Two Multicenter, Randomized Studies of
`the Efficacy and Safety of Cyclosporine
`Ophthalmic Emulsion in Moderate to Severe
`Dry Eye Disease
`
`Kenneth Sall, MD, 1 Onex Dara Stevenson, MD, 2 Thomas K. Mundmf, MD, 3 Brenda L. Reis, PhD4 and
`th~ CsA Phase 3 Study Group
`
`Objective: To compare the efficacy and safety of cyclosporin A ([CsA] 0.05% and 0.1 % ophthalmic
`e1 .1ulsions) to vehicle in patients with moderate to severe dry eye disease.
`Design: Multicenter, randomized, double-masked, parallel-group, 6-month, vehicle-controlled.
`Participants: A total of 877 patients with defined moderate to severe dry eye disease (292 to 293 in each
`trAatment group).
`Methods: Two identical clinical trials; patients were treated twice daily with either CsA, 0.05% or 0.1 %, or
`VEhicle. The results of these two trials were combined for analysis.
`Main Outcome Measures: Efficacy: corneal and interpalpebral dye staining, Schirmer tear test (with and
`w.thout anesthesia), tear break-up time, Ocular Surface Disease Index (OSDI), facial expression, patient subjec(cid:173)
`tive rating scale, symptoms of dry eye, investigator's evaluation of global response to treatment, treatment
`success, and daily use of artificial tears. Safety: occurrence of adverse events, best-corrected visual acuity,
`in~raocular pressure, biomicroscopy, and blood trough CsA concentrations.
`Results: Treatment with CsA, 0.05% or 0.1%, gave significantly (P :S; 0.05) greater improvements than
`vr:.hicle in two objective signs of dry eye disease (corneal staining and categorized Schirmer values). CsA 0.05%
`treatment also gave significantly greater improvements (P < 0.05) in three subjective measures of dry eye disease
`(blurred vision, need for concomitant artificial tears, and the physician's evaluation of global response to
`trPatment). There was no dose-response effect. Both CsA treatments exhibited an excellent safety profile, and
`there were no significant topical or systemic adverse safety findings.
`Conclusions: The novel ophthalmic formulations CsA 0.05% and 0.1 % were safe and effective in the
`treatment of moderate to severe dry eye disease yielding improvements in both objective and subjective
`measures. Topical CsA represents a new pharmacologically based treatment for dry eye disease that may
`provide significant patient benefits. Ophthalmology 2000;107:631-639 © 2000 by the American Academy of
`Ophthalmology.
`
`D~,;spite the millions of individuals who have dry eye dis(cid:173)
`ea~e, 1
`4 there is currently no therapeutic treatment for this
`-
`condition. Individuals plagued by the discomf01t, burning,
`irritation, photophobia, and other symptoms of dry eye
`disease may also have blurred vision, gradual contact lens
`in,olerance, and the inability to produce emotional tears, as
`
`Manuscript no. 993 16.
`
`O ~oginally received June 16, 1999.
`Accepted: December 8, 1999.
`1 ~all Eye Surgery Center, Bellflower, California.
`2 ~Ievenson Medical and Surgical Eye Center, New Orleans, Louisiana.
`3 Charlotte, North Carolina.
`4 h llergan, Inc., Irvine, California.
`Repri nt requests to: Linda Lewis, 575 Anton Blvd, Suite 900, Costa Mesa,
`CA 92626.
`Su?ported by a grant from Allergan Inc.
`Dr. Reis is an employee of Allergan, lnc. None of the other authors has a
`fir .. mcial interest in any of the products mentioned in this paper.
`
`well as an increased risk of ocular swf ace damage and
`ocular infection. 5- 9 Yet, at this time, the only treatments
`available are palliative in nature, consisting of lubricating
`eyedrops or puncta! occlusion procedures that attempt to
`supplement the patient's natural tears or improve the resi(cid:173)
`dence time of the limited quantity of tears that the patient
`can produce.
