Preclinical Toxicity Evaluation of Tepoxalin, a Dual Inhibitor of Cyclooxygenase and 5Lipoxygenase, in Sprague—Dawley Rats and Beagle Dogs

FUNDAMENTAL AND APPLIED TOXICOLOGY 33 , 3 8 - 4 8 (1996)ARTICLE NO. 0141

Preclinical Toxicity Evaluation of Tepoxalin, a Dual Inhibitor ofCyclooxygenase and 5-Lipoxygenase, in Sprague-Dawley

Rats and Beagle Dogs1

E. V. KNIGHT,2 J. P. KIMBALL, C. M. KEENAN, I. L. SMITH,* F. A. WONG,* D. S. BARRETT, A. M. DEMPSTER,

W. G. LIEUALLEN, D. PANIGRAHI, W. J. POWERS, AND R. J. SZOT

Departments of Drug Safety Evaluation and *Drug Metabolism, The R. W. Johnson Pharmaceutical Research Institute,Raman, New Jersey 08869-0602

Received October 10, 1995; accepted Apnl 26, 1996

Preclinical Toxicity Evaluation of Tepoxalin, a Dual Inhibitorof Cyclooxygenase and 5-Lipoxygenase, in Sprague-Dawley Ratsand Beagle Dogs. KNIGHT, E. V., KIMBALL, J. P., KEENAN, C. M.,

SMITH, I. L., WONG, F. A., BARRETT, D. S., DEMPSTER, A. M.,

LIEUALLEN, W. G., PANIGRAHI, D., POWERS, W. J., AND SZOT,

R. J. (1996). Fundam. Appl. Toxicol. 33, 38-48.

Tepoxalin [5-(4-chlorophenyl)-/V-hydroxy- l-(4-methoxyphenyl)-A/-methyl-lH-pyrazole-3-propanamide] is an orally active anti-in-flammatory agent, which inhibits both cyclooxygenase and 5-lipoxy-genase activities. The oral toxicity of tepoxalin was evaluated in 1-and 6-month rat (up to 50 mg/kg/day) and dog (up to 150 mg/kgbid) studies. In rats, increased liver weight, centrilobular hypertrophy,and hepatic necrosis were observed at dosages s=20 mg/kg/day. Renalchanges indicative of analgesic nephropathy syndrome (i.c, papillaryedema or necrosis, cortical tubular dilatation) were seen at s* 15 mg/kg. In rats treated for 1 month, these hepatic and renal effects werelargely reversible after a 1-month recovery period. Gastrointestinalerosions and ulcers were seen in female rats given 40 mg/kg/dayfor 6 months. Changes in clinical pathology parameters includeddecreases in red blood cell count, hemoglobin, and hematocrit meanvalues; elevation in platelet counts; and an increase in prothrombinand activated partial thromboplastin times. Mild increases in alanineaminotransferase, aspartate aminotransferase, and cholesterol werealso noted in rats. Decreased erythrocyte parameters, increased leu-kocyte counts, and decreased total protein, albumin, and/or calciumwere noted in some dogs in the 300 mg/kg/day group following 6months of dosing. Small pyloric ulcerations were seen at 100 and300 mg/kg/day dosages for up to 6 months. In both rats and dogs,no accumulation of tepoxalin or its carboxylic acid metabolite wasdetected in plasma following multiple dosing over a range of 5 to50 mg/kg/day for rats and 20 to 300 mg/kg/day for dogs. Plasmaconcentrations of the carboxylic acid metabolite were severatfoldhigher than those of the parent compound. The no-effect dosages inrats (5 mg/kg/day) and dogs (20 mg/kg/day) were approximately oneand six times the ED50 (3.5 mg/kg), respectively, for inhibition ofinflammatory effects in the adjuvant arthritic rat without gastricmucosal damage. In terms of severity, the relative lack of gastrointes-

' Presented in part at the 34th annual meeting of the Society of Toxicol-ogy, Baltimore, MD. March 5-9, 1995 (Toxicologist 15, 370, 1995).

2 To whom correspondence should be addressed.

0272-0590/96 $18.00 38Copyright O 1996 by the Society of Toxicology.All rights of reproduction in any form reserved.

tinal side effects, within the estimated therapeutic dose range, distin-guishes tepoxalin from most marketed anti-inflammatory drugs.C 19% Soday of Toidcology

Nonsteroidal anti-inflammatory drugs (NSAIDs) are aclass of therapeutic compounds that are widely used to pre-vent inflammation (Xie et al, 1992; Greaves, 1987). Gastro-intestinal toxicity is a common adverse effect of NSAIDs,causing symptoms of gastrointestinal bleeding, erosions, andulcers (Schoen and Vender, 1989; Khokhar, 1984), in addi-tion to nephrotoxicity and cutaneous reactions. It is alsoknown that NSAID sensitivity to gastrointestinal effects dif-fers between species (i.e., dog > rat > monkey), at compara-ble dose level and duration of treatment (Brooks et al, 1993).The ability of NSAIDs to inhibit prostaglandin synthesis,via the cyclooxygenase (CO)-dependent pathway, was sug-gested as the underlying mechanism for both the anti-in-flammatory effects and the gastrointestinal ulceration (Vane,1971; Gyires, 1994). In some inflammatory diseases, suchas rheumatoid arthritis, significant concentrations of the 5-lipoxygenase (LO) pathway metabolites, 5-s-hydroperoxy-eicosatetraenoic acid (5-HPETE) and leukotriene B4 (LTB4),have been detected at the site of inflammation (Weinblatt etal, 1992). The failure of most NSAIDs to block the produc-tion of LO-derived inflammatory mediators may account fortheir limited efficacy seen in diseases such as rheumatoidarthritis as well as their side effects.

By virtue of its dual CO and LO inhibitory activity, tepox-alin is distinguished from most marketed NSAIDs, whichare pure CO inhibitors (Wallace et al., 1993). Similar toNSAIDs, tepoxalin inhibits prostaglandin, thromboxane, andprostacyclin production and, thus, has anti-inflammatoryproperties (Anderson et al., 1990). Moreover, tepoxalin'sinhibition of production of LTB4 not only contributes toits anti-inflammatory activity, but may prevent further jointdestruction in rheumatoid arthritis models and lessen thegastrointestinal side effects associated with gastric ulcerationwithin the therapeutic dose range (Wallace et al., 1993; An-

by guest on July 13, 2011toxsci.oxfordjournals.org

Dow

nloaded from

TOXICITY OF TEPOXALIN IN RATS AND DOGS 39

CH,O.

FIG. 1. Structures of (a) tepoxalin and (b) carboxylic acid metabolite

derson et al., 1990). Tepoxalin's inhibition of CO and LOhas been shown in several in vitro and ex vivo assays. Tepox-alin produced dose-related inhibition of the edematous re-sponse in the adjuvant arthritic rat assay (ED50 = 3.5 mg/kg, po), a model predictive of clinical efficacy (Argentieriet al., 1990). Tepoxalin also inhibited edema influx of in-flammatory cells into the knee joint space of arthritic rabbitsand dogs with a corresponding reduction of LTB4 productionin synovial fluid (Argentieri et al, 1990).

This report describes the oral toxicity of tepoxalin in 1-and 6-month studies in Sprague-Dawley rats (up to 50 mg/kg/day) and beagle dogs (up to 150 mg/kg bid). The relativeeffects of tepoxalin on clinical signs, pharmacokinetics, andhematological, uroanalytical, biochemical, and histopatho-logical findings are presented for these two species.

