Firocoxib, 3-(cyclopropylmethoxy)-4-(4-(methylsulfonyl)phenyl)-5,5-dimethylfuranone, is a coxib-class NSAID approved for use in horses to control pain and inflammation associated with osteoarthritis. In general, NSAIDs are indiscriminate inhibitors of COX, an enzyme involved in the synthesis of prostaglandins that has at least 2 major isoforms, COX-1 and COX-2. In contrast, coxib-class drugs selectively inhibit the COX-2 isozyme. Inhibition of the COX-2 isozyme prevents the production of prostaglandins that function primarily as inflammatory mediators; this inhibition leads to the therapeutic effects of NSAIDs.1–3 However, inhibition of COX-1 may lead to undesirable effects, such as gastric irritation and blood thinning, because COX-1 is involved in processes designed to protect gastric and renal function (such as mucosal blood flow) and it aids in hemostasis.4,5 Development of the new coxib-class NSAIDs began in the 1990s after the discovery of a mitogen-inducible form of COX (also known as prostaglandin G/H synthase).6 Two groups of researchers each discovered that this second COX gene, on the basis of its pattern of regulation and expression, appeared to be the sole isozyme that produced prostaglandins responsible for potentiating inflammatory processes.7 The recognition that inhibition of COX-2 may be sufficient to achieve the anti-inflammatory benefits of NSAIDs and that the adverse effects commonly associated with NSAIDs were attributable, at least in part, to the indiscriminate inhibition of COX-1 resulted in the development of new drugs that would selectively inhibit COX-2. This class of drugs (ie, coxibs), which includes firocoxib, has now evolved to second-generation drugs8,9 that are highly selective for COX-2, thus sparing COX-1 at therapeutic concentrations and potentially reducing the incidence of adverse effects associated with COX-1 inhibition, such as gastrointestinal irritation.10–12,a This hypothesis was supported by studies13,14 involving laboratory animals with experimentally induced disease and by human clinical trials,15–17 in which the gastrointestinal toxicity of coxibs and traditional NSAIDs were compared. It has also been suggested that inhibition of both COX-1 and COX-2, and not inhibition of COX-1 alone, is required to cause gastric irritation and damage.5 It is inferred from the outcomes of these studies that there will be substantially fewer clinically important gastrointestinal events in horses.
On the other hand, when there are preexisting conditions (such as ulcers), the use of COX-2–selective inhibitors may not be as advantageous. Studies have indicated a protective or housekeeping function of COX-2 expression with regard to the healing process of ulcers. Suppression of this constitutive COX-2 expression by selective inhibitors may delay gastric healing via reductions in gastric blood flow and suppression of the mucosal production of prostaglandin E2. Additionally, constitutive expression of COX-2 has also been identified in a variety of noninflammatory tissues, such as the kidneys and bones.18,19 A third COX isozyme (ie, COX-3), which is derived from the COX-1 gene and found mainly in the cerebral cortex, is also inhibited by some NSAIDs and may represent another mode of action, although COX activity of this isozyme is evident only after N-linked glycosylation.20
The metabolism and pharmacokinetics of firocoxib after administration of a single dose to horses have been reported.21 Similar to most NSAIDs, firocoxib is rapidly and nearly completely absorbed after oral administration in horses. Total body clearance in horses is relatively low and similar in magnitude to that for other NSAIDs, such as meloxicam, flunixin meglumine, phenylbutazone, and tolfenamic acid.22–25 The main elimination pathway is hepatic via dealkylation and glucuronidation of firocoxib; hydroxylation is a minor metabolic pathway. Both the parent drug and metabolites have been detected in the urine. Firocoxib is widely distributed in tissues, partially as a result of its physicochemical properties, such as high lipophilicity and low ionizability at physiologic pH. The volume of distribution is much higher than that for many NSAIDs, which results in a mean ± SD t1/2 of 33.8 ± 11.2 hours after IV administraion21–25 that is longer than the value determined for other NSAIDs. Protein binding is generally high for NSAIDs, and > 97% of firocoxib is bound to plasma proteins at concentrations up to 1.0 μg/mL. Dose proportionality of pharmacokinetic variables was established over the range of 0.5 to 2 times the recommended dosage of 0.1 mg/kg. Because of the high potency and long half-life, once-daily administration is sufficient for prolonged treatment of animals with chronic pain and inflammation. Single-dose pharmacokinetic parameters and variables have been determined.21
Associations that govern equine performance and racing events have differing rules regarding the use of NSAIDs preceding competition. Humane consideration of athletic horses requires medical treatment for injury or illness on the basis that persistent untreated pain can lead to hormonal, nervous, and psychologic abnormalities in animals.26 To allow for treatment of horses prior to racing or performance events, withdrawal periods or allowable drug concentrations for each regulated drug may be set by the various associations, although in some equine sports (such as endurance racing), no detectable concentrations of a drug are allowed. As part of this drug-monitoring process, analytic methods must be available to determine drug concentrations in plasma and urine. Because analytic methods are often highly sensitive, physiologically relevant concentrations of NSAIDs must be determined to set threshold drug concentrations in plasma and urine that are to be used for drug monitoring, instead of relying on the detection limits of the analytic methods.27,28 Because urine may be easier to obtain than plasma, it is important to determine the relationship between drug concentrations in urine and plasma after a single dose and at a steady state. One of the objectives of the study reported here was to gain an understanding of the pharmacokinetics after administration of multiple doses of firocoxib in horses by determining plasma and urine concentrations of firocoxib. We also wanted to determine whether plasma concentrations after administration of multiple doses could be predicted from single-dose pharmacokinetic parameters and whether the pharmacokinetic behavior was governed via first-order processes at a steady state.
