Jpn J Ophthalmol (2015) 59:364–371 DOI 10.1007/s10384-015-0389-x
LABORATORY INVESTIGATION
Determination of nonsteroidal anti-inflammatory drugs in human tear and plasma samples using ultra-fast liquid chromatography-tandem mass spectrometry Makiko Hirosawa1,2 • Takehiko Sambe2 • Naoki Uchida2,3 • Xiao-Pen Lee4 Keizo Sato4 • Shinichi Kobayashi3
•
Received: 28 November 2014 / Accepted: 10 April 2015 / Published online: 19 June 2015 Japanese Ophthalmological Society 2015
Abstract Purpose To evaluate a new rapid and sensitive method for analyzing human tears and plasma for nonsteroidal antiinflammatory drugs (NSAIDs) and investigate the influence of the transfer of NSAIDs in an ocular lesion. Methods In this cross-over study, a single dose of 200 mg of ibuprofen and 60 mg of loxoprofen sodium were orally administered to six healthy Japanese subjects. Collected samples were analyzed by ultra-fast liquid chromatography tandem mass spectrometry (UFLC-MS/MS). Results Recoveries of the two drugs spiked in the tears and plasma were 96.0–117.0 % in the tears and 99.0–105.7 % in the plasma. Regression equations for both NSAIDs showed excellent linearity from 0.02–1.0 lg/ml for the tears and 0.1–5.0 lg/ml for the plasma, with the limits of detection at 0.02 lg/ml for tears and 0.1 lg/ml for plasma. Conclusion This new high-throughput NSAID determination method only requires a small tear amount (10 ll) and plasma volume (20 ll) and thus will be useful in clinical and toxicological analyses. Analytical results also showed the presence of ibuprofen and loxoprofen in the
& Makiko Hirosawa [email protected] 1
Department of Ophthalmology, Showa University School of Medicine, 1-5-8 Hatanodai, Shinagawa-ku, Tokyo 142-8555, Japan
2
Department of Pharmacology, Showa University School of Medicine, Tokyo 142-8555, Japan
3
Clinial Research Institute of Clinical Pharmacology and Therapeutics, Showa University, Tokyo 157-8577, Japan
4
Department of Legal Medicine, Showa University School of Medicine, Tokyo 142-8555, Japan
123
actual tears and plasma, which confirms the transition of NSAIDs from the tears to the plasma. Keywords Nonsteroidal anti-inflammatory drugs (NSAIDs) Tandem mass spectrometry (MS/MS) Human tears and plasma
Introduction Nonsteroidal anti-inflammatory drugs (NSAIDs) are widely used for the treatment of pain, fever and inflammation. Ibuprofen is the first member of the propionic acid class of NSAIDs to come into general use [1]. Loxoprofen sodium is a phenylpropionate anti-inflammatory agent with marked analgesic and antipyretic activities. After oral administration, loxoprofen sodium is absorbed as the free acid rather than the sodium salt from the gastrointestinal tract, which causes only weak irritation of the gastric mucosa. Subsequently, it is converted to an active metabolite by reduction of the ketone carbonyl to the transOH form. Therefore, loxoprofen sodium is thought to have a relatively weak gastrointestinal ulcerogenicity compared with ibuprofen [2]. Although NSAIDs are widely used after ocular operations to control post-surgical pain, they are reported to have ocular side effects, such as blurred vision. Tsurumaki et al. report that the ratio of the ocular side effect caused by ibuprofen was 25.1 times higher than that of loxoprofen sodium [3]. Fraufelder et al. propose that some drugs may be secreted in the tears in large amounts and cause a transient inflammation of the conjunctiva, thereby resulting in the need to discontinue the drugs [4]. Monaco et al. additionally report finding a relationship between the concentration in tears and in plasma [5]. However, to the best of
Determination of nonsteroidal anti-inflammatory drugs in human tear and plasma samples using…
Ibuprofen CH3 O CH3 OH H3C
Loxoprofen
O CH3
Ketoprofen-d3 D
O
blank matrix for real human tears in the present study [6]. The ATS used was Soft Santear from Santen Pharmaceutical Co. (Osaka, Japan). The acetonitrile and water from Wako Pure Chemical Industries, Inc. (Osaka, Japan), were HPLC-MS grade. All the other common chemicals used were of the highest purity commercially available. Structures of ibuprofen, loxoprofen and ketoprofen-d3 are shown in Fig. 1. Preparation of standard solutions and quality control (QC) samples
O HO
365
D
D COOH
Fig. 1 Structures of ibuprofen, loxoprofen and ketoprofen-d3
our knowledge, no reports have investigated the tear fluid drug concentration of NSAIDs after oral administration. Therefore, the relationship between the drug concentration in tears and ocular side effects remains unknown. Simple and sensitive determination of drug concentrations in human tears and plasma is required in order to investigate how NSAIDs are transferred from the plasma to tears and whether they can influence ocular lesions. We attempted to use ultra-fast liquid chromatography-tandem mass spectrometry (UFLC-MS/MS) in order to establish a new rapid and sensitive method for analyzing NSAIDs in human tears and plasma after single dose administrations of ibuprofen and loxoprofen sodium. We then conducted a clinical study that was designed to confirm that our new analytical method could successfully measure drug concentrations in human tears and plasma after administration of NSAIDs in Japanese subjects.
Materials and methods Materials Ibuprofen and ketoprofen-d3 were purchased from SigmaAldrich, Inc. (St. Louis, MO, USA). Loxoprofen was purchased from LKT Laboratories (St. Paul, MN, USA). Artificial tear solution (ATS) was used as the surrogate
Stock standard solutions of ibuprofen, loxoprofen and an internal standard (IS, ketoprofen-d3) were prepared by dissolving an accurately weighed quantity of each compound in methanol to obtain final concentrations of 1 mg/ ml. Solutions were stored at 4 C. Working standard solutions were prepared by serial dilution of the stock solutions using the initial UFLC mobile phase (10 mM ammonium acetate in 50 % acetonitrile). All working solutions were made fresh each week and stored at 4 C. Calibration standards were prepared by mixing appropriate amounts of working standard solutions to achieve concentrations of ibuprofen and loxoprofen that ranged from 0.02 to 1 lg/ml for the ATS and 0.1–5 lg/ml for the drugfree plasma. The IS concentrations were 0.05 lg/ml for ATS and 0.5 lg/ml for the plasma. QC samples for both the ATS and plasma were prepared by the same procedure used for the calibration standards. A 10-ll volume of ATS containing each of the test drugs was mixed with 40 ll each of acetonitrile and 10 mM ammonium acetate. After centrifugation of the mixture at 19,6009g for 3 min, a 10-ll aliquot of clear supernatant was placed in an autosampler vial for injection into the UFLC-MS/MS system. For plasma samples, drug-free whole blood samples were obtained intravenously from healthy volunteers and mixed with heparin sodium anticoagulant. Drug-free plasma samples were prepared by centrifuging heparinized whole blood at 17009g at 4 C for 10 min, decanting the plasma into a clean centrifuge tube and then storing it at -80 C until use. A 20-ll volume of human plasma containing each of the test drugs was mixed with 80 ll of 10 mM ammonium acetate solution and 400 ll of acetonitrile. After centrifugation of the mixture at 19,6009g for 3 min, the 100 ll of supernatant was diluted with 100 ll of 10 mM ammonium acetate. A 10-ll aliquot of the mixture was placed in an autosampler vial for injection into the UFLC-MS/MS system. UFLC-MS/MS conditions The UFLC system (Shimadzu Corp., Kyoto, Japan) consists of two LC-20AD pumps, a SIL-20AC HT
123
366
autosampler, CTO-20AC column oven, CBM-20A communications bus module and SPD-20A UV detector. UFLC separations were performed using a Unison UK-C18 HT column (100 mm 9 3 mm i.d., particle size 3 lm, Imtakt Corp., Kyoto, Japan). The mobile phases were 10 mM ammonium acetate solution (pH 6.8, solvent A) and acetonitrile (solvent B). The flow rate was 0.4 ml/min, and the mobile phase gradient ranged from 60 % solvent A/40 % solvent B to 1 % solvent A/99 % solvent B over a period of 5 min. A column temperature of 40 C and sample size of 10 ll were used for all injections. An in-line filter (pore size 0.2 lm, Jasco Corp., Tokyo, Japan) was installed between the autosampler and UK-C18 HT column for protection of the column and mass spectrometer. Contamination of the mass spectrometer with matrix impurities was avoided by directing the eluent to a waste pool via a diverter valve during the early stage of elution. After 0.8 min, the diverter valve was switched to the injection position, and the analytes were introduced into the mass spectrometer for content detection. Mass spectrometric measurements were conducted using a Hybrid API 4000 QTrap mass spectrometer (Applied Biosystems Sciex, Framingham, MA, USA) equipped with a Turbo V Source for TurboIonSprayTM and an electric tenport diverter valve, and set in the triple quadrupole mode. Analyses were carried out using the ESI setting in the negative ion mode: TurboIonSprayTM temperature, 500 C; ion source voltage, -4500 V. Method for recovery, quantitation and linearity Recoveries were calculated by comparing the chromatographic peak areas of the analyte in the QC samples with those obtained by direct injection of the standard analytes dissolved in the initial mobile phase (60 % solvent A/40 % solvent B) at three different concentrations. Regression equations for ibuprofen and loxoprofen in the ATS and plasma were obtained by fitting the ratio of the peak areas of the analytes to those of the IS (0.05 lg/ml for ATS and 0.5 lg/ml for plasma) versus the concentrations of the analyte. The acceptance criterion for the correlation coefficient was [0.9990, and the limit of detection (LOD) was defined as the lowest concentration of analyte spiked in the ATS and plasma that could be detected with a signal-to-noise ratio of at least 3. The lower limit of quantitation (LLOQ) was determined as the lowest concentration on the calibration curve that could be detected with a signal-to-noise ratio of at least 10. The upper limit of quantitation (ULOQ) for the two NSAIDs was set at 1 lg/ml for ATS and 5 lg/ml for plasma in order to protect the MS detector from excess ions. The assay precision and method accuracy were estimated by intra- and interday validation. The intraday coefficient of variation (CV) was determined from six
