JOURNAL OF OCULAR PHARMACOLOGY AND THERAPEUTICS

Volume 20, Number 3, 2004

© Mary Ann Liebert, Inc.

Ocular Pharmacokinetics of Fluocinolone

Acetonide After Retisert™ Intravitreal
Implantation in Rabbits Over a 1-Year Period
JEAN-YVES DRIOT,1 GARY D. NOVACK,2 KAY D. RITTENHOUSE,3
CHRIS MILAZZO,4 and P. ANDREW PEARSON5
1Bausch
& Lomb, Montpellier, France
2PharmaLogicDevelopment, Inc., San Rafael, CA
3Bausch & Lomb, Tampa, FL (Present address: Pfizer, Inc., San Diego, CA

4Control Delivery Systems, Watertown, MA

5University of Kentucky, KY

ABSTRACT

Purpose: The present study was designed to examine the pharmacokinetics of a flu-
ocinolone acetonide (FA) intravitreal implant in pigmented rabbits. Methods: Pigmented
rabbits were randomly assigned to receive either a 0.5 mg or 2.0 mg FA intravitreal im-
plant (Retisert™). Four animals were sacrificed per time point (2 hours; 2 weeks; and 3,
6, 9, and 12 months after implantation) for FA intraocular levels determination. Results:
In the vitreous, concentration of FA was relatively constant from the first time point, 2
hours, through 1 year, and dose-related, approximately seven- to eight-fold greater in the
2-mg implant. Concentrations of FA were generally higher in the vitreous (11–18 and
75–146 ng/g) and retina (42–87 and 224–489 ng/g) than in the aqueous humor (0.21–1.1
and 2.6–13.0 ng/g) for the 0.5- and 2-mg implants, respectively. Urine and plasma val-
ues were below the lower limit of quantitation (200 pg/mL) for all observations, indicat-
ing no evidence of systemic absorption. Conclusions: In this rabbit study, the Retisert™
provides relatively constant levels of FA in the posterior pole, which is consistent with
previous reports of its clinical utility.

INTRODUCTION

Treatment approaches for ocular diseases affecting the posterior segment of the eye are limited
by the lack of effective delivery of drug to these regions (1). Conventional approaches for treating
inflammatory diseases affecting or originating in the posterior segment of the globe include topical
administration, subconjunctival injections, ocular injections, and systemic administration of thera-
peutic agents. While potentially effective, there are a host of disadvantages to these treatment ap-
proaches (2). For example, topical administration does not typically provide sufficient delivery of
drug to the posterior regions of the eye. All routes, including topical ones, risk the potential delivery

269
of clinically relevant concentrations of drug to the systemic circulation, with ensuing systemic toxi-
city (2). Direct intraocular injection of drugs, although providing therapeutic intraocular drug con-
centration, necessitates repeated intervention and is not a desirable approach for long-term therapy.
In an effort to provide long-term direct delivery of a therapeutic agent to the posterior pole of
the eye, Ashton and colleagues invented an intravitreal implant, designed to release drug for 8 months
(3). Prepared, using ganciclovir, the implant (Vitrasert™, Bausch & Lomb, Rochester NY) has been
used therapeutically to treat cytomegalovirus retinitis (4,5).
Corticosteroids are used to treat several serious disorders of the posterior pole, including uveitis
(6) and macular edema (7,8). Ashton and colleagues have modified their intravitreal implant tech-
nology for delivery of corticosteroids. Eyes were initially evaluated for dexamethasone (9) and then
for the less-soluble, and thus potentially longer-lasting fluocinolone acetonide (FA). A pharmacoki-
netic and toxicology study was conducted in albino rabbits using pilot implants containing 2 and 15
mg of FA. Aqueous and vitreous levels of FA were measured over approximately a 1-year period.
Mean vitreous levels of FA varied from 0.10 to 0.21 g/g over the experiment, with no statistically
significant difference over 1 year. No aqueous levels were found. The predicted life span of the 2-
mg implant was approximately 3 years (10).
The clinical utility of the FA intravitreal implant was reported in 7 eyes of 5 patients with uveitis
of various origins. Patients were observed an average of 10 months (range, 5–19 months). All eyes
had stabilized or improved visual acuity after device implantation, and 4 of 7 eyes had an improve-
ment of three lines or more (11).
The present study was designed to extend the preliminary pharmacokinetic study in a larger group
of pigmented rabbits with more examinations and with the 0.5-mg and 2-mg intravitreal implants as
currently being evaluated in clinical studies.

