Water-Free Cyclosporine Ophthalmic Solution vs Vehicle for Dry Eye Disease

This randomized clinical trial investigates the safety and effectiveness of a water-free cyclosporine ophthalmic solution, 0.1%, for dry eye disease.

Key Points

Question

Is SHR8028, a water-free cyclosporine ophthalmic solution, 0.1%, effective and safe for dry eye disease (DED)?

Findings

In this multicenter, double-blind, phase 3 randomized clinical trial including 206 participants, water-free cyclosporine ophthalmic solution showed superiority over vehicle at 29 days in improving total corneal fluorescein staining (tCFS). Between-group difference in dryness score (visual analog scale [VAS]) at day 29 was not found, and assessment of central CFS, total conjunctival staining, and blurred vision also favored water-free cyclosporine over vehicle.

Meaning

Compared with vehicle, water-free cyclosporine was more effective in improving tCFS and symptom control of DED, although no superiority in improving dryness score (VAS) was noted.

Abstract

Importance

Dry eye disease (DED) is a prevalent eye disorder. Cyclosporine is an effective immunomodulator that is widely used in DED; however, due to its highly hydrophobic nature, delivery of cyclosporine to the ocular surface is challenging.

Objective

To evaluate the efficacy and safety of SHR8028, a water-free cyclosporine ophthalmic solution, 0.1%, compared with vehicle in Chinese participants with DED.

Design, Setting, and Participants

This was a multicenter, double-blind, vehicle-controlled, phase 3 randomized clinical trial conducted from March 4, 2021, to July 22, 2022. Adult participants with moderate to severe DED were recruited from 12 hospitals in China. Study data were analyzed April 2, 2022, for the primary analysis.

Interventions

Following a 14-day run-in period with an artificial tear, participants were randomly assigned (1:1) to receive water-free cyclosporine or vehicle (1 eye drop in each eye twice daily). After a 29-day treatment, participants of both groups were given the option to receive water-free cyclosporine for an additional 12 weeks for longer-term safety assessment.

Main Outcomes and Measures

The primary end points were changes from baseline in total corneal fluorescein staining (tCFS) using the National Eye Institute scale and in dryness score on a visual analog scale at day 29.

Results

A total of 206 participants (mean [SD] age, 47.8 [14.2] years; 185 female [90%]) were enrolled, with 103 each in the cyclosporine group and the vehicle group. At day 29, the cyclosporine group experienced improved tCFS compared with vehicle (change [Δ] = −1.8; 95% CI, −2.7 to −1.0; P < .001), with a tCFS score decrease from baseline of −4.8 in the cyclosporine group and −3.0 in the vehicle group. Dryness score decreased from baseline in both groups (−19.2 vs −15.4; Δ = −3.8; 95% CI, −9.2 to 1.6; P = .17). During the 29-day treatment, treatment-related adverse events were reported in 15 participants (14.6%) in the cyclosporine group and 11 participants (10.7%) in the vehicle group.

Conclusions And Relevance

Results demonstrated superiority of a water-free cyclosporine, 0.1%, eye solution over vehicle in improving tCFS score at day 29 in Chinese participants with DED. However, dryness score (VAS) was not improved at day 29.

Trial Registration

ClinicalTrials.gov Identifier: NCT05841043

Introduction

Dry eye disease (DED) is characterized by loss of homeostasis of the tear film, which leads to discomfort and impaired visual performance. The traditional first-line treatment is over-the-counter artificial tear products. However, this approach only offers temporary relief. Ocular surface chronic inflammation is a key underlying factor in DED. Thus, control of inflammation is a recommended approach for patients not adequately treated with artificial tear products.

Cyclosporine is a highly effective immunosuppressant widely used for moderate to severe DED. Several cyclosporine eye drops were approved for DED management, including Restasis (AbbVie) and Cequa (Sun Pharmaceuticals) in the US, Ikervis (Santen) in EU, and Zirun (Sinqi) in China.Due to the highly hydrophobic nature and large molecular weight of cyclosporine, as well as the rapid foreign-substances clearance property of the eye, ocular drug delivery is challenging. The aforementioned medications use oil in water emulsions or nanosized micellar formulations to deliver cyclosporine to ocular surface, respectively. However, low bioavailability, ocular intolerance, ocular discomfort, or visual disturbances were reported. There is still an unmet need for novel delivery strategies.

