Voriconazole Increases while Itraconazole Decreases Plasma Meloxicam Concentrations

Transcription

1 ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Feb. 2009, p Vol. 53, No /09/$ doi: /aac Copyright 2009, American Society for Microbiology. All Rights Reserved. Voriconazole Increases while Itraconazole Decreases Plasma Meloxicam Concentrations V. V. Hynninen, 1,2 * K. T. Olkkola, 2 L. Bertilsson, 3 K. J. Kurkinen, 4 T. Korhonen, 1 P. J. Neuvonen, 4 and K. Laine 1,5 Department of Pharmacology, Drug Development and Therapeutics, University of Turku, Turku, Finland 1 ; Department of Anesthesiology, Intensive Care, Emergency Care and Pain Medicine, Turku University Hospital, Turku, Finland 2 ; Department of Laboratory Medicine, Division of Clinical Pharmacology at Karolinska Institute, Karolinska University Hospital, Huddinge, Stockholm, Sweden 3 ; Department of Clinical Pharmacology, University of Helsinki and Helsinki University Hospital, Helsinki, Finland 4 ; and TYKSLAB, Unit of Clinical Pharmacology, Turku, Finland 5 Received 24 April 2008/Returned for modification 18 September 2008/Accepted 8 November 2008 This study investigated the effect of voriconazole, an inhibitor of cytochrome P450 2C9 (CYP2C9) and CYP3A4, and itraconazole, an inhibitor of CYP3A4, on the pharmacokinetics and pharmacodynamics of meloxicam. Twelve healthy volunteers in a crossover study ingested 15 mg of meloxicam without pretreatment (control), after voriconazole pretreatment, and after itraconazole pretreatment. The plasma concentrations of meloxicam, voriconazole, itraconazole, and thromboxane B 2 (TxB 2 ) generation were monitored. Compared to the control phase, voriconazole increased the mean area under the plasma concentration-time curve from 0 to 72 h (AUC 0-72 ) of meloxicam by 47% (P < 0.001) and prolonged its mean half-life (t 1/2 )by51%(p < 0.01), without affecting its mean peak concentration (C max ). In contrast, itraconazole decreased the mean AUC 0-72 and C max of meloxicam by 37% (P < 0.001) and by 64% (P < 0.001), respectively, and prolonged its t 1/2 and time to C max. The plasma protein unbound fraction of meloxicam was unchanged by voriconazole and itraconazole. Lowered plasma meloxicam concentrations during the itraconazole phase were associated with decreased pharmacodymic effects of meloxicam, as observed by weaker inhibition of TxB 2 synthesis compared to the control and voriconazole phases. Voriconazole increases plasma concentrations of meloxicam, whereas itraconazole, unexpectedly, decreases plasma meloxicam concentrations, possibly by impairing its absorption. Meloxicam is a nonsteroidal anti-inflammatory drug (NSAID) of the oxicam class with selectivity toward cyclo-oxygenase 2 (COX-2) compared to COX-1 (8, 19). It is widely used in the treatment of osteoarthritis, rheumatoid arthritis, ankylosing spondylitis, and other rheumatological conditions. Meloxicam has an oral bioavailability of 89%, and its maximum plasma concentrations (C max ) are achieved within 4 to 11 h (10, 26). It is extensively bound to plasma proteins ( 99%), mainly to albumin (24). The elimination half-life (t 1/2 ) of meloxicam ranges from 13 to 20 h, and it is suitable for once-daily dosing (24, 26). Meloxicam is extensively metabolized in the liver, primarily by polymorphic cytochrome P450 2C9 (CYP2C9) enzyme, and to a minor extent by CYP3A4 enzyme, to four pharmacologically inactive metabolites (6, 24). Only negligible amounts of the parent drug are found in urine and in feces (24). The effect of different genotypes on the pharmacokinetics of meloxicam is not known. Voriconazole is a triazole antifungal agent used both intravenously and orally to treat invasive fungal infections. Voriconazole undergoes extensive oxidative metabolism involving CYP enzymes CYP2C19, CYP2C9, and CYP3A4 (13). Voriconazole is also an inhibitor of CYP2C9, CYP3A4, and CYP2C19 catalyzed reactions both in vitro and in vivo (17, 21, 23, 25). Another triazole antifungal, itraconazole, is a potent * Corresponding author. Mailing address: Department of Pharmacology, Drug Development and Therapeutics, University of Turku, Itäinen Pitkäkatu 4B, FIN-20520, Finland. Phone: Fax: vilhyn@utu.fi. Published ahead of print on 17 November inhibitor of CYP3A4 (18, 28), but it is without effect on CYP2C9 in humans (15, 27). Because both voriconazole and itraconazole inhibit CYP enzymes involved in the metabolism of meloxicam, we hypothesized