Materials and Methods
Animals—Eighteen horses were used in 2 experiments. For experiment 1, 6 nonpregnant female horses (5 Paint horses and 1 Quarter Horse) between 2 and 4 years of age and weighing between 445 and 505 kg were used. Each horse was housed separately in a stall in an environmentally controlled building. Horses were fed oats and alfalfa hay twice daily, and tap water was available ad libitum. For experiment 2, 12 healthy male and female horses (predominantly Quarter Horse breed) > 2 years old and weighing between 373 and 447 kg were used. These horses were housed on pasture, except they were housed separately in pens for treatments and blood collections. Horses were fed commercial feedb and grass hay twice daily, and tap water was available ad libitum.
All horses were acclimated to their surroundings for at least 7 days prior to initiation of the experiments. No concurrent medications were administered during the course of the study. Horses were cared for in compliance with procedures approved by the Merial Institutional Animal Care and Use Committee and were observed at least once daily to monitor for health problems.
Study design—The study consisted of 2 experiments. The first experiment involved oral administration of firocoxib. Each of the 6 horses was orally administered firocoxibc (0.1 mg/kg) as a 0.82% (wt/wt) paste once daily for 12 consecutive days. First day of firocoxib administration was designated as day 0. Firocoxib was administered before the morning feeding. Forage and oats were withheld until after paste administration to ensure no complications because of feed boluses remaining in a horse's mouth. No further restrictions on feeding were applied before or after administration.
Blood samples (approx 10 mL) were collected from a jugular vein of each horse into heparinized tubes. On days 0 and 11, blood samples were collected before and 0.25, 1, 2, 4, 6, 12, and 24 hours after the firocoxib treatment. Blood samples were also collected on days 1 through 10 before each daily treatment. Samples were collected immediately before firocoxib treatment on days 7, 8, and 11 to confirm that steady state was achieved. Blood samples were also collected at approximately 32, 48, 72, and 96 hours after administration of the final dose. Plasma was separated from blood samples and stored frozen at approximately −80°C until analysis.
Urine samples were collected via a urinary catheter inserted aseptically into the bladder and connected to a urine collection bag. The catheter and urine collection bag were maintained throughout the collection period. This allowed for accurate determination of urine volumes while avoiding the necessity for multiple catheterization procedures. Urine samples were collected from each horse before treatment on day 0; for an interval of approximately 2 hours immediately before treatment on days 1, 7, 8, and 11; and over the collection intervals of 1 to 3, 3 to 5, 5 to 7, 11 to 13, 23 to 25, 31 to 33, 47 to 49, 71 to 73, and 95 to 97 hours after the final administration. Plasma samples were obtained at approximately the midpoint of the urine collection intervals. Each urine sample was mixed thoroughly to ensure homogeneity and stored frozen at approximately −80°C until analysis.
For experiment 2, each of the 12 horses received IV administration of firocoxib (0.2 mg/kg) as a 2% (wt/vol) solution. Firocoxib was administered immediately before the morning feeding, and there were no restrictions on feeding prior to or after treatment. Treatments were administered once daily in a jugular vein on days 0 to 8. Blood samples (approx 10 mL) were collected into heparinized tubes from the contralateral jugular vein on days 0, 1, 6, 7, and 8 at approximately 5 and 30 minutes and 1, 2, 4, 6, 8, and 24 hours after treatment and on days 2 through 5 at approximately 24 hours after treatment. Blood samples were also collected at approximately 36, 48, and 120 hours after the final treatment. Plasma was separated from blood samples and stored frozen at approximately −80°C until analysis.
