Ocular Changes in Beagle Dogs following Oral Administration of CGS 24565, a Potential Hypolipidemic Agent
D. M. Schiavo1 and
P. Bentley
Toxicology/Pathology, Preclinical Safety, U.S. Pharmaceuticals Development, Novartis Institute for Biomedical Research, Novartis Pharmaceuticals Corporation, 59 Route 10, East Hanover, New Jersey 07936-1080
Received November 25, 1998;
accepted December 2, 1999
 |
ABSTRACT
|
---|
(11R)-N,15-dideoxo-1-deoxy-1,15-epoxy-11-hydroxy-4-0methyl-8-0-(2,2-dimethyl-1-oxopropyl)-3-[4-{(2,4,6-trimethylphenyl)methyl}-1-piperazinyl]rifamycin has been evaluated as a potential hypolipidemic agent. As part of a safety evaluation program, a 3-month oral toxicity study was performed in which CGS 24565 was administered to beagle dogs via gelatin capsules at 10, 50, or 300 mg/kg/day. Ophthalmoscopic examinations (using focal illumination and indirect opthalmoscopy) on day 83 (week 12) revealed bilateral adnexal and corneal changes affecting 5 dogs (3 males, 2 females, 300 mg/kg/day). Ophthalmoscopically, dogs from the 300 mg/kg dose level exhibited the adnexal changes characterized as ptosis, conjunctivitis, episcleritis, and relaxed membrane nictitans, while the corneal changes were characterized as posterior stromal edema (cloudy, diffuse opacity usually accompanied by deep neovascularization; the diffuse edema masked the complete evaluation of other ocular structures) and stromal infiltrates in the area of Decement's membrane (appeared to be multifocal, polymorphic changes/alterations in Decement's membrane, or endothelial swelling). No changes from normal were seen clinically in the eyes of other dogs on this experiment. In those dogs affected by the ocular changes caused by CGS 24565, a visual deficit in acuity was suspected. The corneal changes, as manifested, were suggestive of permanent, irreversible corneal damage. Subsequent ophthalmoscopic examinations performed at established intervals during weeks 15 through 26, revealed abatement of the adnexal changes, while the corneal changes, as described above, remained generally unchanged, confirming irreversibility of the corneal changes within the recovery period of 13 weeks. Light microscopy confirmed irreversible corneal neovascularization, vacuolar degeneration of the keratocytes at 300 mg/kg, and polymorphic infiltrates in the region of Decement's membrane. The results demonstrate that the cornea was the target tissue of toxicity for CGS 24565, and indicated that the findings represent a significant toxic effect. The correlation of histopathological findings support the hypothesis of the diagnosis of interstitial stromal degeneration/atrophy. The potential for a similar result to the cornea of humans does exist. Due to these changes and other toxic effects associated with this class of compound, further development was terminated.
Key Words: CGS 24565; (11R)-N,15-Didehydro-11,15-dideoxo-1-deoxy-1,15-epoxy-11-hydroxy-4-0-methyl-8-0-(2,2-dimethyl-1-oxopropyl)-3-[4-{ (2,4,6-trimethylphenyl)methyl}-1-piperazinyl]rifamycin; hypolipidemic agent; ocular; adnexal; corneal; toxicity; beagle dogs.
 |
INTRODUCTION
|
---|
The objective of this study was to determine the subacute toxicity profile of CGS 24565, a potential hypolipidemic agent, in the dog. The dog was selected for safety testing on the basis of accumulated historical data and prior experience with this species. The use of the dog conforms with the regulatory guidelines for testing in a nonrodent species. The selection of doses (10, 50, or 300 mg/kg) was based on the results obtained in previously conducted pharmacological and toxicological studies with CGS 24565. The high dose was based on the results of an acute toxicity study in dogs that was asymptomatic at 100 mg/kg and produced only apparent compound in the feces at 750 mg/kg. A rising-dose study in dogs at doses of 25, 50, 100, 200, or 400 mg/kg produced moderate to marked reductions in total cholesterol, HDL cholesterol, and triglyceride in one or both sexes at all doses and stomatocytosis of the red blood cells at all doses. The low dose reflects approximately a 30-fold multiple of a pharmacologically active dose in dogs, and is approximately twice the projected maximum human clinical dose. The mid dose (50 mg/kg) reflects approximately an equal multiple between the low and high dose, and was expected to produce intermediate pharmacological effects between these doses. In the previous studies conducted in the dog as well as other species, no ocular changes were seen. The purpose of this paper is to describe ocular changes observed in the cornea of the eyes of dogs given this hypolipidemic agent and to address the potential significance of these changes in the context of safety evaluation. The proposed mechanism of CGS 24565 causing corneal alterations is presented.
 |
MATERIALS AND METHODS
|
---|
Animals.
Thirty-two standard beagle dogs were randomly assigned by body weight to 4 groups (5/sex in Groups 1 and 4, 3/sex in Groups 2 and 3). All animals were housed individually in stainless steel cages and were maintained under standard laboratory conditions in a windowless room with a 12-h photoperiod of fluorescent lighting. Each dog was offered approximately 400 g of Purina Lab Canine Diet No. 5006 (Purina Mills, Inc.) daily. Water was provided ad libitum. The procedures involving the care and use of dogs in this experiment were reviewed and approved by Novartis Animal Care Committee.
Dosing schedule.
CGS 24565, (11R)-N,15-Didehydro-11,15-dideoxo-1-deoxy-1,15-epoxy-11-hydroxy-4-0-methyl-8-0-(2,2-dimethyl-1-oxopropyl)-3-[4-{(2,4,6-trimethylphenyl)methyl}-1-piperazinyl]rifamycin, was administered orally via hand-filled gelatin capsules to groups of purebred beagle dogs (3 or 5/sex/group) at daily doses of 10, 50, or 300 mg/kg for at least 92 consecutive days. Group 1 animals (5/sex) were similarly dosed with empty gelatin capsules.
Necropsy.
Twenty-four dogs (3/sex/group) were necropsied after approximately 13 weeks of dosing; the remaining animals were maintained without compound for an additional 91-day post-dose period and then sacrificed and necropsied.
Ophthalmology.
Ocular examinations consisting of focal illumination and indirect ophthalmoscopy (using a Fison Indirect Ophthalmoscope) were performed on all dogs once during the pre-dose period, during weeks 7 and 12 (dosing period), and during weeks 15, 17, 20, and 26 (recovery period). To facilitate examination, mydriasis was induced by administering Mydriacyl (tropicamide 1% ophthalmic solution; Alcon Laboratories).
Preparation of ocular tissue for light microscopy.
Immediately after euthanasia and exsanguination, both eyes with optic nerves were enucleated, trimmed of adnexal tissues, and fixed in buffered glutaraldehyde. Following fixation, each globe was incised 1 to 2 mm to the temporal side of the optic nerve in a parasagittal plane, perpendicular to the posterior ciliary vessels. The ocular tissues were embedded in paraffin and sections of affected and unaffected areas of the eye were stained with hematoxylin and eosin and examined microscopically.
 |
RESULTS
|
---|
Ophthalmoscopic Findings
Ophthalmoscopic examinations performed on day 83 (week 12 examinations) revealed bilateral, asymmetric adnexal and/or corneal alterations in 5 of 10 dogs from Group 4 (300 mg/kg/day). The ocular structures of all other dogs in this experiment were clinically unaffected. Adnexal changes were noted in 2 male (2/5) and 2 female (2/5) high-dose dogs and consisted of ptosis, conjunctivitis, episcleritis, and/or relaxed membrane nictitans (Tables 13

