To investigate the efficacy and safety of an investigational integrin antagonist (SAR 1118) ophthalmic solution compared to placebo (vehicle) in subjects with dry eye disease.
Multicenter, prospective, double-masked, placebo-controlled trial.
A total of 230 dry eye subjects selected with use of a controlled adverse environment were randomized 1:1:1:1 to receive SAR 1118 (0.1%, 1.0%, 5.0%) or placebo eye drops twice daily for 84 days. Principal eligibility criteria included exacerbation in corneal staining and ocular symptoms with controlled adverse environment exposure, no active lid margin disease, and Schirmer test (mm/5 min) >1 and <10. Ocular signs and symptoms (Ocular Surface Disease Index, OSDI) were assessed at day 14, 42, and 84. No supplemental artificial tears were allowed. Primary outcome measure was inferior corneal staining score at day 84.
A dose response for the corneal staining score ( P = .0566) was observed for SAR 1118 at day 84 compared to placebo. Mean change from baseline to day 84 showed significant improvements ( P < .05) in corneal staining score, total OSDI, and visual-related function OSDI scores for SAR 1118 compared to placebo; improvements in tear production and symptoms were observed as early as day 14 ( P < .05). Adverse events were mild and transient in nature with no serious ocular adverse events. SAR 1118 5.0% showed increased instillation site adverse events relative to placebo but were limited to the initial dose.
SAR 1118 demonstrated improvements in signs and symptoms of dry eye compared to placebo and appears safe when administered over 84 days.
Dry eye disease is a multifactorial disease of the tears and ocular surface that results in symptoms of ocular discomfort. An estimated 20 million people in the United States alone suffer from dry eye and the prevalence is increasing with the overall aging population. In addition to chronic ocular discomfort, dry eye disease can significantly impair visual-related quality of life such as reading, driving, working at computer monitors, and carrying out professional work. While the etiology of dry eye is complex and multifaceted, emerging evidence supports that chronic inflammation is a critical element in the pathogenesis of dry eye disease. Lymphocyte activation and trafficking to inflamed tissue is an essential step in the chronic inflammatory response. Increased lymphocytic infiltration has been observed in the conjunctiva and/or lacrimal glands in animal models of Sjögren disease, canine dry eye, and patients with dry eye disease. Elevated levels of inflammatory cytokines expressed by T cells have been demonstrated in tear film in both animal models and human dry eye patients.
The ability of lymphocytes to activate and home to the site of inflammation is influenced by the interaction of lymphocyte function–associated antigen-1 (LFA-1; CD11a/CD18; αLβ2) and intercellular adhesion molecule-1 (ICAM-1; CD54). LFA-1 is a heterodimeric protein integrin located on the surface of CD4+ lymphocytes (T cells). ICAM-1, the cognate ligand to LFA-1, is a transmembrane protein found on the surface of endothelial cells, epithelial cells, and immune function cells. The binding of LFA-1/ICAM-1 is essential in the formation of the immunologic synapse between antigen presenting cells and lymphocytes, a critical step in T-cell activation in both normal immune response and inflammation. LFA-1/ICAM-1 binding is also a central mechanism in T-cell adhesion, migration, proliferation, and cytokine release. ICAM-1 is normally expressed at low levels but is elevated in the setting of inflammation. Conjunctival epithelial cells and lacrimal gland acinar cells have demonstrated increased ICAM-1 levels in animal models and patients with dry eye disease.
Since the interaction of LFA-1/ICAM-1 mediates several key roles of lymphocyte activation and mobility, we hypothesize that blockade of LFA-1/ICAM-1 interaction may confer a therapeutic benefit in patients with dry eye disease by breaking the chronic cycle of T cell–mediated inflammation (activation, homing, cytokine release) and enabling the ocular surface tissues to recover. LFA-1/ICAM-1 inhibition as a therapeutic target in chronic ocular T cell–mediated inflammation has been validated in preclinical models of dry eye, diabetic retinopathy, and uveitis.