Analytic procedures—Plasma samples obtained from the horses after IV administration were analyzed by use of a validated method29 that involved solid-phase extraction, liquid chromatography, and UV detection, whereas urine and plasma samples obtained from the horses after oral administration were analyzed by use of a liquid chromatography–mass spectrometry–mass spectrometry method.30 At least 2 standard calibration curves ranging from 1 to 3,000 ng/mL (plasma) or 5 to 3,000 ng/mL (urine) for firocoxib extracted in an identical manner to the procedures used for the samples were included with each analytic set. A minimum of 6 quality-control samples at 3 concentrations and 2 control samples of plasma (or urine) were also analyzed with each set. Firocoxib peak areas from the fortified matrix standards were used to generate a 1/x weighted regression for calculating the quality-control and unknown concentrations of firocoxib in plasma or urine samples.
Pharmacokinetic analysis—Results from single-dose experiments were used for comparison and prediction of multiple-dose concentrations and are described elsewhere.21 Pharmacokinetic parameters for the multiple-dose experiments were calculated with commercial softwared by use of a noncompartmental model–independent method. The Cmax, time to Cmax, and Cmin for each dosing interval were obtained directly from the plasma concentration data. The value for λz was estimated via linear regression of the logarithmic plasma concentration-versus-time curve. The plasma t1/2was calculated by (ln 2)/λz. The notation AUC(n,τ) indicated that the AUC was calculated over the nth dosing interval from time 0 (ie, time of administration of the nth dose) to the time τ after dose n (τ = 24 hours). The AUCs were calculated by use of the linear-logarithmic trapezoidal method. Mean concentration at steady state after oral administration of multiple doses was calculated by use of (AUC[n,τ])/τ, where n equals 12 corresponds to the last dose. The theoretic accumulation index based on linear pharmacokinetics was calculated by use of the equation (1 − exp[−λz • τ])−1, where exp is the exponential function, and the value for λz was obtained from a single-dose study. For oral administration of multiple doses, the experimental accumulation index was determined by use of the equation (AUC[12,τ])/(AUC[1,τ]), where 12 and 1 corresponded to the last dose and first dose, respectively.
In the experience of one of the investigators (VJK), the plasma concentration–time curves were highly variable during the absorption phase for the first 4 hours after oral administration of a single dose of firocoxib. The shape of the plasma concentration–time curves also varied greatly among animals. For these reasons, use of a compartmental model that would adequately fit the profile of all of the horses while making assumptions about absorption kinetics was not practical. Therefore, a nonparametric superposition of plasma concentrations after oral administration of a single dose was chosen to predict the plasma concentrations after oral administration of multiple doses. A superposition for the plasma concentration–time curve of each animal in the single-dose study21 was performed, and the predicted multiple-dose curves were then averaged. For each horse, the predicted curve was calculated by use of the following equation:
To simulate the data after IV administration of multiple doses, an equation was obtained from fitting the plasma concentrations after a single dose administered IV by use of an open 3-compartment model with bolus IV input. The 3-compartment model was defined by use of the following triexponential equation31:
The amount of firocoxib in each urine sample was determined by multiplying the measured concentration by the total volume collected. Excretion rate was the amount of firocoxib in each urine sample divided by the collection interval (approx 2 hours). Renal clearance was then determined for each steady-state urine sample by dividing the excretion rate by the plasma concentration determined at the midpoint of the collection interval.21 Renal clearance was standardized on the basis of the body weight of the horse and averaged over each time point for that horse.
Statistical analysis—Arithmetic means and SDs were determined. Data were analyzed for a normal distribution, and a paired Student t test was then performed with commercial statistical softwaree to compare trough plasma concentrations at steady state. Mean and SD of the t1/2 in plasma, Cmax, time to Cmax, Cmin, mean concentration at steady state, AUC(n,τ) for the first dose and at steady state, experimental accumulation index, renal and total clearance, volume of distribution, urine-to-plasma ratio, and half-life in urine were determined.
Results
Animals—All horses appeared to be healthy during the course of the study, and no treatment-related health problems were evident. There were no circumstances that adversely affected the quality or integrity of the data.
