Evaluation of Triamcinolone Acetonide Following Intravitreal Injection in New Zealand White Rabbits

Abstract

The safety of intravitreally injected triamcinolone acetonide suspension (TA) was evaluated in rabbits. Each animal received 0.1 ml (1) balanced salt solution (BSS) vehicle, (2) formulation vehicle, (3) 4% TA (4-mg dose), (4) 16% TrAc (16-mg dose) or (5) 25% TA (25-mg dose) as a single intravitreal injection into the right eye. The left eyes served as untreated controls. All animals were observed for 1 month following treatment. In-life evaluations included clinical signs, body weights, slit-lamp biomicroscopic and indirect ophthalmoscopic examinations, intraocular pressure and corneal thickness measurements, and electroretinograms (ERGs). Ocular tissues were harvested following a 1-month post-treatment observation period, fixed, processed, and evaluated by light microscopy. No significant or treatment-related clinical signs were observed for any animals during the study. The opaque white test article was clearly visible in the eye for all TrAc-treated groups, and remained so throughout the study. No statistically significant differences in mean body weights were present between the control and treatment groups, though changes in body weight varied. Corneal thickness was slightly reduced for some treated groups. Intraocular pressures were not statistically significantly different from controls for any treatment group. No significant changes in ERG were evident between treatment groups or from baseline readings. Microscopically, basophilic material (presumed to be drug) was seen in the vitreous of all or most treated eyes, with accumulations in the vitreous or in clumps adjacent to the retinal surface. No pathological changes were observed in the retina or other ocular structures. Triamcinolone acetonide suspension was safe and well tolerated following intravitreal injection in New Zealand white rabbits.

Triamcinolone acetonide (TA) is a synthetic glucocorticoid with anti-inflammatory activity estimated to be eight (8) times as potent as prednisone and up to 100 times more potent than hydrocortisone acetate in some animal models. TA has found widespread off-label use in ophthalmology by intravitreal injection to treat intraocular inflammatory conditions. Clinical doses are commonly 4 mg but may range to 20 mg or more. The safety of TA has not been established by standard toxicological assessments, and published reports of intraocular TA administration in animals are few, and with mixed results. McCuen et al. (1981) administered 1 mg (0.1 ml, in balanced salt solution [BSS]) to pigmented rabbits intravitreally, and found no adverse effects, including slit-lamp and indirect ophthalmoscopy, intraocular pressure (IOP), and electroretinogram (ERGs), light or electron microscopy, followed for 3 months. Intravitreal pretreatment with TA reduced retinal edema induced by photodynamic therapy in rabbits and monkeys (Burke et al., 2004). Robinson et al. (2004) compared intravitreal KENALOG-40 (TA, 40 mg/ml, in a preservative vehicle, Bristol-Myers Squibb), with benzyl alcohol to preservative free TA (4 mg or 16 mg) in rabbits, and reported no adverse effects on ERG or retinal histopathology. Contrastingly, albino rabbits receiving TA by intravitreal injection were reported to have retinal toxicity and changes in ERGs (Perlman et al. 2003). These rabbits were treated by intravitreal injection with 0.1 mL KENALOG “pre-incubated at room temperature” and KENALOG “pre-incubated at 37°C for two weeks.” ERG changes and retinal histopathology were observed, and it was concluded that KENALOG is retinotoxic in albino rabbits, with toxicity related to storage conditions, including the elevated intraocular temperature. Subretinal administration of TA was found to improve choroidal neovascularization in the rat model (Lai et al. 2004), though retinal toxicity (ERG) was observed at high doses. Subretinal delivery of TA (25 to 50 mul, 3 mg/ml) to cynomolgus monkeys did not appear to have any toxic effects on the cellular organization of the retina (Tu et al. 2004). Recently, Wills et al. (2005) reported that TA (KENALOG–40, 2 mg/eye) was well tolerated and did not result in any toxic effects on the ocular structures in general or to the retina in particular when administered by intravitreal injection to male cynomolgus monkeys every 4 weeks for 3 months. Differences in TA dose, product composition, and species may influence effects of TA on the eye.

