collection transformation of Pi-LAC into its prodrug. The solutions used for test were ammonium acetate buffer solution (0.6 M; pH 4.2), ammonium acetate buffer solution (0.6 M; pH 2.0) and hydrochloric acid (1 M; pH 0), respectively. After being treated with 20 L different solutions of different pH values, and maintaining at room temperature for 1.5 h, the high QC samples containing both Pi and Pi-LAC were assessed by LC–MS/MS, and the conversion ratios (CR) of Pi-LAC to Pi (or Pi to Pi-LAC) were calculated. It is defined as: CR (%) = (b − a)/a × 100, in which a and b represent the values of peak areas of the analytes obtained from the freshly prepared QC samples and the QC samples kept at room temperature for 1.5 h, respectively.
3. Results and discussion 3.1. Method development 3.1.1. LC–MS–MS optimization To maximize the MS responses to Pi and Pi-LAC, some instrument parameters were investigated and optimized. Positive ion monitoring mode was chosen because of its better sensitivity than negative ion monitoring mode. Additionally, reducing drying gas temperature from 350 ◦ C (the recommended value of the instrument manufacturer) to 300 ◦ C is a matter of choice which presented a significant benefit of reproducibility. Acetonitrile was selected as the organic portion of the mobile phase, because it showed lower background noise. The test results showed that higher column temperature could improve the peak shape and symmetry of Pi. The acceptable peak shape and symmetry of Pi was achieved at the column temperature of 43 ◦ C for the plasma sample analysis.
3.1.2. Sample pretreatment Characters of interconversion confirmed that Pi-LAC was very unstable in plasma, and the interconversion of Pi-LAC to Pi in plasma catalyzed by esterase. As the rate of hydrolysis of Pi-LAC to Pi might be dependent on pH, ionic strength and protein concentration, adding the ammonium acetate buffer was chosen to guarantee the stabilities of the analytes in plasma in the whole sample treatment process. However the instability problem of Pi-LAC did not occur in human urine, for there was no esterase in the urine.
3.2. Method validation 3.2.1. LC-ESI-MS/MS characteristics No endogenous interferences were observed at the retention times of the analytes, see Fig. 2. Pi, Pi-LAC and the IS exhibited sharp peaks at 1.7, 2.0 and 2.2 min, respectively, while the full run time was 3.2 min. The matrix effects (ME) at concentrations of 0.25, 10, 180 ng/mL were (98.9 ± 5.4)%, (102.5 ± 2.9)% and (96.6 ± 3.7)% for Pi; and (97.8 ± 6.8)%, (99.4 ± 4.2)% and (100.7 ± 1.4)% for Pi-LAC. The matrix effect for the IS (251.5 ng/mL) was (99.4 ± 3.9)%. The extraction recoveries were (96.2 ± 4.1)%, (98.8 ± 3.6)% and (96.3 ± 2.7)% for Pi, and (99.3 ± 6.5)%, (94.9 ± 2.4)% and (96.8 ± 3.1)% for Pi-LAC (n = 5) at concentration levels of 0.25, 10 and 180 ng/mL, respectively.
X. Qi et al. / Journal of Pharmaceutical and Biomedical Analysis 72 (2013) 8–15
11
3.2.3. Stability Stability tests were designed to assess the stability of the analytes under the conditions expected during handling of clinical samples. The stability results of the analytes in plasma and urine were summarized in Tables 1 and 2, respectively. Pi, Pi-LAC and the IS were found to be stable in their working solutions after being placed at room temperature for 4.5 h and stored at −20 ◦ C for a period of 50 d. 3.2.4. Carry-over effect The carry-over effect was tested by instantly analyzing blank samples following upper limit of quantification (ULOQ). No residual was observed at the retention times of Pi, Pi-LAC and the IS in the chromatograms of blank plasma and urine samples. 3.3. Interconversion characters of Pi-LAC and Pi The immediate transformation of the unstable metabolite into its parent drug after sample collection was a major concern. The main purpose was to explore the characters of the interconversion of the two analytes under several different conditions and to solve the instability problem.