Oral administration—Limit of detection of the analytic method was 1 ng/mL in equine urine and 0.25 ng/mL in equine plasma. The LOQ of the analytic method was 5 ng/mL in equine urine and 1 ng/mL in equine plasma. Plasma quality-control samples at concentrations of 3, 250, and 2,500 ng/mL were included with each set; mean accuracy was between 91% and 103% (6 replicates), and coefficients of variation were ≤ 9%. Concentrations in all plasma samples were higher than the LOQ at all sampling times and ranged from 2.8 to 217 ng/mL. Mean accuracy for urine quality-control samples was between 90% and 102% (4 replicates), and coefficients of variation were ≤ 11%. Concentrations in urine samples were higher than the LOQ for all samples (range, 6 to 67 ng/mL), except for a single sample collected at 96 hours after administration of the final dose. The R2 of each standard curve was > 0.99, and the accuracy of each standard was between 85% and 115%.
IV administration—The limit of detection and LOQ of the analytic method were 10 ng/mL and 25 ng/mL, respectively. Concentrations in plasma samples ranged from 41 to 677 ng/mL, except for a few samples that had concentrations less than the LOQ of the analytic method. Several quality-control samples (concentrations of 50 or 100 ng/mL) were included with each set. Accuracy of plasma quality-control samples ranged between 85% and 110%, and coefficients of variation were < 10%. The R2 of each standard curve was > 0.99.
Pharmacokinetics—The plasma pharmacokinetic parameters from the multiple-dose experiments conducted here and single-dose experiments reported elsewhere21 were summarized (Table 1). A semilogarithmic plot was generated of the mean plasma concentration-versus-time and urine concentration-versus-time curves after oral administration of 12 consecutive daily doses (Figure 1). Steady state was achieved by the seventh daily dose, as predicted for consecutive daily oral administration of a drug with a mean ± SD single-dose plasma t1/2 of 29.6 ± 7.5 hours. Plasma trough concentrations did not differ significantly (P > 0.05) in samples obtained immediately before dosing on days 7, 8, and 11 of the experiment, which confirmed that a steady state had been reached by day 7. Mean plasma Cmin at steady state was 104 ± 33 ng/mL.
Mean ± SD values for pharmacokinetic parameters for a single dose of firocoxib administered to horses in another study,21 multiple doses of firocoxib (0.2 mg/kg) administered IV daily to 12 horses for 9 consecutive days in the study reported here, and multiple doses of firocoxib (0.1 mg/kg) administered orally daily to 6 horses for 12 consecutive days in the study reported here.
| Parameter | Single dose* | Multiple dose IV | Multiple dose orally | |||
|---|---|---|---|---|---|---|
| IV | Oral | After first dose | After last dose | After first dose | After last dose | |
| Plasma t1/2(h) | 33.8 ± 11.2 | 29.6 ± 7.5 | ND | 44.2 ± 21.6 | ND | 36.5 ± 9.5 |
| Tmax (h) | — | 3.90 ± 4.40 | — | — | 7.80 ± 4.80 | 0.79 ± 0.70 |
| C3(ng/mL) | 239 ± 65 | NA | 310 ± 56 | 523 ± 126 | NA | NA |
| Cmax (ng/mL) | — | 75.0 ± 33.0 | — | — | 45.0 ± 11.3 | 173.0 ± 44.0 |
| Cmin (ng/mL) | — | — | 55.4 ± 9.3 | 181.0 ± 74.0 | 30.2 ± 5.6 | 104.0 ± 33.0 |
| Css mean(ng/mL) | NA | NA | NA | 229 ± 73 | NA | 130 ± 36 |
| AUC (n,τ) (h•μg)/mL | 1.22 ± 0.26 | 0.96 ± 0.26 | 2.00 ± 0.30 | 5.50 ± 1.70 | 0.83 ± 0.17 | 3.12 ± 0.86 |
| Accumulation index | NA | NA | 2.8 ± 1.1† | 3.8 ± 0.7† | ||
| Bioavailability (%) | — | 79 ± 31 | — | — | — | — |
| Clearancess (mL/h/kg) | 36.7 ± 13.3 | — | — | 40.5 ± 14.7 | — | — |
| Vdss (L/kg) | 1.7 ± 0.4 | — | — | 2.3 ± 0.7 | — | — |
Values determined in another study,21 except for Vdss and C0, which were calculated from unpublished experiments performed at Merial Ltd by LTL.
Mean ± SD expected accumulation index (assuming linear kinetics) is 2.7 ± 0.6.