123
M. Hirosawa et al.
replicate analyses of QC samples spiked with the two NSAIDs at three different concentrations on the same day. The same procedure was repeated for 6 days to determine the interday CV. The analyte concentrations in the QC samples were calculated by the regression equations and accuracy expressed as percentage of the mean calculated concentration relative to the nominal concentration. Evaluations of precision and accuracy were based on previously published criteria [7]. Clinical study To confirm the utility of the present method in real samples of human tears and plasma, we performed a clinical study in healthy Japanese subjects. This study was reviewed and approved by the Institutional Review Board at Showa University Karasuyama Hospital (approval date: 06/17/ 2014). The study was conducted at Showa University Clinical Research Institute for Clinical Pharmacology and Therapeutics (Tokyo, Japan) in compliance with the Declaration of Helsinki (World Medical Association) and Ethical Guidelines for Clinical Study (Ministry of Health, Labour and Welfare). Prior to starting the investigation, this study was registered at the official clinical study database (UMIN-CTR, UMIN ID: UMIN000013575). This cross-over designed study was conducted from 26 June 2014 to 4 July 2014, with a 1-week washout period. All participants provided written informed consent before participation. After orally administering a single therapeutic dose of ibuprofen and loxoprofen (200 and 60 mg, respectively) to six healthy Japanese subjects, we evaluated the pharmacokinetics in tear fluid and plasma and performed a safety assessment. The recruited subjects were healthy volunteers aged from 20 to 45 and had no relevant ocular abnormalities of the cornea, conjunctiva, lacrimal system, or meibomian glands. All subjects were further assessed by the Schirmer test paper method (C10 mm/5 min). Tear and blood samples were collected at 0.5, 1, 2, 3 and 4 h before and after each dose. Tear samples (10 ll) were carefully collected using Microcaps Disposable Micropipets (Drummond Scientific Co., PA, USA). Blood samples (5 ml) were collected and transferred to centrifuge tubes containing heparin sodium, and the resultant plasma samples were then subjected to the above-described procedure. All tear and plasma samples were stored at -80 C until analysis. All analyses were performed at the Department of Legal Medicine, Showa University. Concentrations at each time point are expressed as the mean value ± standard deviation. The area under the drug concentrationtime curve from the pre-dose to 4 h after the dose area under the curve (AUC)0?4 was calculated using the trapezoidal rule.
Determination of nonsteroidal anti-inflammatory drugs in human tear and plasma samples using…
Results Method validation In the ATS, the recovery and extraction efficiencies of the two NSAIDs were determined at three different concentrations that ranged from 96.0 to 117.0. The ATS samples (10 ll) were spiked with LLOQ concentrations of each drug and 0.05 lg of IS. Good separation and peak shapes were achieved within an analysis time of 3 min. Regression equations for the two NSAIDs exhibited good linearities, with a correlation coefficient of 0.9999. The LOD, LLOQ and ULOQ values for the two drugs spiked in the ATS under optimal conditions were 0.005, 0.02 and 1.0 lg/ml, respectively. Intra- and interday precisions and accuracies were evaluated by assessing QC samples prepared from the ATS (Table 1). The intraday CVs were not greater than 3.1 %, and the accuracies ranged from 100.5 to 116.5 % for all concentrations. The interday CVs were not greater than 6.8 %, and the accuracies ranged from 100.0 to 118.3 % for all concentrations. In human plasma, the recovery and extraction efficiencies for the two NSAIDs were determined at three different concentrations that ranged from 99.0 to 105.7. Plasma samples (20 ll) were spiked with LLOQ concentrations of each drug and 0.5 lg of IS. Good separation and peak shapes were achieved within an analysis time of 3 min. Regression equations for the two NSAIDs exhibited good linearities, with correlation coefficients of at least 0.9997. The LOD, LLOQ and ULOQ values for the two drugs spiked in the
367
plasma under optimal conditions were 0.01, 0.1 and 5.0 lg/ ml, respectively. Intra- and interday precisions and accuracies were evaluated by assessing the QC samples prepared from human plasma (Table 1). The intraday CVs were not greater than 2.3 %, and the accuracies ranged from 89.1 to 100.3 % for all of the concentrations. The interday CVs were not greater than 3.9 %, and the accuracies ranged from 90.8 to 100.8 % for all of the concentrations. Clinical study Six healthy volunteers (5 males, 1 female) ranging in age from 26 to 45 years (mean age ± SD, 32.2 ± 6.7) participated in this clinical study. The mean BMI ± SD was 21.8 ± 2.5, ranging from 18.0 to 24.9 kg/m2. No adverse events were observed, and all participants completed the study. Concentrations of each drug in the tears and plasma at each sampling point after a single oral administration of either ibuprofen (200 mg) or loxoprofen (60 mg) are shown in Table 2. The tear concentration profile of ibuprofen exhibited similar pharmacokinetics to that observed for the plasma. After the oral administration, tear and plasma concentrations increased. Maximum concentrations were 0.12 ± 0.05 lg/ml in the tears and 14.01 ± 4.47 lg/ml in the plasma at 3 h. At the maximum concentration time (3 h after the dose), the tear/plasma ratio was 0.0088 fold. For loxoprofen, the tear and plasma concentration profiles were similar. Compared to ibuprofen, the tear and
Table 1 Precision and accuracy for ibuprofen and loxoprofen in quality control samples Compound
Concentration added (lg/ml)
Intraday (n = 6)
Interday (n = 6)
Concentration detected (lg/ml)
Precision (CV, %)
Accuracy (%)
Concentration detected (lg/ml)
Precision (CV, %)
Accuracy (%)
