Scott M. Whitcup, Chi-Chao Chan, Debbie A. Luyo, Peng Bo, and Qian Li
Purpose. Allergic conjunctivitis is a common condition caused by a mast cell-mediated hyper-sensitivity reaction to immunoglobulin E-bound allergens. The purpose of this study was toinvestigate the effect of topical cyclosporine A on the development of mast cell-mediatedconjunctivitis in mice.
Methods. Allergic conjunctivitis was induced in C57BL/6 mice by topical applications of com-pound 48/80, a mast cell degranulating agent. In two separate experiments, mice were treatedwith topical cyclosporine A (0.05%, 0.2%, or 0.4%), prednisolone acetate 1%, or phosphate-buffered saline. Twenty-four hours after compound 48/80 instillation, the number of neutro-phils, eosinophils, lymphocytes, macrophages, and the number of preserved goblet cells andundegranulated mast cells in the conjunctiva were counted by a masked observer.
Results. In both experiments, treatment with all three doses of cyclosporine A resulted in astatistically significant reduction in the number of infiltrating neutrophils and eosinophilscompared to saline-treated controls. There was no significant difference in the treatmenteffect of cyclosporine and prednisolone acetate. In addition, there was increased preservationof goblet cells in the cyclosporine A-treated animals. Immunohistochemical staining showeda reduction in infiltrating lymphocytes and a smaller reduction in infiltrating macrophagesin animals treated with cyclosporine compared to saline-treated controls.
Conclusions. Topical cyclosporine A was effective in inhibiting the development of mast cell-mediated allergic conjunctivitis in mice. This study suggests that topical cyclosporine A maybe effective in treating allergic conjunctivitis in humans. Invest Ophthalmol Vis Sci.1996; 37:2686-2693.
Allergic conjunctivitis is a common condition charac-terized by itching, hyperemia, and chemosis. The dis-order usually is triggered by airborne allergens suchas pollens and molds. Studies on the pathogenesis ofallergic conjunctivitis suggest that the mast cell is criti-cal for the development of the condition.1'2 Allergensdissolve in the tear film and bind to the immunoglobu-lin E attached to mast cells. This leads to a type Ihypersensitivity reaction induced by mast cell degran-ulation and a release of inflammatory mediators, in-cluding histamine, platelet-activating factor, eosino-phil chemotactic factors, and prostaglandins. Treat-ment for this disorder includes antihistamines,vasoconstrictors, corticosteroids, and mast cell stabiliz-ing agents; however, many patients are intolerant ofor resistant to these therapies.
From the National Eye Institute, National Institutes of Health, Bethesda, Maryland.Submitted for publication April 16, 1996; revised June 13, 1996; accepted July 26,1996.Proprietary interest category: Cc5.Reprint requests: Scott M. Whitcup, National Eye Institute, 10 Center Drive,Building 10, Room ION 202, Bethesda, MD 20S92-1S5S.
Recently, cyclosporine has been shown not only toinhibit the release of mast cell mediators, such as hista-mine, but to suppress mast cell-leukocyte cytokine cas-cades.34 The purpose of this study was to investigate theeffect of topical cyclosporine A (CsA) on an animalmodel of allergic conjunctivitis induced by compound48/80. Compound 48/80 is a condensation product offormaldehyde with paramethoxyphenylethylamine thatreliably induces the release of chemical mediators intrie mast cell granules. Clinically, conjunctival erythema,chemosis, and mucous discharge develop from 15 to 60minutes after the instillation of compound 48/80, andthese clinical findings consistently correlate with histo-logic findings of conjunctival infiltration with neutro-phils, macrophages, lymphocytes, and eosinophils.5
METHODS
AnimalsFemale C57BL/6 mice, 6 to 8 weeks of age, were ob-tained from Charles River (Raleigh Durham, NC) and
PBS = phosphate-buffered saline; CsA = cyclosporine A; Pred = prednisolone acetate 1%.* Data are mean ± standard error.
were kept under standard pathogen-free conditions.All studies adhered to the ARVO Statement for theUse of Animals in Ophthalmic and Vision Research.
