Comparative Articular Toxicity of Garenoxacin, a Novel Quinolone Antimicrobial Agent, in Juvenile Beagle Dogs

The Journal of Toxicological Sciences, 219 Vol.27, No.3, 219-228, 2002

COMPARATIVE ARTICULAR TOXICITY OF GARENOXACIN, A NOVEL QUINOLONE ANTIMICROBIAL AGENT, IN JUVENILE BEAGLE DOGS

Akio NAGAI, Miyono MIYAZAKI, Teiichi MORITA, Shinichi FURUBO, Kazuo KIZAWA, Hiroyuki FUKUMOTO, Takahiro SANZEN, Hiroyoshi HAYAKAWA and Yasuhito KAWAMURA

Research Laboratories, Toyama Chemical Co., Ltd., 2-4-1 Shimookui, Toyama 930-8508, Japan

(Received February 19, 2002; Accepted June 3, 2002)

ABSTRACT — The articular toxicity of garenoxacin (formerly T-3811 or BMS-284756) was experimen- tally examined utilizing juvenile beagle dogs. Garenoxacin and two other reference quinolones were administered at intravenous dosages of 30 and 60 mg/kg. Each group consisted of 3 male dogs (Experi- ment I). Oral dosages of 50 mg/kg of 3 compounds were also given daily to male only and female only groups (Experiment II) over a period of 7 days. We evaluated the articular toxicity of garenoxacin compared to ciprofloxacin and norfloxacin. In Experiment I, no articular toxicity was detected in the 30 mg/ kg garenoxacin group. One animal from the 60 mg/kg garenoxacin group developed detectable histopathological lesions in the articular cartilages of the shoulder, elbow and knee joints. In the 30 mg/kg ciprofloxacin group and the 30 and 60 mg/kg norfloxacin groups, histopathological articular cartilage lesions of the shoulder, elbow, carpus, hip, knee and tarsus joints were observed in all of the dogs. The area under the plasma concentration-time curve (AUC0→∞) values, after the first dose was administered, for the 30 mg/kg groups given garenoxacin, ciprofloxacin and norfloxacin were 164, 68.1 and 65.7 µg⋅hr/mL, respectively. In Experiment II, the degree of histopathological change was most significant in the ciprofloxacin group, followed by the norfloxacin group, and with comparatively the least changes in the garenoxacin group. The AUC0→∞ values, obtained after the 6th day of antimicrobial administration, were 202 and 173 µg⋅hr/mL for male and female dogs, respectively, from the 50 mg/kg garenoxacin group. The AUC0→∞ values for the garenoxacin group after the 6th daily administration were 7.8 to 17.0 times greater for male dogs and 3.8 to 13.2 times greater for female dogs than those obtained from the ciprofloxacin and norfloxacin groups. The concentrations of garenoxacin in the synovia, articular cartilage and the synovialis 4 hr following the last garenoxacin administration were 2.0 to 6.5 times higher for male dogs and 1.5 to 3.3 times higher for female dogs than the antimicrobial levels measured in the ciprofloxacin and norfloxacin groups. As discussed above, although the garenoxacin concentrations in plasma and joint tissue were higher than those for ciprofloxacin and norfloxacin, however, the articular toxicity of garenoxacin was much less than that of the other two antimicrobials.

KEY WORDS: Garenoxacin, Articular toxicity, Juvenile dog, Quinolone

INTRODUCTION inhibiting tertiary negative supercoiling of bacterial DNA (Crumplin and Smith, 1976; Gellert et al., 1977; Garenoxacin (formerly T-3811 or BMS-284756) Wang, 1985; Gootz and Brighty, 1996). is a novel des-F(6)-quinolone that lacks fluorine at the Articular toxicity in juvenile animals has been C-6 position typical of existing fluoroquinolones. observed following treatment with all of the quinolones Garenoxacin is characterized by improved activity (Gough et al., 1992; Stahlmann et al., 1993; Stahlmann against gram-positive bacteria, while retaining most of and Lode, 1999). It has been reported and documented its efficacy against gram-negative bacteria (Takahata et that the articular toxicity of quinolone antimicrobial al., 1999; Fung-Tomc et al., 2000). Quinolone anti- agents is more severe in dogs, especially juvenile dogs, microbial agents act by inhibition of bacterial topoi- than in other experimental animals (Andriole, 1988; somerase II (DNA gyrase) and topoisomerase IV, thus Fukuda et al., 1990; Schaad, 2000). This chondrotoxic Correspondence: Akio NAGAI

