Research Article Special Focus: Dried Blood Spots For reprint orders, please contact
[email protected]
Determination of naproxen using DBS: evaluation & pharmacokinetic comparison of human plasma versus human blood DBS Background: Dried blood spots (DBS) sampling is a well-known technology for qualitative determination such as DNA analysis and screening of newborn metabolic disorders. The scientific community has recently expressed interest in applying the DBS technique for quantitative determination of drugs in biological fluid. Results: Two new bioanalytical assays were developed and validated for the determination of naproxen in human plasma and in DBS samples using liquid chromatography coupled with tandem MS. Furthermore, plasma and DBS clinical samples were collected from four subjects enrolled as part of a bioequivalence study. Concentration data for plasma and DBS samples were determined and pharmacokinetic (PK) profiles in plasma and in DBS samples were compared. Conclusions: A strong correlation between PK data obtained by the DBS and conventional plasma method was observed, which makes DBS a valuable technique for further naproxen bioavailability and PK investigations and studies.
Dried blood spots (DBS) sampling technique
is well established and widely used for DNA analysis and screening of newborn metabolic disorders [1] . The DBS technique has been used successfully in neonatal screening for metabolic defects such as phenylketonuria, for which normal and abnormal results are significantly different [2] . In recent years, the DBS sam‑ pling technique raised the interest of the scien‑ tific community for application in therapeutic drug monitoring, pharmacokinetic (PK) and bioavailability studies [3–5] . The advantages of DBS are: n Collection of small blood volumes (10–40 µl); Lower storage and transport costs compared with plasma samples and safer sample han‑ dling due to filter paper pretreatment, which lyses cell membranes;
n
Denatures proteins;
n
technique have been published [6–15] . However, only a limited number of the published articles describe complete clinical procedures, analyti‑ cal validation with parallel analysis of the same incurred samples as well as comparison and cor‑ relation of the data obtained from DBS and con‑ ventional analytical methods utilizing plasma or whole blood. Furthermore, from the published literature, minimal data on the PK profile and/ or limited number of subjects are available. In this study, we describe the analytical vali‑ dation of two LC–MS/MS methods for quan‑ tification of naproxen using DBS and plasma protein precipitation techniques. Furthermore, a comparison of the PK profiles from four sub‑ jects obtained after single dose administration of naproxen sodium 440 mg generated from DBS and plasma samples using LC–MS/MS methods is reported.
Inactivates enzymes and all pathogens such as bacteria and viruses. In addition, for preclinical studies, the num‑ ber of animals could be reduced as no satellite group would be required. For clinical studies, the lower volume of blood collected would cer‑ tainly facilitate the recruitment of patients or healthy volunteers mainly in trials where PK assessment is the main objective. A number of validated methods for quantitative analysis of pharmaceutical compounds using the DBS
Experimental Chemicals, reagents & materials Naproxen (chemical purity 99.9%) was purchased from USP. Naproxen‑D3 (Internal standard [IS], isotopic purity 99.0%) and naproxen acylglucuronide (chemical purity 99.4%) were supplied by Toronto Research Chemicals (Canada), and Nanjing Jinglong PharmaTech Inc. (China), respectively (Figure 1) . Methanol (MeOH) and propionic acid were supplied by Fisher Scientific, UK. Human plasma and human whole blood con‑ taining K 2EDTA as anticoagulant were obtained
10.4155/BIO.10.51 © 2010 Future Science Ltd
Bioanalysis (2010) 2(8), 1501–1513
n
Nikolay Youhnovski1, Josée Michon1, Sylvain Latour1, Jean-Nicholas Mess1, Annik Bergeron1, Milton Furtado1, Marianne Rufiange1, Pascal Guibord1, Marc Lefebvre1, Robert B MacArthur2 & Fabio Garofolo†1 Algorithme Pharma Inc., 575 Armand-Frappier, Laval, Quebec, H7V 4B3, Canada 2 Clinical Horizons Research, Inc., 296 Engle Street, Tenafly, NJ 07670, USA † Author for correspondence: Tel.: +1 450 973 6077 Ext. 2301 Fax: +1 450 973 2446 E-mail:
[email protected] 1
ISSN 1757-6180
1501
Research Article | Youhnovski, Michon, Latour et al. Key Terms Dried blood spots:
Technique where an amount of blood is aliquotted to a filter paper and used for qualitative or quantitative determination.