0.02 ± 0.0003a
1.5
110
0.02 ± 0.0014
6.8
118
Tears Ibuprofen
Loxoprofen
0.02 0.2
0.23 ± 0.0072
3.1
117
0.20 ± 0.0042
2.1
100
1.0
1.03 ± 0.0160
1.6
103
1.02 ± 0.0320
3.1
102
0.02
0.02 ± 0.0006
3.0
101
0.02 ± 0.0011
5.3
111
0.2
0.23 ± 0.0029
1.3
115
0.20 ± 0.0057
2.8
101
1.0
1.04 ± 0.0130
1.3
104
1.02 ± 0.0310
3.0
103
Plasma Ibuprofen
Loxoprofen
0.1
0.09 ± 0.0017
1.9
89
0.09 ± 0.0029
3.1
92
1.0
0.99 ± 0.0100
1.1
99
0.97 ± 0.0330
3.5
97
5.0
4.64 ± 0.1100
2.3
93
4.85 ± 0.1900
3.9
97
0.1
0.09 ± 0.0009
1.0
90
0.09 ± 0.0015
1.7
91
1.0
1.00 ± 0.0028
0.3
100
1.01 ± 0.0021
2.1
101
5.0
4.90 ± 0.0370
0.7
98
5.04 ± 0.0930
1.9
101
CV coefficient of variation a
The values are mean ± SD
123
368
M. Hirosawa et al.
Table 2 Concentration of each drug in the tears and plasma at each sampling point after a single oral administration of ibuprofen (200 mg) and loxoprofen (60 mg) Time after administration (h)
Ibuprofen
Loxoprofen
Tears (lg/ml)
Plasma (lg/ml)
Tear/plasma ratio
Tears (lg/ml)
Plasma (lg/ml)
Tear/plasma ratio
0
0
0
0
0
0
0
0.5
0.04 ± 0.03
6.3 ± 5.47
0.0064
0.31 ± 0.21
5.51 ± 1.82
0.0562
1 2
0.09 ± 0.07 0.11 ± 0.05
11.5 ± 7.08 14.0 ± 4.02
0.0075 0.0075
0.21 ± 0.14 0.10 ± 0.04
3.14 ± 0.89 1.52 ± 0.28
0.0678 0.0687
3
0.12 ± 0.05
14.0 ± 4.47
0.0088
0.07 ± 0.09
1.04 ± 0.33
0.0623
4
0.08 ± 0.03
10.9 ± 3.83
0.0070
0.05 ± 0.02
0.68 ± 0.08
0.0675
The concentration values are mean ± SD (n = 6). The tear/plasma ratio was expressed as mean value. The mean ratio value at each time point was calculated from the individual tear and plasma concentration at the time point
plasma concentrations increased rapidly. Maximum concentrations at 0.5 h were 0.31 ± 0.21 lg/ml in the tears and 5.51 ± 1.82 lg/ml in the plasma. At the maximum concentration time (0.5 h after the dose), the tear/plasma ratio was 0.0562 fold. Figure 2 shows the time course for the concentrations in the plasma after a single oral administration of ibuprofen (200 mg) and loxoprofen (60 mg). The concentration of loxoprofen reached a maximum at 0.5 h and then decreased. The concentration of ibuprofen increased until 3 h and after reaching a maximum concentration then gradually decreased. Figure 3 shows the time course for the concentrations in the tears after a single oral administration of the study drugs. The maximum concentration of loxoprofen was reached at 0.5 h and then decreased, similar to what was seen in the plasma. The time course for the concentration of ibuprofen was similar to that observed in the plasma. The maximum concentration occurred at 3 h and then gradually decreased. Area under the drug concentration-time curves for the study drugs that were calculated until 4 h after the dose (AUC0?4) are shown in Table 3. The AUC0?4 for ibuprofen was 21.2 lg h/ml in the tears and 2768 lg h/ml in the plasma. Therefore, the ibuprofen tear/plasma ratio was 0.0076 fold. In contrast, the AUC0?4 for loxoprofen in the tears and plasma was 23.8 and 479 lg h/ml, respectively. The loxoprofen tear/plasma ratio was 0.05 fold.
Discussion Due to their relatively good safety profile, NSAIDs are widely used in clinical practice. However, ocular side effects such as blurred vision are reported in users of NSAIDs. It has been speculated that some medications are secreted in the tears and cause a transient inflammation of the conjunctiva that resolves upon discontinuing the drug,
123
as has been reported for diazepam or isotretinoin [8, 9]. Fraunfelder et al. report that some NSAIDs, such as celecoxib and rofecoxib, caused conjunctivitis after oral administration and also suggest that these side effects might be attributed to drug secretion into the tears [4]. Concentrations in the tears are reported to increase after an oral administration of a drug [10, 11]. However, since the volume of tears is normally only a few ll (typically 5–10 ll), sampling and analysis are very difficult, although several methods for determining the concentration of NSAIDs in the plasma are reported [12–19]. In previous studies that analyzed tear concentrations, large volumes of tears were required. In order to obtain such large volumes, stimulation using onion vapors or Schirmer strips was performed [10, 11, 20, 21]. Although the Schirmer strip method is quite simple and noninvasive, it cannot be used to perform an accurate quantitative analysis as it is not possible to measure the exact amount of tears that permeate the strip. The use of the UFLC-MS/MS technique for analysis makes it possible to perform a high-throughput and sensitive determination. Using this method, we attempted to create a new rapid and sensitive analysis that could determine ibuprofen and loxoprofen concentrations in human tears and plasma. With this method, only 10 ll of tears is required. After collection, samples are diluted with a small volume of mobile phase and centrifuged. The clear supernatant extract is then directly injected into the UFLC-MS/MS. Hence, we were able to achieve a good separation and peak shapes for ibuprofen, loxoprofen and ketoprofen-d3 (IS) within a 3-min analysis time. For plasma samples, previous NSAID extraction methods employed large volumes of plasma (0.05–1 ml) and complex sample pretreatment procedures that involved liquid and liquid extraction or solid-phase extraction [14– 19]. The new method used in the current study requires only a simple pretreatment step that involves a small volume of acetonitrile (400 ll), which makes it possible for
Fig. 2 Plasma concentration of study drugs after a single oral administration of loxoprofen and ibuprofen. Data are expressed as mean ± SD
369
Plasma concentration ( g/mL)
Determination of nonsteroidal anti-inflammatory drugs in human tear and plasma samples using…
Time after administration (h)
Fig. 3 Tear concentration of study drugs after a single oral administration of loxoprofen and ibuprofen. Data are expressed as mean ± SD
Time after administration (h)
Table 3 Area under the drug concentration-time curve for ibuprofen (200 mg) and loxoprofen (60 mg)
AUC0?4 of tears (lg h/ml)
AUC0?4 of plasma (lg h/ml)
Tear/plasma ratio
Ibuprofen
21.1
2768
0.0076
Loxoprofen
23.8
479
0.050
The area under the drug concentration-time curve from pre-dose to 4 h after the dose (AUC0?4) was calculated based on the drug concentration in tears and plasma of individual study subjects. The tear/plasma ratio was expressed as the mean value, and these ratios were calculated from the tear and plasma AUC value in each individual subject. Data were expressed as mean value (n = 6)
protein precipitation from the plasma. Furthermore, with this method it is possible to perform a chromatographic separation of the tear and plasma samples on the same column and under the same conditions. Thus, the current results demonstrate that our method is a simple, rapid and sensitive analysis that can be used to determine tear and plasma ibuprofen and loxoprofen concentrations. To confirm the utility of this new analytical method, we conducted a clinical study that examined the determination of tear and plasma concentrations in healthy subjects. In addition, we investigated the transfer of NSAIDs from the plasma to the tears and then considered the potential influence of substances in the tears when subjects took an
over-the-counter tablet consisting of other substances in addition to the active ingredient. In our current study, we showed that the maximum concentration of ibuprofen in the plasma at 3 h was 14.01 ± 4.47 lg/ml, while for loxoprofen at 0.5 h it was 5.51 ± 1.82 lg/ml. These concentrations are consistent with the concentrations reported in other studies [15–19, 22]. Thus, this confirms that our new technique is as valid as other reported methods, in addition to being both consistent and reproducible. Tsurumaki et al. [3] report that the frequency of ocular side effects caused by ibuprofen was higher than that of loxoprofen. We presumed that the difference in the
123
370
frequency could be attributed to the difference in the tear secretion profiles between the two drugs, i.e., the concentration of ibuprofen in the tears would be higher than that of loxoprofen. However, our current results showed that the concentration of loxoprofen was higher than that of ibuprofen, with a tear/plasma ratio of 0.0562 fold for loxoprofen and 0.0088 fold for ibuprofen. These results demonstrate that larger amounts of loxoprofen are secreted into the tears, which is inconsistent with our previously reported findings. In the current study, the last sampling point was set at 4 h after the oral administration. The concentration of loxoprofen in the tears and plasma decreased to levels near the LOD. In contrast, the levels of ibuprofen were much higher, even at the last sampling point. Elimination halflife (t1/2) of loxoprofen calculated from the values in the current study showed the t1/2 to be 1.3 h in the plasma and 1.5 h in the tears. These values suggest that the pharmacokinetics of loxoprofen should be similar in the tears and the plasma. A previous report found the t1/2 of ibuprofen to be 2 h [22]. Ibuprofen is lipid-soluble while loxoprofen is water-soluble; obviously this solubility difference would change the pharmacokinetics of the drugs. Therefore, careful consideration is required when attempting to adapt plasma pharmacokinetic profiles to tissue areas, such as the tears. If we assume that the pharmacokinetics of ibuprofen in the tears and in the plasma are similar and that the t1/2 of the ibuprofen is similar in the tears and plasma (about 2 h), then ibuprofen should remain on the ocular surface longer than loxoprofen. In reality, the maximum concentration of ibuprofen in the tears occurred at 3 h. At this time point, the loxoprofen in the tears had almost completely disappeared. This demonstrates that the presence of ibuprofen on the ocular surface will continue for more than 3 h after the initial dose. Because of insufficient sampling points in the current study, we could not determine the total amount of ibuprofen and loxoprofen secretion into the tears. Thus, it was not possible to fully examine the total amount of drug exposure that occurred in this study. In order to definitively clarify this point, information on not only the drug exposure time, which can be taken from this study, but also the total amount of drug exposure that actually occurs will be required. A further study with much longer sampling times will need to be undertaken in order to determine the elimination of ibuprofen from the tears and plasma. The current study investigated a new high-throughput method for determining NSAIDs and found it to be very useful in clinical and toxicological analyses, as it only requires a small amount of tears (10 ll) and plasma volume (20 ll). This report also presents data obtained from actual tear and plasma determinations of ibuprofen and