Fig. 2. MRM chromatograms of (A) blank plasma, (B) plasma spiked with Pi and Pi-LAC at 0.1 ng/mL (LLOQ) and the IS, (C) a plasma sample collected after 0.5 h of single 2 mg oral dose, and (D) a urine sample collected in the interval 2–4 h after single 4 mg oral dose.
3.2.2. Linearity, lower limit of quantification, precision and accuracy The calibration curves were obtained by weighted linear regression, the peak area ratio (analyte/IS) was plotted vs the analyte concentrations. The calibration curve was linear over the range of 0.1–200 ng/mL in human plasma and urine with coefficient of correlation (r2 ) > 0.997 for both analytes. At the LLOQ (0.1 ng/mL), the RSD (%) < 8.3% (n = 5) and the RE (%) ranged from −0.1% to 7.3%. Validation samples of five replicates of the QC samples were prepared and analyzed in three separate analytical batches to evaluate the accuracy and intra-batch and inter-batch precisions of the methods. The intra-batch and inter-batch precisions were measured to be less than 8.3% and 9.8%, respectively, and the accuracy to be within ±15%.
3.3.1. Interconversion of Pi-LAC to Pi Pi-LAC is unstable in methanol because of methyl esterification. A stability test showed that 17% of Pi-LAC was converted to its methyl ester (Fig. 1) in its methanol solution after keeping at −20 ◦ C for 20 d. Therefore, acetonitrile was selected as the solvent to prevent the methyl esterification of Pi-LAC. In this study, an acetic acid–ammonium acetate buffer solution (0.6 M; pH 4.2) was used as stabilizing agent, which was promptly added to the freshly collected plasma samples. The interconversion profiles of the two analytes were presented by plotting the MS response intensity of them vs time curves (see Fig. 3). In Fig. 3, the tendency of the interconversion of the two analytes is observed clearly. The test results of the sets 1 and 2 indicated a prominent degradation of Pi-LAC, and over 55% of Pi-LAC degraded and mainly converted to Pi in the first 1.5 h. At the same time, a significant increase of Pi in the plasma samples was observed. The presence of Pi in plasma may inhibit the conversion of Pi-LAC to Pi to some extent. For the initial presence of a large amount of Pi, the decreasing process of Pi-LAC in set 2 was slower than that in set 1. Interestingly, Pi was stable in human plasma and no Pi-LAC was found in the samples of the set 3 in the whole test period, see Fig. 3C. It demonstrated that Pi was stable and would not convert to Pi-LAC in plasma under normal conditions, but Pi-LAC was susceptible to hydrolysis. 3.3.2. Influence of the different stabilizers on the interconversion The test results of the influence of the different stabilizers on the interconversion are summarized in Fig. 4. The ordinate of Fig. 4 represents the residual amounts of Pi-LAC in the samples treated with different stabilizers. It was observed that the esterase inhibitors EDTA (2 mol/L) and KF (3.4 mol/L) did not play a key role in the inhibition of the conversion of Pi-LAC, but the addition of the pH 4.2 buffer solution can inhibit the conversion efficiently. 3.3.3. Influence of the pH values of the buffer solutions on the interconversion The influence of the different pH values of the added buffer solutions on the interconversion of the analytes was investigated. Fig. 5 shows that the different solutions added to the plasma could significantly affect the interconversion of Pi and Pi-LAC. The conversion of Pi to Pi-LAC would occur when the pH value of the plasma was adjusted too low. But, when the pH value of the plasma was not controlled, much amount of Pi-LAC would easily convert to Pi. So it