that they might interact with meloxicam, leading to its increased concentrations in plasma. Therefore, we found it important to investigate the effects of voriconazole and itraconazole on the pharmacokinetics of meloxicam. In addition, we wanted to study whether possibly changed plasma meloxicam concentrations are reflected in its effects on the generation of thromboxane B 2 (TxB 2 ) during whole blood clotting, which is a validated index of COX-1 activity ex vivo (3). MATERIALS AND METHODS Subjects. Twelve healthy male volunteers (age range, 19 to 39 years; weight range, 66 to 82 kg) participated in the study. Each subject was ascertained to be healthy as assessed by medical history, clinical examination, and routine laboratory tests. All subjects were nonsmokers and used no continuous medication. The subjects received both verbal and written information on the study, and written informed consent was obtained. The study protocol was approved by the Ethics Committee of the Hospital District of Southwest Finland as well as by the National Agency for Medicines of Finland. Study design. The study was carried out in a Latin-square, open-label, randomized, three-phase crossover design, with a washout period of 4 weeks between the phases. The volunteers were given either no pretreatment (control phase) or oral voriconazole (voriconazole phase) for 2 days or oral itraconazole (itraconazole phase) for 4 days in a randomized order. The dose of voriconazole (Vfend, 200-mg tablet; Pfizer/Heinrich Mack Nachf. GmbH & Co., Illertissen, Germany) was 400 mg every 12 h for 1 day and then 200 mg every 12 h for one additional day. The dose of itraconazole (Sporanox, 100-mg capsule; Janssen- Cilag, Latina, Italy) was 200 mg once a day at 8 a.m. for 4 days. Voriconazole and itraconazole were self-administered by subjects except for the last doses, which 587

2 588 HYNNINEN ET AL. ANTIMICROB. AGENTS CHEMOTHER. TABLE 1. Pharmacokinetic variables of meloxicam in 12 healthy subjects after a single oral dose of 15 mg of meloxicam without pretreatment (control) or after pretreatment with voriconazole or after pretreatment with itraconazole Parameter Mean SD % of control (95% CI) a Control Voriconazole Itraconazole AUC 0-72 (mg h/liter) (9.0 to 20.7) ( 17.7 to 6.4) C max (mg/liter) ( 0.14 to 0.23) ( 0.98 to 0.60) t 1/2 (h) (2.8 to 15.8) (3.1 to 16.1) T max (h) 4 (4 to 7) 6 (4 to 24) 24 (4 to 48) Free fraction (%) a The results are means the SD (or median with range for T max ). The percentage of the control value was calculated individually for each subject. The free fraction of plasma meloxicam was determined from plasma samples obtained 5 h after meloxicam ingestion., Significantly (P 0.01) different from control;, significantly (P 0.001) different from control. were administered by the study personnel. The intake of the premedication by the subjects was verified by use of mobile phone short message service. One hour after the last dose of voriconazole or itraconazole was ingested, all volunteers received a 15-mg oral dose of meloxicam (Mobic, 15-mg tablet; Boehringer Ingelheim, Ingelheim, Germany) at 9 a.m. with 150 ml of water. During all phases, the subjects fasted overnight before the administration of meloxicam and continued fasting until a standardized lunch was served 4 h after meloxicam ingestion. The subjects were forbidden to use any other medication for 14 days before and during the study and any drugs known to cause enzyme induction or inhibition for a period of 30 days prior the study. Caffeine, grapefruit juice, and alcohol containing beverages were not allowed during the study. Blood sampling and drug analysis. Venous blood samples were drawn from the cannulated forearm vein immediately before and at 1, 2, 3, 4, 5, 6, 7, 8, 10, 12, 24, 48, and 72 h after meloxicam administration. Plasma was separated within 30 min and stored at 70 C until analysis. To study the protein binding of meloxicam, protein-free ultrafiltrates were prepared with a Centrifree filter YM30 (Amicon, Beverly, MA) at room temperature from the plasma samples of each subject taken 5 h after meloxicam ingestion, in each of three phases. The free fraction of plasma meloxicam was prepared from 1 ml of plasma (taken 5 h after meloxicam ingestion) by centrifugation using a fixed-angle rotor at 1,200 g and 25 C for 30 min. There was no detectable binding of meloxicam to filter even when a low concentration (2.5 ng/ml) of meloxicam