Vo l . 2 7 N o . 3 220

A. NAGAI et al. potential has led to a widespread restriction on the use garenoxacin mesilate is shown in Fig. 1. Ciprofloxacin of fluoroquinolones. Their application has been con- hydrochloride was extracted from the commercial sidered to be contraindicated in children, in growing product (Ciproxan Tablets, Bayer Yakuhin, Ltd., adolescents, and during pregnancy and lactation. How- Osaka, Japan) by Toyama Chemical Co., Ltd., and was ever, ciprofloxacin has been approved recently for assayed to be 99.9% pure (lot no. 970819) and 99.8% administration, by injection and orally, for treatment of pure (lot no. KK980131A). Norfloxacin (lot no. cystic fibrosis in children in many countries KCJ3441) was purchased from Wako Pure Chemical (Stahlmann et al.,1993; Karande and Kshirsagar, 1996; Industries, Ltd. (Osaka, Japan). Hampel et al.,1997; Richard et al.,1997; Church et al.,1997; Warren,1997; Jick, 1997; Heggers et Animals al.,1998; Lipsky and Baker,1999). Surveillance of Two- or 3-month-old male and female beagle ciprofloxacin usage and effect in the pediatric popula- dogs were purchased from CSK Research Park Inc. tion have yet to uncover examples of serious articular (Nagano, Japan). Our experimental protocol was initi- toxicity (Jick, 1997, Schaad, 2000), although ciprof- ated following a successful 1- to 3- week quarantine loxacin use can induce acute reversible articular toxic- period. Twenty-seven male and 12 female, 3-month- ity (Schaad et al., 1997; Karande and Kshirsagar, 1996; old, beagle dogs were employed as our experimental Warren, 1997; Lipsky and Baker, 1999). In Japan, nor- animal. Their body weight values for males ranged floxacin is employed as an oral drug treatment in pedi- from 5.3 to 6.6 kg and for females from 4.8 to 5.6 kg. atrics (Fujii, 1990; Takasugi, 1991; Bhattacharya et al., These animals were individually housed in metal cages 1997). To evaluate the articular toxicity of garenoxacin in an animal room which was maintained within a tem- via different clinical routes of administration, garenox- perature range of 21-25°C, and a humidity from 40- acin was given intravenously or orally to 3-month-old 60%. This animal room was also ventilated 10 or more beagle dogs, once daily, over a treatment period of 7 times per hr and artificial light was provided for 12 hr days. We also evaluated the articular toxicity effect of per day (6:00 to 18:00). These dogs were fed 200 g of garenoxacin with toxicity effects generated by use of pellet diet for dogs (CD-5, CLEA Japan Inc., Tokyo, ciprofloxacin and norfloxacin. Japan) daily and water which was filtered, chlorinated and sterilized using ultraviolet radiation, ad libitum. MATERIALS AND METHODS All of the subject animals were individually identified by both cage labels and ear tattoos. After completing Test substance and reference quinolone their quarantine period, the beagle dogs were catego- Garenoxacin mesilate (1-cyclopropyl-8-(difluo- rized and segregated using body weight values, date of romethoxy)-7-[(1R)-1-methyl-2,3-dihydro-1H-5- birth and blood relationship data. Next they were sepa- isoindolyl]-4-oxo-1,4-dihydro-3-quinolinecarboxylic rated into experimental groups by the random sam- acid methanesulfonate monohydrate) was synthesized pling method. by Toyama Chemical Co., Ltd. (Tokyo, Japan) and assayed to be 100% pure (lot no. AY970819) and Dosage and experimental group treatment 99.7% pure (lot no. 804T-1). The molecular formula is 1. Experiment I (intravenous administration) C24H26F2N208S, which corresponds to a molecular It has been reported that ciprofloxacin induces weight of 540.53 (salt, monohydrate). The free base articular cartilage damage in 3-month-old beagle dogs molecular weight is 426.41. The chemical structure of following a one-week repetitive exposure period to subcutaneous administration of 30 mg/kg (New Drug Symposium Data, 1993). In Experiment I, each dosage group consisted of three male juvenile beagle dogs. There were a total of five dosage groups that received intravenous administration daily. The antimicrobial protocol for these five dosage groups was 30 mg/kg (dosage of the corresponding free base) and 60 mg/kg of garenoxacin, 30 mg/kg of ciprofloxacin, and 30 mg/ kg and 60 mg/kg of norfloxacin. The garenoxacin mesilate and ciprofloxacin hydrochloride were first Fig. 1. Chemical structure of garenoxacin mesilate. dissolved in a 5% glucose solution (Otsuka Pharma-

Vo l . 2 7 N o . 3 221

Articular toxicity of garenoxacin in juvenile beagle dogs. ceuticals, Tokyo, Japan) to dilute them to concentra- Clinical observations and pathological examinations of 0.6% (as a free base) and then sterilized tions utilizing a membrane-filter (Millipore Filter GV, Milli- Clinical observation was performed and recorded pore Japan). Norfloxacin was dissolved in 1 mol/L daily for all of the subject beagles. Their body weight NaOH (equal mole of norfloxacin) and diluted with a and diet consumption were also measured and 5% glucose solution to obtain a concentration of 0.4% recorded prior to drug administration. All of the ani- (as a free base) and then sterilized as described above. mals were sacrificed by exsanguination under pento- The garenoxacin and ciprofloxacin at a dose of 30 mg/ barbital sodium (NEMBUTAL Injection, Abbott kg were injected through the cephalic vein at a rate of Laboratories, Inc., U.S.A.) anesthesia. For Experiment 10 mL/min on the first day. One of the 3 beagles in our I, the sacrifices were performed 24 hr after the last dose ciprofloxacin group exhibited shock-like symptoms, was administered on Day 7, and for Experiment II, the specifically lying on his side. Therefore, the intrave- sacrificewereperformed4hrafterthelastdosewas nous infusion rate was adjusted lower to 5 mL/min, given on Day 7. The shoulder joint (scapular fossa and beginning on the second day of this experiment. humerus proximal portion), elbow joint (humerus dis- Garenoxacin at a dose of 60 mg/kg and norfloxacin at tal portion and radioulnaris proximal portion), carpus doses of 30 and 60 mg/kg were infused through the joint (distal portion of the radioulnaris), hip joint (ace- cephalic vein at a rate of 5 mL/min throughout the tabular fossa and proximal portion of the femur), knee experimental period of 7 days. The volume infused for joint (distal portion of the femur and proximal portion the 60 mg/kg garenoxacin group, the 30 and 60 mg/kg of the tibia) and tarsus joint (distal portion of the tibia) norfloxacin groups was 10, 7.5 and 15 mL/kg, respec- were removed. All of these samples were labeled, fixed tively. All of the animal groups were given their spe- in 10% formalin and stored. The joint samples cific intravenous antimicrobial treatment once daily obtained at necropsy were decalcified utilizing the over a period of 7 days. Plank-Rychlo method and embedded in paraffin. They were then stained with hematoxylin and eosin and 2. Experiment II (oral administration) examined histopathologically. It has been reported that articular toxicity for juvenile beagle dogs was observed at a dosage level of Determination of the concentrations of garenoxa- 50 mg/kg, with repeated oral administration of nor- cin, ciprofloxacin and norfloxacin in the plasma, floxacin over a period of one week (Machida et al., synovia, articular cartilage and synovialis speci- 1990). That is why the dosage of garenoxacin was mens selected at 50 mg/kg (dosage of the corresponding free In Experiment I, blood samples were collected base) allowing it to be impartially compared with the from the cephalic vein of all of the dogs into sodium documented articular toxicity of norfloxacin at the heparin. The schedule for this blood collection was just same dosage level. Ciprofloxacin was also selected to prior to quinolone administration, and then 5 min, 30 be evaluated at a dosage of 50 mg/kg (dosage of the min, 2 hr, 6 hr and 24 hr after the first antimicrobial corresponding free base) for its ability to induce artic- dose was infused. Blood was also collected 24 hr after ular toxicity. Garenoxacin mesilate, ciprofloxacin final administration. Plasma was separated from these hydrochloride and norfloxacin were placed into gelatin samples and assayed for determination of blood anti- capsules (No.0, Warner-Lambert K.K., Kanagawa, microbial quinolone levels. Japan) and forcibly administered orally to each of the 3 In Experiment II, the schedule for blood collec- male and 3 female dogs in each antimicrobial treat- tion from beagles was just prior to quinolone adminis- ment group. Administration was followed with inges- tration,andthen30min,1,2,4,6,12and24hrafter tion of 5 mL of water. All of the animals were orally the first antimicrobial was given. Blood was also col- administered their quinolone antimicrobial once daily lected 4 hr after final administration. Plasma samples over a period of 7 days. The control group received an were also separated and analyzed as in Experiment I. empty gelatin capsule in the same manner and under Synovia samples were obtained from the hip and knee the same schedule as the test substance groups. Three joints (right and left) of all the animals collected at dogs of each sex were assigned to each treatment necropsy (4 hr after final quinolone administration). group, with a total of 12 dogs of each sex treated in this Cartilage was obtained from the proximal and distal experimental evaluation. portions of the femur (right and left) of all the animals at autopsy (4 hr after final quinolone administration).