). Three male (3/5) and two female (2/5) high-dose dogs manifested corneal alterations of posterior stromal edema, in which a cloudy, diffuse opacity was observed usually accompanied by deep neovascularization, and/or multifocal, polymorphic stromal infiltrates or endothelial swelling in the region of Decement's membrane (Figs. 13

). No evidence of ocular inflammation was apparent. A visual deficit was suspected in those dogs exhibiting ocular changes; however, specific tests for visual acuity were not performed. Subsequent ophthalmoscopic examinations performed during the recovery period (weeks 15, 17, 20, and 26) revealed abatement of the adnexal changes, while the corneal lesions were generally unchanged, indicating irreversible damage (Tables 4 and 5
).

View larger version (44K):
[in this window]
[in a new window]
|
FIG. 3. Left panel, beagle cornea showing corneal alterations of posterior stromal edema. Right panel, polymorphic stromal infiltrates or endothelial swelling in region of Decement's membrane (arrows).
|
|
Gross Pathological Observations
Gross pathological examination of the eyes revealed ocular opacities in one male (1/3) each from Groups 3 and 4 (50 and 300 mg/kg, respectively) that were sacrificed at the conclusion of the dosing period. All high-dose dogs (2/2 males and 2/2 females) sacrificed after the recovery period exhibited similar findings of white discoloration of the eyes, indicating irreversible damage.
Microscopic Observations
Histological evaluation revealed minimal to moderate corneal neovascularization and vacuolar degeneration of the resident corneal cells or keratocytes in one male (1/3) each from groups 3 and 4 (50 and 300 mg/kg/day, respectively) that were sacrificed after the completion of the dosing period. The majority of the degenerative keratocytes were vacuolated, many containing pyknotic or karyorrhectic nuclei indicative of cell death (Figs. 4 and 5
). Similar microscopic findings were detected in all high-dose dogs (2/2 males and 2/2 females) sacrificed after the recovery period, indicating irreversibility. The severity of the changes appeared to be unrelated to dose, sex, or the extended recovery period. There was no microscopic evidence of corneal inflammation in any terminal- or recovery-sacrifice animal.