Induction of Allergic Conjunctivitis
Mice were anesthetized with methoxyflurane inhala-tion (Metofane; Pitman-Moore, Mundelein, IL). Al-lergic conjunctivitis was induced by the instillation of
2 mg of compound 48/80 (Sigma Chemical, St. Louis,MO) dissolved in 5 //I of phosphate-buffered saline(PBS) into both eyes as previously detailed.6
Treatment Protocol
In two separate experiments (n = 48 for experiment1, and n = 14 for experiment 2), animals were treatedevery 6 hours with topical CsA oil-water emulsion
80-
OCD
CDQ.(0
200
150-
100-
ow. 5 0 -3CD
T•III
*T"
*
T• *T
• • • • •£• : •*•
*
>::;rr;:;:
mo.
1
LOO
d
w
i
d
wO
i
d
wO
i
£Q.
O
Coo<D
CO<DQ.CO
6 0 -
40-
Q.
2 20-
a> W
C/>CD0 .
Bod(0O
CMdM
o
•a
to
oFIGURE l. Number of infiltrating neutrophils. (A) Experiment 1. Compared to phosphate-buffered saline (PBS)-treated controls (n = 10), there was a significant decrease in infiltratingneutrophils into the conjunctiva of animals treated with topical cyclosporine A (CsA) 0.05%(n = 10), CsA 0.2% (n = 10), CsA 0.4% (n = 8), or prednisolone acetate (Pred) (n = 10).*P < 0.0001. (B) Experiment 2. Again, compared to PBS-treated controls (n = 3), therewas a significant decrease in infiltrating neutrophils into the conjunctiva of animals treatedwith topical CsA 0.05% (n = 3), CsA 0.2% (n = 3), CsA 0.4% (n = 2), or pred (n = 3). *P= 0.0006; **P = 0.0015; ***P = 0.0023. Bars = standard error.
FIGURE 2. Histologic section of the conjunctiva from an animal treated with topical phos-phate-buffered saline (A) and cyclosporine 0.4% (CsA) (B). In the cyclosporine-treated eye,there was significantly less neutrophil infiltration into both the conjunctival epithelium (E)and the substantia propia (SP), less chemosis, and better preservation of the conjunctivalarchitecture. Hematoxylin and eosin. Original magnification, X40.
(0.05%, 0.2%, or 0.4%; Allergan, Irvine, CA), predni-solone acetate (pred) 1% (pred forte; Allergan), orPBS; each animal received drops 0, 6,12, and 18 hoursafter compound 48/80 administration.
Histologic Analysis
Mice were killed 24 hours after compound 48/80 ad-ministration. The lower lid tarsal and bulbar conjunc-
tiva attached to the right eye were removed and fixedwith glutaraldehyde 4% for 30 minutes and then trans-ferred into 10% buffered formalin for 24 hours. Theconjunctiva and the globe were embedded in methac-rylate, and 4-/̂ m vertical sections through the pupil-lary-optic nerve head plane were stained with hema-toxylin and eosin for neutrophils and eosinophils, pe-riodic acid-Schiff for goblet cells, and toluidine blue
.2O
oQ.
ao
oUJ
A
1COmQ.
i
usoo
CO
a
d<a
a>
<(0O
C
oo0)
CO<D
Q .
CO
Q.Oc
"35oUJ
HGURE 3. Number of infiltrating eosinophils. (A) Experiment 1. Compared to phosphate-buffered saline (PBS)-treated controls (n = 10), there was a significant decrease in infiltrat-ing eosinophils into the conjunctiva of animals treated with topical cyclosporine A (CsA)0.05% (n = 10), CsA 0.2% (n = 10), CsA 0.4% {n = 8), or prednisolone acetate (Pred) (n= 10). *P < 0.0001; **P = 0.0011. (B) Experiment 2. Again, compared to PBS-treatedcontrols {n = 3), there was a decrease in infiltrating eosinophils into the conjunctiva ofanimals treated with topical CsA 0.05% (n = 3), CsA 0.2% (TO = 3), CsA 0.4% (n = 2), orpred (n = 3). *P = 0.038. Bars = standard error.
FIGURE 4. Number of preserved goblet cells. (A) Experiment 1. Compared to phosphate-buffered saline (PBS)-treated controls (n = 10), there was an increase in preserved gobletcells in the conjunctiva of animals treated with topical cyclosporine A (CsA) 0.05% {n =10), CsA 0.2% (n = 10), CsA 0.4% (n = 8), or prednisolone acetate (Pred) (n = 10). *P= 0.0240. (B) Experiment 2. Again, compared to PBS-treated controls (n = 3), there wasa an increase in preserved goblet cells in the conjunctiva of animals treated with topicalCsA 0.05% (n = 3), CsA 0.2% (n = 3), CsA 0.4% (n = 2), or pred {n = 3). *P = 0.034.Bars = standard error.
for mast cells. The number of infiltrating neutrophils,eosinophils, preserved goblet cells, and undegranu-lated or partially degranulated mast cells were countedby a masked observer.