SHR8028 (CyclASol [Novaliq]), 0.1%, is a water-free cyclosporine ophthalmic formulation using a novel EyeSol (Novaliq) ocular drug delivery technology based on perfluorobutylpentane (F4H5). The product obtained US Food and Drug Administration approval under the trade name Vevye (Novaliq) in May 2023 in the US. The vehicle F4H5 is a colorless liquid free of oils, surfactants, and preservatives, with a high vapor pressure and a low surface tension. A previous phase 2b/3 study showed that the water-free cyclosporine, 0.1%, significantly reduced corneal and conjunctival staining and improved ocular dryness in participants with DED, compared with its vehicle. Later, a phase 3 pivotal study, CyclASol for the Treatment of Signs and Symptoms of Dry Eye Disease (ESSENCE-2), was conducted in the US to further validate the superiority of water-free cyclosporine, 0.1%, over vehicle in participants with moderate to severe DED. This study was designed in parallel to the ESSENCE-2 trial as a bridging study to compare the efficacy and safety of water-free cyclosporine vs vehicle in China.

Methods

Study Design and Assessments

This was a multicenter, double-blind, vehicle-controlled, phase 3 randomized clinical trial conducted at 12 hospitals in China from March 4, 2021, to July 22, 2022. This study was planned in parallel with the ESSENCE-2 trial in the US as a bridging study, and the protocol was developed for both studies. This study was conducted in accordance with the Declaration of Helsinki and the Good Clinical Practice of the International Council for Harmonization. The study protocol (Supplement 1) was reviewed and approved by the independent ethics committee of all sites. Study participants provided written informed consent. No incentive was given for study participants to participate. The Consolidated Standards of Reporting Trials (CONSORT) reporting guidelines have been followed.

This study consisted of a 14-day run-in period, a 29-day treatment period, and a 12-week longer-term follow-up period. During the run-in period, eligible participants received artificial tears (Systane Balance [Alcon Laboratories Inc]) dosed as 1 eye drop in each eye twice daily. On confirmation of the eligibility criteria, participants were randomly assigned (1:1), using an interactive web response system, to receive SHR8028 (hereafter referred to as water-free cyclosporine, 0.1%) or vehicle, 1 single eye drop in each eye twice daily. Randomization was stratified by dryness score on a visual analog scale (VAS; <75 vs ≥75). Block randomization was used, with a block size of 4. All participants, investigators, and other staff involved were masked to assignment. The protocol was amended on October 20, 2021, to add a 12-week longer-term follow-up period to assess the longer-term efficacy and safety of the water-free cyclosporine, 0.1% (Supplement 1). At the end of 29-day treatment, participants in both the cyclosporine and vehicle groups who wished to continue the study entered into the longer-term follow-up period, during which all were given cyclosporine for an additional 12 weeks.

Assessments of efficacy, safety, and tolerability for both eyes were planned at 5 visits: visit 0 (screening; day −14 ± 2), visit 1 (baseline/randomization; day 1), visit 2 (day 15 ± 2), visit 3 (day 29 ± 2), and visit 4 (day 113 ± 3). Total and subregion corneal fluorescein staining (CFS) was measured using the National Eye Institute (NEI) scale. Dryness score was measured with the VAS. Total conjunctival lissamine green staining (LGCS) was assessed using the Oxford scale. Higher scores on all 3 scales (NEI scale, VAS, and Oxford scale) indicated worse conditions. Treatment-emergent adverse events (TEAEs) were documented. Treatment compliance was assessed based on records in the participant diary from which the actual eye drops administered were compared with the anticipated eye drops.

Participants

Eligible participants were adults aged 18 years or older, had a history of DED in both eyes at least 180 days before screening, and were currently using over-the-counter eye drops, eye moisturizing gel, neural stimulation devices, or artificial tears. The majority of the study participants were Han Chinese. Because this was a bridging study assessing the efficacy of a water-free cyclosporine ophthalmic solution, 0.1%, in the Chinese population, no other races were included. Participants had a dryness VAS score of 50 or higher. At least 1 eye (the same eye/study eye) of each participant was required to meet all the following criteria: tCFS score of 10 or higher at screening and randomization, total LGCS score of 2 or higher, and unanesthetized Schirmer I test score of 1 mm or more and 10 mm or less. Participants were excluded if they presented any of the following criteria: clinically relevant abnormal slitlamp findings, active blepharitis or other abnormalities of the eyelid margin, abnormal eyelid anatomy or abnormal blinking, DED secondary to scarring, history of herpetic keratitis, current use of contact lenses, persistent ocular or system infection, topical use of cyclosporine or lifitegrast within 60 days before randomization, intraocular or laser surgery within 180 days before screening, or a plan to have eye or eyelid surgery during the study treatment.

If both eyes met the criteria, the eye with the higher tCFS score at randomization was designated as the study eye. If the tCFS score of both eyes was the same, the right eye was selected as study eye.

