Anal Bioanal Chem (2014) 406:4299–4308 DOI 10.1007/s00216-014-7820-x

RESEARCH PAPER

LC-MS/MS method for the simultaneous quantification of artesunate and its metabolites dihydroartemisinin and dihydroartemisinin glucuronide in human plasma Mirjam C. K. Geditz & Georg Heinkele & Asma Ahmed & Peter G. Kremsner & Reinhold Kerb & Matthias Schwab & Ute Hofmann

Received: 20 January 2014 / Revised: 11 March 2014 / Accepted: 4 April 2014 / Published online: 24 April 2014 # Springer-Verlag Berlin Heidelberg 2014

Abstract Artesunate (AS), a hemisuccinate derivative of artemisinin, is readily soluble in water and can easily be used in formulations for parenteral treatment of severe malaria. AS is rapidly hydrolyzed to the active metabolite dihydroartemisinin (DHA) and primarily eliminated by biliary excretion after glucuronidation. To investigate systematically the AS metabolism and pharmacokinetics, a novel liquid chromatography–tandem mass spectrometry (LC-MS/MS) method for the simultaneous quantification of AS and its metabolites DHA and DHA glucuronide (DHAG) in human plasma samples was developed. Compared to previous methods, our method includes for the first time the quantification of the glucuronide metabolite using a newly synthesized stable isotope-labeled analogue as internal standard. Sample preparation was performed with only 50 μL plasma by high-throughput solid-phase extraction in the 96-well plate format. Separation of the analytes was achieved on a Poroshell 120 EC-C18 column (50*2.1 mm, 2.7 μm, Agilent Technologies, Waldbronn, Germany). The method was validated according to FDA guidelines. Calibration curves were linear M. C. K. Geditz : G. Heinkele : R. Kerb : M. Schwab : U. Hofmann (*) Dr. Margarete Fischer-Bosch Institute of Clinical Pharmacology and University of Tuebingen, Auerbachstrasse 112, 70376 Stuttgart, Germany e-mail: [email protected] A. Ahmed Department of Medicine, Faculty of Medicine, University of Khartoum, Alqasr Street, 102, Khartoum, Sudan P. G. Kremsner Institute of Tropical Medicine, University Tuebingen, Wilhelmstrasse 27, 72074 Tuebingen, Germany R. Kerb : M. Schwab Department of Clinical Pharmacology, University Hospital, Auf der Morgenstelle 8, 72076 Tuebingen, Germany

over the entire range from 1 to 2,500 nM (0.4–961.1 ng/ mL), 165 to 16,500 nM (46.9–4,691.8 ng/mL), and 4 to 10,000 nM (1.8–4,604.7 ng/mL) for AS, DHA, and DHAG, respectively. Intra- and interbatch accuracy, determined as a deviation between nominal and measured values, ranged from −5.7 to 3.5 % and from 2.7 to 5.8 %, respectively. The assay variability ranged from 1.5 to 10.9 % for intra- and interbatch approaches. All analytes showed extraction recoveries above 85 %. The method was successfully applied to plasma samples from patients under AS treatment. Keywords Artesunate . Metabolites . Dihydroartemisinin . Glucuronide . Solid-phase extraction . LC-MS/MS

Introduction Malaria still represents a major global health problem, which caused an estimated 1,238,000 global deaths in 2010, whereof 92 % occurred in African countries. About two-thirds were children under 5 years of age, who suffer from a more rapid and severe course of the disease as compared to adults [1]. Plasmodium falciparum parasites cause the most severe form of malaria with a case fatality rate of approximately 15–20 % [2–4]. Severe malaria is primarily treated with parenteral medication to reduce parasitemia fast and efficiently to a non-life-threatening level, before medication is continued orally. According to WHO guidance, artesunate (AS) is the first-line parenteral treatment for severe malaria, followed by a complete course of an effective artemisinin-based combination therapy (ACT) [5]. AS, dihydroartemisinin-12-alpha-succinate, is a hemisuccinate derivative of the endoperoxide artemisinin, a secondary plant compound of Artemisia annua (sweet wormwood). Artemisinin derivatives are discussed to be the most