Tmax = Time to Cmax. C0= Initial concentration extrapolated back to time Ofor a dose administered IV. Css mean = Mean plasma concentration over the dosing interval at steady state. AUC (n,τ) = The AUC calculated by use of the linear-logarithmic trapezoidal method calculated over the nth dosing interval from time 0 (ie, time of administration of the nth dose) to the time τ after dose n (τ = 24 hours); the AUCs were calculated after administration on day 0 (n = 1) or day 11 (12). Accumulation index was determined at steady state by use of the equation (AUC[12,τ])/(AUC[1,τ]), where 12 and 1 corresponded to the last dose and first dose, respectively. Clearancess = Total body clearance estimated at steady state. Vdss = Volume of distribution at steady state. NA= Not applicable. ND = Not determined. — =Not reported.
Semilogarithmic graph of the mean ± SD firocoxib concentrations for plasma (white diamonds) and urine (black squares) samples obtained before, during, and after once-daily oral administration of 12 doses of firocoxib (0.1 mg/kg) as a paste formulation to 6 horses. Day 0 was the day of administration of the first dose. Urine concentrations are shown at the midpoint of the approximately 2-hour collection interval.
Citation: American Journal of Veterinary Research 69, 11; 10.2460/ajvr.69.11.1399
Mean ± SD plasma Cmax was 45.0 ± 11.3 ng/mL, which was reached 7.80 ± 4.80 hours after the first oral administration. Absorption was slow after the first orally administered dose (Cmax of 173 ± 44 ng/mL was reached 0.79 ± 0.70 hours after administration), compared with absorption after the last dose. Mean plasma t1/2 after the final dose was 36.5 ± 9.5 hours.
Mean ± SD plasma AUC values for the first (AUC[1,τ]) and last (AUC[12,τ]) oral dosing interval were 0.83 ± 0.17 (h • μg)/mL and 3.12 ± 0.86 (h • μg)/mL, respectively, which led to a mean accumulation index of 3.8 ± 0.7. This ratio was slightly higher than the ratio of 2.70 ± 0.60 predicted by use of the plasma t1/2 for single-dose oral administration.21 Mean plasma concentration of firocoxib at steady state was 130 ± 36 ng/mL. A simulated plasma concentration-versustime curve for oral administration of multiple doses was generated by use of data from the nonparametric superposition of the single-dose oral administration data reported elsewhere21 (Figure 2). There was good agreement between simulated and actual steady-state concentrations, which indicated linear pharmacokinetic behavior.
Semilogarithmic graph of the mean ± SD predicted firocoxib plasma concentration–time profile for 12 consecutive daily doses (0.1 mg/kg) of firocoxib administered orally (line) and the plasma concentrations obtained after daily oral administration for 12 days for the 6 horses in the study reported here (white squares and error bars). Day 0 was the day of administration of the first dose. All predicted plasma concentrations were increased by 1 SD to achieve agreement with the experimental plasma concentrations at steady state. The increase was necessary because of the higher-than-expected accumulation ratio.
Citation: American Journal of Veterinary Research 69, 11; 10.2460/ajvr.69.11.1399
The firocoxib concentration in urine samples decreased in parallel to the firocoxib concentrations in the plasma samples. Maximum concentrations in urine were detected during the first 1 to 3 hours after the final dose. Concentrations in urine then decreased exponentially and in parallel to concentrations in plasma, with a mean ± SD t1/2 of 35.3 ± 7.8 hours after the final dose. The concentration of firocoxib in urine was less than that in plasma, and the ratio of the concentration in urine to the concentration in plasma was approximately 0.35 at all time points. Mean renal clearance was 0.26 ± 0.09 mL/kg/h.
Plasma pharmacokinetic parameters were also determined for horses receiving 9 consecutive doses of a firocoxib solution IV. Mean ± SD AUC(1,τ) was 2.0 ± 0.3 (h • μg)/mL, and mean AUC(9,τ) was 5.5 ± 1.7 (h • μg)/mL. This led to an accumulation index of 2.8 ± 1.1, which was comparable to the predicted accumulation index of 2.70 ± 0.60. The mean AUC(9,τ) was also comparable to the AUC(0 to infinity) after a single dose was administered IV (2.98 ± 0.80 [h • μg]/mL for a dose of 0.1 mg/kg),21 which indicated that clearance was not changed after multiple doses. Mean concentration at steady state was 228 ± 72 ng/mL, and mean plasma t1/2 after the final dose was 32.0 ± 14.1 hours. Steady-state clearance (40.5 ± 14.7 mL/h/kg) and volume of distribution (2.3 ± 0.7 L/kg) were comparable to those calculated after administration of a single dose.