METHODS

Test and control materials. Tepoxalin, 3-[5-{4-chlorophenyl>-1 -(4-me-moxypr>enyl>3-pyrazolyl]-/V-hydroxy-/V-rnethylpropanarnide, and 0.5% hydro-xypropyl methylcellulose (HPMC) Premium F4M dosage forms were preparedby The R. W. Johnson Pharmaceutical Research Institute (Rantan, NJ). Thechemical structures of tepoxalin and its carboxylic acid hydrolysis product,3-[5-(4-chlorophenyl)-l-{4-methoxyphenyl)-3-pyrazolyl]propanoic acid, whichwill be referred to as the acid metabolite, are shown in Fig. 1. The test andcontrol articles were assayed for verification of concentration and identity,stability, absence of test article in control article, content, uniformity, density,pH, and particle size and were determined to be acceptable for use in nonclinicalstudies. The test article was micronized (particle size of <2.9 fim) and preparedin 0.5% HPMC Premium F4M at concentratioas necessary for delivery ofappropriate dosages in a volume of 10 ml/kg (rat) or 1-2 ml/kg (dog) bodywt. Dose volumes were adjusted weekly for changes in body weight

Rats. Male and female Crl CD (SD)BR (VAF) rats, approximately 8weeks of age at initiation of dosing, and weighing 151-257 g (females)and 242-373 g (males), were obtained from Charles River Laboratories,Inc. (Kingston, NY). All rats were housed individually in stainless-steelcages. The room was maintained at a temperature of 64-77°F with a relativehumidity of 30-73%, and had a 12-hr light/dark cycle. Rats were used inaccordance with USDA guidelines for humane care. Food (Purina CertifiedRodent Chow Meal No. 5002; Purina Mills Inc., St Louis, MO) and waterwere given ad libitum.

Dogs. Immunized male and female beagle dogs, 7-9 months old atinitiation of dosing and weighing 6.3-9.5 kg, were obtained from MarshallFarms (North Rose, NY). The dogs were individually housed in climate-controlled indoor runs with a 12-hr light/dark cycle. All dogs had water adlibitum and were fed a measured amount of Purina Certified Canine Chow5007 (Purina Mills, Inc.) 1-2 hr after the second daily dose.

Stud}' Design

Rats. Rats were dosed orally daily with tepoxalin for 1 month at dosagesof 0 (vehicle control), 15, 25, 35 and 50 mg/kg and for 6 months (with a3-month interim euthanasia) at 0, 5, 10, 20, and 40 mg/kg. In the 1-monthstudy, 10 rats/sex/group were dosed except for the control and 50 mg/kggroups, which had 15/sex/group. The additional 5 rats/sex from the controland 50 mg/kg groups were maintained for a 1-month recovery period.Ancillary (6 rats/sex) groups were used for evaluation of plasma concentra-tions of tepoxalin and its carboxylic acid metabolite on Days 1 and 28 ofthe study. In the 6-month study with the 3-month interim euthanasia), theinitial 25 rats/sex/group was reduced to 15 rats/sex/group at 3 months fromearly mortality or necropsied at 3 months of dosing. Ancillary (3 rats/sex)groups were included for evaluation of parent drug and its carboxylic acidmetabolite on Days 1, 92, and 184 of study. All remaining rats were necrop-sied after 6 months of dosing.

Dogs. From an initial pilot study, it was observed that there was aplateau in plasma concentrations for both tepoxalin and the acid metaboliteabove a single dose of 200 mg/kg. In addition, at dosages below 200 mg/kg, plasma concentrations did not increase proportionally with increasingdose, indicating that the absorption of tepoxalin in dogs was limited fromoral administration Consequently, dogs were dosed orally twice daily tomaximize exposure during the 1-month study and during the 6-month studywith a 3-month interim euthanasia. Dosages for both studies were 0 (vehiclecontrol), 10, 50, and 150 mg/kg/dose for totals of 0, 20, 100, and 300 mg/kg/day, respectively. In the 1 -month study, the control and 300 mg/kg/daygroups consisted of 5 dogs/sex/group, and the other two groups (20 and100 mg/kg/day) consisted of 3 dogs/sex/group. The additional 2 dogs/sex/group in the control and 300 mg/kg/day groups were maintained for a 1-month recovery period. Plasma concentrations of the parent compound andits acid metabolite were determined on Days 1 and 22 of the study. In the6-month study (4/sex/group) with a 3-month interim euthanasia (3/sex/group), plasma concentrations of tepoxalin and its acid metabolite weredetermined on Days 1, 80, and 179.

For all rat and dog studies, clinical signs of toxicity, body weights, andfood consumption were determined at scheduled intervals throughout thestudy. All animals were observed daily for survival.

Ophthalmoscopic examinations were performed predose and at termina-tion of dosing in the 1 -month rat and predose, prior to the interim euthanasia,and prior to termination in the 6-month rat study with 3-month interimeuthanasia. Clinical pathology specimens were collected prior to terminationof dosing and after the recovery period for the 1-month rat study In the6-month rat study, specimens (blood or urine) were collected from 20 rats/sex/group prior to the 3-month interim euthanasia, and from all survivorsat study termination for hematology, coagulation, clinical chemistry, andurinalysis measurements.

In the dog studies, electrocardiographic, ophthalmoscopic, and physicalexaminations were performed on all dogs predose and prior to terminationof dosing for the 1-month study and predose, prior to interim euthanasia,and prior to termination for die 6-month study with the 3-month interimeuthanasia. Clinical pathology specimens were collected for the evaluation

by guest on July 13, 2011toxsci.oxfordjournals.org

Dow

nloaded from

40 KNIGHT ET AL.

of hematology,3 coagulation,4 clinical chemistry,3 and urinalysis6 parameterspredose and prior to termination of dosing and recovery periods for the 1 -month study. For the 6-month study, specimens were collected predose andat 1.5, 3, and 6 months.

Blood was collected from the retroorbital sinus of fasted rats underCO2:O2 (70:30% v/v) anesthesia or from the jugular vein of fasted awakedogs at periodic intervals prior to, during, or postdosing for clinical pathol-ogy analyses. Urine specimens were collected overnight in the presence ofthymol preservative.

Clinical pathology. Hematology, coagulation, and clinical chemistryparameters were analyzed by routine laboratory procedures and includeddifferential (percent and absolute), hematocrit (HCT), hemoglobin (HB),mean corpuscular hemoglobin concentration (MCHC), mean corpuscularhemoglobin (MCH), mean corpuscular volume (MCV), platelet count(PLT), red blood cell count (RBC), white blood cell count (WBC), activatedpartial thromboplastin time (APTT), prothrombin time (FT), fibrinogen(only in 6-month dog study), alanine aminotransferase (ALT), albumin(ALB), alkaline phosphatase (ALP), aspartate aminotransferase (AST), totalbilirubin (T. Bili), calcium (Ca), chloride (Cl), cholesterol (CHOL), creati-nine (CREAT), -y-glutamyltransferase (GGT), glucose, phosphorus (P), po-tassium (K), sodium (Na), thyroxine (T4), total protein (T. PROT), triglycer-ldes (TRIG), urea nitrogen (UREA N), and uric acid ([UA] only in 6-monthdog study).