Vo l . 2 7 N o . 3 222

A. NAGAI et al.

The synovialis specimens were secured from the knee From the assayed plasma quinolone concentra- joint (right and left) of all the animals collected at tion specific for each animal in our study, the area necropsy (4 hr after final quinolone administration). under the plasma concentration-time curve (AUC0→∞) The concentration of garenoxacin in the plasma and the terminal half life (T1/2) were calculated using a and synovia was determined using LC/MS/MS analysis. moment analysis program (Analysis Soft: Win Nonlin To 0.2 mL of the plasma or synovia sample, 40 µLofthe Standard Ver.1.5, Pharsight, U.S.A.). The maximum internal standard stock solution (50 µg/mL) and 1.0 mL plasma concentration (Cmax) and the elapsed time of acetonitrile were added, and the mixture was shaken. needed to attain the maximum quinolone plasma con- After centrifugation at 40,000 g for 10 min (10°C, centration (Tmax) were determined from the measured SIGMA 3K30C), the supernatant layers were evapo- values. rated to dryness. The dried extracts were re-dissolved in 1.0 mL of the mobile phase for HPLC. After centrifuga- Statistical analysis tion at 1,500 g for 10 min (10°C, KUBOTA 5900), the Data were expressed as the mean ± S.D.. Statisti- supernatant was injected into the LC/MS/MS system. cal evaluation of the data was not performed. LC/MS/MS analysis was performed with a mass spectrometer detector (TSQ-7000, Finnigan, U.S.A.), a RESULTS 626LC System pump (Waters, Tokyo, Japan) and a 717plus auto-sampler (Waters). The column was a Experiment I (intravenous administration) Symmetry C18/5 µm (150 mm × 2.1 mm i.d.; Waters). Physical signs detected in the 30 mg/kg garenox- The guard column was a Develosil ODS-HG-S (10 mm acin group included flushing and vomiting. Signs × 1.5 mm i.d.; Nomura Chemical Co., Ltd., Seto, observed in the 60 mg/kg garenoxacin group were Japan). The mobile phase was acetonitrile-0.25 mol/L flushing, vomiting, excess salivation and defecation. formate buffer (pH 2.8)-distilled water (350 mL : 200 In the 30 mg/kg ciprofloxacin group, a decrease in mL : 450 mL). The flow rate was 0.2 mL/min. The locomotor activity, assumption of lateral position, quantitative range for this method was 0.03-25 µg/mL. flushing, vomiting, defecation, urinary incontinence, The concentration of garenoxacin in the cartilage nasal discharge and increased salivation were wit- and synovialis was determined using HPLC analysis. nessed. In the 30 and 60 mg/kg norfloxacin groups, a The cartilage (about 20 mg) was incubated with 70% decrease in locomotor activity, assumption of a lateral formate (500 µL) overnight at 60°C. After incubation, position, flushing, head edema, vomiting and increased the dissolved solution was diluted with the mobile salivation were recorded. There were no abnormalities phase and 50 µL was injected into the HPLC system. in either body weight change or diet consumption. The synovialis (about 20 mg) was incubated with 1 The results obtained from the histopathological mol/L NaOH (100 µL) for 3 hr at 60°C. Acetonitrile examination of the canine articular cartilage are listed (100 µL) and 70% formate (300 µL) were added to the in Table 1. No histopathological abnormalities were dissolved solution. After centrifugation at 12,000 rpm detected in the 30 mg/kg garenoxacin group. In one for 10 min (4°C, KUBOTA 1700), the supernatant was animal in the 60 mg/kg garenoxacin group, the obser- diluted with the mobile phase and 50 µL was injected vation of a chondrocyte cluster, a degenerative change into the HPLC system. in the chondrocytes, a rarefaction of the cartilage HPLC analysis was performed with an RF-10A matrix, cavitation at the cartilage layer and erosion at detector (Shimadzu, Kyoto, Japan), a L-7100 pump the cartilage layer were judged to be very slight to (Hitachi, Tokyo, Japan), a D-7500 integrator (Hitachi) slight in the shoulder joint, elbow joint and hip joint. In and a 717plus auto injector (Waters). The column was a all 3 dogs from the 30 mg/kg ciprofloxacin group, his- Develosil ODS-HG-5 (150 mm × 4.6 mm i.d.; Nomura topathological findings very similar to the 60 mg/kg Chemical Co., Ltd., Seto, Japan). The guard column garenoxacin group were detected in the shoulder joint, was a Develosil ODS-HG-5 (10 mm × 4.0 mm i.d.; elbow joint, carpus joint, hip joint and knee joint. Nomura Chemical Co., Ltd.). The mobile phase was These changes ranged from very slight to severe. Only acetonitrile-citrate buffer (pH 3.5)-distilled water (280 the tarsus joint did not display signs of arthropathy. In mL : 150 mL : 570 mL). The flow rate was 1.0 mL/min. all 3 animals from the 30 mg/kg norfloxacin group, The concentrations of the reference quinolones in histopathological findings very similar to the 60 mg/kg the plasma, synovia, cartilage and synovialis were garenoxacin group were observed in the shoulder joint, assayed using HPLC analysis. hip joint, knee joint and tarsus joint. These changes