12
X. Qi et al. / Journal of Pharmaceutical and Biomedical Analysis 72 (2013) 8–15
Table 1 Stability of Pi and Pi-LAC in human plasma at three QC levels (n = 3). Storage conditions
Analytes
Short-term stability (6 h, room temperature)
Pi
Pi-LAC
Pre-preparative stability at 10 ◦ C for 7 h (Autosampler)
Pi
Pi-LAC
Freeze/thaw stability (3 cycles)
Pi
Pi-LAC
35 d long-term stability
Pi
Pi-LAC
Concentration levels (ng/mL)
RSD %
RE %
0.2415 9.908 183.6 0.2616 9.931 180.9
5.5 3.7 3.1 6.7 2.4 3.5
−3.6 −1.1 1.8 3.6 −1.7 −0.5
0.2505 10.02 180.4 0.2525 10.10 181.8
0.2571 10.81 186.9 0.2618 10.89 189.2
4.9 2.2 2.6 6.4 3.5 3.1
2.6 7.9 3.6 3.7 7.8 4.1
0.2505 10.02 180.4 0.2525 10.10 181.8
0.2477 10.07 180.9 0.2431 10.33 171.3
4.2 3.3 3.0 4.1 2.0 2.8
−1.1 0.5 0.3 −3.7 2.3 −5.8
0.2505 10.02 180.4 0.2525 10.10 181.8
0.2439 9.789 184.1 0.2437 9.597 181.7
5.3 2.8 3.1 4.3 4.5 3.1
−2.6 −2.3 2.1 −3.5 −5.0 −0.1
Added
Found
0.2505 10.02 180.4 0.2525 10.10 181.8
RSD, relative standard deviation; RE %, [(found − added)/added] × 100.
is important to find an optimal plasma pH value to avoid the interconversion of the analytes. Interestingly, Pi is stable and would not convert to its lactone in plasma samples under normal conditions. That is why the instability problem of the Pi plasma samples, which also containing Pi-LAC in them, was ignored by those bioanalysis methods reported in the literatures [7–15]. Commonly, we develop a method for the quantification of a parent drug using the spiked blank plasma (in vitro) but not the incurred plasma (in vivo). Care must be taken, however, as validations in vitro are not absolutely equal to those in vivo for some drugs. Sometimes, the metabolites
in plasma samples may convert to their parent drugs. However, the instability problem of the plasma samples caused by the coexisting metabolites has been largely ignored, for QC samples used for the validation of an analytical method, which are prepared in the same blank plasma as the calibration samples, will not reveal metabolite-related instability observed in the incurred samples. For unstable analytes, the greatest challenge is how to keep them stable before analyzing them. Tian et al. [19] and Ashwini et al. [20] reported that Pi-LAC in human plasma was stable in the freezethaw cycles. In fact, Pi-LAC degrades rapidly in plasma at room
Table 2 Stability of Pi and Pi-LAC in human urine at three QC levels (n = 3). Storage conditions
Short-term stability (3.5 h, room temperature)
Analytes
Pi
Pi-LAC
Pre-preparative stability at 10 ◦ C for 4 h (Autosampler)
Pi
Pi-LAC
Freeze/thaw stability (3 cycles)
Pi
Pi-LAC
15 d long-term stability
Pi
Pi-LAC
Concentration levels (ng/mL)
RSD %
RE %
0.2538 9.654 185.3 0. 2702 10.32 186.7
3.2 4.8 2.7 6.4 3.5 3.2
1.3 −3.7 2.7 7.0 2.2 2.7
0.2505 10.02 180.4 0.2525 10.10 181.8
0.2489 9.973 165.8 0.2449 9.915 174.3
5.1 3.7 2.9 3.3 4.5 3.6
−0.6 −0.5 −8.1 −3.0 −1.8 4.1
0.2505 10.02 180.4 0.2525 10.10 181.8
0.2491 10.22 170.5 0.2456 9.406 188.9
4.6 6.7 3.1 5.7 4.1 3.4
−0.6 2.0 −5.5 −2.7 −6.9 3.9
0.2505 10.02 180.4 0.2525 10.10 181.8
0.2441 9.398 185.4 0.2433 10.47 177.6
4.8 4.0 2.6 4.3 2.7 2.6
−2.6 −6.2 2.8 −3.6 3.7 −2.3
Added
Found
0.2505 10.02 180.4 0.2525 10.10 181.8
X. Qi et al. / Journal of Pharmaceutical and Biomedical Analysis 72 (2013) 8–15
13
Fig. 4. The residual amounts of Pi-LAC in the QC samples treated with different stabilizers after keeping at room temperature for 1 h.