Curves generated after IV administration of a single dose in another study21 were plotted, and the parameters calculated were used to simulate the plasma concentration-versus-time curve after daily administration of multiple doses (Figure 3). There was good agreement between simulated and experimental data after both IV and oral administration.
Semilogarithmic graph of the mean ± SD predicted firocoxib plasma concentration–time profile for 9 consecutive daily doses (0.2 mg/kg) of firocoxib administered IV (line) and the plasma concentrations obtained after daily IV administration for 9 days for the 12 horses in the study reported here (white triangles and error bars). Day 0 was the day of administration of the first dose.
Citation: American Journal of Veterinary Research 69, 11; 10.2460/ajvr.69.11.1399
Discussion
Accumulation of a drug in plasma after multiple doses depends on both the half-life and the dosing interval. When the pharmacokinetic behavior of the drug is linear after multiple doses and the half-life is known, plasma concentrations at any time during the dosing regimen as well as the accumulation can be predicted accurately. The accumulation index relates the amount of drug in the plasma at any time after steady state is reached to the corresponding values after a single dose. Mean ± SD expected accumulation index of firocoxib in plasma for this dosing regimen was 2.7 ± 0.6, and the observed accumulation index after oral administration was 3.8 ± 0.7. The accumulation index was higher than expected, but it was not because of nonlinearity of the processes governing the clearance mechanisms or distribution behavior of firocoxib, as indicated by the good agreement of the theoretic and experimental accumulation ratios after IV administration and the good correlation between the experimental and simulated plasma concentration–time curves after IV administration of multiple doses (Figure 3). The higher-than-expected accumulation ratio was more likely attributable to a difference in the extent or rate of absorption between the first and last dose (the predicted curve was increased by 1 SD to account for the increased accumulation ratio). The time to Cmax was 7.80 ± 4.80 hours after the first dose, but it was 0.79 ± 0.70 hours after the last dose. The more rapid or more complete absorption resulted in a greater amount of drug absorbed during the last 24-hour period than after the first dose and a higher accumulation ratio than predicted on the basis of the t1/2. The reason for the increased rate or extent of absorption of the last dose was not determined in the study reported here. One possibility is a difference in the feeding time relative to administration of the paste. Differences in mean absorption time with changes in feeding conditions for horses have been reported for NSAIDs22,24,25 and are attributed to a delay in absorption because of binding of the drug to cellulose in the feed. Food was not withheld from horses in the study to more accurately replicate actual use of firocoxib in field conditions. The extent of absorption could also have been influenced by a saturation of first-pass metabolism. Despite the differences in absorption during the first and last period, the plasma t1/2 after discontinuation of daily dosing was comparable to that after IV administration of a single dose and therefore was not influenced by continued absorption.
Mean ± SD total systemic clearance of firocoxib in horses (27.9 ± 11.3 mL/kg/h) is similar to that for other NSAIDs, and firocoxib has a low extraction ratio (< 0.05). Mean renal clearance of 0.26 ± 0.09 mL/kg/h was far less than the total body clearance, which confirmed that hepatic clearance via metabolism of firocoxib is the main pathway for elimination of firocoxib in horses. The low renal clearance of the parent drug is consistent with the low urine-to-plasma ratio of 0.35. Analysis of urine and plasma samples from a subset of the horses in the oral dosing experiment (samples from 3/6 horses were analyzed) was performed by use of mass spectrometry–mass spectrometry transitions corresponding to hydroxylation, dealkylation, and glucuronidation of firocoxib. Because standards of the metabolites were not available, definitive identification and quantitation of metabolites were not made. The suspected metabolites had similar concentration-time profiles to that of firocoxib (data not shown), with a peak during the first collection interval followed by an exponential decrease; therefore, analogous information can be obtained from firocoxib and its metabolites in urine. Because analytic standards for firocoxib are readily availablef and sensitivity is sufficient to quantitate firocoxib at low concentrations (≥ 5 ng/mL) in urine, firocoxib may be used for determination of concentrations in urine.
Although plasma concentrations may not be representative of tissue concentrations for all drugs, a correlation between the 2 compartments is believed to exist for NSAIDs.3,32 On the basis of in vitro experiments,21 the concentration of firocoxib that inhibits 50% of COX-2 activity (determined in equine whole blood incubated with lipopolysaccharide to induce COX-2) is approximately 30 ng/mL,a with a corresponding concentration of firocoxib that inhibits 80% of COX-2 activity of 67 ng/mL. Plasma concentrations were < 30 ng/mL by 96 hours after the final dose of firocoxib administered orally during a steady state. Plasma concentrations exceeded 67 ng/mL at steady state and decreased to less than that concentration by 72 hours after the final dose of firocoxib administered orally during a steady state.