At study termination, rats were euthanized by carbon dioxide asphyxia-tion and dogs were exsanguinated while under barbiturate (thiamylal so-dium) anesthesia. Complete necropsies were performed on all animals.Absolute organ weights were recorded for adrenal glands, brain, heart,kidneys, liver, ovaries, pituitary, testes, thyroid/parathyroid and prostate/seminal vesicles (dogs). All collected tissues were fixed in 10% neutralbuffered formalin. Tissues to be examined were embedded in paraffin,sectioned, and stained with hematoxylin—eosin—phloxine. The followingtissues were examined microscopically in all control and high-dosagegroups with the exception of the 1 -month dog study with 1 -month recoveryin which all animals were examined: adrenal glands, aorta (thoracic), bone(femur/stifle), bone marrow (femur/stifle), brain, epididymis, esophagus,eyes, gall bladder (dog), heart, intestine (duodenum, jejunum, ileum, cecum,colon) kidneys, lacrimal gland, liver, lung, lymph node (mesentenc andmandibular), mammary gland (inguinal), sciatic nerve, ovary, pancreas,parathyroid gland, pituitary gland, prostate, salivary gland (submaxillary),skeletal muscle, skin (inguinal), spinal cord (thoracic), spleen, stomach,testis, thymus, thyroid gland, tongue, trachea, urinary bladder, uterus, andvagina. In addition, selected organs (kjdney, liver, prostate, and stomach)were examined in all remaining animals.

Plasma Drug/Metabolite Concentration Determinations

Plasma concentrations of both tcpoxalin and its acid metabolite weredetermined during the 1- and 6-month dog and rat studies. In the rat studies,blood was collected into heparinized syringes approximately 2 hr afterdosing from 3 rats/sex/dosage. The collections were obtained from theinferior vena cava using ether anesthesia and the blood was transferred tomicrocentrifuge tubes. In the dog studies, 5-ml blood samples were collectedin heparinized tubes from the jugular vein immediately prior to (0 hr) andapproximately 1, 2, 4.5, 6, 7, and 24 hr following the first of the two dailydoses which were separated by a 5-hr interval. During analytical method

3Technicon H-1E Hematology Analyzer, Miles Inc Diagnostics Divi-sion, Tarrytown, NY.

4 ACL-300 Plus Coagulation Analyzer, Instrumentation Laboratories,Lexington, MA.

3 Hitachi 717 Chemistry Analyzer, Boehringer Mannheim Corp., India-napolis, IN.

6 Ames Clinitek 200 Urine Chemistry Analyzer, Miles Inc., DiagnosticsDivision, Elkhart, IN.

validation, tepoxalin was found to be less stable in rat plasma than in dogplasma. Thus, samples were processed for analysis immediately on collec-tion from rats. For dogs, the plasma was stored at -20°C prior to analysis.

Plasma concentrations of both tepoxalin and its acid metabolite weredetermined using either of two validated high-performance liquid chromato-graphic (HPLC) methods. A single liquid-liquid extraction with methylenechloride was used to isolate the analytes from plasma samples. The HPLCsystem consisted of a Hitachi (Danbury, CT) pump and a Shimadzu (Colum-bia, MD) autosampler.

In the rat studies, detection was accomplished with a MacPherson (Acton,MA) 7750 fluorescence detector equipped with a high-sensitivity attachmentat 290 nm (excitation) and 420 ran (emission). In the dog studies, detectionwas accomplished with a Shimadzu SPD-10A UV-Vis detector at 254 nm.Data were integrated using a HP 335OA Laboratory Automation System(Hewlett Packard, Avondale, PA). A 10-/im Inertsil (Keystone, Bellefonte,PA) C18 analytical column (4.6 X 15 mm) was used to separate the analytesfrom endogenous coextractants. The compounds were eluted isocraticallyat ambient temperature with a mobile phase consisting of 0.01 M 1-octane-sulfonic acid buffer (pH adjusted to 5.0 with phosphoric acid):acetoni-trile.methanohtetrahydrofuran (43:17:20:20, v:v:v:v). The flow rate was 0.9ml/min. Using this procedure, the retention times of tepoxalin and its acidmetabolite were 5.4 and 3.8 min, respectively. The detection limit for bothcompounds was 0.25 /Jg/ml.

Data Analyses

All quantitative data were checked for aberrant values. A test to deter-mine variance homogeneity was performed. If the data passed that test,then one-way analysis of variance was used to assess overall differencesamong group means. If overall differences were indicated, Dunnett'scomparison to control was performed (Dunnett, 1964). The Jonckheere-Terpstra test was performed to test for an increasing/decreasing trendin response with increased dose. All tests were conducted at the 1 and5% two-sided risk levels. Incidence tables were generated to summarizequalitative data (clinical observations and clinical pathology, gross pa-thology, and histopathology).

RESULTS

Mortality and Clinical Observations

In the 1-month rat study, one treatment-related deathoccurred (Day 22) in the 50 mg/kg group intended for themeasurement of plasma concentration. One death occurredin the 15 mg/kg group 1-month rat study, and 12 deathsoccurred across all dosage groups in the 6-month rat study.These deaths were attributed to either dosing or bloodcollection accidents and were not considered to be drugrelated. No deaths occurred in the dog studies. In the ratand dog studies, there were no drug-related changes inbody weight, food consumption, or ophthalmoscopic andelectrocardiographic (dog only) examinations; however,administration of tepoxalin to beagle dogs resulted inwhitish to yellowish discoloration of the feces mainly inthe 100 and 300 mg/kg/day groups (at 3 months) andmainly in the 300 mg/kg/day group (at 6 months) whichwas attributed to unabsorbed drug.

Clinical Pathology

Rat studies. The effect of tepoxalin administration after1 or 6 months of treatment on the clinical pathology parame-

by guest on July 13, 2011toxsci.oxfordjournals.org

Dow

nloaded from

TOXICITY OF TEPOXALIN IN RATS AND DOGS 41

TABLE 1Selected Hematology and Clinical Chemistry Parameters (Mean ± SE) for Rats Receiving Tepoxalin for 1 Month"

RBC (XlOl2/liter)Males

Females

HB (g/dl)Males

Females

HCT(%)Males

Females

CHOL (mg/dl)Males

Females

Month*

12 (Recovery)12 (Recovery)

12 (Recovery)12 (Recovery)

12 (Recovery)12 (Recovery)

12 (Recovery)12 (Recovery)

Control

7.82 ± 0.098.05 ± 0.217.52 ±0 .137.54 ±0 .15

16.39 ± 0.2116.15 ± 0.3416.17 ± 0.2415.97 ± 0 07

50.90 ± 0 7649.93 ± 0.8249.88 ± 0.7849.50 ± 0.75

56.70 ± 2.8070.40 ± 5.8070.60 ± 4.2080.20 ± 8.40

15 mg/kg

7.83 ±0.15NA

7.33 ± 0.18NA

16.40 ± 0.28NA

15.80 ± 0.16NA

50.53 ± 0.71NA

48.82 ± 0.57NA

67 50 ± 3.50NA

73.80 ± 3.30NA

Treatment

25 mg/kg

7.60 ± 0.14NA

7.22 ± 0.08NA

16.05 ± 0.23NA

15.38 ± 0.12NA

49.23 ± 0.61NA

47.99 ± 0.41NA

65.70 ± 5.00NA

88.90 ± 5.70*NA

35 mg/kg

7.52 ±0.12NA

7.03 ± 0.07**NA

15.91 ± 0.17NA

15.07 ± 0.15**NA

49.38 ± 0.51NA

47.07 ± 0.38**NA

67.20 ± 4.70NA

105.40 ± 6.10**NA

50 mg/kg

7.52 ±0.118.05 ± 0.216.88 ± 0.09**7.98 ± 0.10*

16.02 ± 0.1616.15 ± 0.3414.90 ± 0.26**16.34 ± 0.16

48.74 ± 0.4649.78 ± 0.5546.47 ± 0.70**51.10 ± 0.37

73.10 ± 4.1057.60 ± 4.6099.90 ± 4.70**69.60 ± 4.90

"Significant difference from control (Dunnett's test): *p * 0.05, **p =£0.01.b n = 10/sex at I month; n = 5/sex at 2 months (Recovery)