Fig. 3. The interconversion profiles (the MS response intensity vs time curves) of Pi and Pi-LAC in the plasma samples of set 1 (A), set 2 (B) and set 3 (C).
temperature, and even during the procedure of freezing-thawing. Enzymes have a catalytic activity and some chemical reactions cannot happen in the absence of enzymes. But according to our test results, the interconversion of Pi-LAC to Pi was still going on with a lower speed even after the plasma samples were treated with deproteinization (data were not given). It is supposed that apart from enzymes, there may be some other influencing factors. Hence it is vital to prevent the degradation of Pi-LAC in plasma by using the method described in this article. Both analytes can be well stabilized by immediate addition of the buffer solution, and there is no need to use an ice bath [19,21] or derivatization [16–18] for the sample preparation. 3.4. Pharmacokinetic study The established methods were applied in a pharmacokinetic study. Thirty healthy Chinese volunteers, 15 males and 15 females,
Fig. 5. The CR (%) of Pi (A) and Pi-LAC (B) in the plasma samples added with the solutions of different pH values.
14
X. Qi et al. / Journal of Pharmaceutical and Biomedical Analysis 72 (2013) 8–15
Table 3 Pharmacokinetic parameters of Pi and Pi-LAC in Chinese volunteers (n = 30). Dose (mg)
Single dose 1
Multi-dose 2
4
2
Pi
AUC (ng h/mL) Cmax (ng/mL) Tmax (h) t1/2 (h) CL (L/h) Vd (L) R
59.60 ± 23.27 ± 0.6 ± 11.6 ± 16.1 ± 264.9 ± –
12.30 2.799 0.1 1.6 3.1 54.12
116.7 ± 57.65 ± 0.6 ± 10.1 ± 26.0 ± 369.3 ± –
89.8 40.58 0.2 1.8 17.1 252.4
213.0 ± 88.78 ± 0.6 ± 12.8 ± 18.5 ± 339.0 ± –
29.17 27.81 0.1 1.2 2.5 65.00
142.0 53.67 0.8 11.6 20.7 312.4 1.2
± ± ± ± ± ± ±
126.2 45.33 0.1 3.0 11.2 159.2 0.3
Pi-LAC
AUC (ng h/mL) Cmax (ng/mL) Tmax (h) t1/2 (h) CL (L/h) Vd (L) R
137.4 ± 20.21 ± 1.1 ± 10.8 ± 7.1 ± 108.3 ± –
18.69 4.348 0.4 1.8 1.0 19.27
234.7 ± 47.78 ± 1.0 ± 11.7 ± 9.5 ± 158.2 ± –
143.2 32.60 0.3 1.3 3.7 74.55
438.6 ± 84.09 ± 1.1 ± 12.1 ± 9.3 ± 159.9 ± –
115.8 28.45 0.3 1.4 2.4 31.16
270.9 47.38 1.2 12.3 8.9 153.6 1.1
± ± ± ± ± ± ±
201.2 33.87 0.3 1.2 4.8 87.98 0.3
AUC, area under the drug concentration–time curve; Cmax , maximum concentration; Tmax , time to reach peak concentration; t1/2 , elimination half life; R, the accumulation ratio was calculated from AUC /AUC0− where AUC is the AUC in a dosing interval ( = 12) during multi-dose administration and AUC0− is the AUC within time span 0–12 h after single dose of 2 mg. The volume of distribution (Vd) and clearance (CL) were calculated from the formulas: Vd = CL × t1/2 /0.693 and CL = Dose/AUC0−∞ , respectively.