METHODS

Animals

The study was carried out at Huntingdon Life Sciences Inc. (East Millstone, NJ) Two groups of
24 (plus an additional 6) New Zealand Black/Satin cross rabbits (Robinson Services, Clemmons NC),
approximately 20-weeks old, were randomly assigned to receive either a 0.5 mg- or 2.0-mg FA in-
travitreal implant (Retisert,™ Control Delivery Systems, Watertown, MA; Figs. 1 and 2), bilaterally.
In the 0.5-mg implant, the diffusion orifice was in the center, whereas in the 2.0-mg implant, the dif-
fusion orifice was on the bottom, with diffusion occurring through the suture tab. Four animals were
sacrificed per time point (2 hours; 2 weeks; and 3, 6, 9, and 12 months after implantation) for FA in-
traocular levels determination. Laboratory management was under good animal husbandry, the facil-
ity was fully accredited by the Association for the Assessment and Accreditation of Laboratory An-
imal Care International, and the study was conducted in compliance with Part 58 of 21 CFR (FDA
Good Laboratory Practice Standards).

Retisert™ Implantation

Rabbits were anesthetized (35 mg/kg ketamine and 5 mg/kg xylazine, i.m.), dilated (tropicamide
drops), and the perioperative area scrubbed with povidone-iodine. A 90o superior temporal peritomy
was performed using scissors followed by a blunt posterior dissection at the episcleral plane between
the insertions of the superior and lateral rectus muscles. A scleral incision of up to 6 mm was made
parallel to the limbus. The eye was entered through the sclera with a MicroVitreoRetinal blade. The
sterile container with the intravitreal implant was opened and the implant removed. An 8-0 nylon su-
ture was passed through the base of the suture tab of the implant. The implant with the delivery cap-

270
 FIGURE 2. 2-Mg Fluocinolone Ace-
FIGURE 1. 0.5-Mg Fluocinolone Acetonide Intravitreal tonide Intravitreal Implant (Retisert™;
Implant (Retisert™; Bausch & Lomb, Rochester, NY). Bausch & Lomb, Rochester, NY).

sule was placed with the delivery capsule facing posterior (to avoid contact with the lens) and se-
cured with 8-0 nylon suture. The sclerotomy was closed with 9-0 nylon suture, and the peritomy was
closed with a 6-0 plain suture. An intramuscular injection of gentamicin sulfate (5 mg/kg) was ad-
ministered immediately after surgery, and 1% tropicamide instilled. A topical bacitracin-neomycin-
polymyxin B with 1% hydrocortisone suspension was instilled ocularly, q.i.d., for 7 days after surgery.

Examinations and Tissue Sampling

Rabbits had regular general condition and ocular examinations. Immediately after euthanasia (in-
travenous sodium pentobarbital and exsanguination) at selected intervals, a sample (100 L) of
aqueous humor from the anterior chamber was aspirated via a 25-gauge or smaller needle, then frozen
at 70°C. The globes were removed and dissected into the following tissues: lens; iris/ciliary body;
vitreous humor; retina; and retinal pigment epithelium/choroid. These tissues were also frozen at
70°C. The implants were also removed for FA level analysis at the same intervals. Blood and urine
samples were collected to assess the systemic uptake consisting of the determination of FA levels at
2 weeks; and 3, 6, 9, and 12 months after implantation.

Analytical Methods

All rabbit ocular tissues, with the exception of vitreous humor, were homogenized prior to ex-
traction. FA was determined using a validated LC/MS method. The limit of quantitation (LOQ) in
ocular tissue, plasma and urine was 200 pg/mL and, for the implant, 20 ng/g. Eyes were averaged for
analysis (n  8 at nearly all time points, n  7 at others).