`Until recently, attempts to develop therapeutic treatments
`for dry eye disease were hampered by a limited understand(cid:173)
`ing of the underlying pathophysiologic processes. Although
`the exact mechanism is still not completely understood,
`there is now sufficient evidence to suggest that dry eye
`disease is the result of an underlying cytokine and receptor(cid:173)
`mediated inflammatory process affecting botl1 the lacrimal
`14 Tllis hypothesis is fur(cid:173)
`gland and the ocular surface.2• 10 -
`18 and human 20
`2 1
`ther supported by results from animal 15
`-
`-
`(Foulks et al, Invest Ophthalmol Vis Sci 1996;37:S646;
`Helms et al, Invest Ophthalmol Vis Sci 1996;37; S646;
`Kunert et al, Invest Ophthalmol Vis Sci 1999;40(4):877 1;
`
`© 2000 by the American Academy of Ophthalmology
`Pu'1lished by Elsevier Science Inc.
`
`ISSN 0 161·6420/00/$-sce front matter
`PH SOI61 -6420(99)00176-1
`
`631
`
` EXHIBIT 1007
`
`

`
`Oph thalm ology Volume 1 07, Number 4, April 2000
`
`Turner et al, Invest Ophthalmol Vis Sci 1999;40[4]:S558)
`studies that demonstrate that topical treatment with the
`immunomodulatory agent cyclosporin A (CsA) can have
`beneficial effects on the underlying inflammatory patho(cid:173)
`logic condition of dry eye disease, as well as improve the
`signs and symptoms of this condition.
`Evidence of the inflammatory nature of dry eye disease
`and the success of preliminary investigations into the effects
`of topical CsA treatment led to a large-scale multicenter,
`vehicle-controlled, dose-ranging (phase 2) clinical trial of
`the efficacy and safety of CsA treatment in moderate to
`severe dry eye disease. This study demonstrated that such
`treatment was safe and resulted in significant improvements
`in the signs and symptoms of the disease.2 1
`The objective of the studies described here was to com(cid:173)
`pare the efficacy and safety of twice daily CsA, 0.05 % and
`0.1 %, ophthalmic emulsions to vehicle in the treatment of
`moderate to severe dry eye disease in patients with or
`without Sjogren's syndrome. In addition, specialized assays
`were performed on conjunctival biopsy specimens from a
`subset of patients to investigate the levels of several inflam(cid:173)
`matory markers and the density of conjunctival goblet cells
`before and after treatment. The results of these specialized
`studies will be reported separately; preliminary reports have
`been presented (Kunert et al, Invest Ophthalmol Vis Sci
`1999;40[4]:S771 ; Turner et al, Invest Ophthalmol Vis Sci
`1999;40[ 4]:S558).
`
`Materials and Methods
`
`Study Design
`
`Two identical multicenter, randomized, double-masked, parallel(cid:173)
`group clinical trials were conducted to compare two concentrations
`of CsA ophthalmic emulsion to its vehicle. The 6-month treatment
`phase was preceded by a 2-week run-in phase to standardize all
`patients to a common regimen of artificial tear use. The results of
`these two trials were combined for analysis.
`Both trials were conducted in compliance with G ood Clinical
`Practices, investigational site Institutional Review Board Regula(cid:173)
`tions, Sponsor and Investigator Obligations, Informed Consent
`Regu lations, and the Declaration of Helsinki. Potential patients
`signed a prescreening informed consent. A second wri tten in(cid:173)
`formed consent was obtained from all patients before actual en(cid:173)
`rollment, and a separate informed consent was obtained for the
`subset of patients who participated in the pharmacokinetic evalu(cid:173)
`ations.
`Patients. Adult patients of either sex were eligible for partic(cid:173)
`ipation if they had a diagnosis of moderate to severe dry eye
`disease as defined by the following criteria (detailed descriptions
`of each parameter are given under " Outcome Measures"): ( 1)
`Schirmer test without anesthesia of :s 5 mm/5 min in at least one
`eye (if Schi rmer test without anesthesia of = 0 nun/5 min, then
`Schirmer with nasal stimulation had to be > 3 mm/ 5 min in the
`same eye); (2) sum of corneal and interpalpebral conjunctival
`staining of~ +5 in the same eye where corneal staining was ~
`+2; (3) a baseline Ocular Slllface Disease Index ([OSDI]; see
`later) score of 0. 1 with no more than three responses of "not
`applicable"; and (4) a score of ~ 3 on the Subjective Facial
`Expression Scale (see later). Signs and symptoms must have been
`present despite conventional management, which may have in(cid:173)
`cluded artificial tear drops, gels and ointments, sympathomi metic
`
`agents, parasympathomimetic agents, and puncta! occlusion. Eli(cid:173)
`gible patients were enrolled if they were deemed capable of
`following the study protocol and considered likely to complet-: the
`treatment period and return for all scheduled visits; if they had
`normal lid position and closure, and a best-corrected ETDRS
`visual acuity score of +0.7 LogMar or better in each eye.