SAR 1118 is a first-in-class investigational small-molecule LFA-1 antagonist that mimics the binding epitope of ICAM-1 and is engineered for topical ophthalmic delivery. SAR 1118 is a potent inhibitor of T-cell activation, adhesion, migration, proliferation, and cytokine release. In preclinical animal studies, SAR 1118 ophthalmic solution administered topically to canines with dry eye demonstrated increased tear production as early as 2 weeks after onset of treatment. In diabetic retinopathy models, SAR 1118 demonstrated a dose-dependent reduction in retinal vascular leukocyte adherence and vascular leakage. A phase 1 safety and pharmacokinetic (PK, tear and plasma) dose-escalation study of SAR 1118 ophthalmic solution in normal healthy adult subjects showed that administered doses up to 5.0% 3 times daily appeared safe and well tolerated with no detectable drug accumulation in plasma or tears over time.
For the dry eye patient, there are few therapeutic options available. While artificial tear substitutes and punctal plugs are commonly used, the United States Food and Drug Administration (FDA) has approved only 1 pharmacologic treatment. Cyclosporine 0.05% emulsion (Restasis; Allergan Inc, Irvine, California, USA) is an immunomodulator approved for increasing tear production but is not approved for improving the symptoms associated with dry eye disease. The onset of action of cyclosporine is approximately 24 weeks, and many patients are intolerant because of drug-related ocular burning (17%) and require induction treatment with topical steroids to reduce the persistent stinging. A significant unmet medical need exists for a second-generation pharmacologic agent with faster onset of action, an enhanced tolerability profile, and the ability to significantly improve patients’ symptomatic quality of life.
The purpose of this study was to determine the therapeutic potential of LFA-1 inhibition by evaluating the efficacy and safety of 3 different concentrations (0.1%, 1.0%, 5.0%) of SAR 1118 ophthalmic solution compared to placebo when administered twice daily over 84 days (12 weeks) in patients with dry eye disease.
This was a phase 2 prospective, randomized, double-masked, placebo-controlled parallel-arm design study with block enrollment conducted between August 3, 2009 and February 17, 2010 at 5 sites in the United States. The study was approximately 100 days in duration and consisted of 3 periods: screening (day −14 to 0), treatment (day 0 to 84), and a follow-up (day 85 or 86) telephone safety interview 1 to 2 days after the last dose of study treatment. A total of 5 study visits were scheduled: 2 during screening (visits 1 and 2) and 3 during treatment (visits 3, 4, and 5).
At visit 1 (day −14), following written informed consent, subjects underwent data collection by study personnel that included demographic data, medical and medication history, urine pregnancy testing (as appropriate), and initial inclusion and exclusion criteria evaluation. A summary of scheduled assessments and study visits is provided in Table 1 . Inclusion criteria included adult subjects ≥18 years of age, established history of bilateral dry eye disease, use or desire to use artificial tear substitutes within the past 6 months, presence of conjunctival redness in any eye, corneal fluorescein staining score of ≥2.0 points in any field (0−4 point Ora scale, 0 = none to 4 = confluent) in any eye, unanesthetized Schirmer test (mm/5 min) of >1 and <10 in any eye, and best-corrected visual acuity (BCVA) of ≥0.7 logMAR (Early Treatment of Diabetic Retinopathy Study [ETDRS] eye chart) in both eyes. Exclusion criteria included known contraindications or hypersensitivity to the study drug or its components, active ocular inflammation (including active lid margin disease), active ocular infection, any ocular surgery within the past 12 months including laser-assisted in situ keratomileusis, the requirement for contact lens use during the study, and pregnancy. Prohibited medications during the study included topical cyclosporine (discontinued >6 weeks prior to visit 1) or use of any other ophthalmic medication (eg, glaucoma medication, topical anti-inflammatory eye drops) for the duration of the study. Subjects actively taking artificial tears had to discontinue use 72 hours prior to visit 1.