ters for rats included decreases for RBC, hemoglobin, andhematocrit mean values and increased cholesterol mean val-ues in the 35 and 50 mg/kg dosage groups after 1 month ofdosing (Table 1). These changes were not observed in ratsfrom the 50 mg/kg recovery group at the end of the 1-month recovery period. Cholesterol values for the 50 mg/kgrecovery group returned to normal values comparable tothose of controls. The apparent decrease (Table 1) comparedwith controls was attributed to mild elevation for single maleand female control rats. In the 6-month (with a 3-monthinterim) rat study, decreases in mean RBC, HB, and HCTvalues for females at the interim and at termination and anincrease in prothrombin and activated partial thromboplastintimes for males at termination were observed (Table 2).Although group means were comparable, elevations in plate-let counts for some individual female rats were noted at theinterim and at termination in the 10, 20, and 40 mg/kggroups. Slight increases in alanine aminotransferase, aspar-tate aminotransferase, and cholesterol occurred in some indi-vidual female rats in the 20 and 40 mg/kg groups. Urinalysisrevealed an increase in urinary protein content for someindividual 20 mg/kg females and for rats of both sexes inthe 40 mg/kg group at the 6-month scheduled termination(Table 3).

Dog studies. No drug-related changes were observed inclinical pathology parameters in the 1-month dog study. Inthe 6-month dog study, treatment with tepoxalin resulted in

two females in the 300 mg/kg/day group with mild decreasesin RBC, HB, and HCT values and increases in WBC andneutrophil counts after 6 months of treatment (Table 4).These female dogs also had mild to moderate decreases inserum total protein, albumin, and/or Ca values at Week 6and 3 and 6 months which were considered to be related todrug treatment (Table 5). No urinalysis parameters wereaffected by tepoxalin administration in either the 1- or 6-month dog studies.

Drug Absorption

In mammals including rats and dogs, tepoxalin undergoesboth rapid and extensive in vivo conversion to its carboxylicacid hydrolysis product (see Fig. 1). Consequently, for bothof these species, plasma concentrations of the acid metabo-lite were substantially higher and remained in the systemiccirculation for a longer period than the parent compound.As might be expected for a compound metabolized in thisfashion, intrasubject variability was high and, due to thelimited number of animals sampled in these studies, datafrom female and male groups were combined. In both the1 - and 6-month studies, plasma tepoxalin and acid metaboliteconcentrations at the completion of the studies appeared tobe similar to those seen on the first day of dosing acrossthe dosage range evaluated. Therefore, representative kineticdata from these studies are reported to show the exposureof the animals to tepoxalin and metabolite. Although the

by guest on July 13, 2011toxsci.oxfordjournals.org

Dow

nloaded from

42 KNIGHT ET AL.

TABLE 2Selected Hematology Parameters (Mean ± SE) for Rats Receiving Tepoxalin for 3 and 6 Months"

RBC (XlO'Vliter)Males

Females

HB (g/dl)Males

Females

HCT (%)Males

Females

PT (sec)Males

Females

APTT (sec)Males

Females

Month*

3 (Interim)63 (Interim)6

3 (Interim)63 (Interim)6

3 (Interim)63 (Interim)6

3 (Interim)63 (Interim)6

3 (Interim)63 (Interim)6

Control

8.09 ± 0.098.73 ± 0.117.90 ± 0.138.12 ± 0.13

14.97 ± 0.1315.78 ± 0.1515.49 ± 0.2015.69 ± 0.24

45.21 ± 0.4947.89 ± 0.4647.59 ± 0.6047.43 ± 0.74

15.75 ± 0.6613.53 ± 0.1912.81 ± 0.3812.59 ± 0.20

19.44 ± 1.0514.20 ± 0.4615.81 ± 0.6513.22 ± 0.34

5 mg/kg

8.52 ± 0.09**8.72 ± 0.147.90 ± 0.138.12 ± 0.12

15.44 ± 0.1715.44 ± 0.1815.30 ± 0.1515.73 ± 0 17

46.08 ± 0.5246.77 ± 0.5146.99 ± 0.5147 99 ± 0.55

14.67 ± 0.6013.45 ± 0.1913.85 ± 1.2612.72 ± 0.18

18.38 ± 1.1713.38 ± 0.4518.16 ± 2.8612.48 ± 0.32

Treatment

10 mg/kg

8.36 ±0 .128.% ± 0.147.79 ± 0.068.04 ± 0.10

15.30 ± 0.1516.01 ± 0.2015.29 ± 0.1015.41 ± 0.20

45.50 ± 0.5448.07 ± 0.5147.05 ± 0.3046.67 ± 0.60

16.29 ± 0.8613.71 ± 0.2213.00 ± 0.4812.48 ± 0.23

19.06 ± 1.0915.08 ± 0.6115.73 ± 0.9512.92 ± 0.40

20 mg/kg

8 26 ± 0.078.62 ±0 .147.76 ± 0.087.89 ±0.16

15.40 ± 0.1315.84 ± 0.1815 30 ± 0.1415.18 ± 0.28

46.20 ± 0.4847.18 ± 0.5647.11 ± 0.4545.81 ± 0.79

15.60 ± 0.9313.99 ± 0.3413.07 ± 0.4312.23 ± 0.13

18.61 ± 0.8414.95 ± 0.6516.16 ± 0.7213.11 ± 0.37

40 mg/kg

8.35 ± 0.098.65 ±0 .127.47 ± 0.09**7.60 ± 0.17*

15.34 ± 0.1315.49 ± 0.1714.80 ± 0.20*14.54 ± 0.30**

45.98 ± 0.3746.42 ± 0.6045.69 ± 0.6044.52 ± 0.89*

16.95 ± 1.0314.94 ± 0.29**12.21 ± 0.1811.85 ± 0.19*

19.97 ± 1.1216.42 ± 0.32**15.67 ± 0.5012.98 ± 0.36

° Significant difference from control (Dunnett's test): * p «; 0.05, **p =s 0.01.*n = 20/sex at 3 months (interim); n = 15/sex at 6 months.

absolute bioavailability of tepoxalin was not determined, thissuggests that the absorption of tepoxalin from oral dosingremains constant over time (for at least out to 1 year ofdosing; unpublished findings).

Rat studies. Plasma concentration data for the 6-monthstudy are indicated in Table 6. Mean plasma tepoxalin con-centrations 2 hr after dosing were below the limits of quanti-fication (0.25 ^tg/ml) in the 5 and 10 mg/kg dosage groupsand did not exceed 0.81 //g/ml at 40 mg/kg. Mean plasmaconcentrations of the acid metabolite were detectable at alldosage levels and ranged from 1.39 ± 0.49 to 10.70 ± 3.68^g/ml.

Dog studies. Oral administration of tepoxalin to dogsresulted in dose-related increases in plasma concentrationsof both parent compound and its acid metabolite (Fig. 2).Mean peak concentrations normally occurred within 1 - 2and 1-3 hr following the first of the two daily doses fortepoxalin and metabolite, respectively. Maximum plasmaconcentrations for the acid metabolite were around fourfoldgreater than those of tepoxalin for all three dosage levels.A rapid decline in tepoxalin concentrations was observedafter the peak absorption had been achieved with no detect-able levels at 24 hr, whereas the acid metabolite was elimi-

nated at a much slower rate and could still be detected atlow concentrations in the plasma 24 hr postdose. However,no unexpected accumulation of the acid metabolite occurredon multiple dosing.