4300

effective antimalarials currently available [6]. AS is watersoluble and thus the only artemisinin derivative, which is available as parenteral formulation for intravenous or intramuscular administration. After injection, AS is rapidly hydrolyzed in vivo to the active metabolite dihydroartemisinin (DHA) and primarily excreted into bile after glucuronidation (Fig. 1). Notably, Ramu and Baker determined some residual antimalarial activity in vitro of the DHA glucuronide (DHAG) [7]. The glucuronidation of dihydroartemisinin is catalyzed predominately by uridine diphosphate (UDP)glucuronosyltransferases (UGT), in particular UGT1A9 and UGT2B7 [8], showing both a broad range of genetic variants. Very recently, delayed hemolytic anemia has been observed as a relatively frequent complication after treatment of severe malaria with AS [9, 10]. A possible relation between AS dose Fig. 1 Chemical structures and product ion spectra (positive ESI) of AS (1) and its metabolites, DHA (2), and the glucuronide DHAG (3)

M.C.K. Geditz et al.

and delayed hemolysis could not be proved by KreeftmeijerVegter et al. [11]; however, data on plasma concentrations were not available in these studies. It has been suggested that interindividual variability in pharmacokinetic profiles of antimalarial drugs might be responsible for variations in drug response or toxicity [12]. To elucidate systematically the impact of AS metabolism and pharmacokinetics on drug response and adverse effects, the plasma concentrations of the parent drug and its two metabolites have to be determined. To date, several LC-MS methods for the determination of artemisinin derivatives and metabolites have been established. However, most methods just determine AS and DHA [13, 14], whereas the simultaneous determination of AS, DHA, and the glucuronide has been described for sheep plasma only [15]. The determination of AS in particular is difficult inasmuch as degradation occurs ex vivo after sample collection both by

LC-MS/MS analysis of artesunate and its metabolites DHA and DHAG

chemical hydrolysis and by plasma esterase-mediated hydrolysis. The addition of organic solvents during sample preparation causes further degradation, thus the use of solid-phase extraction (SPE) instead of liquid-liquidextraction (LLE) or protein precipitation was recommended by Lindegardh et al. eliminating the degradation induced by hemolysis [16]. Furthermore, several of the mentioned approaches used artemisinin or artemether as internal standard (IS) for AS, DHA, and DHAG instead of stable isotope-labeled analogues [14, 15]. Due to their almost identical chemical and physical properties, stable isotope-labeled IS reduce variability within sample extraction and LC-MS analysis increasing accuracy of quantitation. As a stable isotope-labeled IS for DHAG was not available, we prepared a deuterated analogue by chemical synthesis. Here, we describe for the first time a sensitive, robust, and high-throughput LC-MS/MS method for the simultaneous determination of AS and its metabolites DHA and DHAG in human plasma samples using SPE. Furthermore, we used stable isotope-labeled analogues as IS for each compound. The method was validated according FDA guidance [17] and successfully applied to plasma samples from patients treated with AS.

Experimental Chemicals and materials Artesunate (AS) and dihydroartemisinin (DHA) were obtained from Tokyo Chemical Industry (TCI) Co. Ltd. (Tokyo, Japan). The stable isotope-labeled internal standards (IS), 2 H4-AS and 13C1,2H4-DHA, were purchased from Alsachim (Illkirch, France). 12α,1′β-DHA glucuronide (DHAG) and 2 H3-12α,1′β-DHAG were obtained by chemical synthesis. Formic acid and ammonium formate (LC-MS grade) were obtained from Sigma-Aldrich (Steinheim, Germany). LCMS-grade acetonitrile and methanol were purchased from Carl Roth GmbH & Co. KG (Karlsruhe, Germany). All other chemicals were of analytical grade.