|Visit 1 Day -14 ± 2 CAE#1||Day -13–1||Visit 2 Day 0 CAE#2||Days 1−13||Visit 3 Day 14 ± 2 CAE#3||Days 15−41||Visit 4 Day 42 ± 4 CAE#4||Days 43−83||Visit 5 Day 84 ± 6 CAE#5||Days 85,86|
|Demographic data/medical history||X|
|Enrollment and randomization||X|
|Ocular discomfort score||X||X||X||X||X||X||X||X||X||X|
|Ocular Surface Disease Index||X||X||X||X||X|
|Visual analogue scale||X||X||X||X||X||X||X||X||X||X|
|Best-corrected visual acuity||X||X||X||X||X|
|Conjunctival redness score||X||X||X||X||X||X||X||X||X||X|
|Tear film break-up time||X||X||X||X||X||X||X||X||X||X|
|Corneal staining (fluorescein)||X||X||X||X||X||X||X||X||X||X|
|Conjunctival staining (lissamine green)||X||X||X||X||X||X||X||X||X||X|
|Schirmer test (without anesthesia)||X||X||X||X||X|
|Study drug administration||X||X||X||X||X||X||X||X||X||X|
|Placebo dispensation (open-label)||X||X|
|Adverse event assessment||X||X||X||X||X||X||X||X||X||X||X||X||X||X||X|
If subjects met initial screening requirements, additional assessments of symptoms and signs were conducted before and after exposure to acute desiccating stress using the controlled adverse environment model. The controlled adverse environment procedure has been described previously and was used to enrich the study population at screening by identifying subjects with the ability to exacerbate both signs (inferior cornea staining) and symptoms (ocular discomfort score) following 90 minutes of exposure.
Symptom assessments included the Ocular Surface Disease Index (OSDI), ocular discomfort score (0−4 point Ora scale, 0 = none to 4 = severe), and visual analogue scale (7 items: burning/stinging, itching, foreign body sensation, blurred vision, eye dryness, photophobia, pain). The OSDI is a validated patient questionnaire consisting of 12 items in 3 subscales (visual-related function, triggers, and symptoms) and was obtained only pre–controlled adverse environment at every study visit. The ocular discomfort score was obtained before, during (every 5 minutes), and after the controlled adverse environment. Sign assessments included slit-lamp biomicroscopy, corneal fluorescein staining score, conjunctival lissamine green staining score (0−4 point Ora scale, 0 = none to 4 = confluent), tear film break-up time, Schirmer test, and ocular blink rate.
The designated study eye had to demonstrate a Schirmer test (mm/5 min) >1 and <10 mm (pre–controlled adverse environment) and >1.0 point increase in both inferior corneal staining (0−4 scale) and ocular discomfort score (0−4 scale) from baseline at visit 1 following the controlled adverse environment exposure. If both eyes met these requirements, the worst eye was designated the eligible study eye; if both eyes were equal, the right eye (OD) was considered the eligible study eye. Subjects with pre–controlled adverse environment corneal staining score and/or ocular discomfort score of 4.0 were excluded.
All subjects having a positive screening response and meeting all other eligibility criteria after visit 1 (day -14) were initiated on self-administered open-label placebo (vehicle) twice daily (morning and evening) for 14 days until visit 2 (day 0). At visit 2, the sequence of pre–, intra–, and post–controlled adverse environment assessments were repeated similar to visit 1. The eligible study eye identified at visit 1 had to meet the same qualification criteria at visit 2. All subjects meeting the study eligibility criteria were randomized to 1 of 4 treatment arms (1:1:1:1; placebo, and 0.1%, 1.0%, 5.0% SAR 1118) and instructed to instill 1 drop per eye twice daily (morning and evening) for the entire 84-day study treatment period. Subjects were dispensed sufficient study treatment until the next scheduled study visit. Subjects returned to the clinical site at day 14 (visit 3), day 42 (visit 4), and day 84 (visit 5) for scheduled study assessments. Exposure to the controlled adverse environment was conducted at visits 3 through 5 to obtain additional exploratory data as summarized in Table 1 . No supplemental artificial tears or rescue therapy were allowed throughout the duration of the study.
The treating physician, study and site personnel, and subjects were masked to the treatment assignment. All investigators used standardized procedures for objective measures summarized in Table 1 . Packaging of study drug and vehicle (placebo) were identical and contained in low-density polyethylene unit dose vials in numbered study kits that contained the appropriate supply of assigned study treatment until the next scheduled study visit. Both study drug and vehicle were clear and colorless aqueous solutions. Subjects were randomized by assigning each consecutive subject the lowest numbered study kit provided to each study site. The kit numbers were assigned according to a block randomization list generated by an independent statistician.