Organ Weights and Anatomic Pathology

Rat studies. In the 1 -month study, dose-related increasesin liver weights up to 28% over control weights occurred inall dosage groups except 15 mg/kg males. Although hepaticnecrosis was present in some rats, neither this nor any othermicroscopic findings appeared to be related to liver weightdifferences. Necropsy observations, some of which wereconfirmed microscopically, revealed an increased incidenceof discolored foci in the mucosa of the stomach of rats inthe 50 mg/kg dosage group. Although microscopically, theincidence of stomach erosion(s) appeared to be slightlyhigher in the 25 (females) and 50 mg/kg dosage groups, theirrelationship to tepoxalin administration is uncertain becausethere was no qualitative difference between the lesions inthe vehicle control and drug-treated rats.

Microscopic evaluation of rats in the 1-month study re-vealed drug-related changes in the liver and kidney. Femalesexhibited hepatic necrosis in 1 of 10 and 4 of 10 rats in

by guest on July 13, 2011toxsci.oxfordjournals.org

Dow

nloaded from

TOXICITY OF TEPOXALIN IN RATS AND DOGS 43

TABLE 3Selected Clinical Chemistry and Urinalysis Parameters (Mean ± SE) for Rats Receiving Tepoxalin for 3 and 6 Months"

ALT (U/liter)Males

Females

AST (U/liter)Males

Females

CHOL (mg/dl)Males

Females

Unnalysis: Protein (mg/dl)Males

Females

Month"

3 (Interim)63 (Interim)6

3 (Interim)63 (Interim)6

3 (Interim)63 (Interim)6

3 (Interim)63 (Interim)6

Control

42.543.943.359.2

103.1103.7105.3114.5

65.278.792.0

107.3

66.693.625.356.9

± 1.3± 2.6± 3.9± 7.2

± 3.9± 4.1± 4.7± 9.3

± 3.6± 4.2± 4.0± 6.7

± 69± 14.6± 13.8± 3 9 8

5 mg/kg

40.1 ± 2 246.0 ± 2 963.6 ± 1 4 759.9 ± 8.4

106.3 ± 6.092.4 ± 2 8

115.5 ± 1 4 4122.8 ± 15.9

67.1 ± 4 079 9 ± 6 493 7 ± 5.3

l l l . 0 ± 72

78 3 ± 8.8133.5 ± 26 1196 ± 3843.7 ±17 2

Treatment

10 mg/kg

37.5 ±409 ±53 2 ±665 ±

103.1 ±99.4 ±

114.5 ±133 2 ±

62 0 ±77 7 ±86.8 ±969 ±

82.9 ±79.1 ±21 8 ±17.6 ±

1 52 3

12.210.3

4.35 8

12.5137

2.84 65.25.7

7.57 39.05.6

20 mg/kg

43.4 ± 1.842.0 ± 1 346.9 ± 7 461.1 ± 6.8

98.9 ± 3.495.6 ± 5.7

103.3 ± 9.0116.1 ± 9.2

68.9 ± 4.776.9 ± 4.198.3 ± 5.1

127.5 ± 10.8

78.8 ± 9.782.0 ± 9.8130± 1.7

111.4 ± 68.1

40 mg/kg

43.4 ± 1.842.0 ± 1.335 4 ± 2.775.2 ± 12.8

103.0 ± 5.792 6 ± 5.089.9 ± 6.2*

153.5 ± 26.3

66 9 ± 3.178 4 ± 5.0

103.5 ± 4.9145 1 ± 7.3**

91.9 ± 9.0123 7 ± 20.124.3 ± 5.9

1384 ±73 3

1 Significant difference from control (Dunnett's test): *p >s 0.05, **p s 0.01'/i = 20/sex at 3 months (Interim), n = 15/sex at 6 months.

the 25 and 50 mg/kg groups, respectively; however, liversappeared to be normal after a 1-month recovery period. Kid-ney changes included tubular dilatation in the cortex, chronicprogressive nephropathy, and renal papillary edema or ne-crosis in both males and females. Renal papillary edema ornecrosis was observed in 0 of 10, 1 of 10, 2 of 10, and 5of 10 male rats and 1 of 10, 2 of 10, 1 of 10, and 3 of 10female rats treated with 15, 25, 35, and 50 mg/kg tepoxalin,respectively. The severity of chronic progressive nephropa-thy in both male and female rats, a lesion usually associatedwith aging, was also slightly accentuated in tepoxalin-treatedrats. One female rat in the 50 mg/kg dosage group developedmarked chronic interstitial nephritis and papillary necrosiswhich may be drug related. Renal papillary edema was stillseen in 1 of 5 rats/sex (50 mg/kg) after the recovery period.

Liver weights were significantly increased by approxi-mately 20% in female rats dosed at 20 and 40 mg/kg for 3and 6 months. The increased weight was associated withcentrilobular hypertrophy of hepatocytes observed in 3 of10 and 6 of 10 female rats dosed with 20 and 40 mg/kg,respectively, for 3 months. After 6 months, 1 of 15, 2 of 15,and 7 of 15 female rats dosed with 10, 20, or 40 mg/kgrespectively, had this change in the liver.

Drug-related changes were observed in the kidney. Papil-lary edema was observed in 2 of 10 males dosed with 20mg/kg and 7 of 10 male and 2 of 10 female rats dosed with40 mg/kg for 3 months. After 6 months, only 1 of 15 males

dosed with 40 mg/kg had papillary edema. Compared withcontrols, an increased incidence of chronic progressive ne-phropathy was observed in 40 mg/kg females at 3 monthsand in male rats at this dosage level after 6 months.

Gastric erosions were observed in one female rat dosed with40 mg/kg for 6 months. Ulcers of the cecum were observed inone 40 mg/kg female after 3 months and two 40 mg/kg femalesafter 6 months of treatment. One additional 40 mg/kg femalerat exhibited acute inflammation of the cecum. One 40 mg/kgfemaJe rat had an ulcer in the ileum as well as in the cecum.

Dog studies. In the 1-month study, there were no micro-scopic findings in any organs that were considered to berelated to drug treatment. In the 6-month study, however,one male dog dosed with 300 mg/kg/day had two mucosalulcers and associated inflammation in the pylorus after 3months of dosing. In addition, pyloric ulceration occurredin two females dosed with 300 mg/kg/day after 6 months ofdosing. One female dosed with 100 mg/kg/day and one maledosed with 300 mg/kg/day showed gross evidence of ulcer-ation, but this was not confirmed microscopically. A numberof microscopic changes were seen sporadically and wereconsidered to be incidental findings, unrelated to administra-tion of tepoxalin.