Vo l . 2 7 N o . 3 223

Articular toxicity of garenoxacin in juvenile beagle dogs.

Table 1. Incidence of arthropathy observed in juvenile male beagle dogs that were administered garenoxacin, ciprofloxacin and norfloxacin intravenously, once daily, over an experimental period of 7 days. Incidence of Arthropathy Joint Garenoxacin Dose Ciprofloxacin Dose Norfloxacin Dose (mg/kg) (mg/kg) (mg/kg) 30 60 30 30 60 Shoulder 0/3 1/3 3/3 3/3 3/3 Elbow 0/3 1/3 1/3 0/3 3/3 Carpus 0/3 0/3 1/3 0/3 3/3 Hip 0/3 1/3 3/3 3/3 3/3 Knee 0/3 0/3 2/3 3/3 3/3 Tarsus 0/3 0/3 0/3 2/3 0/3 The quinolone drugs were intravenously administered to each treatment group, consisting of 3 male dogs, once daily, over a period of 7 days. The histopathological examinations were focused primarily on the condition of the articular cartilages collected 24 hr after final dose was given on Day 7. were judged as ranging from very slight to moderate. severe were detected in the shoulder joint and hip joint In all 3 animals from the 60 mg/kg norfloxacin group, of one male and in all of the female dogs. In the cipro- histopathological findings very similar to the 60 mg/kg floxacin group, histopathological changes that were garenoxacin group were detected in the shoulder joint, judged as being from very slight to severe were pre- elbow joint, carpus joint, hip joint and knee joint. sented in the shoulder joint, elbow joint, carpus joint, These alterations were evaluated as ranging from very hip joint and knee joint for all of the male and female slight to severe. dogs. In the norfloxacin group, histopathological The results obtained from the analysis of the qui- changes that were determined to be slight to severe in nolone concentrations in the canine plasma are listed in intensity were observed in the shoulder joint, elbow Table 2. The AUC0→∞ values after receiving the first joint, hip joint and knee joint in 2 of the male and 2 of infusion dose of the quinolone were as follows: 164 thefemaledogs.Thedegreeofintensityofthehisto- and 321 µg⋅hr/mL for the 30 and 60 mg/kg garenoxa- pathological change was most significant and most cin groups, 68.1 µg⋅hr/mL for the 30 mg/kg ciproflox- severe in the ciprofloxacin group, followed by the nor- acin group, 65.7 and 116 µg⋅hr/mL for the 30 and 60 floxacin group. The garenoxacin group demonstrated mg/kg norfloxacin groups, respectively. the least significant histopathological changes of all of the quinolones treated in this experiment. Experiment II (oral administration) The results from our analysis of the test substance Vomiting was witnessed, beginning immediately concentrations in plasma after the 1st and 6th oral and for up to 4 hr following oral administration for all administration are listed in Table 4. After the first oral 3 quinolone treatment groups. Arthralgia (pain when administration, the Cmax values calculated for the bending the forelimb or hindlimb) was also detected in garenoxacin, ciprofloxacin and norfloxacin treatment all of the male dogs (after Day 3) and in all of the groups were 15.1, 5.64 and 3.52 µg/mL for the male female dogs (after Day 4) in the ciprofloxacin group dogs and 19.1, 3.93 and 4.19 µg/mL for the female and in 2 male dogs (after Day 3) and 2 female dogs dogs, respectively. The AUC0→∞ values for each treat- (after Day 4) in the norfloxacin group. There were no ment group were 157, 40.8 and 24.2 µg⋅hr/mL for the abnormalities in either body weight or diet consump- male dogs and 176, 24.9 and 25.0 µg⋅hr/mL for the tion. female dogs, respectively. The Tmax values for each The results of the histopathological examination treatment group were 4.67, 3.33 and 2.33 hr for the of canine articular cartilage are listed in Table 3. In the male dogs and 5.33, 2.67 and 2.00 hr for the female garenoxacin group, histopathological changes (erosion dogs, respectively. The T1/2 values for each treatment in the cartilage layer, cavitation in the cartilage layer, group were 4.83, 2.83 and 3.15 hr for the male dogs rarefaction of the cartilage matrix, the appearance of and 4.23, 3.70 and 3.22 hr for the female dogs, respec- chondrocyte clusters and degeneration and necrosis of tively. After the 6th oral administration, the Cmax val- chondrocytes) that were graded from very slight to ues for the garenoxacin, ciprofloxacin and norfloxacin