123
M. Hirosawa et al.
loxoprofen in human subjects and, based on the findings of this analysis, confirmed the transition of NSAIDs from the plasma to the tears. Conflicts of interest M. Hirosawa, None; T. Sambe, None; N. Uchida, None; X.-P. Lee, None; K. Sato, None; S. Kobayashi, None.
References 1. Bushra R, Aslam N. An overview of clinical pharmacology of Ibuprofen. Oman Med J. 2010;25:155–1661. 2. Terada A, Naruto S, Wachi K, Tanaka S, Iizuka Y, Misaka E. Synthesis and antiinflammatory activity of [(Cycloalkylmethyl)phenyl]acetic acids and related compounds. J Med Chem. 1984;404:212–6. 3. Tsurumaki N, Tsuchihashi H. The application of adverse effect database of drug 30. Comparison between ibuprofen and loxoprofen sodium hydrate in efficacy. Safety and economy. Japanese J Med Pharm Sci. 2013;69:275–81. 4. Fraunfelder FW, Solomon J, Mehelas TJ. Ocular adverse effects associated with cyclooxygenase-2 inhibitors. Arch Ophthalmol. 2006;124:277–9. 5. Monaco F, Piredda S, Mutani R, Mastropaolo C, Tondi M. The free fraction of valproic acid in tears, saliva, and cerebrospinal fluid. Epilepsia. 1982;23:23–6. 6. Mirejovsky D, Patel AS, Rodriguez DD, Hunt TJ. Lipid adsorption onto hydrogel contact lens materials. Advantages of Nile red over oil red O in visualization of lipids. Optom Vis Sci. 1991;68:858–64. 7. U.S.Food and Drug Administration. Guidance for industry analytical procedures and methods validation for drugs and biologics. FDA. 2014. http://www.fda.gov/downloads/Drugs/ GuidanceComplianceRegulatoryInformation/Guidances/UCM38 6366.pdf. 8. Lutz EG. Allergic conjunctivitis due to Diazepam. Am J Psychiatry. 1975;132:548. 9. Fraunfelder FT, LaBraico JM, Meyer SM. Adverse ocular reactions possibly associated with isotretinoin. Am J Ophthalmol. 1985;100:534–7. 10. Nakajima M, Sato S, Yamato S, Shimada K, Kitagawa S, Honda A, et al. Assessment of tear concentrations on therapeutic drug monitoring. III. Determination of theophylline in tears by gas chromatography/mass spectrometry with electron ionization mode. Drug Metab Pharmacokinet. 2003;18:139–45. 11. Mora P, Ceglarek U, Manzotti F, Zavota L, Carta A, Aldigeri R, et al. Cyclosporin A in the ocular fluids of uveitis patients following long-term systemic administration. Graefes Arch Clin Exp Ophthalmol. 2008;246:1047–52. 12. Lee X-P, Kumazawa T, Hasegawa C, Arinobu T, Kato A, Seno H, et al. Determination of nonsteroidal anti-inflammatory drugs in human plasma by LC–MS-MS with a hydrophilic polymer column. Forensic Toxicol. 2010;28:96–104. 13. Nemoto T, Lee X-P, Kumazawa T, Hasegawa C, Fujishiro M, Marumo A, et al. High-throughput determination of nonsteroidal anti-inflammatory drugs in human plasma by HILIC–MS/MS. J Pharm Biomed Anal. Elsevier B.V.; 2014;88:71–80. 14. Canaparo R, Muntoni E, Zara GP, Della Pepa C, Berno E, Costa M, et al. Determination of Ibuprofen in human plasma by highperformance liquid chromatography: validation and application in pharmacokinetic study. Biomed Chromatogr. 2000;14:219–26. 15. Choo K-S, Kim I-W, Jung J-K, Suh Y-G, Chung S-J, Lee M-H, et al. Simultaneous determination of loxoprofen and its
Determination of nonsteroidal anti-inflammatory drugs in human tear and plasma samples using… diastereomeric alcohol metabolites in human plasma and urine by a simple HPLC-UV detection method. J Pharm Biomed Anal. 2001;25:639–50. 16. Lee HW, Ji HY, Sohn DH, Kim S-M, Lee YB, Lee HS. Liquid chromatography-tandem mass spectrometry method of loxoprofen in human plasma. Biomed Chromatogr. 2009;23:714–8. 17. Suenami K, Lim LW, Takeuchi T, Sasajima Y, Sato K, Takekoshi Y, et al. Rapid and simultaneous determination of nonsteroidal anti-inflammatory drugs in human plasma by LC-MS with solidphase extraction. Anal Bioanal Chem. 2006;384:1501–5. 18. Cho H-Y, Park C-H, Lee Y-B. Direct and simultaneous analysis of loxoprofen and its diastereometric alcohol metabolites in human serum by on-line column switching liquid chromatography and its application to a pharmacokinetic study. J Chromatogr B Analyt Technol Biomed Life Sci. 2006;835:27–34.
371
19. Farrar H, Letzig L, Gill M. Validation of a liquid chromatographic method for the determination of ibuprofen in human plasma. J Chromatogr B. 2002;780:341–8. 20. Arnold DR, Granvil CP, Ward KW, Proksch JW. Quantitative determination of besifloxacin, a novel fluoroquinolone antimicrobial agent, in human tears by liquid chromatography-tandem mass spectrometry. J Chromatogr B Analyt Technol Biomed Life Sci. 2008;867:105–10. 21. Chiambaretta F, Garraffo R, Elena P, Pouliquen P, Delval L, Rigal D, et al. Tear concentrations of azithromycin following topical administration of a single dose of (T1225) in healthy volunteers. Eur J Opthalmology. 2008;18:13–20. 22. Davies NM. Clinical pharmacokinetics of ibuprofen. The first 30 years. Clin Pharmacokinet. 1998;34:101–54.