DISCUSSION

The results of these studies indicate that tepoxalin waswell tolerated for up to 6 months in rats and dogs at dosages

by guest on July 13, 2011toxsci.oxfordjournals.org

Dow

nloaded from

44 KNIGHT ET AL.

TABLE 4Selected Hematology Parameters (Mean ± SE) for Dogs Receiving Tepoxalin for 3 and 6 Months"

RBC (XlOl2/liter)Males

Females

HB (g/dl)Males

Females

HCT (%)Males

Females

WBC (XlO'/liter)Males

Females

Month''

Predose1.5 (Interim)3 (Interim)6Predose1 5 (Interim)3 (Interim)6

Predose1.5 (Interim)3 (Interim)6Predose1.5 (Interim)3 (Interim)6

Predose1.5 (Interim)3 (Interim)6Predose1.5 (Interim)3 (Interim)6

Predose1.5 (Interim)3 (Interim)6Predose1 5 (Interim)3 (Interim)6

Control

6.41 ± 0 026.51 ± 0.147 03 ± 0.267 11 ± 0.426.25 ± 0.246.95 ± 0.207.03 ± 0.206.47 ± 0.32

14.17 ± 0.4514.73 ± 0.3015.49 ± 0.5416.58 ± 0.9414.33 ± 0.4516.17 ± 0.4715 99 ± 0.3415 88 ± 0.83

40 99 ± 1.3940.80 ± 0.9747.27 ± 1.6649.00 ± 2.3441.19 ± 1.2045.11 ± 1.3849.31 ± 0.9647.85 ± 2.34

11.89 ± 0.909.40 ± 0.569.81 ± 0419.30 ± 0.889.67 ± 0.66

10.09 ± 1.0610.53 ± 0.7210.55 ± 1.24

20 mg/kg/day

6.21 ± 0.156.80 ±0.117.01 ± 0.166.43 ± 0.286 59 ± 0.146.87 ± 0.266 29 ± 0.63644 ± 0.31

13.90 ± 0.3115.67 ± 0.3415.70 ± 0.3115.45 ± 0.3014.73 ± 0.3315.83 ± 0.5714.13 ± 1.3415.45 ± 0.76

40.00 ± 0.9743.37 ± 0.9147.81 ± 0.9345.80 ± 2.1542.44 ± 0.9143.79 ± 1.6143.23 ± 4.1146.25 ± 2.06

11.89 ± 0.8410.41 ± 0.939.41 ± 0.72

10.05 ± 0 579.66 ± 1.17

10.30 ± 1 1410 86 ± 1 158.35 ± 1.07

Treatment

100 mg/kg/day

6.34 ± 0.246.47 ± 0 186.64 ± 0.156.15 ± 0.087.00 ± 0.196.84 ± 0.246.69 ± 0.386.40 ± 0.15

14.29 ± 0.5815.01 ± 0.3215.19 ± 0.4014.55 ± 0.3015.24 ± 0.2615.24 ± 0.2914.71 ± 0.8414.73 ± 0.46

40.99 ± 1.8241.50 ± 0.8746.34 ± 1.2043.45 ± 0.9243.26 ± 0.8741.89 ± 0.8545.47 ± 2.4944.73 ± 1.06

11.34 ± 0.7912.66 ± 1.7010.43 ± 1 1211 55 ± 0.6310.54 ± 0.8111 00 ± 0 558 69 ± 0.839.85 ± 0.95

300 mg/kg/day

6.50 ±0 .136 84 ± 0.167 29 ± 0.106.37 ±0.176.60 ± 0.196.55 ± 0.466.98 ± 0.266.25 ± 0.54

14.56 ± 0.4015.66 ± 0.3616.26 ± 0.2615.58 ± 0.2714.83 ± 0.3515.26 ± 0.2915.73 ± 0.6614.55 ± 1.53

42.24 ± 1.1743.30 ± 0.9150.26 ± 0 7846.78 ± 0.6342.81 ± 0.9342.03 ± 3.0648.53 ±1.9144.33 ± 4.39

11.93 ± 0.7511.54 ± 0.9312.16 ± 1.2410.88 ± 0.759.80 ± 0.62

12.31 ± 0.9911.50 ± 1.4515 85 ± 3.89

' There were no significant differences from control.l n = 7/sex at 1.5 and 3 months (Interim); n = 4/sex at 6 months.

in excess of the ED50 (3.5 mg/kg) for inhibition of inflam-matory effects in the adjuvant arthritic rat, with reducedgastric mucosal damage. In the rat studies, dosage-relatedhistomorphological changes were observed in the liver, kid-neys, stomach, ileum, and/or cecum at dosages of 10 mg/kgand higher, whereas in the dog studies gastrointestinal effectswere observed at 100 mg/kg and higher dosages. In the 1-month rat study, dose-related increased liver weights wereobserved in all tepoxalin groups except 15 mg/kg males.Hepatic necrosis was observed in female rats at dosage levelsof 25 and 50 mg/kg after 1 month of dosing; however, thelivers of rats dosed with 50 mg/kg appeared normal after 4weeks of recovery. Administration of tepoxalin for 6 monthsresulted in minimal to mild hypertrophy of centrilobular he-

patocytes in female rats given 10, 20, or 40 mg/kg. Thismorphological change was associated with a significant in-crease in liver weights at the 20 and 40 mg/kg dosage levels,an approximately 20% increase over controls. Two 40 mg/kg females dosed for 6 months had minimal centrilobularnecrosis, which was not associated with increased AST orALT activities. Generally, centrilobular hypertrophy is char-acterized by an increase in one or more subcellular organ-elles and is considered to be an adaptive rather than a toxicresponse (Popp and Cattley, 1991). It was the only drug-induced histomorphological change at the 10 mg/kg dosagelevel. Some anti-inflammatory drugs have been shown toalter aryl hydrocarbon hydroxylase activity and cytochromeP450 content of rodent liver microsomes (Mostafa et ai,

by guest on July 13, 2011toxsci.oxfordjournals.org

Dow

nloaded from

TOX1CITY OF TEPOXALIN IN RATS AND DOGS 45

TABLE 5Selected Clinical Chemistry Parameters (Mean ± SE) for Dogs Receiving Tepoxalin for 3 and 6 Months0

CA (mg/dl)Males

Females

T. PROT (g/dl)Males

Females

ALB (g/dl)Males

Females

Month*

Predose1.5 (Interim)3 (Interim)6Predose1.5 (Interim)3 (Interim)6

Predose1.5 (Interim)3 (Interim)6Predose1 5 (Interim)3 (Interim)6

Predose1.5 (Interim)3 (Interim)6Predose1.5 (Interim)3 (Interim)6

Control

10.51 ± 0.0710.89 ± 0.1310.27 ± 0.1410.33 ± 0.0510.80 ± 0.1711.06 ± 0.0910.40 ± 0.1210.08 ± 0.09

5.54 ± 0.085.73 ± 0.135 69 ± 0.115.58 ± 0 095.30 ± 0.065 56 ± 0 065.53 ± 0.095 30 ± 0.13

3.42 ± 0.053.70 ± 0 043.53 ± 0.063.48 ± 0.093.44 ± 0.093.79 ± 0.103.63 ± 0.093.40 ± 0.11

20 mg/kg/day

10.50 ± 0 1010.86 ± 0 1410.29 ± 0.1410.23 ± 0.1510.73 ± 0 0810.91 ± 0.2310.07 ± 0.2310.10 ± 0.12

5.56 ± 0 105.99 ± 0.215.74 ±0 .125.93 ±0 .155.37 ± 0.075 63 ± 0.125.36 ± 0.165.33 ± 0.06

3.37 ± 0.053 86 ± 0.043.46 ± 0.063.33 ± 0.093 39 ± 0.053 76 ± 0.083.19 ± 0.153 30 ± 0.07

Treatment

100 mg/kg/day

10.90 ± 0.121047 ± 0.141029 ± 0.1210.03 ±0.1110.73 ± 0.1110.83 ± 0.2210.14 ± 0 1610.20 ± 0.08

5.44 ± 0.115.34 ± 0 125.29 ±0 .185.23 ±0 .145 64 ± 0 06**5.64 ± 0.145.27 ± 0 125.40 ± 0.07

3 40 ± 0 073.46 ± 0.06**3.29 ±0 .143 10 ± 0093.46 ± 0.053.69 ± 0.06327 ± 0.133.20 ± 0.08

300 mg/kg/day

11.04 ± 0.18*10.94 ± 0.1910.47 ± 0.1310.15 ± 0.1610.87 ± 0.1110.69 ± 0.2710.19 ± 0.209.60 ± 0 42

5.63 ± 0.075.79 ± 0.145.59 ± 0.085.38 ± 0.055.56 ± 0.08*5.41 ± 0.225.24 ±0 .134.83 ± 0.32

3.51 ± 0.063.71 ± 0.063.46 ± 0.063.28 ± 0.063.57 ± 0.063.54 ± 0.213.34 ± 0.122.93 ± 0.28

"Significant difference from control (Dunnett's test): *p ss 0.05, **p* n = 7/sex at 1.5 and 3 months (Interim), n = 4/sex at 6 months.