`Patients were excluded from the study if they had partici!Jated
`in an earlier clinical trial with CsA ophthalmic emulsion or had
`used systemic or topical ophthalmic cyclosporine within 90 days
`before the study. Other exclusion criteria included the preser re or
`history of any systemic or ocular disorder or condition (including
`ocular surgery, trauma, and disease) that could possibly intn fere
`with the interpretation of the study resul ts; current or recent use of
`topical ophthalmic or systemic medications that could affect 1 dry
`eye condition; known hypersensiti vity to any component of the
`study or procedural medications; requi red contact lens wear u ring
`the study; recent (within I month) or anticipated use of temporary
`puncta! plugs during the study; permanent occlusion of laLdmal
`puncta within 3 months of the study; or if they were pregnant,
`lactating, or planning a pregnancy. Patients were also excluJed if
`they appeared to have end-stage lacrimal gland di sease (Schirmer
`reading with nasal stimulation of< 3 mm/5 min) or if their dry eye
`disease was the result of destruction of conjunctival goblet C~'lls or
`scarring. Any patient who no longer met the criteria for moderate
`to severe dry eye (as defined previously) after completir g the
`2-week run-in phase was excluded from enrollment in the tTeat(cid:173)
`ment phase of the study.
`Patients could be discontinued before the completion of the
`study because of adverse events, pregnancy, protocol viol;,tions,
`lack of efficacy, or administrative or personal reasons.
`Study Medications. CsA, 0.05% and 0. 1 %, ophthalmic emul(cid:173)
`sions and the vehicle of CsA ophthalmic emulsion were the study
`medications, with indi vidually packaged preservative-free anificial
`tears (REFRESH Lubricant Eye Drops, Allergan, Inc.) provided as
`an adjuncti ve treatment to be used as frequently as needeu. The
`doses of CsA used were based on the results of an earlier dose(cid:173)
`ranging study? ' Both the CsA emulsions and vehicle were sterile,
`nonpreserved castor oil in water emulsions whose precise rormu(cid:173)
`lation is proprietary. All the study medications were supplied in
`unit dose via ls.
`Concomitant Medications. Any therapy considered necessary
`for the maintenance of patient welfare was given at the dis~ retion
`of the investigator and noted on the case report form. lf the
`medication would not interfere with the response to study medi(cid:173)
`cation, the patient was kept in the study. During the study, all
`concomitant medication treatment regimens were kept as constant
`as pennitted by accepted medical practice. Systemic and topical
`ophthalmic medications that could interfere with the response to
`study medications or the interpretation of the study result~ were
`prohibited during the study. This included cyclosporine, other
`immunomodulatory agents, general anesthetics, antihist~-nines,
`cholinergic agents, antimuscarinics, 13-blocki ng agents, tricyclic
`antidepressants, phenothiazines, topical ophthalmic steroid~ estro(cid:173)
`gen-progesterone (or other estrogen derivatives), or any topical
`ocular medi cations other than the assigned study medicali..Jn and
`assigned artificial tears.
`Study P rotocol. During the 2-week run-in phase, all patients
`were instructed to use the assigned artificial tears only (as needed).
`During the treatment phase, patients were instructed to illstill I
`drop of study medication twice dai ly in each eye for 6 PlOilths;
`once on waking in the morning and once at bedtime. Patients were
`allowed to use the assigned artificial tears as needed up 10 mo~th
`4. To avoid dilution of the study medication, patients were Ill"
`structed not to use artificial tears within 30 minutes before ' r after
`use of the study medication. To minimize the habitual conco1~1it~n:
`use of artificial tears, patients were asked to stop using a ttficJa
`
`632
`
`

`
`Sall et al · CsA in Dry Eye Disease
`
`ttJJS 1 week before their month 4 study visit. At this visit, patients
`were encouraged to use artificial tears less than eight times a clay
`fa the rest of the study period.
`During the treatment phase, patients returned for evaluation
`al£er I, 3, 4, and 6 months of treatment.