Vo l . 2 7 N o . 3 224

A. NAGAI et al.

Table 2. Plasma concentrations and pharmacokinetic parameters for garenoxacin, ciprofloxacin and norfloxacin in juvenile male beagle dogs that were administered these drugs intravenously, once daily, over an experimental period of 7 days. First administration (1st) Last administration (7th) Treatment group Dose AUC0-∞ T1/2(hr) Plasma concentration( g/mL) ( mg/kg ) ( g hr/mL) 24 hr 30 164 ± 11.6 4.53 ± 0.27 0.56 ± 0.13 Garenoxacin 60 321 ± 24.6 4.91 ± 0.15 1.08 ± 0.23 Ciprofloxacin 30 68.1 ± 3.04 3.72 ± 0.27 0.07 ± 0.06 30 65.7 ± 11.9 3.34 ± 0.21 ND Norfloxacin 60 116 ± 23.8 2.98 ± 0.15 ND Data are expressed as the Mean ± S.D. of three male beagle dogs in each group. N.D.: Not detected (below the analytical limit of detection).

Table 3. Incidence of arthropathy in juvenile male and female beagle dogs administered a 50 mg/kg dose of garenoxacin, ciprofloxacin and norfloxacin orally, once daily, over an experimental period of 7 days. Incidence of arthropathy Control Garenoxacin Ciprofloxacin Norfloxacin Joint (gelatin capsule) Dose Dose Dose (mg/kg) (mg/kg) (mg/kg) –505050 male female male female male female male female Shoulder 0/3 0/3 1/3 1/3 3/3 3/3 2/3 1/3 Elbow 0/3 0/3 0/3 0/3 3/3 2/3 1/3 1/3 Carpus 0/3 0/3 0/3 0/3 1/3 2/3 0/3 0/3 Hip 0/3 0/3 1/3 3/3 3/3 3/3 1/3 2/3 Knee 0/3 0/3 0/3 0/3 3/3 3/3 1/3 1/3 Tarsus 0/3 0/3 0/3 0/3 0/3 0/3 0/3 0/3 Each group consisted of 3 dogs. Each antimicrobial and control had two treatment groups, male and female. The quinolone drugs were orally administered to each group, once daily, over an experimental period of 7 days. The histopathological examinations were focused primarily on the condition of the articular cartilage collected 4 hr after final administration on Day 7. treatment groups were 19.6, 4.25 and 2.04 µg/mL for from 7.8 to 17.0 times higher for the male dogs and the male dogs and 17.8, 6.24 and 1.72 µg/mL for the from 3.8 to 13.2 times higher for the female dogs than female dogs. The AUC0→∞ values for each treatment the values obtained from the reference quinolone group were 202, 25.9 and 11.9 µg⋅hr/mL for the male (ciprofloxacin and norfloxacin) groups. dogs and 173, 45.4 and 13.1 µg⋅hr/mL for the female The results of the analysis for the garenoxacin, dogs, respectively. The Tmax values for each treatment ciprofloxacin and norfloxacin concentration, 4 hr after group were 4.67, 3.33 and 1.00 hr for the male dogs the last oral administration are depicted in Fig. 2. The and 4.67, 2.67 and 3.33 hr for the female dogs, respec- concentrations are expressed as µg/mL or µg/g. The tively. The T1/2 values for each treatment group were plasma levels for garenoxacin, ciprofloxacin and nor- 4.67, 2.49 and 6.19 hr for the male dogs and 3.96, 2.73 floxacin groups were 14.4, 3.72 and 1.71 µg/mL for and 3.99 hr for the female dogs, respectively. The the male dogs and 13.7, 5.67 and 1.90 µg/mL for the AUC0→∞ values calculated for the garenoxacin group female dogs, respectively. The synovia concentrations, following the first administration were from 3.8 to 6.5 from the knee joint, were for the garenoxacin, ciprof- times higher for the male dogs and from 7.0 to 7.1 loxacin and norfloxacin: 9.88, 4.96 and 1.97 µg/mL for times higher for the female dogs than the values the male dogs and 9.92, 6.43 and 3.03 µg/mL for the obtained from the reference quinolone (ciprofloxacin female dogs, respectively. The cartilage concentra- and norfloxacin) groups. The AUC0→∞ values for the tions, from the knee joints, for the garenoxacin, ciprof- garenoxacin group after the 6th administration were loxacin and norfloxacin groups were 18.1, 8.61 and

Vo l . 2 7 N o . 3 225

Articular toxicity of garenoxacin in juvenile beagle dogs.