Pharmacokinetic (PK):
Study of the absorption, distribution, metabolism and elimination of drugs.
Quantification or Quantitation: Measuring the quantity of a compound.
Validation: Evaluation of
different parameters to perform before analysis of samples for a new analytical method.
Naproxen: Anti-inflammatory
agent with analgesic and antipyretic properties. Used in the treatment of rheumatoid arthritis and other rheumatic or musculoskeletal disorders, dysmenorrhea, and acute gout.
Naproxen glucuronide or naproxen acylglucuronide: Metabolite of naproxen.
from Biological Specialty Corporation (PA, USA) and water was dispensed in-house by a Millipore water distribution system. Harris punch (3 mm), cutting mat, Whatman® FTA Classic Card and Whatman DMPK B Cards were purchased from Whatman (GE Healthcare). Vacutainers (6 ml) containing K 2EDTA were supplied by Becton Dickinson Diagnostic (NJ, USA), and desicca‑ tors were provided by VWR International Ltd (Leicestershire, UK). Solutions
preparation Naproxen stock and intermediate solutions at con‑ centrations of 10, 1, 0.1 and 0.01 mg/ml were pre‑ pared in MeOH. Naproxen-D3 IS stock solution was prepared at a concentration of 100 µg/ml in MeOH and used to prepare the IS working solu‑ tion (ISWS) for plasma and DBS methods at a concentration of 3 µg/ml and 0.30 µg/ml in MeOH, respectively. Naproxen acylglucuronide was dissolved in MeOH:H2O 50:50 v/v at a con‑ centration of 1 mg/ml. All solutions were stored at the nominal temperature of 4°C. Preparation
of calibration standards & quality control samples The human whole blood and human plasma, collected using K 2EDTA as anticoagulant, were fortified at a nominal temperature of 4°C CH3
O
CD3
Naproxen plasma samples
OH
O O Naproxen-D3
O
O O Naproxen acylglucuronide
COOH
The blood samples were collected into K 2EDTA vacutainers prior to and 0.25, 0.5, 0.67, 0.83, 1, 1.25, 1.5, 2, 2.5, 3, 4, 6 and 8 h after drug admin‑ istration. Blood was centrifuged at 1500 × g for 10 min at the nominal temperature of 4°C to sep‑ arate the plasma from cells. The plasma samples were stored at ‑20°C until analysis. The human plasma samples of naproxen were extracted according to the procedure outlined below in the section titled ‘Sample preparation’.
OH
HO
Naproxen DBS samples OH
Figure 1. Naproxen, naproxen-D3 and naproxen acylglucuronide.
1502
sample collection & preparation Blood samples were collected from four differ‑ ent subjects randomly selected from Algorithme Pharma Inc. during a clinical trial sponsored by Pharmaceutics International Inc., (MD, USA). This trial was performed by plasma method on 26 male and female healthy subjects to evaluate the bioequivalence of naproxen sodium 220 mg liq‑ uid gel capsules manufactured by Pharmaceutics International Inc. and naproxen sodium 220 mg liquid gel capsules (equivalent to 200 mg naproxen) from Bayer Health Care following a 440 mg oral dose under fasting conditions. Incurred sample reproducibility (ISR) was performed separately using plasma as well as the DBS method on the selected four sub‑ jects at all collected timepoints (0.25, 0.5, 0.67, 0.83, 1, 1.25, 1.5, 2, 2.5, 3, 4, 6 and 8 h post-drug administration).
O
O
Incurred
Incurred sample reproducibility
OH
Naproxen
by adding naproxen stock or intermediate solu‑ tions. Calibration standards at concentrations of 0.50, 1, 2, 5, 15, 25, 40, 70, 85 and 100 µg/ml, and quality control (QC) samples at 0.50, 1.50, 20 and 75 µg/ml were prepared. A second set of QC samples was also prepared with naproxen and naproxen acylglucuronide for stability evaluations. The naproxen acylglucuronide was fortified at a 1:1 molar proportion of naproxen. The plasma calibration and QC samples were aliquoted in polypropylene tubes and stored at -20°C. For DBS analysis, 20 µl from spiked whole blood calibration standards and QC samples were immediately applied on DBS cards. The cards were dried overnight and stored at room temperature in plastic bags.