0.01.

1990). Interestingly, increased liver weight of approximately20% was observed in female rats after 1 year of treatmentwith tepoxalin and the increased weight was not associatedwith an increase in total cytochrome P450 content or peroxi-

some proliferation (unpublished data). No adverse hepaticeffects were noted in any of the dog studies.

Renal papillary edema or necrosis related to tepoxalinadministration was seen in female rats at 15, 25, 35, and 50

TABLE 6Tepoxalin and Carboxylic Acid Metabolite Plasma Concentrations (//g/ml) in Rat 2 hr Postdose

TepoxalinDay 1Day 92Day 184

Acid metaboliteDay 1Day 92Day 184

5

BQL-BQLBQL

1.39 ± 0.492.80 ± 0.931.81 ± 0.78

Dosage

10

BQLBQLBQL

1.40 ± 0.493.51 ± 1.353.13 ± 1.32

(mg/kg/day)

20

0.70 ± 0.530.52 ± 0.580.34 ± 0 52

5.24 ± 2.166.22 ± 3.834.49 ± 3.00

40

0.72 ± 0.400.30 ± 0.170.81 ± 0.58

5 04 ± 2.585.85 ± 2 64

10.70 ± 3.68

Note. Each value represents the group mean ± SD, combined sex, for six rats." Below quantitation limit (0.25 £ig/ml).

by guest on July 13, 2011toxsci.oxfordjournals.org

Dow

nloaded from

46 KNIGHT ET AL.

a 6

20 mg/kg/day

lOOmg/kg/day

300 mg/kg/day

10

Hour (post-dose)

20 mg/kg/day

100 mg/kg/day

300 mg/kg/day

20 250 5 10 15

Hour (post-dose)

FIG. 2. Representative mean profiles in dog plasma for (a) tepoxalin and (b) carboxylic acid metabolite

mg/kg and in male rats at 25, 35, and 50 mg/kg after 1month of treatment. At the end of the 1-month recoveryperiod, evidence of renal papillary edema was still presentin 1 of 5 rats/sex dosed with 50 mg/kg. Renal papillaryedema was also noted in male and female rats at 20 and 40mg/kg after 6 months of treatment. This lesion is compatiblewith the analgesic nephropathy characteristic of NSAIDs(Alden and Frith, 1991). Rats are particularly susceptible tothe development of renal papillary edema and/or necrosis.An increased incidence of nephropathy was noted for 40mg/kg female rats after 3 months. The increased urinaryprotein levels noted at the 40 mg/kg dosage level after 6months may be associated with the increased incidence of

chronic progressive nephropathy observed in male rats. Noadverse renal effects were noted in any of the dog studies.This is considered very unique in view of the sensitivity ofthe canine species to many of the marketed NSAIDS.

Erosions and ulcers of the gastrointestinal tract are com-monly observed with NSAIDs (Walker, 1985; Levi andShaw-Smith, 1994; Bjarson and Macpherson, 1989); how-ever, in the 1 -month rat study, there was no qualitative differ-ence between stomach erosions in the vehicle control andtepoxalin-treated groups. With continuous tepoxalin treat-ment over 6 months, only 1 of 15 female rats at 40 mg/kghad stomach erosions. In the intestinal tract, at 40 mg/kg, 1of 10 female rats dosed for 3 months and 1 of 15 female

by guest on July 13, 2011toxsci.oxfordjournals.org

Dow

nloaded from

TOXICITY OF TEPOXALIN IN RATS AND DOGS 47

rats dosed for 6 months had an ulcer of the ileum. Two ofthe 15 females dosed for 6 months with 40 mg/kg had ulcersin the cecum and one additional female rat had a focal areaof acute inflammation. While the usual location for NSAID-induced changes is the upper gastrointestinal tract, someNSAIDs have been noted to produce lesions in the lowerintestinal tract (Walker, 1985; Bjarson and Macpherson.1989); however, in an earlier dose range-finding study, at ahigher dosage of 75 mg/kg tepoxalin, there was a higherincidence of ulcers of the stomach, which is a primary targetorgan in the rat (unpublished data). This was possibly attrib-uted to the level of the acid metabolite, which has beencharacterized as a pure CO inhibitor. There were no consis-tent histomorphological correlates for the anemia noted clini-cally in the 6-month rat study. Administration of NSAIDscan result in anemia associated with gastrointestinal bloodloss (Bjarson and Macpherson, 1989). In this study, however,only 4 of 25 female rats given 40 mg/kg had morphologicalevidence of gastrointestinal erosions or ulcerations. Never- .theless, similar changes in erythrocyte parameters were ob-served in the 1-month rat study at dosages of 35 and 50 mg/kg but were not seen in the 50 mg/kg dosage group after 1month of recovery. In the 6-month dog study, small pyloriculcers were seen (either grossly or microscopically) in one100 mg/kg/day (one female) and three 300 mg/kg/day (onemale and two females) group dogs. In addition, a similarulcer was seen in one 300 mg/kg/day group dog at the 3-month interim euthanasia. Although not directly correlatedwith histopathological findings, the hematological changesobserved for high-dosage female dogs were suggestive ofpotential gastrointestinal blood loss secondary to gastric irri-tation. Cyclooxygenase-inhibiting anti-inflammatory agentsare known to cause extensive gastric irritation in the dog(Bertram, 1991); however, the mild and sporadic nature ofthe changes seen in the stomachs of dogs in this study inthe presence of compound absorption is considered notewor-thy. In fasted rats, tepoxalin induced lesions in 50% [ulcero-genic dose 50 (UD50)] of the animals tested at a dose of173 mg/kg. In contrast, the UD50 for naproxen was 1.0mg/kg at 3 hr postdose and 5.0 mg/kg at 16 hr postdose(unpublished data). BF-389, a dual CO/LO inhibitor, had areported UD50 of 520 mg/kg/day in rats (Wong et ai, 1992).CI-986, another CO/LO dual inhibitor, showed low ulcero-genic potential relative to other NSAIDs when tested in dogs,monkeys, and rats (Robertson et al., 1993). It has been sug-gested that the reduced ulcerogenic activity of dual CO/LOinhibitors may be due to the ability to inhibit leukotrienesynthesis (Rainsford, 1987; Guslandi, 1987); however,NSAID-induced gastric mucosal damage is not consistentlyaccompanied by stimulation of leukotrienes formation. Leeand Feldman (1992) reported no significant changes in mu-cosal leukotriene C4 synthesis and content and no correlationbetween changes in mucosal LTB4 synthesis and the extentof mucosal damage in aspirin-induced acute gastric mucosal

injury in rats. Also, MK-571, a leukotriene D4 receptor an-tagonist, and MK-886, a leukotriene biosynthesis inhibitor,did not reduce mucosal lesions induced by aspirin, sug-gesting that leukotriene production is not a crucial factor inthe development of indomethacin-induced gastric mucosalerosion (Lee and Feldman, 1992). Mediators of NSAID-induced mucosal injury are probably different from thoseinvolved in mucosal damage induced by various necrotizingagents such as ethanol (Peskar, 1991). Peskar (1991) alsoreported that stimulation of leukotriene C4 formation in therat stomach, induced by ethanol, could be blocked by MK-886 without any evidence for inhibition of gastric damage.Ford-Hutchison et al. (1993) showed that MK-886 had nosignificant effect on indomethacin-induced gastrointestinallesions in the rat.