The lower lid tarsal and bulbar conjunctivaattached to the left eyes from animals treated withtopical CsA, pred, or PBS were removed, embeddedin optimum cutting temperature (OCT) compound(Miles Laboratory, Naperville, IL), and immediatelysnap frozen in a dry ice and methylbutane bath. Six-micrometer vertical (pupillary-optic ) frozen sectionswere prepared on lysine-coated glass slides, and immu-nohistochemical staining was performed using an avi-din-biotin-complex technique.7 Primary antibodiesincluded mouse monoclonal antibodies against Lyt2(CD8+ T lymphocytes), L3T4 (CD4+ T lymphocytes)(Becton Dickinson, Mountain View, CA), and M l /70.15 (macrophages) (Sera Laboratory, Westbury,NY). Rat immunoglobulin G (Sigma Chemical, St.Louis, MO) was used as the control primary antibody,and biotin-conjugated goat anti-rat immunoglobulinG (American Qualex, La Mirada, CA) was used as thesecondary antibody. Avidin-biotin-peroxidase com-plex (Vector Laboratory, Burlingame, CA) was ap-plied, and sections were developed in diaminobenzi-
dine. The number of infiltrating lymphocytes andmacrophages per histologic section were counted bya masked observer.
Statistical Analysis
To control for multiple comparisons, the number ofinfiltrating neutrophils, eosinophils, preserved gobletcells, and undegranulated mast cells in the conjunc-tiva of animals treated with the three doses of topicalCsA, pred, or PBS were compared using analysis ofvariance and the Fisher's protected least significantdifference test. The number of infiltrating lympho-cytes and macrophages between animals treated withtopical CsA 0.4% or PBS were compared using anunpaired Hest. The null hypothesis that there was nodifference in the number of cells between groups wasrejected at a P < 0.05.
RESULTS
The numbers of infiltrating neutrophils and eosino-phils, goblet cells, and undegranulated mast cells arelisted in Table 1. Figure 1 shows the effect of topicalCsA therapy on neutrophil infiltration into the con-junctiva. In the first experiment, compared to PBS-
oFIGURE 5. Number of undegranulated or partially degranulated mast cells. (A) Experiment1. Compared to phosphate-buffered saline (PBS)-treated controls (n = 10), there was nosignificant difference in the number of mast cells counted in the conjunctiva of animalstreated with topical cyclosporine A (CsA) 0.05% (n = 10), CsA 0.2% (n = 10), CsA 0.4%(n = 8), or prednisolone acetate (Pred) (n = 10). (B) Experiment 2. Again, compared toPBS-treated controls (n = 3), there was no significant difference in undegranulated orpartially degranulated mast cells in the conjunctiva of animals treated with topical CsA0.05% (n = 3), CsA 0.2% (n = 3), CsA 0.4% (n = 2), or pred (n = 3). Bars = standarderror.
treated animals, treatment with topical CsA at all threedoses or with pred significantly reduced the numberof infiltrating neutrophils into the conjunctiva 24hours after compound 48/80 instillation (P < 0.0001for each treatment group) (Fig. 1A). There was nostatistically significant difference in the number of in-filtrating neutrophils between animals treated withany dose of topical CsA and animals treated with pred.In addition, there was no statistically significant differ-ence in the number of infiltrating neutrophils amongthe three doses of topical CsA. These results were con-firmed when the experiment was repeated (Fig. IB).Again, treatment with all three doses of topical CsAor with pred significantly reduced the number of infil-trating neutrophils. Figure 2 shows representative his-tologic sections of the conjunctiva from an animaltreated with PBS (Fig. 2A) and with CsA 0.4% (Fig.2B). There is not only less neutrophil infiltration inthe CsA-treated eye but also less chemosis and betterpreservation of the normal conjunctival architecture.
Similar treatment effects were found on infiltrat-ing eosinophils (Fig. 3). In the first experiment, treat-ment with all three doses of topical CsA and with predsignificantly reduced the number of infiltrating eosin-ophils into the conjunctiva (P < 0.001 for each treat-
ment group) (Fig. 3A). Again, no statistical differenceswere found between the treatment effect of the threedoses of topical CsA compared with each other orwith pred. These results were confirmed when theexperiment was repeated (Fig. 3B), although, becauseof the small numbers of animals, only the decreasein infiltrating eosinophils in the pred-treated groupreached statistical significance (P = 0.038).