and a protein-free solution were used in the ultrafiltration. Determination of drug concentrations. Plasma concentrations of meloxicam were measured, as described earlier (14), using piroxicam as an internal standard and a Q-Trap liquid chromatography-tandem mass spectrometry system (Sciex Division of MDS, Toronto, Ontario, Canada). The mass spectrometer was operated in positive TurboIonSpray (Sciex Division of MDS) mode, and the transition of the [M H] precursor ion to a product ion was monitored. The ion transitions monitored were mass/charge ratio (m/z) to m/z for meloxicam and m/z to m/z 95.1 for the internal standard piroxicam. The interday coefficient of variation (CV) for total meloxicam was 10% at relevant concentrations (n 14). All plasma samples obtained after meloxicam ingestion were quantifiable for their total meloxicam and all samples taken at 5 h for their free (protein unbound) meloxicam concentration. All plasma ultrafiltrate samples (altogether 36 samples from plasma at 5 h) were analyzed in the same assay for their free meloxicam concentration. The meloxicam standard curve was linear over the concentration range used (standards 0.5, 1.0, 2.5, 5.0, 10, 25, and 50 ng/ml), and the intra-assay CV was 9.7% at 10 ng/ml (n 7). Voriconazole and itraconazole did not interfere with the determination of meloxicam. In the voriconazole assay, the internal standard (UK 54373) and 0.1 ml of perchloric acid were added to plasma samples (0.5 ml). The samples were then vortex mixed and centrifuged, and the supernatants were applied to Oasis MCX (30 mg, 1 ml) solid-phase extraction cartridges (Waters, Milford, MA), which had been conditioned with 1 ml of methanol, followed by 1 ml of water. The cartridges were then washed with 1 ml of 0.1 M hydrochloric acid and 1.0 ml of methanol-water (40:60 [vol/vol]). The cartridges were dried by a vacuum prior to the elution step with 1 ml of 5% ammonium hydroxide in methanol. The samples were evaporated to dryness under nitrogen and reconstituted in 0.1 ml of mobile phase (72.5% ammonium phosphate 40 mm [ph 9.0] and 27.5% acetonitrile). The chromatographic conditions were otherwise as described previously (9), but we used a higher ph (ph 9.0) of the mobile phase instead of ph 6.0. The plasma concentrations of itraconazole were quantified by high-performance liquid chromatography as described earlier (11). The interday CV for voriconazole was 2.8% at relevant concentrations (n 3). The interday CV for itraconazole was 6.1% at relevant concentrations (n 4). Genotyping. Blood samples were drawn and stored at 20 C. Genomic DNA was extracted from the leukocytes by using QiAmp DNA blood kit (Qiagen, Hilden, Germany). A TaqMan assay was used to determine the CYP2C9*2 and CYP2C9*3 alleles, as previously described (29). Alleles containing no *2 or *3 are named CYP2C9*1. After an internal DNA denaturation at 95 C for 10 min, the samples were run for 40 cycles at 95 C for 15 s and at 60 C for 1 min. Pharmacokinetics. The C max and T max for each subject were obtained directly from the plasma concentration data. The elimination rate constant (k el ) was determined by a linear regression analysis of the log-linear part of the plasma concentration-time curve. The t 1/2 was calculated by the equation t 1/2 ln 2/k el. The area under the plasma concentration-time curve was calculated from 0 to 72 h postdose (AUC 0-72 ) using linear trapezoidal rule for the rising phase of the plasma concentration-time curves and the logarithmic trapezoidal rule for the descending phase. All pharmacokinetic calculations were performed with Win- Nonlin pharmacokinetic program (version 4.1; Pharsight, Mountain View, CA). Pharmacodynamics. The pharmacodynamics of meloxicam were assessed by measuring the TxB 2, a stable metabolite of TxA 2, generation by platelets in spontaneously clotting whole blood (3). The decrease in TxB 2 generation is shown to reflect the degree of COX-1 inhibition. The decrease in TxB 2 generation was calculated by comparingtxb 2 concentration at different time points to the individual baseline value. On study days, blood samples for TxB 2 assay were drawn before, and at 5, 8, 12, 24, and 48 h after meloxicam administration. Blood samples were collected into glass tubes containing no anticoagulant and were immediately incubated for 1hat37 C to stimulate the TxB 2 production in platelets during coagulation. Next, serum was collected and centrifuged and stored