Bioanalysis (2010) 2(8)
Prior to the centrifugation of previously described samples, an aliquot of 20 µl of whole blood samples was applied onto DBS FTA Classic Card. future science group
Determination of naproxen using DBS The cards were dried overnight on an open nonabsorbent surface at room temperature and stored at room temperature in plastic bags. All standard calibration samples used for stabilities of naproxen and naproxen acylglucuronide evaluations were freshly prepared and extracted immediately after drying for approximately 3 h, without additional storage. For some experiments, as indicated and described in ‘Recovery & matrix effect compari‑ son: classic card versus DMPK‑B cards’, DMPK‑B cards were used. For all other experiments, FTA Classic Cards were used. Sample
preparation
Plasma extraction method
An aliquot of 50 µl of human plasma was fortified with 250 µl of naproxen‑D3 ISWS at a concentra‑ tion of 3 µg/ml prepared in MeOH. The samples were mixed for 5 s to precipitate the plasma pro‑ teins. Tubes were centrifuged at 10,000 × g and 50 µl of the supernatant were transferred to a clean 96-well plate containing 250 µl of MeOH:H2O 50:50 v/v. After vortexing, the samples were kept at 4°C until the injection. DBS method extraction
A 3-mm disk was punched out of the DBS cards with a Harris UNICORE™ device and was for‑ tified with 150 µl of naproxen-D3 ISWS at a concentration of 0.3 µg/ml prepared in MeOH. The tubes were vortexed and left for 1 h at room temperature. Next, 150 µl of type 1 water was added to each sample. The tubes were vortexed and the supernatant was transferred into a 96‑well plate and kept at 4°C until the injection. Chromatographic
conditions All analyses were performed using Agilent Technologies HPLC system Series 1100 binary pump, column heater and autosampler, operated at 4°C. The chromatographic separations were per‑ formed using a Waters XBridge™ C18 column, 3.5 µm (30 × 2.1 mm) at a temperature of 35°C. Plasma method chromatographic conditions
For the plasma method, reversed-phase chro‑ matography using isocratic mode with MeOH and 0.1% propionic acid (60:40 v/v) as mobile phase at a flow rate of 0.6 ml/min was used over 1.25 min (HPLC Program 1). DBS method chromatographic conditions
Reversed-phase chromatography with mobile phase A 0.1% propionic acid and mobile phase B MeOH, using a gradient (0.0–0.3 min: 50% future science group
| Research Article
A, 0.3–2.0 min: 40–50% A linear, 2–3 min: 40% A, 3.1–4.5 min: 50% A) at constant flow rate of 0.6 ml/min was used (HPLC Program 2) for the DBS method. Detection
conditions Applied Biosystems/MDS Sciex API3000™ Triple-Quadrupole Mass Spectrometer (Toronto, Canada) equipped with the TurboIonSpray source operated in negative ion mode at a volt‑ age of -2000 V was used. Other MS parameters were: declustering potential: -20.00 eV, colli‑ sion exit potential: -10.00 eV, collision energy: -23.00 eV, focusing potential: ‑100.00 eV and exit potential: -4.00 eV. The source was set at a temperature of 550°C. Multiple reaction moni‑ toring (MRM) mode was used with the mass transitions for naproxen and naproxen-D3 (IS) of 229.09→170.10 m/z and 232.11→173.00 m/z, respectively. For detection of naproxen acylgluc‑ uronide, a transition of 405.10→170.00 was used. Dwell time was set at 100 ms for all transitions.
Key Term Hematocrit: Proportion of
blood volume that is occupied by red blood cells. It is normally approximately 48% for men and 38% for women.