4301

Chemical synthesis of 2,3,4-[2H3] -dihydroartemisinin-12α,1′β-glucuronide The methyl ester (5) of the 2,3,4-[2H3]-12α,1′β-DHAG was obtained from DHA (2) (Fig. 2) using the deuterium-labeled glucuronidation reagent (4) according to O’Neill et al. [18]. Hydrolysis of the methyl ester with lithium hydroxide yielded 2,3,4-[2H3]-12α,1′β-DHAG (6) with an isotopic distribution of 86.2 % of the 2H3-product, 12.9 % of 2H2-DHAG, 0.9 % of 2 H1-DHAG, and 0.08 % of the unlabeled DHAG. This background does not interfere with the performance of the assay at low DHAG concentrations. The chemical purity of intermediate and product was examined by TLC and MS analysis. No impurities with a percentage above 0.5 % could be found. Unlabeled DHAG was obtained by the same synthesis approach using the respective unlabeled glucuronidation reagent with an overall yield of 19.4 %. The purity of the unlabeled DHAG was determined by mass spectrometric comparison with a purchased and well-characterized reference compound (IRIS Biotech, Germany). Stock and working solutions Stock solutions (1 mg/mL) of AS and DHA were prepared in ethanol. DHAG, its IS, and the IS of AS were diluted in methanol/water (1:1 v/v). 13C1,2H4-DHA was diluted in methanol. All solutions were stored at −20 °C. The stability of the stock solutions was confirmed by comparing measurements of freshly prepared stock solutions with stock solutions stored for at least 3 months. Combined working solutions were prepared freshly at the day of assay by diluting the stock solution in methanol/water (1:1 v/v), whereas IS working solutions were prepared in water. Calibration and quality control samples Appropriate volumes of working solutions were diluted in methanol/water (1:1 v/v), whereof 5 μL were added to 50 μL of blank plasma, obtaining eight calibration standards at concentrations from 1 to 2,500 nM (0.4−961.1 ng/mL) for AS, 165 to 16,500 nM (46.9–4,691.8 ng/mL) for DHA, and 4

Fig. 2 Scheme of the chemical synthesis of the internal standard 2,3,4-[2H3]-12α,1′β-DHAG

4302

to 10,000 nM (1.8–4,604.7 ng/mL) for DHAG. The plasma samples were vortexed, centrifuged at 1,660×g for 3 min at 4 °C, and kept on ice until use. Low, medium, and high concentration quality controls (QC) for AS (25/625/1,875 nM, respectively, 9.6/240.3/ 720.8 ng/mL), DHA (165/4,125/12,375 nM, respectively, 46.9/1,172.9/3,518.8 ng/mL), and DHAG (100/2,500/ 7,500 nM, respectively, 46.1/1,151.2/3,453.5 ng/mL) were prepared from separate stock solutions in the same way as the calibration standards.

M.C.K. Geditz et al.

approximately 2 min. Samples were eluted into a 1 mL 96well collection plate (Waters, Eschborn, Germany) with 200 μL methanol/acetonitrile (90:10 v/v) by gravity followed by the application of low pressure for approximately 15 s. The eluate was evaporated to dryness at room temperature under nitrogen gas. The residue was reconstituted with 100 μL of mobile phase, vortex mixed, and centrifuged at 1,660×g for 3 min at 4 °C prior to LC-MS/MS analysis.

LC-MS/MS analysis Study samples The method was applied to two clinical studies. The first set of plasma samples was obtained from African children participating in a clinical study on “Comparative, Open Label, Dose and Regimen Optimization Follow-up Study of Intravenous and Intramuscular Artesunate in African Children with Severe Malaria,” a “Severe Malaria in African Children” (SMAC)trial. The patient cohort was divided in three subgroups. One group was given AS as intramuscular administration (3× 4 mg/kg AS). The second group received the same amount of AS intravenously, and the last group was treated with the standard dosing regimen of five administrations of intramuscular AS with 2.4 mg/kg AS each. The total dose for all three dosing regimens was 12 mg/kg AS. The second study cohort included adult Sudanese patients with uncomplicated P. falciparum malaria and noninfected volunteers treated with a combination of AS, sulfadoxine, and pyrimethamine. Both healthy volunteers and patients were treated with a total dose of 200 mg AS given as single oral dose. The objective of the second study was to compare pharmacokinetic parameters of AS between infected patients and noninfected, healthy volunteers. Written informed consent from the child’s parent/guardian, and the participants of the second study as well as approval from the local ethical committees were obtained.