123
LABORATORY INVESTIGATION
Determination of nonsteroidal anti-inflammatory drugs in human tear and plasma samples using ultra-fast liquid chromatography-tandem mass spectrometry Makiko Hirosawa1,2 • Takehiko Sambe2 • Naoki Uchida2,3 • Xiao-Pen Lee4 Keizo Sato4 • Shinichi Kobayashi3
•
Received: 28 November 2014 / Accepted: 10 April 2015 / Published online: 19 June 2015 Japanese Ophthalmological Society 2015
Abstract Purpose To evaluate a new rapid and sensitive method for analyzing human tears and plasma for nonsteroidal antiinflammatory drugs (NSAIDs) and investigate the influence of the transfer of NSAIDs in an ocular lesion. Methods In this cross-over study, a single dose of 200 mg of ibuprofen and 60 mg of loxoprofen sodium were orally administered to six healthy Japanese subjects. Collected samples were analyzed by ultra-fast liquid chromatography tandem mass spectrometry (UFLC-MS/MS). Results Recoveries of the two drugs spiked in the tears and plasma were 96.0–117.0 % in the tears and 99.0–105.7 % in the plasma. Regression equations for both NSAIDs showed excellent linearity from 0.02–1.0 lg/ml for the tears and 0.1–5.0 lg/ml for the plasma, with the limits of detection at 0.02 lg/ml for tears and 0.1 lg/ml for plasma. Conclusion This new high-throughput NSAID determination method only requires a small tear amount (10 ll) and plasma volume (20 ll) and thus will be useful in clinical and toxicological analyses. Analytical results also showed the presence of ibuprofen and loxoprofen in the
& Makiko Hirosawa [email protected] 1
Department of Ophthalmology, Showa University School of Medicine, 1-5-8 Hatanodai, Shinagawa-ku, Tokyo 142-8555, Japan
2
Department of Pharmacology, Showa University School of Medicine, Tokyo 142-8555, Japan
3
Clinial Research Institute of Clinical Pharmacology and Therapeutics, Showa University, Tokyo 157-8577, Japan
4
Department of Legal Medicine, Showa University School of Medicine, Tokyo 142-8555, Japan
123
actual tears and plasma, which confirms the transition of NSAIDs from the tears to the plasma. Keywords Nonsteroidal anti-inflammatory drugs (NSAIDs) Tandem mass spectrometry (MS/MS) Human tears and plasma
Introduction Nonsteroidal anti-inflammatory drugs (NSAIDs) are widely used for the treatment of pain, fever and inflammation. Ibuprofen is the first member of the propionic acid class of NSAIDs to come into general use [1]. Loxoprofen sodium is a phenylpropionate anti-inflammatory agent with marked analgesic and antipyretic activities. After oral administration, loxoprofen sodium is absorbed as the free acid rather than the sodium salt from the gastrointestinal tract, which causes only weak irritation of the gastric mucosa. Subsequently, it is converted to an active metabolite by reduction of the ketone carbonyl to the transOH form. Therefore, loxoprofen sodium is thought to have a relatively weak gastrointestinal ulcerogenicity compared with ibuprofen [2]. Although NSAIDs are widely used after ocular operations to control post-surgical pain, they are reported to have ocular side effects, such as blurred vision. Tsurumaki et al. report that the ratio of the ocular side effect caused by ibuprofen was 25.1 times higher than that of loxoprofen sodium [3]. Fraufelder et al. propose that some drugs may be secreted in the tears in large amounts and cause a transient inflammation of the conjunctiva, thereby resulting in the need to discontinue the drugs [4]. Monaco et al. additionally report finding a relationship between the concentration in tears and in plasma [5]. However, to the best of
Determination of nonsteroidal anti-inflammatory drugs in human tear and plasma samples using…
Ibuprofen CH3 O CH3 OH H3C
Loxoprofen
O CH3
Ketoprofen-d3 D
O
blank matrix for real human tears in the present study [6]. The ATS used was Soft Santear from Santen Pharmaceutical Co. (Osaka, Japan). The acetonitrile and water from Wako Pure Chemical Industries, Inc. (Osaka, Japan), were HPLC-MS grade. All the other common chemicals used were of the highest purity commercially available. Structures of ibuprofen, loxoprofen and ketoprofen-d3 are shown in Fig. 1. Preparation of standard solutions and quality control (QC) samples
O HO
365
D
D COOH
Fig. 1 Structures of ibuprofen, loxoprofen and ketoprofen-d3
our knowledge, no reports have investigated the tear fluid drug concentration of NSAIDs after oral administration. Therefore, the relationship between the drug concentration in tears and ocular side effects remains unknown. Simple and sensitive determination of drug concentrations in human tears and plasma is required in order to investigate how NSAIDs are transferred from the plasma to tears and whether they can influence ocular lesions. We attempted to use ultra-fast liquid chromatography-tandem mass spectrometry (UFLC-MS/MS) in order to establish a new rapid and sensitive method for analyzing NSAIDs in human tears and plasma after single dose administrations of ibuprofen and loxoprofen sodium. We then conducted a clinical study that was designed to confirm that our new analytical method could successfully measure drug concentrations in human tears and plasma after administration of NSAIDs in Japanese subjects.
Materials and methods Materials Ibuprofen and ketoprofen-d3 were purchased from SigmaAldrich, Inc. (St. Louis, MO, USA). Loxoprofen was purchased from LKT Laboratories (St. Paul, MN, USA). Artificial tear solution (ATS) was used as the surrogate
Stock standard solutions of ibuprofen, loxoprofen and an internal standard (IS, ketoprofen-d3) were prepared by dissolving an accurately weighed quantity of each compound in methanol to obtain final concentrations of 1 mg/ ml. Solutions were stored at 4 C. Working standard solutions were prepared by serial dilution of the stock solutions using the initial UFLC mobile phase (10 mM ammonium acetate in 50 % acetonitrile). All working solutions were made fresh each week and stored at 4 C. Calibration standards were prepared by mixing appropriate amounts of working standard solutions to achieve concentrations of ibuprofen and loxoprofen that ranged from 0.02 to 1 lg/ml for the ATS and 0.1–5 lg/ml for the drugfree plasma. The IS concentrations were 0.05 lg/ml for ATS and 0.5 lg/ml for the plasma. QC samples for both the ATS and plasma were prepared by the same procedure used for the calibration standards. A 10-ll volume of ATS containing each of the test drugs was mixed with 40 ll each of acetonitrile and 10 mM ammonium acetate. After centrifugation of the mixture at 19,6009g for 3 min, a 10-ll aliquot of clear supernatant was placed in an autosampler vial for injection into the UFLC-MS/MS system. For plasma samples, drug-free whole blood samples were obtained intravenously from healthy volunteers and mixed with heparin sodium anticoagulant. Drug-free plasma samples were prepared by centrifuging heparinized whole blood at 17009g at 4 C for 10 min, decanting the plasma into a clean centrifuge tube and then storing it at -80 C until use. A 20-ll volume of human plasma containing each of the test drugs was mixed with 80 ll of 10 mM ammonium acetate solution and 400 ll of acetonitrile. After centrifugation of the mixture at 19,6009g for 3 min, the 100 ll of supernatant was diluted with 100 ll of 10 mM ammonium acetate. A 10-ll aliquot of the mixture was placed in an autosampler vial for injection into the UFLC-MS/MS system. UFLC-MS/MS conditions The UFLC system (Shimadzu Corp., Kyoto, Japan) consists of two LC-20AD pumps, a SIL-20AC HT
123
366
autosampler, CTO-20AC column oven, CBM-20A communications bus module and SPD-20A UV detector. UFLC separations were performed using a Unison UK-C18 HT column (100 mm 9 3 mm i.d., particle size 3 lm, Imtakt Corp., Kyoto, Japan). The mobile phases were 10 mM ammonium acetate solution (pH 6.8, solvent A) and acetonitrile (solvent B). The flow rate was 0.4 ml/min, and the mobile phase gradient ranged from 60 % solvent A/40 % solvent B to 1 % solvent A/99 % solvent B over a period of 5 min. A column temperature of 40 C and sample size of 10 ll were used for all injections. An in-line filter (pore size 0.2 lm, Jasco Corp., Tokyo, Japan) was installed between the autosampler and UK-C18 HT column for protection of the column and mass spectrometer. Contamination of the mass spectrometer with matrix impurities was avoided by directing the eluent to a waste pool via a diverter valve during the early stage of elution. After 0.8 min, the diverter valve was switched to the injection position, and the analytes were introduced into the mass spectrometer for content detection. Mass spectrometric measurements were conducted using a Hybrid API 4000 QTrap mass spectrometer (Applied Biosystems Sciex, Framingham, MA, USA) equipped with a Turbo V Source for TurboIonSprayTM and an electric tenport diverter valve, and set in the triple quadrupole mode. Analyses were carried out using the ESI setting in the negative ion mode: TurboIonSprayTM temperature, 500 C; ion source voltage, -4500 V. Method for recovery, quantitation and linearity Recoveries were calculated by comparing the chromatographic peak areas of the analyte in the QC samples with those obtained by direct injection of the standard analytes dissolved in the initial mobile phase (60 % solvent A/40 % solvent B) at three different concentrations. Regression equations for ibuprofen and loxoprofen in the ATS and plasma were obtained by fitting the ratio of the peak areas of the analytes to those of the IS (0.05 lg/ml for ATS and 0.5 lg/ml for plasma) versus the concentrations of the analyte. The acceptance criterion for the correlation coefficient was [0.9990, and the limit of detection (LOD) was defined as the lowest concentration of analyte spiked in the ATS and plasma that could be detected with a signal-to-noise ratio of at least 3. The lower limit of quantitation (LLOQ) was determined as the lowest concentration on the calibration curve that could be detected with a signal-to-noise ratio of at least 10. The upper limit of quantitation (ULOQ) for the two NSAIDs was set at 1 lg/ml for ATS and 5 lg/ml for plasma in order to protect the MS detector from excess ions. The assay precision and method accuracy were estimated by intra- and interday validation. The intraday coefficient of variation (CV) was determined from six