Concentrations of firocoxib in plasma and urine samples obtained from horses are predictable after single-dose or multiple-dose administration. Use of plasma or urine to ascertain drug concentrations in horses is scientifically valid and robust because the plasma-tourine ratio is stable over time and among animals. The pharmacokinetic behavior of firocoxib was linear after multiple doses administered IV and orally.
ABBREVIATIONS
| AUC | Area under the plasma concentration-versus-time curve |
| Cmax | Peak concentration |
| Cmin | Minimum concentration during the dosing interval |
| COX | Cyclooxygenase |
| λz | First-order rate constant associated with the terminal log-linear portion of the plasma concentration-versus-time curve |
| LOQ | Limit of quantitation |
| t1/2 | Terminal half-life |
McCann ME, Andersen DR, Brideau C, et al. In vitro activity and in vivo efficacy of a novel COX-2 inhibitor in the horse (abstr). J Vet Intern Med 2002;16:355.
Check-R-Mix 14, Kingdom Feed, Fulton, Mo.
Equioxx, Merial Ltd, Duluth, Ga.
WinNonlin, version 5.0.1, Pharsight Corp, Mountain View, Calif.
SAS, version 8.2, SAS Institute Inc, Cary, NC.
GMP Operations, Merial Ltd, North Brunswick, NJ.
References
- 1.
Fitzpatrick FA. Cyclooxygenase enzymes: regulation and function. Curr Pharm Des 2004;10:577–588.
- 2.↑
Wilson JE, Chandrasekharan NV, Westover KD, et al. Determination of expression of cyclooxygenase-1 and -2 isozymes in canine tissues and their differential sensitivity to nonsteroidal anti-inflammatory drugs. Am J Vet Res 2004;65:810–818.
- 3.
Lees P, Landoni MF, Giraudel J, et al. Pharmacodynamics and pharmacokinetics of nonsteroidal anti-inflammatory drugs in species of veterinary interest. J Vet Pharmacol Ther 2004;27:479–490.
- 4.
Twomey BM, Dale MM. Cyclooxygenase-independent effects of non-steroidal anti-inflammatory drugs on the neutrophil respiratory burst. Biochem Pharmacol 1992;43:413–418.
- 5.↑
Simmons DL, Botting RM, Hla T. Cyclooxygenase isozymes: the biology of prostaglandin synthesis and inhibition. Pharmacol Rev 2004;56:387–437.
- 6.↑
Xie WL, Chipman JG, Robertson DL, et al. Expression of a mitogen-responsive gene encoding prostaglandin synthase is regulated by mRNA splicing. Proc Natl Acad Sci U S A 1991;88:2692–2696.
- 7.↑
Kujubu DA, Fletcher BS, Varnum BC, et al. TIS10, a phorbol ester tumor promoter-inducible mRNA from Swiss 3T3 cells, encodes a novel prostaglandin sythase/cyclooxygenase homologue. J Biol Chem 1991;266:12866–12872.
- 8.
Stichtenoth DO, Frolich JC. The second generation of COX-2 inhibitors: what advantages do the newest offer? Drugs 2003;63:33–45.
- 9.
Schnitzer TJ, Hochberg MC. COX-2-selective inhibitors in the treatment of arthritis. Cleve Clin J Med 2002;69 (suppl 1):SI20–SI30.
- 10.
Reimer ME, Johnston SA, Leib MS, et al. The gastroduodenal effects of buffered aspirin, carprofen, and etodolac in healthy dogs. J Vet Intern Med 1999;13:472–477.
- 11.
Forsyth SF, Guilford WG, Lawoko CRO. Endoscopic evaluation of the gastroduodenal mucosa following non-steroidal anti-inflammatory drug administration in the dog. N Z Vet J 1996;44:179–181.
- 12.
Hawkey CJ. Cyclooxygenase inhibition: between the devil and the deep blue sea. Gut 2002;50 (suppl 3):25–30.
- 13.
Futaki N, Yoshikawa K, Hamasaka Y, et al. NS-398, a novel non-steroidal anti-inflammatory drug with potent analgesic and antipyretic effects, which causes minimal stomach lesions. Gen Pharmacol 1993;24:105–110.
- 14.
Masferrer JL, Zweifel BS, Manning PT, et al. Selective inhibition of inducible cyclooxygenase 2 in vivo is antiinflammatory and nonulcerogenic. Proc Natl Acad Sci U S A 1994;91:3228–3232.