were screened to enroll in the study after signing the informed consent form. The study was approved by the Ethics Committee of the First Affiliated Hospital of the Fourth Military Medical University of PLA (Xi’an, China) and conducted at this study center, in accordance with the GCP requirements and the Declaration of Helsinki. Written informed consent was obtained from each subject. A randomized, open-label and parallel-group design was used to evaluate the pharmacokinetic profiles of Pi and Pi-LAC after single oral dosing. Twenty healthy volunteers were randomized into 1 and 4 mg dose groups (10 volunteers in each, half male and half female) for the determination of the Pi and Pi-LAC pharmacokinetic profiles of the single-dose of pitavastatin calcium tablets. Another 10 volunteers, half male and half female, were administrated 2 mg dose of pitavastatin calcium tablets on day 1; received no treatment on day 2; and continued to receive the study drug once daily from days 3 through 7 to assess multiple-dose pharmacokinetics. The volunteers were required to fast overnight (12 h) before drug administration, while standard meals and water intake were provided 4 h post-dose. Blood samples were collected at 0 h (pre-dose), 0.25, 0.5, 0.75, 1, 1.5, 2, 3, 4, 6, 8, 12, 24, 36 and 48 h post-dose for the single-dose study. For the multiple-dose study, blood samples were collected prior to dosing on days 1, 5, 6 and 7 (0 h prior to dosing) and at 0.25, 0.5, 0.75, 1, 1.5, 2, 3, 4, 6, 8, 12, 24, 36 and 48 h post-dose on day 1 and day 7. Plasma was separated and stored at −20 ◦ C until analysis. The urine samples were collected pre-dose and over the intervals 0–2, 2–4, 4–6, 6–8, 8–12, 12–24, 24–36 and 36–48 h post-dose on day 1 of the 4 mg dose group. The collected plasma samples and the urine samples were assessed by the established methods. The concentration levels of Pi and Pi-LAC in each sample were calculated and the pharmacokinetics of Pi and PiLAC were evaluated. The mean plasma concentration–time profiles and the urinary excretion characteristics of Pi and Pi-LAC are illustrated in Figs. 6 and 7, respectively. The urinary excretion profile of Pi-LAC in Chinese volunteers was reported for the first time in this paper. In the interval of 0–2 h post-dose, the urinary excretion rates for both analytes reached the peak values of (5.04 ± 2.53) g/h and (85.3 ± 59.8) g/h, respectively. The average cumulative urinary excretion percentages of Pi and Pi-LAC in 48 h post-dose were (0.41 ± 0.16)% and (6.1 ± 5.0)%, respectively, suggesting that Pi and Pi-LAC may be subject to non-renal elimination in healthy subjects. Pi is a hydrophilic compound. And it was reported for pitavastatin that the biliary excretion of the intact form was the main elimination pathway in some experimental animals including rats, rabbits and dogs [24]. The pharmacokinetic parameters for Pi and Pi-LAC
are presented in Table 3. Both the AUC and Cmax of two analytes were dose-proportional over the dose range of 1–4 mg. No accumulation and gender differences were found. The pharmacokinetic parameters showed that the AUC and Cmax of Pi and Pi-LAC in extensive metabolizers were about four times more than that of poor metabolizers in the 2 mg dose group. This suggested a significant individual difference in the pharmacokinetic characteristics of Pi and its metabolite. Several single-nucleotide polymorphisms (SNPs) of OATP1B1 are known which associate with the transport capacity and significantly alter the disposition of Pi [25]. This may explain the wide inter-individual variation in Pi pharmacokinetics.
Fig. 6. Mean plasma concentration–time profiles of Pi (A) and Pi-LAC (B) in Chinese volunteers after the single dose administration (n = 30).
X. Qi et al. / Journal of Pharmaceutical and Biomedical Analysis 72 (2013) 8–15
Fig. 7. Mean urinary cumulative excretion percentage–time profiles of Pi and Pi-LAC in Chinese volunteers after a single oral administration of 4 mg dose pitavastatin calcium tablets (n = 10).