at 70 C until assayed for TxB 2. Serum TxB 2 concentrations were determined by using an enzyme immunoassay kit (Amersham Thromboxane B 2 Enzymeimmunoassay Biotrak System; GE Healthcare, United Kingdom). The limit of detection was 10 ng/ml, and the interassay CV was 10%. Statistical analysis. The pharmacokinetic and pharmacodynamic variables were compared by use of analysis of variance for repeated measures, and a posteriori testing was performed by use of the Tukey test. The T max was analyzed with Friedman s test and Wilcoxon signed-rank test was used for pairwise comparisons. The correlation between voriconazole or itraconazole AUC and the ratio of meloxicam AUC in the voriconazole and in the itraconazole phase to the meloxicam AUC in the control phase, as well as the correlation between meloxicam C max or AUC values and the decrease in TxB 2 formation (AUC), was tested by using a Pearson correlation test. Statistical analyses were carried out by using statistical program SYSTAT for Windows (version 10.2; Systat Software, Richmond, CA). The chosen statistical significance level was P The results are presented as means the standard deviations (SD) in Table 1 and, for clarity, as the means the standard errors of the mean in the figures. The T max median with the range is shown in the table. The percent differences between treatments were calculated within subjects, and 95% confidence intervals (CIs) are given. RESULTS All 12 subjects completed the study according to the protocol. Mean plasma concentrations of meloxicam as a function of time in 12 volunteers during different phases of the study are illustrated in Fig. 1. Effect of voriconazole. Voriconazole increased the mean AUC 0-72 of meloxicam by 47% (range, 11 to 120%; P 0.001) compared to the control value and prolonged the mean t 1/2 of

3 VOL. 53, 2009 INTERACTIONS OF MELOXICAM WITH TWO AZOLE ANTIMYCOTICS 589 FIG. 1. Plasma concentrations (mean the standard error of the mean) of meloxicam after a single 15-mg oral dose in 12 healthy male subjects in control phase (E) or after pretreatment with voriconazole (f) or after pretreatment with itraconazole (F). The inset depicts the same data on a semilogarithmic scale. meloxicam by 51%, from 17.4 to 26.7 h (P 0.01). The mean C max and median T max of meloxicam were unaffected by voriconazole (Table 1). The increase of the AUC 0-72, as well as the prolongation of t 1/2 of meloxicam, was evident in all 12 subjects (Fig. 2). Effect of itraconazole. During the itraconazole phase, the mean AUC 0-72 of meloxicam was 63% (range, 31 to 90%; P 0.001), and its mean C max was 36% (range, 23 to 76%; P 0.001) of the respective control values (Table 1). The decrease of AUC 0-72 and C max was seen in all subjects (Fig. 2). The median T max of meloxicam was also reached significantly later (24 h in the itraconazole phase versus 4 h in the control phase; P 0.01) and t 1/2 was prolonged from 17.4 to 27 h (P 0.01) (Table 1). Four subjects had meloxicam T max values as late as 48 h after its ingestion, which made it impossible to determine meloxicam t 1/2 in these subjects, because the concentrations FIG. 2. Individual values for AUC, C max, t 1/2, and time to maximum plasma concentrations (T max ), after the administration of 15 mg of meloxicam either alone (control), after pretreatment with voriconazole (Voric), or after pretreatment with itraconazole (Itrac). Symbols: F, CYP2C9*1/*1 genotype; E, CYP2C91*/2* genotype;, CYP2C9*2/*2 genotype;, CYP2C9*1/*3 genotype. FIG. 3. Mean percent inhibition of TxB 2 generation from baseline by 15 mg of meloxicam either given alone (E) or after pretreatment with voriconazole (f) or after pretreatment with itraconazole (F). were measured for up to 72 h and there were no measurements between the 48- and 72-h time points. Protein binding. The plasma protein binding of meloxicam was 99.83% 0.05% (mean the SD), 99.83% 0.04%, and 99.82% 0.07% during the control, voriconazole, and itraconazole phases, respectively, measured from the plasma samples taken 5 h after meloxicam ingestion. Effects of genotype. Two subjects had the *1/*2, one had *2/*2, and one had *1/*3 genotype for CYP2C9; the others were homozygous wild-type *1/*1 for CYP2C9. The meloxicam AUC 0-72 of subjects with *1/*2, *2/*2, and *1/*3 genotype in the control phase were 33.1 mg h/liter (mean of two subjects), 26.5 mg h/liter, and 40.1 mg h/liter, respectively, and were comparable with the mean AUC 0-72 