Figure 4 shows the number of goblet cells identi-fied in the conjunctiva. In the first experiment, therewas a trend toward increased preservation of gobletcells in animals treated with topical CsA or pred (Fig.4A). The greatest number of preserved goblet cellswas found in animals treated with the highest doseof topical CsA (0.4%). The increase in goblet cellscompared to PBS-treated animals was statistically sig-nificant only for the CsA 0.4% group (P = 0.02). In thesecond experiment, there was again a trend towardincreased goblet cell preservation in animals treatedwith topical CsA or pred (Fig. 4B). The increase ingoblet cells compared to PBS-treated animals reachedstatistical significance only for the topical CsA (0.2%)group (P= 0.034).
The number of undegranulated or partially de-granulated mast cells remaining in the conjunctiva
Topical Cyclosporine for Conjunctivitis
25
O0)
CO
20-
0) 1 5 -
a.CO
o 10-
ooa 5-
0
T T
m mCOCQQ.
inod(0
O
CMd(0
o
d
O
•o
£
FIGURE 6. Number of infiltrating lymphocytes. Compared tophosphate-buffered saline (PBS)-treated controls (n = 4),there was a significant decrease in the number of infiltratinglymphocytes in the conjunctiva of animals treated with topi-cal cyclosporine A (CsA) 0.05% (re = 3), CsA 0.2% (n = 5),CsA 0.4% (n = 4), or prednisolone acetate (Pred) (n = 4).*P = 0.0033; **P = 0.0091; ***P = 0.0192; ****P = 0.03.Bars = standard error.
were counted to ensure an equal degranulating effectof compound 48/80 in each group (Fig. 5). In thefirst experiment, there was no difference in mast cellscounted in the conjunctiva of animals treated withPBS, topical CsA, or pred (Fig. 5A). These findingswere confirmed in the second experiment; no statisti-cally significant differences in mast cells were foundin any group (Fig. 5B).
Immunohistochemical staining of the conjunctivashowed a statistically significant decrease in the num-ber of infiltrating lymphocytes in animals treated withtopical CsA or pred when compared to controls (Fig.6). There also appeared to be a decrease in the num-ber of infiltrating macrophages in animals treated withthe higher doses of topical CsA (0.2% and 0.4%) com-pared to pred-treated animals and PBS-treated con-trols (Fig. 7), although the differences were not statis-tically significant.
DISCUSSION
Our data show that the topical application ofcyclosporine effectively inhibits a mast cell-mediatedmodel of allergic conjunctivitis. Cyclosporine reducedthe number of neutrophils, eosinophils, and lympho-
2691
cytes infiltrating the conjunctiva 24 hours after com-pound 48/80 instillation. There was also a trend to-ward decreased infiltration of macrophages in CsA-treated animals, although the results did not reachstatistical significance. Furthermore, CsA appeared toimprove the preservation of goblet cells, consistentwith the finding of decreased inflammation.
We also assessed the number of undegranulatedand partially degranulated mast cells in each sectionof conjunctiva to assure an equal effect of compound48/80. Compound 48/80 has been shown to inducea predictable degranulation response of mast cells.Maximal degranulation occurs within the first hour ofa single topical dose of compound 48/80, leading tofull degranulation in approximately 31% of mastcells.8 In our study, there was no difference in thenumber of undegranulated or partially degranulatedmast cells in any of the treatment groups or controls,suggesting an equal degranulating effect of com-pound 48/80.
Although there was not a well-defined dose-re-sponse relationship between the concentration of CsAused and the number of infiltrating cells, this may beexplained by the efficacy of even the lowest concentra-tion of the drug used in this study (0.05%). Future
300
o'£• 250OCD
COw 200
CO<DO )CO
aoh.oCO
150-
100-
50-
-
-
Tm
:¥:•
TT
COCDQ.