123
M. Hirosawa et al.
replicate analyses of QC samples spiked with the two NSAIDs at three different concentrations on the same day. The same procedure was repeated for 6 days to determine the interday CV. The analyte concentrations in the QC samples were calculated by the regression equations and accuracy expressed as percentage of the mean calculated concentration relative to the nominal concentration. Evaluations of precision and accuracy were based on previously published criteria [7]. Clinical study To confirm the utility of the present method in real samples of human tears and plasma, we performed a clinical study in healthy Japanese subjects. This study was reviewed and approved by the Institutional Review Board at Showa University Karasuyama Hospital (approval date: 06/17/ 2014). The study was conducted at Showa University Clinical Research Institute for Clinical Pharmacology and Therapeutics (Tokyo, Japan) in compliance with the Declaration of Helsinki (World Medical Association) and Ethical Guidelines for Clinical Study (Ministry of Health, Labour and Welfare). Prior to starting the investigation, this study was registered at the official clinical study database (UMIN-CTR, UMIN ID: UMIN000013575). This cross-over designed study was conducted from 26 June 2014 to 4 July 2014, with a 1-week washout period. All participants provided written informed consent before participation. After orally administering a single therapeutic dose of ibuprofen and loxoprofen (200 and 60 mg, respectively) to six healthy Japanese subjects, we evaluated the pharmacokinetics in tear fluid and plasma and performed a safety assessment. The recruited subjects were healthy volunteers aged from 20 to 45 and had no relevant ocular abnormalities of the cornea, conjunctiva, lacrimal system, or meibomian glands. All subjects were further assessed by the Schirmer test paper method (C10 mm/5 min). Tear and blood samples were collected at 0.5, 1, 2, 3 and 4 h before and after each dose. Tear samples (10 ll) were carefully collected using Microcaps Disposable Micropipets (Drummond Scientific Co., PA, USA). Blood samples (5 ml) were collected and transferred to centrifuge tubes containing heparin sodium, and the resultant plasma samples were then subjected to the above-described procedure. All tear and plasma samples were stored at -80 C until analysis. All analyses were performed at the Department of Legal Medicine, Showa University. Concentrations at each time point are expressed as the mean value ± standard deviation. The area under the drug concentrationtime curve from the pre-dose to 4 h after the dose area under the curve (AUC)0?4 was calculated using the trapezoidal rule.
Determination of nonsteroidal anti-inflammatory drugs in human tear and plasma samples using…
Results Method validation In the ATS, the recovery and extraction efficiencies of the two NSAIDs were determined at three different concentrations that ranged from 96.0 to 117.0. The ATS samples (10 ll) were spiked with LLOQ concentrations of each drug and 0.05 lg of IS. Good separation and peak shapes were achieved within an analysis time of 3 min. Regression equations for the two NSAIDs exhibited good linearities, with a correlation coefficient of 0.9999. The LOD, LLOQ and ULOQ values for the two drugs spiked in the ATS under optimal conditions were 0.005, 0.02 and 1.0 lg/ml, respectively. Intra- and interday precisions and accuracies were evaluated by assessing QC samples prepared from the ATS (Table 1). The intraday CVs were not greater than 3.1 %, and the accuracies ranged from 100.5 to 116.5 % for all concentrations. The interday CVs were not greater than 6.8 %, and the accuracies ranged from 100.0 to 118.3 % for all concentrations. In human plasma, the recovery and extraction efficiencies for the two NSAIDs were determined at three different concentrations that ranged from 99.0 to 105.7. Plasma samples (20 ll) were spiked with LLOQ concentrations of each drug and 0.5 lg of IS. Good separation and peak shapes were achieved within an analysis time of 3 min. Regression equations for the two NSAIDs exhibited good linearities, with correlation coefficients of at least 0.9997. The LOD, LLOQ and ULOQ values for the two drugs spiked in the
367
plasma under optimal conditions were 0.01, 0.1 and 5.0 lg/ ml, respectively. Intra- and interday precisions and accuracies were evaluated by assessing the QC samples prepared from human plasma (Table 1). The intraday CVs were not greater than 2.3 %, and the accuracies ranged from 89.1 to 100.3 % for all of the concentrations. The interday CVs were not greater than 3.9 %, and the accuracies ranged from 90.8 to 100.8 % for all of the concentrations. Clinical study Six healthy volunteers (5 males, 1 female) ranging in age from 26 to 45 years (mean age ± SD, 32.2 ± 6.7) participated in this clinical study. The mean BMI ± SD was 21.8 ± 2.5, ranging from 18.0 to 24.9 kg/m2. No adverse events were observed, and all participants completed the study. Concentrations of each drug in the tears and plasma at each sampling point after a single oral administration of either ibuprofen (200 mg) or loxoprofen (60 mg) are shown in Table 2. The tear concentration profile of ibuprofen exhibited similar pharmacokinetics to that observed for the plasma. After the oral administration, tear and plasma concentrations increased. Maximum concentrations were 0.12 ± 0.05 lg/ml in the tears and 14.01 ± 4.47 lg/ml in the plasma at 3 h. At the maximum concentration time (3 h after the dose), the tear/plasma ratio was 0.0088 fold. For loxoprofen, the tear and plasma concentration profiles were similar. Compared to ibuprofen, the tear and
Table 1 Precision and accuracy for ibuprofen and loxoprofen in quality control samples Compound
Concentration added (lg/ml)
Intraday (n = 6)
Interday (n = 6)
Concentration detected (lg/ml)
Precision (CV, %)
Accuracy (%)
Concentration detected (lg/ml)
Precision (CV, %)
Accuracy (%)