Table 4. Pharmacokinetic parameters obtained from data collected on dogs treated with garenoxacin, ciprofloxacin and norfloxacin. Drugs were orally administered, once daily, over a period of 7 days. First administration Treatment Dose Tmax Cmax T1/2 AUC0-∞ Group ( mg/kg ) (hr) ( g/mL) (hr) ( g hr/mL) male 4.67 ± 2.31 15.1 ± 2.3 4.83 ± 0.76 157 ± 41 Garenoxacin 50 female 5.33 ± 1.15 19.1 ± 1.5 4.23 ± 0.69 176 ± 14 male 3.33 ± 2.31 5.64 ± 2.06 2.83 ± 1.03 40.8 ± 11.8 Ciprofloxacin 50 female 2.67 ± 1.15 3.93 ± 1.12 3.70 ± 1.81 24.9 ± 9.9 male 2.33 ± 1.53 3.52 ± 2.05 3.15 ± 0.36 24.2 ± 14.8 Norfloxacin 50 female 2.00 ± 0.00 4.19 ± 2.59 3.22 ± 0.27 25.0 ± 16.1 6th administration Treatment Dose Tmax Cmax T1/2 AUC0-∞ Group ( mg/kg ) (hr) ( g/mL) (hr) ( g hr/mL) male 4.67 ± 1.15 19.6 ± 3.0 4.67 ± 0.99 202 ± 19 Garenoxacin 50 female 4.67 ± 1.15 17.8 ± 1.4 3.96 ± 0.09 173 ± 22 male 3.33 ± 1.15 4.25 ± 2.79 2.49 ± 0.54 25.9 ± 9.2 Ciprofloxacin 50 female 2.67 ± 1.15 6.24 ± 1.55 2.73 ± 0.58 45.4 ± 13.9 male 1.00 ± 0.00 2.04 ± 1.04 6.19 ± 4.86 11.9 ± 3.7 Norfloxacin 50 female 3.33 ± 1.15 1.72 ± 1.10 3.99 ± 0.39 13.1 ± 8.8 Dose for all groups was 50 mg/kg. Data represent the Mean ± S.D. of 3 juvenile dogs. Pharmacokinetic parameters were calculated from the assayed plasma quinolone concentration for each animal after the first and 6th administrations.

Fig. 2. Assayed concentrations of garenoxacin, ciprofloxacin and norfloxacin in the plasma and knee joint tissue collected from juvenile beagle dogs administered garenoxacin, ciprofloxacin and norfloxacin orally, once daily, over a period of 7 days. The plasma and joint tissue samples were obtained 4 hr after final dose on Day 7. The concentrations are expressed as µg/mL or µg/g. Data are shown as the Mean ± S.D. of 3 dogs.

Vo l . 2 7 N o . 3 226

A. NAGAI et al.

4.74 µg/g for the male dogs and 18.0, 12.3 and 6.20 µg/ knee) in the ciprofloxacin group and 4 joints (shoulder, g for the female dogs, respectively. The synovialis con- elbow, hip and knee) in the norfloxacin group. The centrations, from the knee joints, in the garenoxacin, degree or severity of the histopathological change ciprofloxacin and norfloxacin groups were 13.5, 4.38 observed was significantly less for the garenoxacin and 2.08 µg/g for the male dogs and 13.1, 4.63 and group than that found in the ciprofloxacin and norflox- 4.48 µg/g for the female dogs, respectively. The syn- acin groups. In the oral toxicokinetic (TK) study, the ovia and cartilage concentrations, 4 hr after final oral AUC0→∞ values for the garenoxacin group, after the administration, in the hip joint were nearly the same as 6th administration, were from 7.8 to 17.0 times larger for the knee joint. The synovia, cartilage and synovialis for the male dogs and from 3.8 to 13.2 times larger for concentrations from the knee joint of the garenoxacin the female dogs than for the ciprofloxacin and norflox- group were from 2.0 to 6.5 times higher for the male acin treatment groups. For all of the tissues that were dogs and from 1.5 to 3.3 times higher for the female collected and assayed from the knee joints of these dogs than the quinolone values obtained from the other dogs, 4 hr after final oral administration, the garenoxa- 2 reference (ciprofloxacin and norfloxacin) groups. cin concentration was significantly higher than that for the ciprofloxacin and norfloxacin groups. Specifically, DISCUSSION the synovia levels were from 2.0 to 5.0 times higher for the garenoxacin group of the male dogs and from 1.5 to In Experiment I, no beagles from the 30 mg/kg 3.3 times higher for the female dogs; in the articular garenoxacin group exhibited signs of articular toxicity. cartilage samples, the concentrations were from 2.1 to However, in one dog from the 60 mg/kg garenoxacin 3.8 times higher for the garenoxacin group of male dogs group, very slight to slight histopathological changes and from 1.5 to 2.9 times higher for the female dogs; were detected in the articular cartilage of the shoulder, and in the synovialis specimens, the levels were from elbow and knee joints. In the 30 mg/kg ciprofloxacin 3.1 to 6.5 times higher for the garenoxacin group of the group and in the 30 and 60 mg/kg norfloxacin groups, male dogs and from 2.8 to 2.9 times higher for female very slight to severe histopathological changes in the dogs. Although the quinolone concentrations assayed articular cartilage of the shoulder, elbow, carpus, hip, from the plasma and joint tissue were significantly knee and tarsus joints were observed in all of the dogs. higher in the garenoxacin group than for the ciproflox- Signs of articular toxicity were also present in the acin and norfloxacin groups, the severity of the articular shoulder joint, elbow joint and hip joint in the 60 mg/ toxicity for garenoxacin was significantly weaker than kg garenoxacin group and in the shoulder joint, elbow that observed with ciprofloxacin and norfloxacin. joint, carpus joint, hip joint, knee joint and tarsus joint Generally, quinolones have been documented as in dogs from the ciprofloxacin and norfloxacin treat- causing blisters and erosions in the articular cartilage. ment groups. Histopathologically, the presence of Initially, blisters in the articular cartilage were chondrocyte clusters, degenerative changes in the observed and these blisters then progressed into ero- chondrocyte, rarefaction of the cartilage matrix, cavita- sions (Stahlmann and Lode 2000). Typical histopatho- tion at the cartilage layer and erosion at the cartilage logical lesions, after quinolone treatment of juvenile layer were uncovered. The AUC0→∞ values, after the animals, include fluid-filled blisters, fissures, erosions first intravenous dose was administered, for the 30 mg/ and clustering of chondrocytes (Schaad, 2000). The kg garenoxacin, ciprofloxacin and norfloxacin groups garenoxacin, ciprofloxacin and norfloxacin treatment were 164, 68.1 and 65.7 µg⋅hr/mL, respectively. The groups in this study exhibited very similar lesions in amounts of garenoxacin exposure the beagles were the articular cartilage. Therefore, the mechanism of subjected to ranged from 2.4 to 2.5 times more than the garenoxacin-induced arthropathy may be the same as specific antimicrobial exposure levels obtained from that reported previously for other quinolones. the ciprofloxacin and norfloxacin treatment groups. The exact mechanism of quinolone-induced Although the exposure levels were much higher for the arthropathy remains unexplained. However, some data garenoxacin group, the articular toxicity detected in indicate that the affinity of the drugs for magnesium is this group was significantly weaker than that found in probably the crucial initial step. Fluoroquinolones the ciprofloxacin and norfloxacin groups. form chelate complexes with divalent and trivalent cat- In Experiment II, articular cartilage changes were ions, and their affinity for magnesium is more pro- detected in 2 joints (shoulder and hip) in the garenoxa- nounced than for calcium and other minerals cin group, 5 joints (shoulder, elbow, carpal, hip and (Stahlmann and Lode, 2000). Some studies have dem-