- 15.
Bombardier C, Laine L, Reicin A, et al. Comparison of upper gastrointestinal toxicity of rofecoxib and naproxen in patients with rheumatoid arthritis. N Engl J Med 2000;343:1520–1528.
- 16.
Silverstein FE, Faich G, Goldstein JL, et al. Gastrointestinal toxicity with celecoxib vs non-steroidal anti-inflammatory drugs for osteoarthritis and rheumatoid arthritis: the CLASS study: a randomized controlled trial. JAMA 2000;284:1247–1255.
- 17.
Schnitzer TJ, Burmester GR, Mysler E, et al. Comparison of lumiracoxib with naproxen and ibuprofen in the Therapeutic Arthritis Research and Gastrointestinal Event Trial (TARGET), reduction in ulcer complications: randomised controlled trial. Lancet 2004;364:665–674.
- 18.
Brzozowski T, Konturek PC, Konturek SJ, et al. Classic NSAID and selective cyclooxygenase (COX)-1 and COX-2 inhibitors in healing of chronic gastric ulcers. Microsc Res Tech 2001;53:343–353.
- 19.
Brzozowska I, Targosz A, Sliwowski Z, et al. Healing of chronic gastric ulcers in diabetic rats treated with native aspirin, nitric oxide (NO)-derivative of aspirin and cyclooxygenase (COX)-2 inhibitor. J Physiol Pharmacol 2004;55:773–790.
- 20.↑
Chandrasekharan NV, Dai H, Roos KLT, et al. COX-3, a cyclooxygenase-1 variant inhibited by acetaminophen and other analgesic/antipyretic drugs: cloning, structure, and expression. Proc Natl Acad Sci U S A 2002;99:13926–13931.
- 21.↑
Kvaternick V, Pollmeier M, Fischer J, et al. Pharmacokinetics and metabolism of orally administered firocoxib, a novel second generation coxib, in horses. J Vet Pharmacol Ther 2007;30:208–217.
- 22.
Toutain PL, Reymond N, Laroute V, et al. Pharmacokinetics of meloxicam in plasma and urine of horses. Am J Vet Res 2004;65:1542–1547.
- 23.
Toutain PL, Autefage A, Legrand C, et al. Plasma concentrations and therapeutic efficacy of phenylbutazone and flunixin meglumine in the horse: pharmacokinetic/pharmacodynamic modelling. J Vet Pharmacol Ther 1994;17:459–469.
- 24.
Maitho TE, Lees P, Taylor JB. Absorption and pharmacokinetics of phenylbutazone in Welsh Mountain ponies. J Vet Pharmacol Ther 1986;9:26–39.
- 25.
Jaussaud P, Guieu D, Bellon C, et al. Pharmacokinetics of tolfenamic acid in the horse. Equine Vet J Suppl 1992;69–72.
- 26.↑
Kamerling SG. Nonsteroidal anti-inflammatory drugs (NSAIDs) affect pain and inflammation. Louisiana State University Equine Veterinary Research Program Newsletter 1997;6:11.
- 27.
Tobin T, Harkins JD, Sams RA. Testing for therapeutic medications: analytical/pharmacological relationships and ‘limitations' on the sensitivity of testing for certain agents. J Vet Pharmacol Ther 1999;22:220–233.
- 28.
Toutain PL, Lassourd V. Pharmacokinetic/pharmacodynamic approach to assess irrelevant plasma or urine drug concentrations in postcompetition samples for drug control in the horse. Equine Vet J 2002;34:242–249.
- 29.↑
Kvaternick V, Malinski T, Wortmann J, et al. Quantitative HPLC-UV method for the determination of firocoxib from horse and dog plasma. J Chromatogr B Analyt Technol Biomed Life Sci 2007;854:313–319.
- 30.↑
Letendre L, Kvaternick V, Tecle B, et al. Automated liquid chromatography-tandem mass spectrometry method for the analysis of firocoxib in the urine and plasma from horse and dog. J Chromatogr B Analyt Technol Biomed Life Sci 2007;853:333–345.
- 31.↑
Gibaldi M, Perrier D. Multicompartment models. In: Pharmacokinetics. 2nd ed. New York: Marcel Dekker Inc, 1982;45–111.
- 32.
Warner TD, Giuliano F, Vojnovic I, et al. Nonsteroid drug selectivities for cyclo-oxygenase-1 rather than cyclo-oxygenase-2 are associated with human gastrointestinal toxicity: a full in vitro analysis. Proc Natl Acad Sci U S A 1999;96:7563–7568.