The present study was conducted to assess the intraocular safety of triamcinolone acetonide, over a wide dose range, in an unpreserved formulation in New Zealand white rabbits.

MATERIALS AND METHODS

Animals

New Zealand white rabbits obtained from Myrtle’s Rabbitry (Thompson Station, TN), weighing 3.7 to 4.6 kg and approximately 7 to 8 months of age, were used.

Each animal was permanently tattooed in one ear with a unique identification number. The animals were maintained on a measured amount of LabDiet Certified High Fiber Rabbit Chow no. 5325. Drinking water was available, ad libitum, via the automatic watering system.

Animals were maintained and treated in accordance with the National Institutes of Health Guide for Care and Use of Laboratory Animals and with the procedures approved by the Institutional Animal Care and Use Committee. The laboratory is AAALAC (Association for Assessment and Accreditation of Laboratory Animal Care-International) accredited, and animal use was in accordance with the American College of Toxicology’s Policy Statement on the Use of Animals in Toxicology.

Drugs

Triamcinolone acetonide, United States Pharmacopeia (USP) grade (Pharmacia & Upjohn), was used in the preparation of dosing suspensions. The suspensions were prepared as sterile and nonpyrogenic, with concentrations of 40, 160, and 250 mg/ml TA in a proprietary formulation (a polymeric suspension with tonicity and buffering agents), and provided in individual sealed glass vials. No preservative was added. Vehicle groups consisted of the formulation without triamcinolone acetonide, and a BSS (balanced salt solution; Alcon Laboratories) group.

Experimental Design

Forty (40) normal, healthy rabbits (20 males, 20 females) were randomly assigned to one of five control or treatment groups, as presented in Table 1.

Anesthesia was induced with ketamine (30 mg/kg) and xylazine (6 mg/kg) given subcutaneously (SC), and 1–2 drops of ALCAINE (0.5% proparacaine HCl; Alcon Laboratories), were applied topically to provide topical anesthesia of the globe prior to the intravitreal injection procedure. One drop of Tobrex (tobramycin ophthalmic solution; Alcon Laboratories) was applied topically to provide prophylaxis against infection. The fellow eye (OS) was treated topically with BSS or an artificial tear to prevent drying during the procedure.

Intravitreal injections were performed using an operating microscope to visualize the eye. The head was draped for sterile surgery and the eyelids retracted with a wire speculum. The globe was immobilized using ocular forceps, and a hypodermic needle (27 gauge, with syringe) was passed through the sclera 3 to 5 mm posterior to the temporal limbus. The needle is angled posteriorly to avoid contact with the lens, and 0.1 ml of the vehicle or test article was injected into the vitreous. No vitreous humor was aspirated prior to injection.

The control or test articles were administered by a single injection into the vitreous chamber of the right (OD) eye of each animal. Intravitreal injections were done in replicate order by group and sex so that the first male injected is assigned to group 1M, the first female to group 1F, the second male to group 2M, the second female to group 2F, etc.

Study Evaluations

Observations

All animals were observed twice daily (am and pm) throughout the study for morbidity, moribundity, general well being, and overt pharmacotoxic signs. All findings were recorded at the time observed. Each animal was examined in detail at prescreen and twice weekly.

Individual body weights were determined for all animals prior to the first treatment (prescreen), weekly during the study, and at necropsy.

Ophthalmoscopic Evaluations

The eyes, both right and left, of all animals were examined by slit-lamp biomicroscopy at prescreen and weekly thereafter. The slit-lamp was used to evaluate the conjunctiva, cornea, anterior chamber, light reflex, iris and lens. Slit-lamp scores were recorded in accordance with the Hackett-McDonald method (Hackett and McDonald 1996).

Indirect ophthalmoscopic examinations consisted of evaluation of the fundus of each eye with respect to the optic nerve head (ONH) characteristics, vascular pattern (retinal and choroidal), and pigmentation/coloration characteristics, and were conducted prior to the initiation of dosing (prescreen), at study days 15 and 22, and during the final week of the study.