\^
od
CMd0)O
NP
d(0O
•o0)Qu
oFIGURE 7. Number of infiltrating macrophages. Comparedto phosphate-buffered saline (PBS)-treated controls (re =5), there was a small decrease in the number of infiltratingmacrophages in the conjunctiva of animals treated thehigher doses of topical cyclosporine A (CsA):CsA 0.2% (re= 3) and CsA 0.4% (re = 4). There was no difference ininfiltrating macrophages in animals treated with CsA 0.05%(n = 2) or prednisolone acetate (Pred) (re = 4). *P= 0.0448.Bars = standard error.
studies will investigate the therapeutic effect of lowerdoses of CsA. Of note, there was no statistical differ-ence between the effects of CsA and prednisoloneacetate on mast cell-mediated conjunctivitis at allthree doses. Because topical corticosteroids remainthe mainstay of therapy, it is important for any drugtested for allergic ocular disease to have an effect simi-lar to corticosteroids.
However, although corticosteroids are used fre-quently for the treatment of chronic allergic disorders,many patients are resistant or intolerant to corticoste-roid therapy. Topical corticosteroid therapy has beenassociated with cataract formation, increased intraocu-lar pressure, and delayed wound healing. Therefore,other immunosuppressive agents, such as CsA, havebeen tried for the treatment of severe allergic condi-tions. Cyclosporine A has been used to treat patientswith a number of allergic diseases, including asthma.9
Topical CsA also has been used successfully to treatpatients with vernal keratoconjunctivitis, an allergicdisease characterized by the thickening and formationof giant papillae on the upper tarsal conjunctiva.1011
Recently, the mechanisms explaining how CsAacts to inhibit allergic disease have been elucidated.Although many of the immediate mast cell-mediatedeffects in allergic reactions are thought to result fromthe actions of mediators such as histamine, data sug-gest that cytokines produced by mast cells and infil-trating lymphocytes play an important pathogenicrole.12"17 Recently, CsA has been shown to inhibit mastcell-mediated cytokine production and to have a thera-peutic effect on allergic disease. Hatfield and Roehm18
showed that cyclosporine inhibited murine mast cellcytokine production. Specifically, CsA completely in-hibited IL-2, IL-3, IL-4, and granulocyte-macrophagecolony stimulating factor secreted by all cell linestested. Wershil and colleagues4 similarly showed thatCsA blocked mouse mast cell TNF-a production invitro, and they demonstrated that CsA inhibited mastcell-dependent inflammation in vivo. The authors alsosuggested that CsA can inhibit the responsiveness oftarget cells to the produced cytokines. Additional datashow that CsA interferes with the degranulation ofbasophils, another key inflammatory cell in the patho-genesis of allergy.19
Cyclosporine is an immunosuppressive agentwith predominant inhibitory effects against T lym-phocytes by blocking early activation genes20 spe-cifically related to cytokines.21 It is not surprising,therefore, that CsA interferes with both mast cell-and lymphocyte-mediated cytokine production andhas an inhibitory effect on the development of aller-gic disease. Topical application of CsA results ingood levels in ocular tissues,22'23 sparing many of theadverse systemic effects of the medication, whichinclude hypertension and renal disease.24"28 Giventhe fact that allergic ocular disease can threaten vi-
sion and be corticosteroid resistant, topical CsA mayhave an important role in the treatment of allergy-associated eye disease in humans.
1. Allansmith MR, Baird RS, Bloch KJ. Mast cells in ocu-lar tissues of normal rats and rats infected with Nippos-trongylus brasilensis. Invest Ophthalmol Vis Sci. 1979;18:863-867.
2. Abelson MB, Baird RS, Allansmith MR. Tear hista-mine levels in vernal conjunctivitis and other ocularinflammations. Ophthalmology. 1980;87:812-814.
3. Stellato C, de Paulis A, Ciccarelli A, et al. Anti-in-flammatory effect of cyclosporin A on human skinmast cells. JInvest Dermatol. 1992;98:800-804.
5. Li Q, Luyo D, Hikita N, Whitcup SM, Chan CC. Com-pound 48/80-induced conjunctivitis in the mouse: Ki-netics, susceptibility, and mechanism. Int Arch AllergyImmunol. 1996; 109:277-285.
6. Peng B, Li Q, Luyo D, Whitcup SM, Roberge FG, ChanCC. Treatment of allergic conjunctivitis in murinemodel. In: Nummenblatt RB, Whitcup SM, Caspi RR,Gery I, eds. Advances in Ocular Immunology. Amster-dam: Elsevier; 1994:249-252.
7. Hsu SM, Raine L, Fanger H. The use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidasetechnique: A comparison between ABC and unlabeledantibody (PAP) procedure. / Histochem Cytochem.1981;29:557-580.