of meloxicam of wild-type *1/*1 subjects (32.0 mg h/liter) (Fig. 2). Voriconazole and itraconazole seemed to have similar effect on the meloxicam AUC 0-72, C max, and t 1/2 in the subjects with *1/*2, and *2/*2 genotype compared to wild-type subjects. The subject with *1/*3 genotype had the longest t 1/2 in the control phase. In addition, the strongest inhibitory effect on meloxicam t 1/2 by voriconazole and itraconazole was observed with this subject (Fig. 2). Pharmacodynamics. The absolute TxB 2 concentrations and percentage reduction from baseline TxB 2 concentrations at different time points are illustrated in Fig. 3. The baseline serum TxB 2 levels were similar before each treatment and , , and ng/ml (mean the SD) prior to the ingestion of meloxicam, during the control phase, and during the voriconazole phase and itraconazole phase, respectively. Meloxicam alone (control phase) inhibited the synthesis of TxB 2 from baseline significantly, i.e., at 5 h after ingestion by 30% (P 0.05), at8hby37%(p 0.05), at 12 h by 56% (P 0.001), at 24 h by 28% (P 0.05), and at 48 h by 25% (P 0.05). In the voriconazole phase, the corresponding values at the same time points were 37% (P 0.01), 46% (P 0.001), 49% (P 0.001), 31% (P 0.05), and 14% (P 0.05). In the itraconazole phase, meloxicam did not cause any statistically significant inhibition of TxB 2 generation before 48 h after meloxicam ingestion; at 48 h, 29% (P 0.05) decline from baseline was seen. There was a significant correlation between both meloxicam C max values (P 0.01, Pearson r 0.46) and meloxicam AUC 0-48 values (P 0.01; Pearson r 0.47) and

4 590 HYNNINEN ET AL. ANTIMICROB. AGENTS CHEMOTHER. the inhibition of TxB 2 synthesis. No difference was observed in the inhibition of TxB 2 synthesis between the control and the voriconazole phase at any time point, whereas the inhibition of TxB 2 synthesis was significantly greater, from 5 to 12 h after meloxicam ingestion in the control and in the voriconazole phase compared to the itraconazole phase. Plasma voriconazole and itraconazole. The mean C max, AUC (0- ), and t 1/2 of voriconazole were 2,390 ng/ml (range, 1,020 to 4,440 ng/ml), 30,700 ng h/ml (range, 8,100 to 123,000 ng h/ml), and 10 h (range, 6.0 to 26 h), respectively. The corresponding values for itraconazole were 385 ng/ml (range, 250 to 620 ng/ml), 11,200 ng h/ml (range, 6,790 to 18,700 ng h/ml), and 31 h (range, 15 to 49 h), respectively. The voriconazole AUC 0- values of subjects with CYP2C9*1/*2, *2/*2, and *1/*3 genotype were 20,400, 17,700, and 24,500 ng h/ml, respectively, and the itraconazole AUC 0- values of subjects with the *1/*2, *2/*2, and *1/*3 genotype were 17,100, 8,100, and 13,000 ng h/ml, respectively. There were no significant correlations between the pharmacokinetic variables of voriconazole and the extent of interaction between voriconazole and meloxicam or between the pharmacokinetic variables of itraconazole and the extent of interaction between itraconazole and meloxicam. Adverse effects. During the voriconazole pretreatment, 3 of the 12 subjects reported visual disturbances, including photophobia and altered color vision changes. No other clinically relevant adverse effects were observed or reported during the study. DISCUSSION Our results reveal that the interactions of meloxicam with two triazole antifungals, voriconazole and itraconazole, are opposite. Voriconazole significantly increased plasma concentrations of meloxicam, whereas itraconazole unexpectedly caused a notable decrease in the plasma concentrations of meloxicam. This was also associated with decreased pharmacodymic effects of meloxicam, as observed by weaker inhibition of TxB 2 synthesis during whole blood clotting, in the itraconazole phase compared to the control and voriconazole phases. The major primary metabolite of meloxicam is 5 -hydroxymethyl meloxicam, which is further oxidized to 5 -carboxy meloxicam. The oxidative cleavage of the benzothiazine ring of meloxicam creates two additional metabolites (6, 24). Based on in vitro studies, formation of 5 -hydoxymethyl metabolite is catalyzed mainly by CYP2C9 (80%) and CYP3A4 appears to play a minor role (20%) (6). There are only a few studies addressing the influence of other drugs on the pharmacokinetics of meloxicam. Cholestyramine has increased the clearance of intravenous meloxicam by 50%, suggesting that meloxicam undergoes enterohepatic circulation (4), whereas