Intraocular Pressure

Intraocular pressure (IOP) measurements were obtained from both eyes of each animal prior to initiation (prescreen), at study day 9, and during the final week of treatment. All IOP measurements were taken at approximately the same time of day throughout the study. A Mentor Model 30 Classic pneumotonograph was used.

Corneal Pachymetry

A PACHETTE (DGH) ultrasound pachymeter was used to obtain central corneal thickness measurements from both the right and left eye of each animal prior to initiation of dosing (pre-screen) and during the final week of treatment. Three readings were obtained from each eye at each observation. Pachymetry measurements were taken at approximately the same time of day throughout the study.

Electroretinograms

Electroretinograms (ERGs) were obtained for each animal from treated eye (OD) at prescreen and near the end of the study. Rabbits were dark-adapted for 1 h, anesthetized with ketamine (30 mg/kg) and xylazine (6mg/kg), and the pupils were dilated with mydriacyl 1%. Each animal was then placed in a restraining box and the electrodes were positioned as follows: Jet contact lens electrode (active, sterile, disposable) on the anesthetized (alcaine 0.5%) cornea, gold disk electrode contralaterally in the mouth (reference), and platinum needle electrode ipsilaterally on the ear (ground). Gonioscopic solution was used to secure proper connection between the active electrode and the cornea.

ERG data were collected with a computer data acquisition system using Polyview software package (Astromed, Grass Instruments). The amplification of 20,000× and bandpass of 1 to 1000 Hz were used. Mini-Ganzfeld stimulator, MGS-2 (LKC Technologies), served as the stimulator.

Three waveforms were collected. Scotopic recordings consisted of maximal rod response obtained at flash intensity of −1 log cd/m²/s, whereas mixed rods and cones response was gathered at 1 log cd/m²/s. The responses were averaged from five recordings obtained at flash interval of 10 s. Photopic ERG (10-min light adaptation, background light of 30 cd/m²) was an average of 20 recordings collected at flash intensity of 1 log cd/m²/s and interval of 2 s. Retinal function was evaluated based on the amplitude and peak time measurements of four parameters: maximal rod b-wave, a-wave, mixed b-wave, and cone b-wave.

Pathology

Macroscopic

At the completion of the study period, all animals were sacrificed by intravenous injection of a sodium pentobarbital based euthanasia solution. Animals were examined carefully for external abnormalities. The eyes and adnexa (including the lacrimal glands) were examined in situ, excised and examined for abnormalities, fixed in Davidson’s fixative, rinsed, and preserved in 10% neutralbuffered formalin.

Microscopic

The eyes and adnexa from all animals were submitted for standard histologic processing and microscopic evaluation. Paraffin-embedded sections were prepared on glass microscopic slides, and hematoxylin and eosin stained sections of eyes, eyelids, nictitating membranes, harderian glands, and lacrimal glands were examined by light microscopy. Inflammatory, hyperplastic, and degenerative changes were graded on a severity scale of 1 (minimal) to 5 (severe).

RESULTS

Drugs

Prestudy analyses of the strength and identity of the test article, triamcinolone acetonide, showed mean concentrations of 101% to 115% of target concentrations. Triamcinolone acetonide was not present in the vehicle suspension. All prepared lots of vehicle or test article passed global sterility (U.S., European, and Japanese Pharmacopoieae) and the bacterial endotoxin test (LAL), with the exception of the endotoxin test for the 160- and 250-mg/mL suspensions, which could not be tested due to the thickness or opaqueness of these test articles.

Observations

No mortalities occurred during the treatment phase of the study. No significant pharmacotoxic signs or effects on general health were observed. Following intravitreal injection, the opaque white test article was clearly visible for all TA-treated eyes, and remained so throughout the study. General observations included occasional scabs, lesions, scratches, or hairloss on the back or dewlap of some animals. Bruising on ears and pustules in the urogenital area were also observed for a few animals. These general findings were minor in nature, are commonly seen in rabbits, did not occur with a dose related incidence of severity, and are not considered related to the treatment.