0.02 ± 0.0003a
1.5
110
0.02 ± 0.0014
6.8
118
Tears Ibuprofen
Loxoprofen
0.02 0.2
0.23 ± 0.0072
3.1
117
0.20 ± 0.0042
2.1
100
1.0
1.03 ± 0.0160
1.6
103
1.02 ± 0.0320
3.1
102
0.02
0.02 ± 0.0006
3.0
101
0.02 ± 0.0011
5.3
111
0.2
0.23 ± 0.0029
1.3
115
0.20 ± 0.0057
2.8
101
1.0
1.04 ± 0.0130
1.3
104
1.02 ± 0.0310
3.0
103
Plasma Ibuprofen
Loxoprofen
0.1
0.09 ± 0.0017
1.9
89
0.09 ± 0.0029
3.1
92
1.0
0.99 ± 0.0100
1.1
99
0.97 ± 0.0330
3.5
97
5.0
4.64 ± 0.1100
2.3
93
4.85 ± 0.1900
3.9
97
0.1
0.09 ± 0.0009
1.0
90
0.09 ± 0.0015
1.7
91
1.0
1.00 ± 0.0028
0.3
100
1.01 ± 0.0021
2.1
101
5.0
4.90 ± 0.0370
0.7
98
5.04 ± 0.0930
1.9
101
CV coefficient of variation a
The values are mean ± SD
123
368
M. Hirosawa et al.
Table 2 Concentration of each drug in the tears and plasma at each sampling point after a single oral administration of ibuprofen (200 mg) and loxoprofen (60 mg) Time after administration (h)
Ibuprofen
Loxoprofen
Tears (lg/ml)
Plasma (lg/ml)
Tear/plasma ratio
Tears (lg/ml)
Plasma (lg/ml)
Tear/plasma ratio
0
0
0
0
0
0
0
0.5
0.04 ± 0.03
6.3 ± 5.47
0.0064
0.31 ± 0.21
5.51 ± 1.82
0.0562
1 2
0.09 ± 0.07 0.11 ± 0.05
11.5 ± 7.08 14.0 ± 4.02
0.0075 0.0075
0.21 ± 0.14 0.10 ± 0.04
3.14 ± 0.89 1.52 ± 0.28
0.0678 0.0687
3
0.12 ± 0.05
14.0 ± 4.47
0.0088
0.07 ± 0.09
1.04 ± 0.33
0.0623
4
0.08 ± 0.03
10.9 ± 3.83
0.0070
0.05 ± 0.02
0.68 ± 0.08
0.0675
The concentration values are mean ± SD (n = 6). The tear/plasma ratio was expressed as mean value. The mean ratio value at each time point was calculated from the individual tear and plasma concentration at the time point
plasma concentrations increased rapidly. Maximum concentrations at 0.5 h were 0.31 ± 0.21 lg/ml in the tears and 5.51 ± 1.82 lg/ml in the plasma. At the maximum concentration time (0.5 h after the dose), the tear/plasma ratio was 0.0562 fold. Figure 2 shows the time course for the concentrations in the plasma after a single oral administration of ibuprofen (200 mg) and loxoprofen (60 mg). The concentration of loxoprofen reached a maximum at 0.5 h and then decreased. The concentration of ibuprofen increased until 3 h and after reaching a maximum concentration then gradually decreased. Figure 3 shows the time course for the concentrations in the tears after a single oral administration of the study drugs. The maximum concentration of loxoprofen was reached at 0.5 h and then decreased, similar to what was seen in the plasma. The time course for the concentration of ibuprofen was similar to that observed in the plasma. The maximum concentration occurred at 3 h and then gradually decreased. Area under the drug concentration-time curves for the study drugs that were calculated until 4 h after the dose (AUC0?4) are shown in Table 3. The AUC0?4 for ibuprofen was 21.2 lg h/ml in the tears and 2768 lg h/ml in the plasma. Therefore, the ibuprofen tear/plasma ratio was 0.0076 fold. In contrast, the AUC0?4 for loxoprofen in the tears and plasma was 23.8 and 479 lg h/ml, respectively. The loxoprofen tear/plasma ratio was 0.05 fold.
Discussion Due to their relatively good safety profile, NSAIDs are widely used in clinical practice. However, ocular side effects such as blurred vision are reported in users of NSAIDs. It has been speculated that some medications are secreted in the tears and cause a transient inflammation of the conjunctiva that resolves upon discontinuing the drug,
123
as has been reported for diazepam or isotretinoin [8, 9]. Fraunfelder et al. report that some NSAIDs, such as celecoxib and rofecoxib, caused conjunctivitis after oral administration and also suggest that these side effects might be attributed to drug secretion into the tears [4]. Concentrations in the tears are reported to increase after an oral administration of a drug [10, 11]. However, since the volume of tears is normally only a few ll (typically 5–10 ll), sampling and analysis are very difficult, although several methods for determining the concentration of NSAIDs in the plasma are reported [12–19]. In previous studies that analyzed tear concentrations, large volumes of tears were required. In order to obtain such large volumes, stimulation using onion vapors or Schirmer strips was performed [10, 11, 20, 21]. Although the Schirmer strip method is quite simple and noninvasive, it cannot be used to perform an accurate quantitative analysis as it is not possible to measure the exact amount of tears that permeate the strip. The use of the UFLC-MS/MS technique for analysis makes it possible to perform a high-throughput and sensitive determination. Using this method, we attempted to create a new rapid and sensitive analysis that could determine ibuprofen and loxoprofen concentrations in human tears and plasma. With this method, only 10 ll of tears is required. After collection, samples are diluted with a small volume of mobile phase and centrifuged. The clear supernatant extract is then directly injected into the UFLC-MS/MS. Hence, we were able to achieve a good separation and peak shapes for ibuprofen, loxoprofen and ketoprofen-d3 (IS) within a 3-min analysis time. For plasma samples, previous NSAID extraction methods employed large volumes of plasma (0.05–1 ml) and complex sample pretreatment procedures that involved liquid and liquid extraction or solid-phase extraction [14– 19]. The new method used in the current study requires only a simple pretreatment step that involves a small volume of acetonitrile (400 ll), which makes it possible for
Fig. 2 Plasma concentration of study drugs after a single oral administration of loxoprofen and ibuprofen. Data are expressed as mean ± SD
369
Plasma concentration ( g/mL)
Determination of nonsteroidal anti-inflammatory drugs in human tear and plasma samples using…
Time after administration (h)
Fig. 3 Tear concentration of study drugs after a single oral administration of loxoprofen and ibuprofen. Data are expressed as mean ± SD
Time after administration (h)
Table 3 Area under the drug concentration-time curve for ibuprofen (200 mg) and loxoprofen (60 mg)
AUC0?4 of tears (lg h/ml)
AUC0?4 of plasma (lg h/ml)
Tear/plasma ratio
Ibuprofen
21.1
2768
0.0076
Loxoprofen
23.8
479
0.050
The area under the drug concentration-time curve from pre-dose to 4 h after the dose (AUC0?4) was calculated based on the drug concentration in tears and plasma of individual study subjects. The tear/plasma ratio was expressed as the mean value, and these ratios were calculated from the tear and plasma AUC value in each individual subject. Data were expressed as mean value (n = 6)
protein precipitation from the plasma. Furthermore, with this method it is possible to perform a chromatographic separation of the tear and plasma samples on the same column and under the same conditions. Thus, the current results demonstrate that our method is a simple, rapid and sensitive analysis that can be used to determine tear and plasma ibuprofen and loxoprofen concentrations. To confirm the utility of this new analytical method, we conducted a clinical study that examined the determination of tear and plasma concentrations in healthy subjects. In addition, we investigated the transfer of NSAIDs from the plasma to the tears and then considered the potential influence of substances in the tears when subjects took an
over-the-counter tablet consisting of other substances in addition to the active ingredient. In our current study, we showed that the maximum concentration of ibuprofen in the plasma at 3 h was 14.01 ± 4.47 lg/ml, while for loxoprofen at 0.5 h it was 5.51 ± 1.82 lg/ml. These concentrations are consistent with the concentrations reported in other studies [15–19, 22]. Thus, this confirms that our new technique is as valid as other reported methods, in addition to being both consistent and reproducible. Tsurumaki et al. [3] report that the frequency of ocular side effects caused by ibuprofen was higher than that of loxoprofen. We presumed that the difference in the
123
370
frequency could be attributed to the difference in the tear secretion profiles between the two drugs, i.e., the concentration of ibuprofen in the tears would be higher than that of loxoprofen. However, our current results showed that the concentration of loxoprofen was higher than that of ibuprofen, with a tear/plasma ratio of 0.0562 fold for loxoprofen and 0.0088 fold for ibuprofen. These results demonstrate that larger amounts of loxoprofen are secreted into the tears, which is inconsistent with our previously reported findings. In the current study, the last sampling point was set at 4 h after the oral administration. The concentration of loxoprofen in the tears and plasma decreased to levels near the LOD. In contrast, the levels of ibuprofen were much higher, even at the last sampling point. Elimination halflife (t1/2) of loxoprofen calculated from the values in the current study showed the t1/2 to be 1.3 h in the plasma and 1.5 h in the tears. These values suggest that the pharmacokinetics of loxoprofen should be similar in the tears and the plasma. A previous report found the t1/2 of ibuprofen to be 2 h [22]. Ibuprofen is lipid-soluble while loxoprofen is water-soluble; obviously this solubility difference would change the pharmacokinetics of the drugs. Therefore, careful consideration is required when attempting to adapt plasma pharmacokinetic profiles to tissue areas, such as the tears. If we assume that the pharmacokinetics of ibuprofen in the tears and in the plasma are similar and that the t1/2 of the ibuprofen is similar in the tears and plasma (about 2 h), then ibuprofen should remain on the ocular surface longer than loxoprofen. In reality, the maximum concentration of ibuprofen in the tears occurred at 3 h. At this time point, the loxoprofen in the tears had almost completely disappeared. This demonstrates that the presence of ibuprofen on the ocular surface will continue for more than 3 h after the initial dose. Because of insufficient sampling points in the current study, we could not determine the total amount of ibuprofen and loxoprofen secretion into the tears. Thus, it was not possible to fully examine the total amount of drug exposure that occurred in this study. In order to definitively clarify this point, information on not only the drug exposure time, which can be taken from this study, but also the total amount of drug exposure that actually occurs will be required. A further study with much longer sampling times will need to be undertaken in order to determine the elimination of ibuprofen from the tears and plasma. The current study investigated a new high-throughput method for determining NSAIDs and found it to be very useful in clinical and toxicological analyses, as it only requires a small amount of tears (10 ll) and plasma volume (20 ll). This report also presents data obtained from actual tear and plasma determinations of ibuprofen and