Vo l . 2 7 N o . 3 227

Articular toxicity of garenoxacin in juvenile beagle dogs. onstrated that the ultrastructural changes in canine car- and Nair, G.B. (1997): Double-blind, random- tilage are similar after either ciprofloxacin treatment or ized clinical trial for safety and efficacy of nor- magnesium deficiency (Stahlmann, et al., 1995, 2000a, floxacin for shigellosis in children. Acta 2000b; Stahlmann and Lode, 2000). It has been Pediatr., 86, 319-320. reported that the stability constants of metal complexes Church, D.A., Kanga, J.F., Kuhn, R.J., Rubio, T.T., for several fluoroquinolones (ofloxacin, norfloxacin Spohn, W.A., Stevens, J.C., Painter, B.G., and lomefloxacin) gradually increased with increasing Thurberg, B.E., Haverstock, D.C., Perroncel, pKa values of the fluoroquinolones. The stability con- R.Y. and Echols, R.M. and The Cystic Fibrosis stant for a fluoroquinolone drug can be approximately Study Group (1997): Sequential ciprofloxacin estimated from the pKa values of the fluoroquinolone therapy in pediatric cystic fibrosis: Comparative group (Okabayashi et al., 1992). The pKa values for study vs. ceftazidime/tobramycin in the treat- ciprofloxacin and norfloxacin are 6.0 and 6.3, respec- ment of acute pulmonary exacerbations. Pediatr. tively (Kamidono, 1991). The pKa value (5.6) for Infect. Dis. J., 16, 97-105. garenoxacin is smaller than for ciprofloxacin and nor- Crumplin, G.C. and Smith, J.T. (1976): Nalidixic acid floxacin. From the interaction of fluoroquinolones and bacterial chromosome replication. Nature with metals, it might be suggested that garenoxacin has (London), 260, 643-645. a lower affinity for magnesium than ciprofloxacin and Fujii, R. (1990): Evaluation of norfloxacin in the pedi- norfloxacin. atric field. Jap. J. Antibiot., 43, 181. Garenoxacin appears to be a less potent inducer Fukuda, H., Akimoto, T. and Sakaguchi, T. (1990): of articular toxicity than fluorinated quinolones, possi- Principles for Safety Evaluation of Medical bly related to its novel structure, which lacks fluorine Drugs. pp.35-42, Chijin-syokan, Tokyo (in Jap- at the C-6 position typical of existing fluoroquinolo- anese). nes. Preclinical studies of garenoxacin suggest an Fung-Tomc, J.C., Minassian, B., Kolek, B., Huczko, improved safety profile over the closely related F(6)- E., Aleksunes, L., Stickle, T., Washo, T., quinolones. Garenoxacin demonstrated no phototoxic- Gradelski, E., Valera, L. and Bonner, D.P. ity in a guinea-pig model (vs. ciprofloxacin, norfloxa- (2000): Antibacterial spectrum of a novel des- cin and levofloxacin which did), and was a less potent fluoro(6) quinolone, BMS-284756. Antimicrob. inducer of histamine release in dogs than norfloxacin, Agents Chemother., 44, 3351-3356. ofloxacin and ciprofloxacin (Nagai et al., 1997; Furuhata, K., Soumi, K., Matsumoto, J., Horikawa, A., Furuhata et al., 1997). In a mouse model, garenoxacin Nojima, N., Arai, H., Todo, Y., Watanabe, Y. demonstrated weaker potential for central nervous sys- and Narita, H. (1997): T-3811, a novel des-F(6)- tem toxicity than trovafloxacin, ciprofloxacin and nor- quinolone: Comparative studies of the adverse floxacin (Furuhata et al., 1997). Several other reactions associated with quinolones, abstr. properties of garenoxacin suggest that the drug would F161, p.173. In Program and abstracts of the be safe for oral and intravenous administration in both 37th Interscience Conference on Antimicrobial the adult and pediatric populations. Agents and Chemotherapy. American Society In conclusion, although the assayed concentra- for Microbiology, Washington, D.C. tions of garenoxacin in both the plasma and joint tis- Gellert, M., Mizuuchi, K., O’Dea, M.H. and Nash, sues were significantly higher than for the levels of H.A. (1977): DNA gyrase; An enzyme that ciprofloxacin and norfloxacin, the observed articular introduces superhelical turns into DNA. Proc, toxicity of garenoxacin was found to be significantly Natl. Acad. Sci. U.S.A., 73, 3872-3876. weaker than that for ciprofloxacin and norfloxacin. Gootz, T.D. and Brighty, K.E. (1996): Fluoroqui- nolone antibacterials: SAR, mechanism of REFERENCES action, resistance and clinical aspects. Med. Res. Rev., 16, 433-486. Andriole, V.T. (1988): Safety overview: Toxicity, Gough, A.W., Kasali, O.B., Sigler, R.E. and Baragi, V. adverse effects and drug interactions. In the (1992): Quinolone arthropathy: Acute toxicity Quinolones, pp. 201-233, Academic Press, Cali- to immature cartilage. Toxicol. Pathol., 20, 436- fornia. 450. Bhattacharya, S.K., Bhattacharya, M.K., Dutta, D., Hampel, B., Hullmann, R. and Schmidt, H. (1997): Dutta, S., Deb, M., Deb, A., Das, K.P., Koley, H. Ciprofloxacin in pediatrics: Worldwide clinical