antacids, aspirin, furosemide, and cimetidine have not affected its pharmacokinetics (5, 16). In the present study, voriconazole, an inhibitor of both CYP2C9 and CYP3A4 (17), increased the meloxicam AUC 0-72 by 1.5-fold and prolonged the meloxicam t 1/2 from 17 to 27 h but had no effect on the C max or T max of meloxicam. Together, these observations suggest that the voriconazole-meloxicam interaction occurs principally during the meloxicam elimination phase and is likely due to the inhibition of CYP2C9 and, to a lesser extent, CYP3A4-mediated metabolism of meloxicam by voriconazole. The individual with the CYP2C9*1/*3 genotype, thus carrying low basal CYP2C9 activity, had the strongest effect by both the CYP3A4-inhibitory azoles on the meloxicam t 1/2, suggesting a bigger role of CYP3A4 in the metabolism of meloxicam in individuals possessing impaired CYP2C9 activity. In clinical situations, the voriconazole-meloxicam interaction can be stronger than that seen in the present study, where, for ethical reasons (to minimize the common visual disturbances), administration of voriconazole to healthy volunteers was not continued after meloxicam ingestion. Since the halflife of voriconazole was about 10 to 11 h, its plasma concentrations were from 12 h onward lower than during its normal twice-daily clinical use. This may have somewhat underestimated the extent of the interaction during the elimination phase of meloxicam, since meloxicam is eliminated more slowly than voriconazole. The changes in the pharmacokinetics of meloxicam during the itraconazole phase were surprising. The plasma concentrations of meloxicam were clearly lower during the first 24 h after the ingestion of meloxicam in the itraconazole phase compared to the control phase (Fig. 1). The AUC 0-72 and C max were considerably decreased and the T max was greatly prolonged by itraconazole. The plasma protein binding of meloxicam was very high (99.8%), but there was no difference in its percentage binding between different phases of the study. Thus, a displacement of meloxicam from plasma protein by itraconazole does not explain the observed interaction. The present findings strongly suggest that itraconazole decreased the exposure to meloxicam by impairing its gastrointestinal absorption. However, the exact mechanism of impaired absorption remains to be studied. Itraconazole could have inhibited some transport system in the gut wall, which is needed for the absorption of meloxicam. However, this theory is highly speculative, because there is no evidence for the participation of drug transporters in the absorption of meloxicam. Due to the delayed absorption of meloxicam, our blood sampling period appeared to be too short for the reliable determination of the elimination rate constant (k el ) in four subjects in the itraconazole phase. Accordingly, because AUC values could not be extrapolated reliably to infinity in all 12 subjects in the itraconazole phase, the AUC 0-72 was used for comparison of AUC values between the phases. It is therefore possible that the actual effect of itraconazole on the total exposition to meloxicam is somewhat smaller than that calculated in the present study. In the voriconazole phase, the magnitude of interaction between meloxicam and voriconazole remained essentially the same when checked by using the AUC values extrapolated to infinity (data not shown). The t 1/2 of meloxicam, calculated from eight subjects, was significantly prolonged by itraconazole. However, the decline of the plasma meloxicam concentration in the itraconazole phase seems to be determined by its rate of absorption. Therefore, if the absorption of meloxicam is prolonged as in the present study, its apparent t 1/2 also is prolonged (flip-flop phenomenon), and this can occur without a change in the rate of elimination. However, the CYP3A4 inhibition by itraconazole as a partial cause for prolonged t 1/2 cannot be ruled out.