Body Weight

Group mean body weights were generally slightly reduced among groups receiving drug, as compared with the BSS and vehicle controls, though there were no statistically significant differences. Mean body weights are summarized in Table 2. Decreases in individual animal weights were evident for most animals at the day 8 interval, and body weights of animals of the BSS and vehicle-control groups tended to remain stable throughout the study.

This is expected for animals of this age when maintained on restricted feed. Animals of the triamcinolone acetonide-treated groups tended toward slightly reduced body weight at termination, as compared with pretest, with males more affected than females. All males of the triamcinolone acetonide-treated groups (12/12) had lower body weights at day 36 than at pretest, whereas 7/12 females had lower weights, 4 were the same and 1 was increased for the same intervals. Rabbits are highly sensitive to the systemic effects of corticosteroids, and this change in body weight, though slight, is considered to reflect this effect.

Ophthalmic Evaluations

Slit-Lamp Biomicroscopic Examinations. The presence of drug in the vitreous was reported for the right eyes of all animals receiving TA suspension (groups 3 to 5) (Figure 1). This was recorded at examination on study days 8, 15, and 36. The presence of drug in the vitreous changed very little over the course of the study for these animals. A typical description of drug in the treated eye was at slit-lamp biomicroscopic examination is “Drug present in superior nasal and temporal quadrants and in inferior temporal quadrant, OD” at study day 8, and “drug is present scattered throughout anterior vitreous—OD.” Drug was visible grossly upon observation with the unaided eye.

Minimal to moderate conjunctival congestion (score = 1 or 2) was observed in both eyes of all animals of both control and treated group throughout the study. Minimal discharge (score = 1) was observed sporadically among animals of the vehicle control and test article treated groups, but were distributed between treated (OD) and untreated (OS) eyes, and not considered related to treatment. No effects were seen on cornea, aqueous flare, iris or light reflex.

Lens abnormalities were observed in a few animals during the study, including one cataract in the left, untreated eye, and one subcapsular cataract attributed to a lens capsule nick at the time of injection. One animal of the mid-dose group had a posterior capsular cataract in the treated, right eye.

Indirect Ophthalmic Examinations. The examinations performed prior to the start and at the end of the study revealed no abnormalities of the fundus. The presence of drug in the vitreous was recorded at indirect ophthalmic examinations on study day 22, while being recorded at other intervals in association with the slit-lamp examinations at study days 15 and 36. No abnormalities of the fundus were noted in TA-treated or control eyes.

Corneal Thickness

Treatment related differences in corneal thickness were apparent in the TA-treated eyes (Table 3). Corneal thickness in the right eyes of male rabbits of groups 3 and 4 was statistically significantly decreased, as compared with BSS controls. The mean corneal thickness for group 5 males was also decreased, compared with BSS controls, but this difference was not statistically significant. Differences from pretest thickness was also evident for these groups, though these were not compared statistically. Mean corneal thickness increased (about 5%) for the BSS and vehicle-control groups over the study period, whereas mean corneal thickness was reduced in groups treated with TA. The magnitude of this reduction was consistently 5% to 7% for the treated groups, and was not dose related.

Intraocular Pressure

Mean intraocular pressure (IOP) data are presented in Table 4. Treatment-related differences in IOP were not apparent in the TA-treated eyes. Mean IOP for the right eyes of male rabbits of groups 4 were statistically significantly decreased at day 9, as compared with BSS controls. No other statistically significant differences were seen during the study. Mean IOPs did not appear to be significantly affected by intravitreal administration of TA.

Electroretinograms

Electroretinogram (ERG) data are summarized in Table 5. No treatment-related differences in ERGs were apparent in the TA-treated eyes in any of the evaluated parameters. At 1 month, retinal function did not appear to be significantly affected by intravitreal administration of TA as measured with full-field ERG.