123
M. Hirosawa et al.
loxoprofen in human subjects and, based on the findings of this analysis, confirmed the transition of NSAIDs from the plasma to the tears. Conflicts of interest M. Hirosawa, None; T. Sambe, None; N. Uchida, None; X.-P. Lee, None; K. Sato, None; S. Kobayashi, None.
References 1. Bushra R, Aslam N. An overview of clinical pharmacology of Ibuprofen. Oman Med J. 2010;25:155–1661. 2. Terada A, Naruto S, Wachi K, Tanaka S, Iizuka Y, Misaka E. Synthesis and antiinflammatory activity of [(Cycloalkylmethyl)phenyl]acetic acids and related compounds. J Med Chem. 1984;404:212–6. 3. Tsurumaki N, Tsuchihashi H. The application of adverse effect database of drug 30. Comparison between ibuprofen and loxoprofen sodium hydrate in efficacy. Safety and economy. Japanese J Med Pharm Sci. 2013;69:275–81. 4. Fraunfelder FW, Solomon J, Mehelas TJ. Ocular adverse effects associated with cyclooxygenase-2 inhibitors. Arch Ophthalmol. 2006;124:277–9. 5. Monaco F, Piredda S, Mutani R, Mastropaolo C, Tondi M. The free fraction of valproic acid in tears, saliva, and cerebrospinal fluid. Epilepsia. 1982;23:23–6. 6. Mirejovsky D, Patel AS, Rodriguez DD, Hunt TJ. Lipid adsorption onto hydrogel contact lens materials. Advantages of Nile red over oil red O in visualization of lipids. Optom Vis Sci. 1991;68:858–64. 7. U.S.Food and Drug Administration. Guidance for industry analytical procedures and methods validation for drugs and biologics. FDA. 2014. http://www.fda.gov/downloads/Drugs/ GuidanceComplianceRegulatoryInformation/Guidances/UCM38 6366.pdf. 8. Lutz EG. Allergic conjunctivitis due to Diazepam. Am J Psychiatry. 1975;132:548. 9. Fraunfelder FT, LaBraico JM, Meyer SM. Adverse ocular reactions possibly associated with isotretinoin. Am J Ophthalmol. 1985;100:534–7. 10. Nakajima M, Sato S, Yamato S, Shimada K, Kitagawa S, Honda A, et al. Assessment of tear concentrations on therapeutic drug monitoring. III. Determination of theophylline in tears by gas chromatography/mass spectrometry with electron ionization mode. Drug Metab Pharmacokinet. 2003;18:139–45. 11. Mora P, Ceglarek U, Manzotti F, Zavota L, Carta A, Aldigeri R, et al. Cyclosporin A in the ocular fluids of uveitis patients following long-term systemic administration. Graefes Arch Clin Exp Ophthalmol. 2008;246:1047–52. 12. Lee X-P, Kumazawa T, Hasegawa C, Arinobu T, Kato A, Seno H, et al. Determination of nonsteroidal anti-inflammatory drugs in human plasma by LC–MS-MS with a hydrophilic polymer column. Forensic Toxicol. 2010;28:96–104. 13. Nemoto T, Lee X-P, Kumazawa T, Hasegawa C, Fujishiro M, Marumo A, et al. High-throughput determination of nonsteroidal anti-inflammatory drugs in human plasma by HILIC–MS/MS. J Pharm Biomed Anal. Elsevier B.V.; 2014;88:71–80. 14. Canaparo R, Muntoni E, Zara GP, Della Pepa C, Berno E, Costa M, et al. Determination of Ibuprofen in human plasma by highperformance liquid chromatography: validation and application in pharmacokinetic study. Biomed Chromatogr. 2000;14:219–26. 15. Choo K-S, Kim I-W, Jung J-K, Suh Y-G, Chung S-J, Lee M-H, et al. Simultaneous determination of loxoprofen and its
Determination of nonsteroidal anti-inflammatory drugs in human tear and plasma samples using… diastereomeric alcohol metabolites in human plasma and urine by a simple HPLC-UV detection method. J Pharm Biomed Anal. 2001;25:639–50. 16. Lee HW, Ji HY, Sohn DH, Kim S-M, Lee YB, Lee HS. Liquid chromatography-tandem mass spectrometry method of loxoprofen in human plasma. Biomed Chromatogr. 2009;23:714–8. 17. Suenami K, Lim LW, Takeuchi T, Sasajima Y, Sato K, Takekoshi Y, et al. Rapid and simultaneous determination of nonsteroidal anti-inflammatory drugs in human plasma by LC-MS with solidphase extraction. Anal Bioanal Chem. 2006;384:1501–5. 18. Cho H-Y, Park C-H, Lee Y-B. Direct and simultaneous analysis of loxoprofen and its diastereometric alcohol metabolites in human serum by on-line column switching liquid chromatography and its application to a pharmacokinetic study. J Chromatogr B Analyt Technol Biomed Life Sci. 2006;835:27–34.
371
19. Farrar H, Letzig L, Gill M. Validation of a liquid chromatographic method for the determination of ibuprofen in human plasma. J Chromatogr B. 2002;780:341–8. 20. Arnold DR, Granvil CP, Ward KW, Proksch JW. Quantitative determination of besifloxacin, a novel fluoroquinolone antimicrobial agent, in human tears by liquid chromatography-tandem mass spectrometry. J Chromatogr B Analyt Technol Biomed Life Sci. 2008;867:105–10. 21. Chiambaretta F, Garraffo R, Elena P, Pouliquen P, Delval L, Rigal D, et al. Tear concentrations of azithromycin following topical administration of a single dose of (T1225) in healthy volunteers. Eur J Opthalmology. 2008;18:13–20. 22. Davies NM. Clinical pharmacokinetics of ibuprofen. The first 30 years. Clin Pharmacokinet. 1998;34:101–54.
123