View larger version (117K):
[in this window]
[in a new window]
|
FIG. 5. Beagle cornea showing multifocal degeneration/atrophy of the keratocytes (interstitial stromal degeneration, arrows); x 270.
|
|
 |
DISCUSSION
|
---|
The cornea is a continuation of the sclera; there is a very abrupt transition from the dense material of the sclera to the clear tissue of the cornea. From anterior to posterior in the dog, there are 4 distinct layers within the cornea: the epithelium (which is continuous with the epithelium of the conjunctivae), the stroma, Decement's membrane, and the endothelium. The cornea of all experimental animals is subject to several congenital and hereditary anomalies, infections,and acquired abnormalities (e.g., drug-induced changes).
If systemically administered chemical agents are to affect the cornea, they must be excreted in or alter the composition of tears, or alternatively, must readily pass the blood-aqueous barrier to enter the anterior chamber (aqueous humor). The selective entrance of drugs into the aqueous is dependent on molecular size and lipid solubility. Lipid-soluble substances rapidly transverse the blood-aqueous barrier and reach equilibrium (Gerson et al., 1989
; Sakuragawa, 1976
). The corneal endothelium of the dog appears to be sensitive to CGS 24565. This single layer of cells performs several functions vital to the cornea (production of chondroitin and keratin sulfates of the stromal proteoglycans, production of Decement's membrane); however, the primary function of this layer is maintaining corneal detergescence (the state of relative dehydration). Damage (endothelial distress/decompensation) to the endothelial cells results in the cornea taking up aqueous fluid and becoming edematous. Action of the endothelium in maintaining corneal detergescence involves both metabolically active transport function as well as a physical barrier function. The proposed mechanism of CGS 24565 causing corneal alterations is presented.
The ophthalmoscopic findings in this study suggest a local effect of CGS 24565 on the cornea via either the presence of compound in, or its effect on, the aqueous humor. Given the endothelial effects observed, it is postulated that the compound or its metabolite altered the metabolic function of the corneal endothelium (i.e., maintenance of corneal detergescence). Due to this endothelial functional loss (endothelial distress/decompensation), posterior stromal swelling (edema) ensued with the foreign material, CGS 24565, causing multifocal degeneration/atrophy of the keratocytes (i.e., direct impediment of a metabolic function resulting in a lack of substrate, hypoxia, causing vacuolar degeneration; interstitial stromal degeneration). As a sequela to these changes, a new posterior collagen layer to Decement's membrane was secreted by a distressed but persistent endothelial layer (Figs. 6 and 7
). Ophthalmoscopically, these changes were manifested as multifocal, polymorphic infiltrates in the region of Decement's membrane. In the absence of any ocular inflammatory process (uveitis, iritis, etc.) deep neovascularization occurred (derived from the deep ciliary vessels), probably stimulated as a response to the degenerative keratocytes (release of biochemical mediators). There are other corneal conditions which may cause endothelial distress (i.e., keratitis profunda or deep keratitis [infectious canine hepatitis]; anterior lens luxation [mechanical trauma to the endothelium]; transmitted uveitis, keratouveitis [bluish gray, scattered, discoloration of the cornea with chronic irritation of the conjunctival and prolapse of the enlarged gland of the nictitating membrane; in this condition, the corneal edema results from the endothelium, not the epithelium]). Conditions required for the above mechanisms were not observed during ophthalmoscopic examinations conducted prior to initiation of this experiment. The correlation of histopathological findings support this hypothesis of the diagnosis of interstitial stromal degeneration/atrophy, as do the findings of vacuolar degeneration of the keratocytes and deep corneal neovascularization. To our knowledge, this is the first report of an ophthalmoscopically observable corneal alteration attributed to this class of drugs.

View larger version (104K):
[in this window]
[in a new window]
|
FIG. 7. Beagle cornea showing thickening of Decement's membrane (newly secreted posterior collagen layer to Decement's membrane, arrows); x 260.
|
|
 |
ACKNOWLEDGMENTS
|
---|
The authors acknowledge the technical assistance of Mr. Robert Spaet for the histopathology.
 |
NOTES
|
---|
A search of the literature at the time of preparation of this manuscript did not reveal any published data regarding corneal changes in dogs induced by potential hypolipidemic agents.
1 To whom correspondence should be addressed. Fax: (973) 781-5489. E-mail: donald.schiavo{at}pharma.novartis.com. 
 |
REFERENCES
|
---|
Gerson, R. J., MacDonald, J. S., Alberts, A. W., Kornbrust, D. J., Majka, J. A., Stubbs, R. J., and Bokelman, D. L. (1989). Animal safety and toxicology of simvastatin and related hydroxy- methylglutaryl-coenzyme A reductase inhibitors. Am. J. Med. 87, 28S38S.
Sakuragawa, M. (1976). Niemann-Pick disease-like inclusions caused by a hypocholesteremic agent. Invest. Ophthalmol. 15, 10221027.[ISI][Medline]