Microscopic

Evaluation of all ocular tissues from all animals revealed treatment-related “basophilic material” in the vitreous of animals receiving triamcinolone acetonide (Figure 2). The incidence of this finding was 6/8, 8/8, and 8/8 for the 4-, 16-and 25-mg TA groups, respectively. This finding consisted of accumulations of morphologically distinct amorphous to finely granular basophilic material presumed to be test article. Mild to moderate amounts of this material appeared in clumps adjacent to the retinal surface, with clusters sometimes present in the vitreous body. Capsule perforation and mild cataract was observed for one rabbit, for which cataract and possible capsule nick was previously identified. No adverse effects were observed in the retinal structures, or for any tissues of the posterior segment. Other findings were minor and considered spontaneous and common to animals of this age and species, and unrelated to test article administration.

DISCUSSION

Nonclinical and clinical studies have demonstrated that triamcinolone acetonide is effective in reducing ocular inflammation. Reports of the intraocular use of TA for the treatment of ophthalmic conditions such as macular edema, choroidal neovascularization, uveitis and central retinal vein occlusion have increased in recent times (Moshfeghi et al. 2003; National Eye Institute 2004), though no products are currently approved for these indications. Clinically, this commercial product is commonly used at a volume of 0.1 mL, delivering 4 mg of drug into the vitreous, though doses as high as 40 mg have been reported. The commercially available drug product contains a preservative, benzyl alcohol, with potential for retinal toxicity.

Adverse ophthalmic findings associated with clinical corticosteroid use, e.g., increased intraocular pressure and cataract, are widely recognized. Increased IOP is encountered with about 30% of patients using steroids (Bartlet, Wolley, and Adams 1993; Carnahan and Goldstein 2000; Sung et al. 2004), and is reported to develop about 2 months after administration of TA (Jonas et al. 2003). Cataract is reported in about up to 38% of patients receiving systemic corticosteroids (Carnahan and Goldstein 2000), usually manifesting after 2 months to 1 year of exposure. The absence of significant effects of intravitreal injection of TA on intraocular pressure or cataract in rabbits seen with the present study is consistent with the delayed onset of such findings. Longer-term studies in animals will be needed to confirm the relevance of these effects (or lack thereof) in animals and to assess this potential.

Other findings that have been reported, such as retinopathy in the rabbit (Perlman et al. 2003) and endophthalmitis (Moshfeghi et al. 2003), may be related to changes in drug substance (such as degradation) or to components or characteristics of an injected formulation (such as purity, preservatives, pH, osmolarity, surfactants, suspending agents, pyrogenicity, etc.). Quality of the pharmaceutical product, including compatibility of the formulation with the intraocular environment, as well as acceptability of individual ingredients, may be of paramount importance in assuring safe use in patients. Hida et al. (1986) reported retinal toxicity in pigmented rabbits following intravitreal injection of vehicles of some commercially available corticosteroids, and the authors associated damage mostly with preservatives (e.g., benzalkoinium chloride) in product formulations. Benzyl alcohol is present in KENALOG as a preservative (0.99%). Intraocular use of 2% benzyl alcohol in saline in cataract surgery caused severe keratopathy, and intraocular injection (aqueous) of a similar solution was significantly toxic in rabbits, with recovery requiring 2 weeks (Grant 1986). Morrison et al. (2004) intravitreally injected New Zealand rabbits with benzyl alcohol at concentrations of 0.0073% to 0.733%, and reported that higher concentrations were toxic to the retina. Changes were seen largely in the outer retina, including outer segments and photoreceptors. Exudative retinal detachment was seen at higher concentrations.

The slightly reduced corneal thickness observed in eyes of animals treated with triamcinolone acetonide in this study is consistent with studies using topical corticosteroids in this laboratory. A reduction in thickness of 5% to 10% is typically seen in rabbit corneas treated with topical corticosteroids, and mean reductions in triamcinolone-treated eyes in this study ranged from 5.0% to 6.6%.

Published clinical experience suggests triamcinolone acetonide has been safe and effective for intraocular use in patients. This intravitreal study in rabbits conducted with triamcinolone acetonide suspension at doses of up to 25 mg confirms this, though longer-term assessments are appropriate to evaluate the effects on intraocular pressure and the lens, as well as the retina.