The analysis was performed with an Agilent 1200 HPLC system coupled to an Agilent 6460 triple quadrupole mass spectrometer (Agilent Technologies, Waldbronn, Germany) equipped with a jet stream electrospray ion source (ESI). Chromatographic separation was performed at 40 °C using a Poroshell 120 EC-C18 column (50*2.1 mm, 2.7 μm, Agilent Technologies, Waldbronn, Germany). The mobile phases consisted of 10 mM ammonium formate in water (solvent A) and 0.1 % formic acid in acetonitrile (solvent B). Elution was performed with a gradient starting at 10 % B for 0.50 min, increasing to 35 % B within 0.25 min, followed by 2.25 min to 46.5 % B. Solvent B was then increased to 95 % within 0.10 min and kept for 1.30 min before it was decreased to 10 % within 0.10 min. The overall runtime was 6 min at a flow rate of 0.6 mL/min (Fig. 3). The injection volume was 5 μL. MS analysis was performed in the multiple reaction monitoring (MRM) mode. The capillary voltage was set at 3,000 V, and the nozzle voltage at 0 V. The drying gas flow was 12 L/min at a gas temperature of 350 °C. The nebulizer pressure was set at 60 psi, and the sheath gas temperature was 350 °C with a sheath gas flow rate of 11 L/min. MRM transitions, fragmentor voltage, collision energy, and retention times of the analytes and their corresponding IS are summarized in Table 1.

Sample preparation Standardization Plasma sample preparation was performed by solid-phase extraction (SPE) in Oasis® HLB (10 mg) 96-well plates (Waters, Eschborn, Germany) using a positive pressure-96 processor (Waters Corporation, Milford, USA). One hundred fifty microliters of IS-working solution (270/3,000/1,500 nM, respectively, 104.9/868.1/695.2 ng/mL, 2H4-AS/13C1,2H4DHA/2H3-12α,1′β-DHAG in water) were added to 50 μL EDTA plasma in a 96-well plate (Agilent Technologies, Waldbronn, Germany) kept on ice. The SPE-plate was conditioned using 200 μL methanol followed by 200 μL water. Samples were loaded onto the SPE-plate by gravity. The SPE wells were washed with 200 μL water at medium to high pressure, then dried by applying high pressure for

Each batch of study samples was analyzed together with one blank plasma sample (matrix processed without IS), one zero sample (matrix processed with IS), eight calibration standards, and QCs in duplicates at low, medium, and high concentrations worked up the same way as the study samples. Calibration curves were obtained according to the internal standard method by plotting the peak area ratios of target compound to IS versus the analyte concentration. For α,β-DHA, the peak area of the second anomer was used (2.26 min). The curves were fitted using linear regression with 1/x weighting. Data analysis was conducted using MassHunter software (Version B.06.00, Agilent Technologies, Waldbronn, Germany).