`
`Outcome Measures
`Etficacy. The objective signs monitored were corneal and inter(cid:173)
`p~lpebral conjunctival staining, Schirmer tear test (with and with(cid:173)
`out anesthesia), and tear break-up time. The subjective endpoints
`us~d were the OSDI, the facial expression subjective rating scale,
`symptoms of dry eye, investigator's evaluation of global response
`to treatment, treatment success, and artificial tears use. All these
`variables were evaluated at baseline and at each study visit, except
`fo, tear break-up time, which was not evaluated at the ! -month
`visit; the investigator's evaluation of global response to treatment,
`which could not be meaningfully assessed at baseline; and the
`Schirmer tear test with anesthesia, which was only evaluated at
`baseline, month 3, and month 6.
`For corneal fluorescein staining, the entire cornea was exam(cid:173)
`ined using slit-lamp evaluation with a yellow barrier fi lter and
`co11alt blue illumination. Staining was graded using the Oxford
`Scheme 6-point scale (from 0 to 5), with each investigator using
`the same set of photographs (provided by the study sponsor) as a
`guide. Lissamine green was then instilled, and interpalpebral con(cid:173)
`jlll ctival staining was evaluated more than 30 seconds, but less
`than 2 minutes, later. Using wh ite light of moderate intensity, the
`inh.:rpalpebral regions of the temporal and nasal conjunctiva were
`graded using the same Oxford Scheme. The sum of corneal and
`intt:rpalpebral staining was therefore on a 0 to 15 point scale. On
`all scales, a negative change from baseline indicated an improve(cid:173)
`ment.
`The Schirmer tear test was peif'ormed both with and without
`anesthesia and graded on a 5-point scale as follows: I (< 3 mm/5
`min), 2 (3-6 mm/5 min), 3 (7-10 mm/5 min), 4 (1 1- 14 mm/5
`min), and 5 (> 14 mrn/5 min) using the worse eye. A positive
`chrnge from baseline indicated improvement. Categorized values
`were used to reduce overall withi n-patient variability known to
`oclur for Schirmer wetting scores. The categories were chosen a
`priori on the basis of cutoff points suggested by a review of the
`meJ ical literature and consultation with clinical investigators.
`Ti me until random location tear break-up between blinks was
`measured only up to 10 seconds and recorded only if the value was
`less than I 0 seconds.
`The OSDI questionnaire was used to evaluate the impact of a
`patiP-nt's dry eye disease on vision-related functioning.22 It con(cid:173)
`sisted of 12 questions that were each rated from 0 = none of the
`tim•· to 4 = all of the time. An overall score was calculated by
`dividing the sum of the responses fo r all questions answered by the
`tota' possible score. Thus, overall scores ranged from 0 = no
`disability to 1 = complete disability. A negative change fro m
`basdine indicated an improvement in vision-related functioning.
`The Subjective Facial Expression Rating Scale23 consisted of
`nim. facial schematics ranging from I (the happiest face) to 9 (the
`unhappiest face) analyzed in five grades from 1 (pictures 1 and 2)
`to 5 (pictures 8 and 9). A negative change from baseline indicated
`an improvement.
`~ymptoms of ocular discomfort, such as stinging/burning, itch(cid:173)
`ing. sandiness/grittiness, blurred vision, dryness, light sensitivity,
`pain or soreness, were graded using a 5-point scale ranging from
`0 = do not have th is symptom to +4 = always notice this
`symptom). A negative change from baseline indicated an improve(cid:173)
`mer !.
`The investigator completed a global evaluation of the overall
`effe t of the study medication relative to the qualification visit
`
`l
`
`using a 7-point scale in which 0 = completely cleared, I = = 90%
`improvement, 2 = =75% improvement, 3 = = 50% improvement,
`4 = =25% improvement, 5 = condition unchanged, and 6 =
`condition worsened. Treatment success was defined as a global
`evaluation of = 90% improvement or better.
`Patient use of concomi tant artificial tears was also monitored.
`At each study visit, patients were queried about the average num(cid:173)
`ber of times they used artificial tears each day during the past week
`and the number of days during the past week when they did not use
`any artificial tears.
`Safety. The primary safety variable monitored was the occur(cid:173)
`rence of adverse events. The severi ty of each adverse event ob(cid:173)
`served was rated from mi ld (awareness of sign or symptom, but
`easily tolerated) to severe (incapacitating with inability to work or
`do usual activity). The relationship of the event to the study
`medication was assessed by the investigator as none, unlikely,
`possible, probable, or definite.