End Points

The 2 primary end points were change from baseline in tCFS score using the NEI scale and dryness score on the VAS at day 29. The key secondary end points were change from baseline in tCFS score (NEI scale) at day 15, change from baseline in central CFS score (NEI scale) at day 29, change from baseline in total LGCS score (Oxford scale) at day 29, tCFS response rate (NEI scale) at day 29 (defined as the percentage of participants with a 3-grade or higher improvement in the total cornea), central CFS response rate (NEI scale) at day 29 (defined as the percentage of participants with a grade-1 or higher improvement in the central region of the cornea), and change from baseline in blurred vision using the VAS at day 29. The other secondary end points are listed in the eAppendix in Supplement 2. Systemic and ocular adverse events were reported for safety assessment. Other safety measurements included assessments of visual acuity, intraocular pressure, slitlamp biomicroscopy, and dilated ophthalmoscopy. Tolerability was assessed based on the Drop Comfort Scale and Questionnaire at day 1 and the Eye Drops Acceptability Questionnaire at day 29.

Statistical Analyses

The sample size calculation in the present bridging study was based on the effect preservation strategy. Assuming that the efficacy of the water-free cyclosporine in China was the same as that in the region of the ESSENCE-2 study, a sample size of 170 participants in total could provide this bridging study with 90% probability of showing consistency with the results of the ESSENCE-2 study. Taking into account the safety evaluation and participant dropout, 200 participants were required for this trial. Hierarchical testing was used to control multiplicity across the 2 primary end points, with tCFS being tested first. Data from the ESSENCE-2 trial and the current study were pooled for the analysis of bridging success (ie, if the efficacy of the drug was extrapolated to the Chinese population). For cyclosporine to be considered superior to vehicle in the Chinese population, the following criteria were applied: (1) the treatment effect in the entire population (pooled data) reached statistical significance with a 2-sided α of less than .05, and (2) the ratio of the between-group difference in the Chinese population to that in the entire population exceeded 50%. P values for other comparisons were nominal, with no adjustment for multiplicity.

Efficacy and safety were analyzed in all participants who were randomized and received at least 1 dose of study medication. The primary end points were analyzed using a covariance model on the study eye, with baseline values and treatment groups as factors. Least squares (LS) mean for each treatment group and for the difference between groups were provided, as well as 2-sided P values and 95% CIs. The quantitative key secondary end points were analyzed using the same methods as for the primary analysis. The binary key secondary end points were calculated using descriptive statistics and a logistic regression model. Predictive marginal proportion for each group, difference of predictive marginal proportion, as well as 2-sided P values and 95% CIs were presented from the logistic regression model. The tCFS responder analysis as a tool to underline relevance was conducted as an ad hoc analysis. Data were analyzed using SAS software, version 9.4 (SAS Institute) or higher. Study data were analyzed as of April 2, 2022, for the primary analysis, and as of July 22, 2022, for the longer-term follow-up.

Results

Patient Disposition and Baseline Characteristics

The recruitment period was March 18, 2021, to January 26, 2022. A total of 251 participants were screened, and 206 (mean [SD] age, 47.8 [14.2] years; 21 male [10%]; 185 female [90%]) were enrolled, with 103 each in the cyclosporine and vehicle groups. A total of 204 participants (99.0%) completed the 29-day study treatment, and 2 (1.0%) withdrew consents (Figure 1). Two hundred two participants (98.1%) were Han Chinese, and 4 participants (1.9%) were of other ethnic groups.

Figure 1. Study Participant Disposition Over the Study Treatment and Longer-Term Follow-Up.

Participants had a mean (SD) tCFS score of 12.2 (1.8) and a mean (SD) dryness score of 73.0 (12.8). The baseline characteristics were well balanced between groups (Table 1).

Table 1. Baseline Demographics and Disease Characteristics.

Characteristics	Mean (SD)
	Cyclosporine (n = 103)	Vehicle (n = 103)
Age, y	46.7 (14.1)	48.9 (14.4)
Sex, No (%)
Male	14 (13.6)	7 (6.8)
Female	89 (86.4)	96 (93.2)
Dry eye disease duration, mo	61.9 (61.0)	47.3 (51.6)
Baseline ocular characteristics
tCFS (NEI scale)	12.2 (1.9)	12.3 (1.8)
Central CFS (NEI scale)	2.3 (0.9)	2.4 (0.7)
Dryness score (VAS), No. (%)	73.6 (12.2)	72.4 (13.3)
<75	55 (53.4)	56 (54.4)
≥75	48 (46.6)	47 (45.6)
Blurred vision	48.3 (27.3)	51.3 (24.5)
Total conjunctival staining (Oxford scale)	4.4 (2.1)	4.3 (2.3)
Unanesthetized Schirmer test	3.6 (2.6)	4.0 (2.5)
Tear film break-up time (VAS), s	2.4 (1.1)	2.4 (1.1)
OSDI total score	48.7 (21.8)	50.6 (23.6)

Abbreviations: NEI, National Eye Institute; OSDI, Ocular Surface Disease Index; tCFS, total corneal fluorescein staining; VAS, visual analog scale.