Vo l . 2 7 N o . 3 228

A. NAGAI et al.

experience based on compassionate use -safety Schaad, U.B., Wedgwood, J., Ruedeberg, A., Kraemer, report. Pediatr. Infect. Dis. J., 16, 127-129. R. and Hampel, B. (1997): Ciprofloxacin as Heggers, J.P., Villarreal, C., Edgar, P., Wolf, S., Klein, antipseudomonal treatment in patients with G.L., John, S.D., Desai, M. and Herndon, D.N. cystic fibrosis. Pediatr. Infect. Dis. J., 16, 106- (1998): Ciprofloxacin as a therapeutic modality 111. in pediatric burn wound infections: Efficacious Schaad, U.B. (2000): Use of the quinolones in pediat- or contraindicated? Arch. Surg., 133, 1247- rics. In the Quinolones, pp. 455-475, Academic 1250. Press, California. Jick, S. (1997): Ciprofloxacin safety in a pediatric Stahlmann, R., Förster, C. and van Sickle, D. (1993): population. Pediatr. Infect. Dis. J., 16, 130-134. Quinolones in children: Are concerns over Kamidono, S. (1991): Prostatitis, epididymitis. In the arthropathy justified? Drug Saf., 9, 397-403. Quinolones, pp. 182-188, Life Science Publish- Stahlmann, R., Förster, C., Shakibaei, M., Vormann, ing, Tokyo, Japan (in Japanese). J., Günther, T. and Merker, H.-J. (1995): Mag- Karande, S. and Kshirsagar, N.A. (1996): Ciprofloxa- nesium deficiency induces joint cartilage lesions cin use: Acute arthropathy and long-term follow in juvenile rats which are identical to quinolone- up. Indian Pediatr., 33, 910-916. induced arthropathy. Antimicrob. Agents Lipsky, B.A. and Baker, C.A. (1999): Fluoroquinolone Chemother., 39, 2013-2018. toxicity profiles: A review focusing on newer Stahlmann, R. and Lode, H. (1999): Toxicity of quino- agents. Clin. Infect. Dis., 28, 352-364. lones. Drugs, 58 (Suppl. 2), 37-42. Machida, M., Kusajima, H., Aijima, H., Maeda, A., Stahlmann, R., Kühner, S., Shakibaei, M., Schwabe, Ishida, R. and Uchida, H. (1990): Toxicokinetic R., Flores, J., Evander, S.A. and van Sickle, study of norfloxacin-induced arthropathy in D.C. (2000a): Chondrotoxicity of ciprofloxacin juvenile animals. Toxicol. Appl. Pharmacol., in immature Beagle dogs; Immunohistochemis- 105, 403-412. try, electron microscopy and drug plasma con- Nagai, A., Takahata, M., Miyazaki, M., Kawamura, centrations. Arch. Toxicol. 73, 564-572. Y., Kodama, T., Todo, Y., Watanabe, Y. and Nar- Stahlmann, R., Kühner, S., Shakibaei, M., Flores, J., ita, H. (1997): T-3811, a novel des-F(6)-qui- Vormann, J. and van Sickle, D.C. (2000b): nolone: Toxicological evaluation, abstr. F162, Effects of magnesium deficiency on joint carti- p.173. In Program and abstracts of the 37th Inter- lage in immature Beagle dogs: Immunohis- science Conference on Antimicrobial Agents tochemistry, electron microscopy, and mineral and Chemotherapy. American Society for concentrations. Arch. Toxicol. 73, 573-580. Microbiology, Washington, D.C. Stahlmann, R. and Lode, H. (2000): Safety overview. New Drug Symposium Data of Grepafloxacin (1993): In the Quinolones, pp. 397-413, Academic 41th Japanese Society of Chemotherapy (in Jap- Press, California. anese). Takahata, M., Mitsuyama, J., Yamashiro, Y., Okabayashi, Y., Hayashi, F., Terui, Y. and Kitagawa, Yonezawa, M., Araki, H., Todo, Y., Minami, S., T. (1992): Studies on the interaction of pyridone Watanabe, Y. and Narita, H. (1999): In vitro and carboxylic acids with metals. Chem. Pharm. in vivo antimicrobial activities of T-3811ME, a Bull., 40, 692-696. novel des-F(6)-quinolone. Antimicrob. Agents Richard, D.A., Nousia-Arvanitakis, S., Sollich, V., Chemother., 43, 1077-1084. Hampel, B.J., Sommerauer, B., Schaad, U.B. Takasugi, M. (1991): Clinical pharmacology of nor- and The Cystic Fibrosis Study Group (1997): floxacin -little child-. Antibiot. Chemotherap., Oral ciprofloxacin vs. intravenous ceftazidime 7, 1731-1737 (in Japanese). plus tobramycin in pediatric cystic fibrosis Wang, J.C. (1985): DNA topoisomerases. Annu. Rev. patients: Comparison of antipseudomonas effi- Biochem., 54, 665-697. cacy and assessment of safety with ultrasonog- Warren, R.W. (1997): Rheumatologic aspects of pedi- raphy and magnetic resonance imaging. Pediatr. atric cystic fibrosis patients treated with fluoro- Infect. Dis. J., 16, 572-578. quinolones. Pediatr. Infect. Dis. J., 16, 118-122.

Vo l . 2 7 N o . 3