`Other safety variables monitored included best-corrected visual
`acuity (using the ETDRS chart), intraocular pressure (usi ng Gold(cid:173)
`man applanation tonometry), and biom icroscopy (using slit-lamp
`examination without pupil dilation). Pharmacokinetic testing was
`also done on a subset of patients to determine blood trough CsA
`concentrations during treatment.
`Safety variables were evaluated at baseline and at all study
`visits, except for intraocular pressure, which was only evaluated at
`baseline and month 6, and CsA blood levels, which were evaluated
`at baseline and months I and 6.
`
`Other Measures
`
`Blood was collected from all patients for autoantibody testing to
`aid in the prospective identification of individuals with Sjogren's
`syndrome. Frozen samples were sent to a central laboratorx (Co(cid:173)
`vance Central Laboratories Inc, Indianapolis, IN) fo r log in and
`inventory before shipment to the Scripps Reference Laboratory
`(San D iego, CA) for the actual tests. Blood samples were evalu(cid:173)
`ated for the presence of antinuclear antibodies, rheumatoid factor,
`and class SS-A and class SS-B Sjogren's antibodies.
`Blood was also collected from a subset of patients for deter(cid:173)
`mination of mean blood CsA trough concentration. Frozen samples
`were sent to the sponsor (Ailergan, Inc.) for analysis, under Good
`Laboratory Practices regulations, using liquid chromatography(cid:173)
`mass spectrometry mass/mass spectrometry (LC-MS/MS) with a
`quantitation limit of 0.1 ng/m l.
`
`Statistical M ethods
`
`All analyses presented in this report were conducted on the intent(cid:173)
`to-treat patient population (all patients random ized). A last-obser(cid:173)
`vation carried-forward method was used to input missing data,
`although baseline data (before study treatment) were not canied
`forward. For efficacy variables collected on both eyes, only the
`data from the "worse" eye were included in the analyses. The
`"worse" eye was defined as the eye with the worse Schirmer tear
`test value (without anesthesia) and the worse sum of corneal and
`interpalpebral conjunctival staining. If both eyes were comparable,
`then the right eye was used. The "worse" eye was defined o n the
`basis of the baseline measurements, and data from this eye were
`used for all subsequent analyses. However, it should be noted that
`data were collected on both eyes throughout the study. All safety
`analyses included data from both eyes.
`Desc ripti ve s tatistics were used to summarize all continuous
`variables (such as all staining variables and the OSDI score) and
`categorical variables (such as the Subjective Facial Expression
`Rating Scale, symptoms of dry eye disease, and categorized
`Schirmer values). A two-way analysis of variance (ANOY A), with
`
`633
`
`

`
`Ophthalmology Volume 107, Number 4 , April 2000
`
`Table I. Patient Disposition
`
`Enrolled
`Completed
`Discontinued
`Lack of efficacy
`Adverse events
`Lost to follow-up
`Personal reasons
`Protocol or enrollment violations*
`O ther•
`
`CsA 0.05%
`
`293
`235 (80.2%)
`
`I (0.3%)
`19 (6.5%)
`4 (1.4%)
`9(3.1%)
`19 (6.5%)
`6(2.0%)
`
`CsA 0.1 o/o
`
`292
`218 (74.7%)
`
`3( 1.0%)
`29 (9.9%)
`3 ( 1.0%)
`14 (4.8%)
`21 (7.2%)
`4( 1.4%)
`
`Vehicle
`
`292
`2 18 (74.7%)
`
`3 \ 1.0%)
`13 ' 4.5%)
`11 (3.8%)
`9 0.1%)
`30 \ 10.3%)
`8 (2.7%)
`
`CsA = cyclosporin A.
`* Protocol or enrollment violations included improper en try, non-compliance, and use of prohibited medications.
`t Other included pregnancy, relocation, and removal from the study by the sponsor.
`
`group and in vestigator effects and interaction, was used to test fo r
`among-group differences in continuous variables. T he Cochran(cid:173)
`Mantel-Haenszel24 procedure with modified ridits, stratified by
`site, was used to test for differences in change f rom baseline
`among treatment groups in categorical variables and fo r among(cid:173)
`group d ifferences in global response d istributions. The paired 1 test
`method was used to analyze within-group changes from baseli ne in
`conti nuous variables, whereas the Wilcoxon signed rank test was
`used to analyze within-group changes from baseline in categorical
`variables.