Efficacy

At day 29, tCFS score improvement from baseline was higher in the cyclosporine group (LS mean, −4.8) than in the vehicle group (LS mean, −3.0; change [Δ] = −1.8; 95% CI, −2.7 to −1.0; P < .001) (Table 2 and Figure 2). The dryness score improved in both groups, with a decrease of −19.2 and −15.4 points from baseline, respectively (Δ = −3.8; 95% CI, −9.2 to 1.6; P = .17) (Table 2 and Figure 2). Pooled analysis of the entire population showed significantly greater improvement in tCFS score with cyclosporine treatment compared with vehicle (Δ = −0.7; 95% CI, −1.1 to −0.3; P < .001), and the treatment effect in tCFS in the Chinese population was 257% of that in the entire population (−1.8 vs −0.7). In the entire population, the LS mean between-group change from baseline in dryness score was 0.3 (95% CI, −2.5 to 3.1; P = .83).

Table 2. The Primary and Key Secondary End Points in the Chinese Population (FAS)^a.

Measures	Day	Change from baseline, LS mean (95% CI)		Between group difference, LS mean (95% CI)	P valueb
		Cyclosporine (n = 103)	Vehicle (n = 103)
Primary
tCFS	29	−4.8	−3.0	−1.8 (−2.7 to −1.0)	<.001
Dryness score	29	−19.2	−15.4	−3.8 (−9.2 to 1.6)	.17
Key secondary
tCFS	15	−4.0	−2.8	−1.2 (−2.0 to −0.3)	.01
Central CFS	29	−1.0	−0.6	−0.4 (−0.6 to −0.1)	.01
Total LGCS	29	−1.4	−0.8	−0.6 (−1.1 to −0.2)	.003
Blurred vision	29	−9.2	−2.8	−6.4 (−11.1 to −1.7)	.01
Central CFS response rate, %c	29	61.5	52.1	9.5 (−4.2 to 23.1)	.17
tCFS Response rate, %b	29	78.6	47.2	31.4 (18.5 to 44.3)	<.001

Abbreviations: FAS, full analysis set; LGCS, lissamine green conjunctival staining; LS, least squares; tCFS, total corneal fluorescein staining.

^a According to the protocol, if the overall dropout rate for this study was less than 5%, missing values were not imputed. At day 15, 98 participants in the cyclosporine group and 101 participants in the vehicle group were evaluable. At day 29, 99 and 102 participants were evaluable in the 2 groups, respectively.

^b P values for key secondary end points were nominal.

^c Data presented are predictive marginal proportion.

Figure 2. Change From Baseline in Total Corneal Fluorescein Staining (CFS) National Eye Institute (NEI) Scale and Dryness Score (Visual Analog Scale [VAS]) Over the 29-Day Treatment Period Full-Analysis Set.

Data are least squares mean. Error bars show SE.

At day 15, a higher improvement in tCFS score with cyclosporine over vehicle was already observed (Table 2 and Figure 2). At day 29, change from baseline in central CFS (Δ = −0.4; 95% CI, −0.6 to −0.1), total conjunctival staining (Δ = −0.6; 95% CI, −1.1 to −0.2), and blurred vision (Δ = −6.4; 95% CI, −11.1 to −1.7) favored cyclosporine over vehicle (Table 2). At day 29, the proportion of participants who had a grade-3 or higher improvement in tCFS score (considered a tCFS responder) was higher in the cyclosporine group than in the vehicle group (Δ = 31.4%; 95% CI, 18.5-44.3). Sixty-one participants (61.5%) vs 53 (52.1%) achieved a grade-1 or higher improvement in central CFS, respectively (Δ = 9.5; 95% CI, −4.2 to 23.1) (Table 2). Improvements in secondary end points were listed in eTables 1 to 6 in Supplement 2.

In an ad hoc analysis, tCFS responders irrespective of treatment demonstrated higher improvements in various symptoms per the VAS at day 29 than nonresponders (Δ ranged from −0.2 to −9.4), and the improvement between responders and nonresponders reached nominal significance in frequency of dry eye, blurred vision, fluctuating vision, difficulty in reading, difficulty in looking at screens, and awareness of dry eye (eTable 7 in Supplement 2).