RESULTS

FA concentrations are shown in Figure 3 and Table 1. In the vitreous humor, levels were rela-
tively constant from the first time point, 2 hours, through 1 year, and dose-related, approximately 7–8-
fold greater in the 2.0-mg implant. The vitreous levels seen in this study were dose-related, approx-
imately seven to eight-fold greater in the 2.0-mg implant. Concentrations of FA tended to be higher
in the vitreous (11–18 and 75–146 ng/g) and retina (42–87 and 224–489 ng/g) than in the aqueous
humor (0.21–1.1 and 2.6–13.0 ng/g) for the 0.5 and 2.0-mg implants, respectively. In posterior tis-
sues other than the vitreous, the value at 2 weeks was higher than at 2 hours. Tissue concentrations
tended to be relatively constant from 2 weeks through 12 months, corresponding globally to a zero-

271
FIGURE 3. Mean ( Standard Error of
the Mean) Fluocinolone Acetonide Con-
centrations. (A) Aqueous humor. (B)
Iris/ciliary body. (C) Lens. (D) Vitreous
humor. (E) Retina. (F) RPE/Choroid.
(G) Implants.

272
Table 1. Mean ( standard error of the mean [SEM]) Fluocinolone Acetonide Concentrations at 6 Months
and 12 Months in Ocular Tissues (ng/g)

0.5 mg 2.0 mg
Time Tissue Mean SEM Mean SEM

6 months Aqueous humor 0.21 0.10 3.39 0.69

Lens 27.87 7.53 217.34 45.39
Iris-ciliary body 35.05 14.95 72.61 20.90
Vitreous 17.10 3.37 74.54 9.76
Retina 42.28 10.69 248.29 73.72
Retinal pigment epithelium/choroid 21.93 5.21 149.98 43.82
12 months Aqueous humor 0.19 0.13 12.95 6.63
Lens 23.72 4.70 296.64 34.39
Iris-ciliary body 13.89 2.26 176.79 24.81
Vitreous 10.72 2.34 106.48 17.59
Retina 56.01 11.70 335.88 43.42
Retinal pigment epithelium/choroid 24.88 3.48 137.53 26.07

order kinetics, although there was some increase seen in the 2.0-mg treatment group at 9 and 12
months in some tissues. After 1 year, 38% of the FA in the 0.5-mg implant had been delivered,
which was within the range expected. For the 2.0-mg implant, which has since been redesigned, 65%
of the 2.0-mg implant was delivered within 1 year. Not shown are the data on urine and plasma val-
ues, which were below the lower limit of quantitation (200 pg/mL) for all observations, indicating no
evidence of systemic absorption.

DISCUSSION

In this study, we found sustained release of FA throughout the eye for at least 1 year. In the pos-
terior tissues of the eyes (vitreous, retina, and RPE/choroid), concentrations of FA were relatively
constant over 1-year, consistent with a constant release of FA from the implant. There may be an ini-
tial equilibrium period for the first several weeks after implantation. The lack of FA in either the
blood or urine suggests that the goal of local therapy was achieved—minimal systemic exposure to
the corticosteroid—thus minimizing the risk of a clinically important toxicity issue when treating se-
vere ocular inflammation.
In these pigmented rabbits, which received 0.5- or 2.0-mg FA implants, the vitreous levels were
within the range previously reported by Jaffe et al., 100–210 ng/g with the pilot 15-mg implant in al-
bino rabbits (10). Although a reversible binding of FA to melanin was previously reported (12), pig-
mentation does not seem to be an important factor in implant delivery of FA to target ocular tissues.
Within the present study, FA concentrations were similar between pigmented structures (iris/ciliary
body, retinal pigment epithelium, and choroid) and nonpigmented structures (aqueous humor, lens,
vitreous humor, and retina).
The time-course of FA concentration in the iris/ciliary body with the 0.5-mg implant is some-
what perplexing. Initially, concentrations were nearly as high with the 0.5-mg implant as with the
2.0-mg implant. Then, over the course of the year, they decreased. This could be the result of some
type of stimulation of metabolism or to the inherent biologic variability in this sample.
In order to measure FA amounts remaining in the implant, the pellet was manually extracted
from the silicone casing, a procedure that was most likely less than complete, and may have been a
source of variability, possibly resulting in an underestimation of the level of FA remaining in im-
plants at each time point. Once hydrated in vitro, it appears that it takes a month for the implant to