The primary objective efficacy endpoint (sign) in the study eye was the inferior corneal staining score at day 84 (pre–controlled adverse environment) using the 0−4 point Ora scale. Secondary objective (signs) endpoints were Schirmer test, conjunctival staining score, tear film break-up time, and blink rate. Secondary subjective (symptoms) endpoints were the OSDI, ocular discomfort score, and visual analogue scale.
The primary efficacy endpoint was analyzed across all time points by using an analysis of covariance (ANCOVA, least squares means) model accounting for repeated measures within each subject with inferior corneal staining score as the response and allowing each dose concentration independent comparison to placebo. Mean change from baseline (day 0) analysis to day 84 using a 2-sided t test (t test) and Wilcoxon rank sum test (W) were prespecified confirmatory methods. A 2-sided alpha level of 0.050 was used to determine statistical significance. Statistical hypothesis testing of pairwise comparison of discrete variables used the χ 2 and/or Fisher exact test (F). A sample size of 230 subjects was empirically derived based upon an assumed treatment difference in inferior corneal staining score of 0.45 points between SAR 1118 and placebo, with a common standard deviation of 0.8 units and a correlation of 0.4 between time points. Under these assumptions, a repeated measures analysis would have 80% power to show a significant difference at an alpha of 0.050.
The primary and secondary analyses of the endpoints for the study were conducted in the intent-to-treat (ITT) population (all randomized subjects) and using last-observation-carried-forward (LOCF) data including baseline data. All data reported are the ITT population, except where noted, using the pre–controlled adverse environment data. The per-protocol population was defined prior to unmasking of the database and excluded subjects who withdrew prior to study completion, subjects who violated protocol by taking disallowed medications, subjects who were noncompliant with the study drug, and subjects lost to follow-up.
Safety assessments and recording of adverse events were conducted at all study visits and included BCVA, slit-lamp examination, and dilated funduscopy. All adverse events were classified according to the Medical Dictionary for Regulatory Activities (MedDRA version 11).
The study screened 546 adult subjects to achieve an enrolled population of 230 adult subjects (51 men, 179 women; Table 2 ). Of the 230 subjects, 201 subjects (87%) completed the study and 29 subjects (13%) were discontinued ( Figure 1 ).
|All Subjects (N = 230)||Placebo (N = 58)||SAR 1118|
|0.1% (N = 57)||1.0% (N = 57)||5.0% (N = 58)|
|Intent-to-treat population||230 (100%)||58 (100%)||57 (100%)||57 (100%)||58 (100%)|
|Per-protocol population||186 (81%)||46 (79%)||49 (86%)||47 (83%)||44 (76%)|
|Safety population||230 (100%)||58 (100%)||57 (100%)||57 (100%)||58 (100%)|
|ITT completed||201 (87%)||48 (83%)||54 (95%)||51 (90%)||48 (83%)|
|Adverse event||12 (5%)||1 (2%)||2 (4%)||3 (5%)||6 (10%)|
|Noncompliance||4 (2%)||1 (2%)||0||2 (4%)||1 (2%)|
|Lost to follow-up||3 (1%)||2 (3%)||0||0||1 (2%)|
|Other||10 (4%)||6 (10%)||1 (2%)||1 (2%)||2 (3%)|
|Mean age, years (SD)||62.3 (12.5)||60.4 (12.9)||63.1 (13.1)||63.6 (11.9)||62.3 (12.2)|
|Sex, N (%)|
|Male||51 (22.2)||13 (22.4)||10 (17.5)||17 (29.8)||11 (19.0)|
|Female||179 (77.8)||45 (77.6)||47 (82.5)||40 (70.2)||47 (81.0)|
|Race, N (%)|
|White||213 (92.6)||54 (93.1)||53 (93.0)||53 (93.0)||53 (91.4)|
|Asian||8 (3.5)||1 (1.7)||2 (3.5)||2 (3.5)||3 (5.2)|
|Black||7 (3.0)||3 (5.2)||1 (1.8)||1 (1.8)||2 (3.4)|
|Other||2 (0.9)||0||1 (1.8)||1 (1.8)||0|