5 VOL. 53, 2009 INTERACTIONS OF MELOXICAM WITH TWO AZOLE ANTIMYCOTICS 591 In our study, voriconazole dosing was based on a previous study indicating that the steady-state concentration of voriconazole can be achieved in 2 days, using a loading dose of 400 mg twice daily on the first day followed by 200 mg twice daily on the second day (20). In contrast, 4-day pretreatment with itraconazole was too short to achieve steady-state concentrations of itraconazole, which are reached only 10 to 14 days after the beginning of treatment (12). Four-day pretreatment was selected, because it is not desirable to expose healthy volunteers to a 2-week pretreatment of itraconazole and also because 4-day pretreatment with similar doses of itraconazole as used in our study has been shown to produce a strong inhibition of CYP3A4 (1). Typical adverse effects of NSAIDs include gastrointestinal damage, inhibition of platelet function, and renal impairment. The first two effects are attributed mainly to the inhibition of COX-1, whereas renal impairment is also a prevalent side effect of COX-2-selective NSAIDs. Meloxicam is a preferential COX-2 inhibitor, which also dose dependently inhibits COX-1 (2, 19, 22). In the present study, a 47% increase in the exposure to meloxicam by voriconazole was not associated with increased COX-1 inhibition. The maximum decline of the synthesis of TxB 2 was 56 and 49% in the control and voriconazole phases, respectively, which is comparable with other studies, in which TxB 2 synthesis was inhibited 35 to 66% by 15 mg of meloxicam (7, 19, 22). Accordingly, it is unlikely that a shortterm concomitant use of voriconazole or other CYP2C9 inhibitors with meloxicam would increase its risk for gastrointestinal adverse effects or bleeding. However, increased clinical alertness is recommended with the long-term coadministration of meloxicam with CYP2C9 inhibitors, especially in the elderly or in patients with an increased risk for NSAID-related adverse effects. In the itraconazole phase, the decreased meloxicam concentrations were associated with a clearly reduced pharmacodynamic effect. No reduction on TxB 2 synthesis was observed by 24 h after ingestion of 15 mg of meloxicam. At 48 h, TxB 2 synthesis was inhibited by 29%, which corresponds to the degree of inhibition reported previously 5 h after ingestion of 7.5 mg of meloxicam, without itraconazole (2). These findings suggest that the interaction between itraconazole and meloxicam is clinically significant and probably will cause a reduced meloxicam effect, at least in the short term. Because both the AUC 0-72 and the C max of meloxicam correlated similarly with the degree of inhibition of TxB 2 formation, it is difficult to evaluate whether it is lowered exposure or lowered the peak concentration that is relevant for the interaction between meloxicam and itraconazole. In conclusion, voriconazole moderately increases the plasma concentrations of meloxicam, probably by inhibiting its metabolism via CYP2C9. In contrast to voriconazole, itraconazole causes a substantial decrease in plasma concentrations, especially in the C max, of meloxicam, probably by impairing its gastrointestinal absorption. The clinical efficacy of meloxicam can be reduced, at least in the short term, when given during itraconazole treatment, but the efficacy is unchanged or increased when given during voriconazole treatment. Further studies on the effect of itraconazole on the pharmacokinetics and pharmacodynamics of meloxicam during their longer concomitant administration are needed. ACKNOWLEDGMENTS This study was supported by Turku University Hospital research funds EVO (K.L.) and EVO (K.T.O.) and by Swedish Research Council grant 3902 (L.B.). We thank Elina Kahra for skillful assistance in organizing the study. REFERENCES 1. Backman, J. T., K. T. Kivisto, K. T. Olkkola, and P. J. Neuvonen The area under the plasma concentration-time curve for oral midazolam is 400- fold larger during treatment with itraconazole than with rifampicin. Eur. J. Clin. Pharmacol. 54: Blain, H., C. Boileau, F. Lapicque, E. Nedelec, D. Loeuille, C. Guillaume, A. Gaucher, C. Jeandel, P. Netter, and J. Y. Jouzeau Limitation of the in vitro whole blood assay for predicting the COX selectivity of NSAIDs in clinical use. Br. J. Clin. Pharmacol. 53: Brideau, C., S. Kargman, S. Liu, A. L. Dallob, E. W. Ehrich, I. W. Rodger, and C. C. Chan A human whole blood assay for clinical evaluation of biochemical efficacy of cyclooxygenase inhibitors. Inflamm. Res. 45: Busch, U., G. Heinzel, and H. Narjes The effect of cholestyramine on the pharmacokinetics of meloxicam, a new non-steroidal anti-inflammatory drug (NSAID), in man. Eur. J. Clin. Pharmacol. 48: Busch, U., G. Heinzel, H. Narjes, and G. Nehmiz Interaction of meloxicam with cimetidine, Maalox, or aspirin. J. Clin. Pharmacol. 