Hematocrit
determination The hematocrit values were measured with COULTER® LH 780 Hematology Analyzer (Beck man-Coulter, Mississauga, [ON, Canada]). The hematocrit determinations were performed for all subjects before the study. Blood
spots area determination Method 1: The DBS surface area was calculated from the average of 10 DBS (FTA Classic Card) diameters measured with precise instrument (±0.1 mm). The average DBS on FTA Classic Card was estimated to be 7.42 mm. To ensure that the DBS diameters were properly measured, a second method was used. Method 2: 10 DBS were carefully excised by scissors and weighed on an analytical balance. The average was cal‑ culated (% coefficient of variation [CV] = 4.8). Ten of the 3-mm punches were also individually weighed on an analytical balance. The average was calculated (% CV = 1.2). The average sur‑ face of the DBS was then back calculated and was estimated to be 7.29 mm. The difference between the results, obtained by the described two methods, was less than 5%. Results & discussion Analytical method validations for quantification of naproxen in human plasma and DBS were per‑ formed according to Algorithme Pharma stand‑ ard operating procedures (SOP) and according to guidelines from regulatory agencies. www.future-science.com
1503
Research Article | Youhnovski, Michon, Latour et al. Internal
standard In order to minimize the variations induced during the extraction and injection process and from potential matrix effect on the naproxen quantification, a stable isotopically labeled IS, naproxen-D3, was used for plasma method. This compound has similar physicochemical properties as the analyte, which lead to close chromatographic and ionization behavior. These characteristics minimize the variations coming from sample preparation, extraction and/or matrix effect [16,17] . Since the used DBS extrac‑ tion procedure does not allow the use of IS, the naproxen‑D3 was used as an external standard, which compensates the influence of the co-elut‑ ing signal suppressors or enhancers coming from the matrix and/or DBS card. Chromatography
The representative chromatograms of blank, LLOQ (0.50 µg/ml) and ULQ (100.00 µg/ml) samples obtained using the plasma (Figure 2A–C) and DBS (Figure 2D–F) methods are presented in Figure 2 . The naproxen retention time dif‑ ferences result from the use of two different HPLC mobile phase compositions used to sep‑ arate the different suppressors extracted from plasma (HPLC Program 1) and DBS (HPLC Program 2). This gradient program (HPLC Program 2) was used to avoid the suppressors coming from the matrix or the DBS card at the retention time of naproxen. Both the blank samples (Figure 2A & D) were clean from interfer‑ ence and the signal-to-noise ratio of the LLOQ samples were more than five (Figure 2B & E) . Linearity
& regression type analysis Peak area ratios of naproxen/IS were plotted ver‑ sus their corresponding human plasma concentra‑ tions. The linearity of the calibration curve over a range of 0.5 to 100.0 µg/ml was determined by a weighted (1/x2) linear least squares regression for naproxen in the human plasma and DBS assays:
Selectivity
in presence of concomitantly administered compounds The evaluation of potential interference from concomitant medication/active ingredients was performed. The interference from the following compounds due to their ubiquity were investi‑ gated with plasma and DBS assays: acetamino‑ phen, acetylsalicylic acid/salicylic acid, caffeine, chlorpheniramine dextromethorphan, dimenhy‑ drinate, ibuprofen, nicotine/cotinine, phenira‑ mine, phenylephrine, pseudoephedrine, diphen‑ hydramine, ethinyl estradiol, etonogestrel, norelgestromin, norgestrel and norethindrone, Blank samples were spiked with the individual concomitantly administered compound (CAC) solutions to give a final concentration in plasma and blood that would represent approximately twice the expected maximum concentration (Cmax) usually observed from a typical dosage strength administered. These samples were evaluated for interference at the naproxen and naproxen-D3 mass transitions and retention times. Interference at the drug must not exceed 20% of the LLOQ for the analyte. Interference at the IS must not exceed 5.0% of the mean IS response. Low and high QC (QC1 and QC3) samples that were spiked with CAC solutions at the aforementioned concentrations were also extracted and evaluated to ensure no impact on the percent deviations (% dev). None of the CACs tested demonstrated a significant interference according to the above mentioned requirements for both the plasma and DBS assays. Accuracy
y = mx + b Equation 1 Selectivity
The selectivity for plasma and DBS methods was evaluated using blank matrix from ten different human plasma lots and six different human whole blood lots, respectively. These samples were processed according to previ‑ ously described methods with addition of the IS (plasma method) and external standard (DBS 1504
method). All plasma and whole blood lots were found to be free of significant interference at the mass transition and retention times of naproxen (