8. Udell IJ, Kenyon KR, Hanninen LA, Abelson MB.Time course of human conjunctival mast cell degranu-lation in response to compound 48/80. Ada Ophthal-mol. 1989;192(suppl):154-161.
9. Triggiani M, Cirillo R, Lichtenstein LM, Marone G.Inhibition of histamine and prostaglandin D2 releasefrom human lung mast cells by ciclosporin A. Int ArchAllergy Appl Immunol. 1989;88:253-255.
10. BenEzra D, Pe'er J, Brodsky M, et al. Cyclosporineeyedrops for the treatment of severe vernal keratocon-junctivitis. Am J Ophthalmol. 1986; 101:278-282.
11. Kaan G, Ozden O. Therapeutic use of topicalcyclosporine. Ann Ophthalmol. 1993;25:182-186.
12. Plaut M, Pierce JH, Watson CJ, Hanley-Hyde J, Nor-dan RP, Paul WE. Mast cells produce lymphokinesin response to cross-linkage of FceRI or to calciumionophores. Nature. 1989;339:64-67.
13. Wodnar-Filipowicz A, Heusser CH, Moroni C. Pro-duction of the haemopoietic growth factors GM-CSFand interleukin-3 by mast cells in response to IgE re-ceptor-mediated activation. Nature. 1989; 339:150-152.
14. Burd PR, Cost JJ, Metcalfe DD, Siebenlist U. Cytokinegene expression occurs via multiple pathways in pri-
Topical Cyclosporine for Conjunctivitis 2693
mary murine bone marrow mast cells (BMMC). FASEBJ. 199O;4:A17O5.
15. Burd PR, Rogers HW, Gordon JR, et al. Interleukin3-dependent and -independent mast cells stimulatedwith IgE and antigen express multiple cytokines./£x/?Med. 1989; 170:245-257.
16. Gordon JR, Galli SJ. Mast cells as a source of bothpreformed and immunologically inducible TNF-a/cachectin. Nature. 1990;346:274-276.
17. Ohno I, Tanno Y, Yamauchi K, Takishima T. Geneexpression and production of tumor necrosis factorby a rat basophilic leukaemia cell line (RBL-2H3) withIgE receptor triggering. Immunology. 1990;70:88-93.
18. Hatfield SM, Roehm NW. Cyclosporine and FK506inhibition of murine mast cell cytokine production. /Pharmacol Exp Ther. 1992;260:680-688.
19. Hultsch T, Rodriguez JL, Kaliner MA, Hohman RJ.Cyclosporin A inhibits degranulation of rat basophilicleukemia cells and human basophils. Inhibition of me-diator release without affecting PI hydrolysis or Ca2+
fluxes. / Immunol. 1990; 144:2659-2664.20. Schreiber SL, Crabtree GR. The mechanism of action
of cyclosporin A and FK506. Immunol Today.1992;13:136-142.
21. Cross SL, Halden NF, Lenardo MJ, Leonard WJ. Func-tionally distinct NF-kappa B binding sites in the immu-
noglobulin kappa and IL-2 receptor alpha chaingenes. Science. 1989; 244:466-469.
22. Bell TA, Hunnisett AG. Cyclosporin A: Tissue levelsfollowing topical and systemic administration to rab-bits. Br JOphthalmol. 1986; 70:852-855.
23. Wiederholt M, Kossendrup D, Schulz W, Hoffman F.Pharmacokinetic of topical cyclosporin A in the rabbiteye. Invest Ophthalmol Vis Sci. 1986;27:519-524.
24. Nussenblatt RB, Palestine AG, Chan CC. Cyclosporinetherapy for uveitis: Long-term followup./OculPharma-col. 1985; 1:369-381.
25. Nussenblatt RB, de Smet MD, Rubin B, et al. A maskedrandomized, dose response study betweencyclosporine A and G in the treatment of sight-threat-ening uveitis of non-infectious etiology. AmJ Ophthal-mol. 1993; 115:583-591.
26. Palestine AG, Austin HA, Balow JE, et al. Renal histo-pathologic alterations in patients treated withcyclosporine for uveitis. NEnglJMed. 1986;314:1293-1298.
27. Whitcup SM, Nussenblatt RB. Treatment of autoim-mune uveitis. Ann NY Acad Sci. 1993;696:307-318.
28. Whitcup SM, Salvo EC, Jr, Nussenblatt RB. Combinedcyclosporine and corticosteroid therapy for sight-threatening uveitis in Behcet's disease. AmJOpthahnol1994; 118:39-45.