273
reach steady-state release rates (Data on File, Bausch & Lomb) and this finding was also observed
after in vivo intravitreal implantation in most ocular tissues. This phenomenon suggests that the amount
of FA remaining in the implant at 2 weeks was probably underestimated. Nonetheless, there is some
suggestion of more rapid release from the 2.0-mg implant than desired, it has been redesigned with
two smaller orifices, instead of one large orifice.
In conjunction with the present pharmacokinetic study, separate sets of pigmented rabbits re-
ceived 0.5-, 2.0-, or 6.0-mg FA intravitreal implants and were followed for 1 year. Ophthalmoscopic
examinations demonstrated only a mild postoperative reaction to the implant containing FA. There
was no evidence of cataracts, in spite of the presence of FA in the lens nor were there any observed
abnormalities in a host of measures, including electroretinograms and histopathology of the eye (Data
on File, Bausch & Lomb).
In this rabbit study, the Retisert™ provided constant levels of FA in the posterior pole, which is
consistent with previous reports of its clinical utility (11). Additional investigations in steroid-
responsive posterior pole inflammation are ongoing.

ACKNOWLEDGMENTS

The authors acknowledge Carol S. Auletta, M.B.A., D.A.B.T., R.A.C., of Huntingdon Life Sci-
ences for study direction; Carl W. Baker, M.D., for surgical implantation; and R. Stephens Crockett,
Ph.D., and Timothy Comstock, O.D., for scientific input.

REFERENCES

1. Stern, A.L., Taylor, D.M., Dalburg, L.A., and Cosentino, R.T. Pseudophakic cystoid macu-
lopathy: A study of 50 cases. Ophthalmology 88:942–946, 1981.

2. Leopold, I.H., and Gaster, R. Ocular inflammation and anti-inflammatory drugs. In: Kaufman
HE, Barron BA, McDonald MB, Waltman SR, eds. The Cornea. New York: Churchill Living-
stone, 1987, p. 70.

3. Ashton, P., Blandford, D.L., Pearson, P.A., Jaffe, G.J., Martin, D.F., and Nussenblatt, R.B. Re-
view: Implants. J. Ocul. Pharmacol. 10:691–701, 1994.

4. Musch, D.C., Martin, D.F., Gordon, J.F., Davis, M.D., and Kuppermann, B.D. Treatment of cy-
tomegalovirus retinitis with a sustained-release ganciclovir implant: The Ganciclovir Implant
Study Group. N. Engl. J. Med. 337:83–90, 1997.

5. Martin, D.F., Parks, D.J., Mellow, S.D., et al. Treatment of cytomegalovirus retinitis with an in-
traocular sustained-release ganciclovir implant. Arch. Ophthalmol. 112:1531–1539, 1994.

6. Cunningham, E.T., Jr., and Nozik, R.A. Uveitis: Diagnostic approach and ancillary analysis. In:
Tasman W, Jaeger EA, eds. Schwab, I.R., and Tessler, H.H., Volume eds. Duane’s Clinical
Ophthalmology. Philadelphia: J.B. Lippincott, 1998, pp. 1–25.

7. Jampol, L.M. Pharmacologic therapy of aphakic cystoid macular edema: A review. Ophthal-
mology 89:891–897, 1982.

8. Martidis, A., Duker, J.S., Greenberg, P.B., Rogers, A.H., Puliafito, C.A., Reichel, E., and Bau-
mal, C. Intravitreal triamcinolone for refractory diabetic macular edema. Ophthalmology
109:920–927, 2002.

274
 9. Hainsworth, D.P., Pearson, P.A., Conklin, J.D., and Ashton, P. Sustained release intravitreal
dexamethasone. J. Ocul. Pharmacol. Ther. 12:57–63, 1996.

10. Jaffe, G.J., Yang, C.H., Guo, H., Denny, J.P., Lima, C., and Ashton, P. Safety and pharmaco-
kinetics of an intraocular fluocinolone acetonide sustained delivery device. Invest. Ophthalmol.
Vis. Sci. 41:3569–3575, 2000.

11. Jaffe, G.J., Ben Nun, J., Guo, H., Dunn, J.P., and Ashton, P. Fluocinolone acetonide sustained
drug delivery device to treat severe uveitis. Ophthalmology 107:2024–2033, 2000.

12. Blois, M.S., and Taskovich, L. The reversible binding of some aromatic and cyclic compounds
to biopolymers in vitro. J. Invest. Dermatol. 53:344–350, 1969.

Received: November 17, 2003

Accepted for Publication: February 2, 2004

Reprint Requests: Jean-Yves Driot

Bausch & Lomb
416, rue Samuel Morse, CS 99535
34961 Montpellier, Cedex 2
France
E-Mail: jean-yves.driot@bausch.com

275