LC-MS/MS analysis of artesunate and its metabolites DHA and DHAG

4303

Fig. 3 Combined MRM chromatograms of the analytes of a medium quality control: DHAG ((1) 2,500 nM), DHA (4,125 nM, α-DHA (2a), β-DHA (2b)), and AS ((3) 625 nM)

Method validation Validation experiments were performed according to the FDA guidelines [17]. The parameters selectivity, accuracy, precision, recovery, matrix effects, and stability were evaluated with QC samples at low, medium, and high concentration. Selectivity of the method was tested by the analysis of six randomly selected blank plasma samples in comparison to plasma samples spiked with IS. Intrabatch accuracy and precision of the method were monitored by repeating the analysis of each QC standard six times. This experiment was performed at least in three different batches (interbatch) and the deviation should be within 15 % of the true value, except at the lower limit of quantification (LLOQ), where it should not deviate by more than 20 %. Recovery and matrix effects were determined at low, medium, and high concentrations in triplicates by comparing the response of the analyte in plasma samples spiked before or after extraction to the response of unextracted standards. Short-term temperature stability was investigated by the analysis of QC samples in triplicates, after storage for 2 h at room temperature before sample processing. The postpreparative stability was examined after storage of processed QC samples for 16 h at 8 °C. A corresponding set of

samples was kept at −20 °C for 3 months before sample preparation to evaluate the long-term storage stability. Furthermore, a set of QCs was completely thawed and refrozen for 24 h at −20 °C. After three freeze-thaw cycles, the samples were compared to samples thawed only once before sample preparation.

Results and discussion Chromatographic separation Our aim was to develop a high-throughput LC-MS/MS method for the simultaneous quantification of AS, its primary metabolite DHA, and DHAG in small volumes of human plasma. The chromatographic separation of AS, DHA, and DHAG, achieved by gradient elution on a Poroshell 120 EC-C 18 column (50*2.1 mm, 2.7 μm, Agilent Technologies, Waldbronn, Germany), was optimized to a shortened overall runtime of 6 min compared to 23 min as described by Duthaler et al. [15]. Figure 3 shows combined MRM chromatograms of the analytes of a medium quality control: DHAG ((1) 2,500 nM), DHA (4,125 nM, α-DHA (2a), β-DHA (2b)), and AS ((3) 625 nM). DHA occurs as two anomers, which can

Table 1 Main HPLC and MS parameters Compound

Retention time (min)

Polarity

MRM (m/z)

Dwell (ms)

Fragmentor (V)

Collision energy (V)

Internal standard

ASa

2.97

Positive

407.2→261.1

100

100

12

2

2.26 1.52 2.97 2.26 1.52

Positive Positive Positive Positive Positive

307.2→261.2 478.2→267.2 411.2→261.1 312.2→266.1 481.2→267.2

100 100 100 100 100

85 100 100 85 100

4 8 12 4 8

13

b

DHA DHAG 2 H4-AS 13 C1,2H4-DHA 2 H3-DHAG a

AS artesunate

b

DHA dihydroartemisinin

c

12α,1′β-DHAG dihydroartemisinin 12α,1′β-glucuronide

H4-AS C1,2H4-DHA H3-DHAG

2

4304

be separated chromatographically (Fig 3, peaks 2a and 2b). The equilibrium between α- and β-anomer depends on temperature, solvents, and chromatographic conditions [13] leading to different α/β-ratios in the individual samples. As always, the same α/β-ratio could be observed both for the unlabeled compound and the internal standard, only the second peak at 2.26 min was used for quantification of DHA.

M.C.K. Geditz et al.

Validation Sensitivity and specificity Figure 4 shows chromatograms of a blank plasma sample spiked with IS, the lowest calibration point (LLOQ), and a patient plasma sample after extraction. As shown in Fig. 4a for AS and DHAG, negligible interferences were observed

Fig. 4 MRM chromatograms for AS, DHA, and DHAG of a blank plasma sample spiked with IS, b the lowest calibration standard (LLOQ), and c plasma sample obtained from a patient 10 min after second administration of AS (intramuscular injection (i.m.) 4 mg/kg)

LC-MS/MS analysis of artesunate and its metabolites DHA and DHAG

4305

(

MS method for the simultaneous quantification of artesunate and its metabolites dihydroartemisinin and dihydroartemisinin glucuronide in human plasma.

Artesunate (AS), a hemisuccinate derivative of artemisinin, is readily soluble in water and can easily be used in formulations for parenteral treatmen...
586KB Sizes 0 Downloads 3 Views