`Statistical adjustment of P values for multiple comparisons
`have been addressed by use of the Fisher's least significant differ(cid:173)
`ence protected tests to ensure that the experimental-wise error rate
`was 0.05. Specifically, a pair wise comparison between either
`cyclosporine group and vehicle groups is considered statisticall y
`sig nificant if and only if ( I) the overall comparison among the
`three groups is significant at the 0.05 level, and (2) the pair wise
`comparison between cyclosporine and vehicle is significant at the
`0.05 level. In addition, althoug h patients were evaluated at multi(cid:173)
`ple time points throughout the study, the primary endpoint is on the
`last observation while on treatment during the 6-month study
`period, and no further adjustments need to be made.
`Each trial was powered separately, whereby given an expected
`sample size of 100 per group, the power to detect a three-grade
`differe nce between treatment groups in the change from baseline
`for the sum of corneal and interpalpebral conjunctival staining was
`greater than 0.86, using a two-sided Wilcoxon rank s um test with
`an estimated standard deviation of 6.49.
`A two-sided test with a P value ::;; 0.05 was cons idered statis(cid:173)
`tically significant.
`
`Patient Treatment Assignment
`
`Qualified patients were randomly assigned to receive one of the
`three study med ications in a I: I: I ratio using a block size of 3.
`Only a sing le randomization was used per trial. There was no
`stratification by either site or baseline factors.
`
`Study Masking
`
`Results
`
`Participant Flow and Follow-up
`
`A total of 877 patients was enrolled in approx imately eqll tl num(cid:173)
`bers in the three treatment groups and more than 76% (671/877)
`completed the study (Table 1). T he first patient was enu lled in
`July 1997 and the last patient completed the 6-month treatment
`period and evaluation in September 1998. M ost of the patiL.lts who
`exited the study prematurely did so as a result of protocol or
`enro ll ment violations, personal reasons, lost to follow-up, J r other
`no ntreatment-related reaso ns. Only 0.8% of patients (7/877) were
`discontinued because of lack of efficacy; I of 293 (0.3%) in the
`CsA 0.05% group, 3 of 292 ( I .0%) in the CsA 0.1 % g rouo, and 3
`of 292 ( 1.0%) in the vehicle gro up. Only 7.0% (61/877) were
`discontinued because of adverse events; 19 of 293 (6.5 o/· ) in the
`CsA 0.05% group, 29 of 292 (9.9%) in the CsA 0. 1% group, and
`13 of 292 ( 4.5%) in the vehicle group.
`
`Patient Demographics
`
`The demographic characteristics of the study population hre listed
`in Table 2. O verall, most of the patients were women (82%;
`7 15/877) and Caucasian (84%; 740/847). There were no statisti(cid:173)
`cally signi ficant differences between the treatment groups in any of
`the demographic variables measured. The use of prior therapy, or
`the types of concomitant medications used during the o•udy for
`other medical conditions, were similar among the treatment
`groups .
`Sjogren's syndrome was defined as the presence of ocular
`symptoms, oral symptoms, and a Schirmer test ::;; 5 mm, <. well as
`the presence of at least one of the following autoantibodi~s:
`antinuc lear antibodies, rheumatoid factor, and Sjog ren's at.tibodtes
`class SS-A and SS-B. Of the 877 enrolled patients, 270 (30.8%)
`were determined to have Sjogren 's syndrome on the basi~ of the~e
`criteria. The nu mber of patients with Sjogren's syndrome identt·
`tied may underestimate the number in the study popul ation be·
`cause some patients left the studies without having auto1ntibodY
`data collected and therefore could not meet the criteria for
`Sjogren's syndrome.
`
`The study medication was packaged, labeled, and m asked in a
`manner consistent with Good Manufac turing Practices for inves(cid:173)
`tigational supplies. Ide ntical unit-dose vials were used to hold the
`study treatments, which were each of an identical milky color.
`
`Efficacy Analysis
`Corneal Staining. At baseline, the mean values for corn• .1! sta~;
`ing in the d ifferent treatment groups ranged between 2.6 1 an~ 2 .. f.