In conclusion, tepoxalin was remarkably well toleratedfor up to 6 months in rats and dogs at dosages in excess of theED50 for inhibition of inflammatory effects in the adjuvantarthritic rat without gastric mucosal damage. The fact thatonly a mild effect of tepoxalin on the gastric mucosa andno other indication of progression in toxicity was observedbetween 3 and 6 months, in both rats and dogs, has beenparticularly impressive when compared with most NSAIDs.Further work is required to clarify the implications of thesefindings and the possible risk of human exposure.

ACKNOWLEDGMENTS

The authors thank the Drug Safety Evaluation and Drug MetabolismAssociate Staff for their technical support Special thanks to Basil McKenzieand Harris Mosher for critically reviewing the manuscript.

REFERENCES

Alden, C. L., and Frith, C. H. (1991). Urinary system. In Handbook ofToxicohgic Pathology (W. M. Haschek and C. G. Rousseaux, Eds.), pp.347-348. Academic Press, New York.

Anderson, D. W., Argentien, D. C , Ritchie, D. M., Katz, L. B., Shriver,D A , Rosenthale, M. E., and Capetola, R. J. (1990). Gastrointestinal(GI) profile of tepoxalin (TX), an orally active dual cyclooxygenase(CO)/lipoxygenase (LO) inhibitor with potent antiinflammatory activity.FASEB J 4, Al 122.

Argentieri, D. C , Anderson, D. W., Ritchie, D. M , Rosenthale, M. E., andCapetola, R. J. (1990). Tepoxalin (RWJ 20485) inhibits prostaglandin(PG) and leukotriene (LT) production in adjuvant arthritic rats and indog knee joints challenged with sodium urate and immune complexes.FASEB J. 4, Al 122.

Bertram, T. A. (1991). Gastrointestinal tract. In Handbook of ToxicologicPathology (W. M. Haschek and C. G. Rousseaux, Eds.), p. 217. AcademicPress, New York

Bjarson, I., and Macpherson, A. (1989). The changing gastrointestinal side-effect profile of non-steroidal anti-inflammatory drugs. Scand. J. Gas-troenterol. 24(Suppl. 163), 56-64.

Brooks, R. R., Moorehead, T. J., and Pong, S. F. (1993). Gastric toxicityand prostaglandin content in rats dosed with two chemically similar non-steroidal anti-inflammatory agents. Proc. Soc. Exp. Biol. Med. 202, 233 -238.

by guest on July 13, 2011toxsci.oxfordjournals.org

Dow

nloaded from

48 KNIGHT ET AL.

Dunnett, C. W. (1964). New tables for multiple comparison with a control.Biometrics 20, 482-491.

Ford-Hutchison, A. W., Tagari, P., Ching, S. V., Anderson, C. A., Coleman,J. B., and Peter, C. P. (1993). Chronic leukotriene inhibition in the ratfails to modify the toxicological effects of a cyclooxygenase inhibitor.Can. J. Physiol. Pharmacol. 71, 806-810.

Greaves, M. W. (1987). Pharmacology and significance of nonsteroidalanti-inflammatory drugs in the treatment of skin diseases. J Am. Acad.Dermawl. 16(4), 751-764.

Guslandi, M (1987). Gastric effects of leukotrienes. Prostaglandins Leuko-trienes Med. 26(3), 203-208.

Gyires, K. (1994). Some of the factors thai may mediate or modify thegastrointestinal mucosal damage induced by non-steroidal anti-inflam-matory drugs. Agents Actions 41, 73-79.

Khokhar, N. (1984). Nephrotoxicity of nonsteroidal anti-inflammatorydrugs. Am Fam. Physician 30, 123-128.

Lee, M., and Feldman, M (1992). Nonessential role of leukotrienes asmediators of acute gastric mucosal injury induced by aspirin in rats Dig.Dis. Set. 37, 1282-1287.

Levi, S., and Shaw-Smith, C. (1994). Non-steroidal anti-inflammatorydrugs: How do they damage the guf Br J. Rheum. 33, 605-612.

Mostafa, M. H., Skeweita, S. A., and Abdel-Moneam, N. M. (1990). Influ-ence of some anti-inflammatory drugs on the activity of aryl hydrocarbonhydroxylase and the cytochrome P450 content. Environ. Res. 52, 77-82.

Peskar, B. M. (1991). Role of leukotriene C4 in mucosal damage causedby necrotizing agents and indomethacin in the rat stomach. Gastroenterol-ogy 100,619-626.

Popp, J. A., and Cattley, R. C. (1991). Hepatobiliary system In Handbookof Toxicologic Pathology (W M. Haschek and C. G. Rousseaux, Eds.),pp. 293-294. Academic Press, New York.

Rainsford, K. D. (1987). The effects of 5-lipoxygenase inhibitors and leuko-triene antagonists on the development of gastric lesions induced by non-steroidal antiinflammatory drugs in mice. Agents Actions 21, 316-319.

Robertson, D. G., Loewen, G., Walsh, K. M., Dethloff, L. A., Sigler,R. S., Dominick, M. A., and Urda, E. R. (1993). Subacute and subchronictoxicology studies of CI-986, a novel antiinflammatory compound. Fun-dam. Appl. Toxicol 20, 446-455.

Schoen, R. T., and Vender, R. J. (1989). Mechanisms of nonsteroidal anti-inflammatory drug-induced gastric damage. Am. J. Med. 86, 449-458.

Vane, J. R. (1971). Inhibition of prostaglandin synthesis as a mechanismof action for aspirin-like drugs. Nature (London) 231, 232-235.

Walker, F. S. (1985). Azapropazone and related benzotriazines In Anti-inflammatory and Anti-rheumatic Drugs. Vol. II: Newer Anti-inflamma-tory Drugs (K. D. Rainsford, Ed.), pp. 1-32. CRC Press, Boca Raton,FL.

Wallace, J. L., McCafferty, D., Carter, L., McKnight, W., and Argentieri,D. (1993). Tissue selective inhibition of prostaglandin synthesis in ratby tepoxahn: Anti-inflammatory without gastropathy? Gastroenlerology105, 1630-1636.

Weinblatt, M E., Kremer, J. M , Coblyn, J. S., Helfgott, S., Maier, A L.,Petnllo, G., Henson, B., Rubin, P., and Sperling, R. (1992) Zileuton, a5-lipoxygenase inhibitor in rheumatoid arthritis. J. Rheum. 19, 1537-1541.

Wong, S., Lee, S. J., Frierson, M. R., Ill, Proch, J., Miskowski, T. A.,Rigby, B. S., Schmolka, S. J., Naismith, R. W., Kreutzer, D. C , andLindquist, R. (1992). Antiarthritic profile of BF-389: A novel antiin-flammatory agent with low ulcerogenic liability. Agents Actions 37, 9 0 -98.

Xie, W., Robertson, D. L., and Simmons, D. L. (1992). Mitogen-inducibleprostaglandin G/H synthase: A new target for nonsteroidal antiinflamma-tory drugs Drug. Dev. Res. 25, 249-265.

by guest on July 13, 2011toxsci.oxfordjournals.org

Dow

nloaded from