4. Conclusion The interconversion of Pi-LAC to Pi in plasma and its influencing factors were evaluated. The studies indicate that Pi-LAC undergoes fast degradation by hydrolysis in plasma. Addition of the buffer solution (pH 4.2) to the freshly collected plasma samples can prevent the interconversion of Pi-LAC to Pi. The simple and rapid LC-ESI-MS/MS methods were developed for the simultaneous determination of Pi and Pi-LAC in human plasma and urine. The methods were used to measure the concentration of Pi and Pi-LAC in bio-samples from healthy volunteers after oral administration of pitavastatin calcium tablets. The simple method totally solved the instability problems of Pi and Pi-LAC in plasma, and was successfully used in the clinical pharmacokinetic studies. The urinary excretion profile of Pi-LAC in human was reported for the first time in this paper. References [1] Y. Saito, N. Yamada, T. Teramoto, H. Itakura, Y. Hata, N. Nakaya, H. Mabuchi, M. Tushima, J. Sasaki, N. Ogawa, Y. Goto, A randomized, double-blind trial comparing the efficacy and safety of pitavastatin versus pravastatin in patients with primary hypercholesterolaemia, Atherosclerosis 162 (2002) 373–379. [2] A. Tonkin, P. Aylward, D. Colquhoun, P. Glasziou, P. Harris, S. MacMahon, P. Magnus, D. Newel, P. Nestel, N. Sharpe, D. Hunt, J. Shaw, R.J. Simes, P. Thompson, A. Thomson, M. West, H. White, S. Simes, W. Hague, S. Caleo, J. Hall, A. Martin, S. Mulray, P. Barter, L. Beilin, R. Collins, J. McNeil, P. Meier, H. Willimott, D. Smithers, P. Wallace, D. Sullivan, A. Keech, Prevention of cardiovascular events and death with pravastatin in patients with coronary heart disease and a broad range of initial cholesterol levels, N. Engl. J. Med. 339 (1998) 1349–1357. [3] R.J. Havel, E. Rapaport, Management of primary hyperlipidemia, N. Engl. J. Med. 332 (1995) 1491–1498. [4] I. Yamada, H. Fujino, S. Shimada, J. Kojima, Metabolic fate of pitavastatin, a new inhibitor of HMG-CoA reductase: similarities and difference in the metabolism of pitavastatin in monkeys and humans, Xenobiotica 33 (2003) 789–803. [5] H. Fujino, I. Yamada, S. Shimada, M. Yoneda, J. Kojima, Metabolic fate of pitavastatin, a new inhibitor of HMG-CoA reductase: human UDPglucuronosyl-transferase enzymes involved in lactonization, Xenobiotica 33 (2003) 27–41.
15
[6] H. Fujino, I. Yamada, S. Shimada, J. Kojima, Metabolic fate of pitavastatin, a new inhibitor of HMG-CoA reductase-effect of Cmoat deficiency on hepatobiliary excretion in rats and of mdr 1 a/b gene disruption on tissue distribution in mice, Drug. Metab. Pharmacokinet. 17 (2002) 449–456. [7] H. Lv, J.G. Sun, G.J. Wang, X.Y. Zhu, Y. Zhang, S.H. Gu, Y. Liang, J. Sun, Determination of pitavastatin in human plasma via HPLC-ESI-MS/MS and subsequent application to a clinical study in healthy Chinese volunteers, Clin. Chim. Acta 386 (2007) 25–30. [8] B. Di, M.X. Su, F. Yu, L.J. Qu, L.P. Zhao, M.C. Cheng, L.P. He, Solid-phase extraction and liquid chromatography/tandem mass spectrometry assay for the determination of pitavastatin in human plasma and urine for application to phase I clinical pharmacokinetic studies, J. Chromatogr. B 868 (2008) 95–101. [9] J.H. Wen, Y.Q. Xiong, Determination of the concentration of pitavastatin in human plasma by LC–MS and subsequent application to a clinical study, Chin. JMAP 27 (2010) 62–65. [10] C. Qiu, M. Zhang, X. Liu, Y.G. Wen, The