`(usi ng the Oxford scale fro m 0 to 5), with no statistical!_. stgnt
`
`634
`
`

`
`Salt et al · CsA in Dry Eye Disease
`
`T able 2. Patient Demographics
`
`CsA 0.05%
`(n = 293)
`58.7 ± 13.9
`
`49 (16. 7%)
`244 (83.3%)
`
`253 (86.3%)
`8 (2.7%)
`8 (2 .7%)
`23 (7.8% )
`1 (0.3% )
`
`97 (33.1%)
`126 (43.0%)
`20 (6.8%)
`0(0.0%)
`48 (16.4%)
`2 (0.7%)
`96 (32.8%)
`
`CsA 0.1 %
`(n = 292 )
`60. 1 ± 13.3
`
`54 (18.5%)
`238 (81.5%)
`
`243 (83.2%)
`16 (5.5%)
`6 (2.1%)
`26 (8.9%)
`I (0.3%)
`
`95 (32.5%)
`127 (43.5%)
`26 (8.9%)
`2 (0.7%)
`38 (1 3.0%)
`4 ( 1.4%)
`83 (28.4%)
`
`Vehicle
`(n = 292)
`59.9 ± 14.3
`
`59 (20.2%)
`233 (79.8%)
`
`244 (83.6%)
`15 (5.1 %)
`6 (2.1%)
`26 (8.9%)
`1 (0.3%)
`
`109 (37.3%)
`116 (39.7%)
`18 (6.2%)
`0(0.0%)
`46 (15.8%)
`3 ( 1.0%)
`91 (31.2%)
`
`Age (mean± SO)
`Sn
`Male
`Female
`Race
`Caucasian
`African-American
`Asian
`Hispanic
`O ther
`lrb color
`Blue
`J rown
`G reen
`J lack
`Hazel
`.)ther
`Sjogren's syndrome patients
`
`Cs '\ = cyclosporin A .
`
`ic~ nt differe nces among the groups . There was a statistically
`significant improvement fro m baseline in corneal stain ing (de(cid:173)
`crease in mean score) within all treatment groups at all follow-up
`visits (P < 0.00 I ). The improvement in corneal staining was
`sirn ificantly greater in both CsA groups than the vehicle group (P
`s: 0.044) at month 4, and in the CsA 0.05% group at month 6 (P
`= 0.008). There was also a trend (P = 0.062) toward a signifi(cid:173)
`cantly greater improvement in the CsA 0.1 % group than the
`ve11icle group at month 6 (Fig I).
`Conjunctival staining. At baseline, the mean values for tem(cid:173)
`po. a) interpalpebral conjunctival staining ranged between 2.22 and
`2.26 (using the Oxford scale from 0 to 5), with no stati stically
`significant differences among the groups . Similarly, the mean
`ba~eline values for nasal interpalpebral conjunctival staining
`ranged betwee n 2.42 and 2.45, with no statistically signi ficant
`diffe rences among the gro ups. Statistically significant improve(cid:173)
`ments from baseline were seen in both temporal and nasal con(cid:173)
`jur ctival staining within all study g roups at all follow- up visits
`(P < 0.00 I), but there were no statistically significant among(cid:173)
`grc. up differences.
`
`Change From Baseline in Corneal Staining
`
`Month 1
`
`Month 3
`
`Month 4
`
`Month 6
`
`.0.1 -
`
`.().2
`
`.0.3 -t
`
`Change .0.4
`.n
`Staining .0.5 -
`Sl""ore
`.0.6
`
`1
`
`·0.7
`
`·0.8
`
`.0.9
`· I J
`
`• P s. .05 compared to vehicle
`1 P = .62 compared to vehicle
`
`l• c sA 0.05% r.;iCsA 0.1% OVehlcle :
`
`Figure 1. Change from baseline in corneal staining. Mean value ±
`stat.dard error. Graded on a scale from 0 to 5.
`
`the mean categorized
`Schirmer Tear Test. At baseline,
`Schirmer values (obtained with anesthesia) in the different treat(cid:173)
`ment groups ranged between 1.94 and 2.1 1 (on a scale from 0 to
`5 with I < 3 mm/5 min, 2 = 3- 6 mm/5 min, 3 = 7-10 mm/ 5 min,
`4 = 11-14 mm/5 min, and 5 > 14mm/5 min), with no statistically
`significant differences among the gro ups. At month 3, there was a
`significant worsening with the vehicle group (P = 0.01 4) and a
`significant difference among the treatment g roups, with the CsA
`0.05% group significantly greater than the vehicle group (P =
`0.009). At mo nth 6, there was a statistically significant improve(cid:173)
`ment fro m baseline within both CsA groups. Moreover, the
`changes in Schirmer values in each of the CsA groups was signif(cid:173)
`icantly better than in the vehicle gro up (P < -0.007; Fig 2).