36: Chesne, C., C. Guyomard, A. Guillouzo, J. Schmid, E. Ludwig, and T. Sauter Metabolism of Meloxicam in human liver involves cytochromes P4502C9 and 3A4. Xenobiotica 28: de Meijer, A., H. Vollaard, M. de Metz, B. Verbruggen, C. Thomas, and I. Novakova Meloxicam, 15 mg/day, spares platelet function in healthy volunteers. Clin. Pharmacol. Ther. 66: Engelhardt, G., R. Bogel, C. Schnitzer, and R. Utzmann Meloxicam: influence on arachidonic acid metabolism. 1. In vitro findings. Biochem. Pharmacol. 51: Gage, R., and D. A. Stopher A rapid HPLC assay for voriconazole in human plasma. J. Pharm. Biomed. Anal. 17: Gates, B. J., T. T. Nguyen, S. M. Setter, and N. M. Davies Meloxicam: a reappraisal of pharmacokinetics, efficacy and safety. Expert Opin. Pharmacother. 6: Gubbins, P. O., B. J. Gurley, and J. Bowman Rapid and sensitive high performance liquid chromatographic method for the determination of itraconazole and its hydroxy-metabolite in human serum. J. Pharm. Biomed. Anal. 16: Hardin, T. C., J. R. Graybill, R. Fetchick, R. Woestenborghs, M. G. Rinaldi, and J. G. Kuhn Pharmacokinetics of itraconazole following oral administration to normal volunteers. Antimicrob. Agents Chemother. 32: Hyland, R., B. C. Jones, and D. A. Smith Identification of the cytochrome P450 enzymes involved in the N-oxidation of voriconazole. Drug Metab. Dispos. 31: Ji, H. Y., H. W. Lee, Y. H. Kim, D. W. Jeong, and H. S. Lee Simultaneous determination of piroxicam, meloxicam, and tenoxicam in human plasma by liquid chromatography with tandem mass spectrometry. J. Chromatogr. B Analyt. Technol. Biomed. Life. Sci. 826: Kaukonen, K. M., K. T. Olkkola, and P. J. Neuvonen Fluconazole but not itraconazole decreases the metabolism of losartan to E Eur. J. Clin. Pharmacol. 53: Muller, F. O., R. Schall, A. C. de Vaal, G. Groenewoud, H. K. Hundt, and M. V. Middle Influence of meloxicam on furosemide pharmacokinetics and pharmacodynamics in healthy volunteers. Eur. J. Clin. Pharmacol. 48: Niwa, T., T. Shiraga, and A. Takagi Effect of antifungal drugs on cytochrome P450 (CYP) 2C9, CYP2C19, and CYP3A4 activities in human liver microsomes. Biol. Pharm. Bull. 28: Olkkola, K. T., J. Ahonen, and P. J. Neuvonen The effects of the systemic antimycotics, itraconazole and fluconazole, on the pharmacokinetics and pharmacodynamics of intravenous and oral midazolam. Anesth. Analg. 82: Panara, M. R., G. Renda, M. G. Sciulli, G. Santini, M. Di Giamberardino, M. T. Rotondo, S. Tacconelli, F. Seta, C. Patrono, and P. Patrignani Dose-dependent inhibition of platelet cyclooxygenase-1 and monocyte cyclooxygenase-2 by meloxicam in healthy subjects. J. Pharmacol. Exp. Ther. 290: Purkins, L., N. Wood, K. Greenhalgh, M. D. Eve, S. D. Oliver, and D. Nichols The pharmacokinetics and safety of intravenous voriconazole: a novel wide-spectrum antifungal agent. Br. J. Clin. Pharmacol. 56(Suppl. 1): Purkins, L., N. Wood, D. Kleinermans, and D. Nichols Voriconazole potentiates warfarin-induced prothrombin time prolongation. Br. J. Clin. Pharmacol. 56(Suppl. 1): Rinder, H. M., J. B. Tracey, M. Souhrada, C. Wang, R. P. Gagnier, and C. C. Wood Effects of meloxicam on platelet function in healthy adults: a

6 592 HYNNINEN ET AL. ANTIMICROB. AGENTS CHEMOTHER. randomized, double-blind, placebo-controlled trial. J. Clin. Pharmacol. 42: Saari, T. I., K. Laine, K. Leino, M. Valtonen, P. J. Neuvonen, and K. T. Olkkola Effect of voriconazole on the pharmacokinetics and pharmacodynamics of intravenous and oral midazolam. Clin. Pharmacol. Ther. 79: Schmid, J., U. Busch, G. Heinzel, G. Bozler, S. Kaschke, and M. Kummer Pharmacokinetics and metabolic pattern after intravenous infusion and oral administration to healthy subjects. Drug Metab. Dispos. 23: Theuretzbacher, U., F. Ihle, and H. Derendorf Pharmacokinetic/pharmacodynamic profile of voriconazole. Clin. Pharmacokinet. 45: Turck, D., U. Busch, G. Heinzel, and H. Narjes Clinical pharmacokinetics of meloxicam. Arzneimittelforschung 47: Venkatakrishnan, K., L. L. von Moltke, and D. J. Greenblatt Effects of the antifungal agents on oxidative drug metabolism: clinical relevance. Clin. Pharmacokinet. 38: von Moltke, L. L., D. J. Greenblatt, J. Schmider, S. X. Duan, C. E. Wright, J. S. Harmatz, and R. I. Shader Midazolam hydroxylation by human liver microsomes in vitro: inhibition by fluoxetine, norfluoxetine, and by azole antifungal agents. J. Clin. Pharmacol. 36: Yasar, U., S. Lundgren, E. Eliasson, A. Bennet, B. Wiman, U. de Faire, and A. Rane Linkage between the CYP2C8 and CYP2C9 genetic polymorphisms. Biochem. Biophys. Res. Commun. 299:25 28.