research of pharmacokinetics of pitavastatin in healthy volunteers, Pharm. Care Res. 9 (2009) 225–227. [11] C. Deng, T. Yi, L. Tang, Y.J. Li, A simple and sensitive LC–MS/MS method for determining the concentration of pitavastatin in human plasma, Chin. J. New Drug 18 (2009) 1113–1115. [12] C.X. Chen, G.W. Li, Y.G. Wen, Rapid determination of pitavastatin in plasma by LC-ESI-MS/MS, Chin. JMAP 27 (2010) 642–645. [13] J.Z. Shen-Tu, X. Xu, J. Liu, X.J. Hu, J.C. Chen, L.H. Wu, M.Z. Huang, H.L. Zhou, Determination of pitavastatin in human plasma by LC–MS–MS, J. Chromatogr. 69 (2009) 1041–1047. [14] Y. Chen, Z.R. Tan, G. Zhou, D.S. Ou-Yang, J. Zhang, H.H. Zhou, LC–MS determination of pitavastatin in human plasma and its application, J. Clin. Pharmacol. 46 (2011) 134–137. [15] J.W. Deng, K.B. Kim, I.S. Song, J.H. Shon, H.H. Zhou, K.H. Liu, J.G. Shin, Determination of two HMG-CoA reductase inhibitors, pravastatin and pitavastatin, in plasma samples using liquid chromatography–tandem mass spectrometry for pharmaceutical study, Biomed. Chromatogr. 22 (2008) 131–135. [16] H. Fujino, J. Kojima, Y. Yamada, H. Kanda, H. Kimata, Studies on the metabolic fate of NK-104, a new inhibitor of HMG-CoA Reductase (4): interspecies variation in laboratory animals and humans, Xenobio. Metabol. Dispos. 14 (1999) 79–91. [17] J. Kojima, H. Fujino, M. Yosimura, H. Morikawa, H. Kimata, Simultaneous determination of NK-104 and its lactone in biological samples by column-switching high-performance liquid chromatography with ultraviolet detection, J. Chromatogr. B 724 (1999) 173–180. [18] C.K. Hui, B.M.Y. Cheung, G.K.K. Lau, Pharmacokinetics of pitavastatin in subjects with child – Pugh A and B cirrhosis, Br. J. Clin. Pharmacol. 59 (2004) 291–297. [19] L. Tian, Y.L. Huang, Y.H. Jia, L. Hua, Y.S. Li, Development and validation of a liquid chromatography-tandem mass spectrometric assay for pitavastatin and its lactone in human plasma and urine, J. Chromatogr. B 865 (2008) 127–132. [20] O. Ashwini, G. Swati, V. Chintan, P. Harish, Determination of pitavastatin from human plasma using high performance liquid chromatography with fluorescence detection, Chin. J. Chromatogr. 25 (2007) 715–718. [21] L.L. Wang, X.Y. Chen, B. Song, Y.F. Zhang, D.F. Zhong, Determination of pitavastatin in human plasma by liquid chromatography–tandem mass spectrometry, Chin. J. New Drugs Clin. Rem. 26 (2007) 805–808. [22] H. Zhu, L. Ding, S. Shakya, X.M. Qi, L.L. Hu, X.L. Yang, Z.L. Yang, Simultaneous determination of asperosaponin VI and its active metabolite hederagenin in rat plasma by liquid chromatography–tandem mass spectrometry with positive/negative ion-switching electrospray ionization and its application in pharmacokinetic study, J. Chromatogr. B 30 (2011) 3407–3414. [23] Y.R. Zhao, L. Ding, H.W. Fan, Y. Yu, X.M. Qi, Y. Leng, Y.K. Rao, Determination of the unstable drug otilonium bromide in human plasma by LC-ESI-MS and its application to a pharmacokinetic study, J. Chromatogr. B 878 (2010) 2896–2900. [24] J. Kojima, H. Fujino, H. Abe, M. Yosimura, H. Kanda, H. Kimata, Identification of metabolites of NK-104, an HMG-CoA reductase inhibitor, in rat, rabbit and dog bile, Biol. Pharm. Bull. 22 (1999) 142–150. [25] J. Wen, Y. Xiong, OATP1B1 388A>G polymorphism and pharmacokinetics of pitavastatin in Chinese healthy volunteers, J. Clin. Pharm. Ther. 35 (2010) 99–104.