`Schirmer tear tests were also conducted without anesthesia. At
`baseline, the mean categorized Schirmer values (obtai ned without
`anesthesia) in the different treatment groups ranged between 1.47
`and 1.50 (on the same scale as was used for the Schirmer test with
`anesthesia), wi th no statistically significant differences among the
`
`Change From Baseline in Categorized Schirmer Values
`Measured With Anesthesia
`!ICsA 0.05% 1!:1 CsA 0.1 % 0 Vehicle
`
`0.6 j riCsA
`
`0.5
`
`0.4
`
`0.3 1
`
`Change 0.2 ~
`
`Sch~~mer 0
`
`Score
`
`1
`
`'
`
`.().1
`
`.0.2
`
`·0.3
`
`·0.4
`
`• P s .05 compared to vehicle
`
`Month 3
`
`Month 6
`
`Figure 2. C hange from baseline in categorized Schirmer values (measured
`with anesthesia). Mean value ± standard error. Categorized Schirmer
`values were graded on a 5-point scale as follows: 1 ( < 3 mm/5 min), 2 (3-6
`mm/5 min ), 3 (7-10 mm/5 min), 4 ( 11-14 mm/5 min), and 5 (> 14 mm/5
`min) using the worse eye.
`
`635
`
`

`
`Ophthalmology Volume 107, Number 4 , April 2000
`
`Change From Baseline in Blurred Vision
`
`Month 1
`
`Month 3
`
`Month4
`
`Month 6
`
`Change From Baseline In Average Dally Use of Aruficial
`Tears
`
`Month!
`
`Month 3
`
`Month 4
`
`Montr, 6
`
`Change
`In
`Units per
`Day
`
`·0.2
`
`·0.3
`
`·0.4 -
`
`·0.6 ~
`
`• p.::, .05 compared to vehicle
`
`' • csAO.OS% f'J CsA0.1% DV~
`
`·0.7
`
`·0.5 -
`
`·0.6
`
`• P ~ .05 compared to vehicle
`J l• csA o.OS% f'J CsA 0.1% DVehic~
`
`Figure 3. Change from baseline in blurred vision. Mean value ::t standard
`error. Graded on a scale from 0 to 4.
`
`Figure 4. Change from baseline in assigned lubricating eyedrop u~· Mean
`value ::t standard error.
`
`groups. Statistically significant improvements from baseline were
`seen in these Schi rmer val ues within all treatment groups at all
`follow-up visits, but there were no significant among-group dif(cid:173)
`ferences.
`Symptoms of Dry Eye Disease. Of the subjective ocular
`symptoms monitored, the most notable changes were seen in
`blurred vision. At baseline, the mean score for blurred vision in the
`different treatment groups ranged between 1.83 and 2.04 (on a
`scale from 0 to 4 ), with no statistically significant differences
`among the groups. The decrease from baseline in blurred visio n
`was stati sticall y significa nt w ithin both CsA groups at all fol(cid:173)
`low-up visits (P s 0.0 12) but only at month 6 within the vehicle
`group (P = 0.039). Moreover, the improvement in blurred vision
`was signi fica ntly greater in the CsA 0.05% group than the vehicle
`group at all follow-up v isits (P s 0.0 14; Fig 3_).
`Statistically significant changes from baselme we re observed
`withjn all treatment groups at all time points in dryness (P <
`0.001 ), sandy/gri tty feeling (P < 0.001 ), and itching ( P s 0.038).
`Statistically significant improvements from baseline were seen in
`photophobia within both CsA groups at all follow-up visits (P s
`0.028) and within the vehicle group at months 3, 4, and 6 (P s
`0.023). Statistically significant improvements from baseline were
`also seen in burning and stinging within both CsA groups at
`months 4 and 6 (P s 0.024) and within the vehicle group at
`months 3, 4, and 6 (P s 0.0 19). There was also a statistically
`significant improvement from baseline in pain within the CsA
`0.1 % group at all follow-up visi ts (P s 0.0 12) and the C sA 0.05%
`group and vehicle group at month 6 (P s 0.0 12). There were no
`statistically significant among-group differences in any of