Safety and Tolerability

The mean (SD) treatment compliance was 98.2 (9.7) and 97.5 (7.6) in the cyclosporine and vehicle groups, respectively. The distribution and severity of TEAEs were balanced between the 2 groups. A total of 35 participants (34.0%) in the cyclosporine group and 31 (30.1%) in the vehicle group reported TEAEs. No participants delayed or discontinued study treatment due to TEAEs. No serious TEAEs were reported. Treatment-related AEs occurred in 15 participants (14.6%) and 11 participants (10.7%) in the cyclosporine and vehicle groups, respectively, and all were mild in severity. The most common TEAEs were reduced visual acuity (13 participants [12.6%] in the cyclosporine group vs 18 [17.5%] in the vehicle group) (Table 3). The comfort and treatment satisfaction levels were high among participants in both groups, whereas participants in the cyclosporine group were more willing to continue treatment than those in the vehicle group (eTables 8 and 9 in Supplement 2).

Table 3. Summary of Treatment-Emergent Adverse Events (TEAEs) Over the 29-Day Treatment Duration.

TEAEs	No. (%)
	Cyclosporine (n = 103)	Vehicle (n = 103)
All causality
TEAEs, No.	56	41
Participants with at least 1 TEAE	35 (34.0)	31 (30.1)
Ocular TEAEs	42	27
Participants with at least 1 ocular TEAE	26 (25.2)	22 (21.4)
Treatment-related
TEAEs, No.	21	11
Participants with at least 1 TEAE	15 (14.6)	11 (10.7)
Ocular TEAEs	21	11
Participants with at least 1 ocular TEAE	15 (14.6)	11 (10.7)
Ocular TEAEs that occurred in >2% of study participants in either group
Visual acuity reduced	13 (12.6)	18 (17.5)
Ocular discomfort	4 (3.9)	2 (1.9)
Eye pain	3 (2.9)	2 (1.9)
Vision blurred	3 (2.9)	2 (1.9)
Instillation site reactions	3 (2.9)	0

Longer-Term Follow-Up

Efficacy

At the end of the 29-day treatment, 38 participants from the cyclosporine group and 39 from the vehicle group voluntarily proceeded to receive cyclosporine (Figure 1). Therefore, the data presented for longer-term follow-up were not derived from randomized comparison. Over the following 12-week follow-up period, tCFS and dryness score continued to decrease in all 77 participants. The mean (SD) change in tCFS score from day 29 was −1.3 (3.0) and −3.2 (3.5) in the cyclosporine group and the vehicle switched to cyclosporine (hereafter referred to as vehicle-cyclosporine) group, respectively. The mean (SD) change in dryness score from day 29 in the cyclosporine and vehicle-cyclosporin groups was −11.9 (20.4) and −15.4 (22.1), respectively.

Safety

Over the 12-week follow-up duration, TEAEs were reported in 10 of 38 participants (26.3%) from the cyclosporine group and 9 of 39 participants (23.1%) from the vehicle-cyclosporine group. Of them, 2 participants (5.3%) from the cyclosporine group and 5 participants (12.8%) from the vehicle-cyclosporine group had TEAEs deemed related to study treatment. No participants discontinued study treatment because of TEAEs or experienced serious TEAEs. Six participants (15.8%) and 8 participants (20.5%) from the cyclosporine and vehicle-cyclosporine group had ocular-related TEAEs, with the most common ones being reduced visual acuity in both groups (7.9% and 10.3%, respectively) (eTable 10 in Supplement 2).

Discussion

This randomized clinical trial assessed a water-free cyclosporine, 0.1%, in Chinese participants with moderate to severe DED. At day 29, tCFS score improvement was significantly higher with cyclosporine vs vehicle. Dryness score showed treatment benefits in both groups at day 29. After an additional 12-week treatment with cyclosporine, tCFS and dryness scores continued to improve.

Cyclosporine is a highly hydrophobic molecule with large molecular weight, and the unique anatomy and physiology features of the eye further complicate the delivery of cyclosporine to the eye. The EyeSol (Novaliq) delivery platform was developed to overcome this barrier. The surface tension of the vehicle is low, allowing it to spread rapidly over the ocular surface without activating the blinking and tearing reflex. The vehicle has a similar refractive index to water. Given these properties, cyclosporine showed high bioavailability without compromising visual function. The vehicle itself markedly alleviates symptoms and signs of DED, which partly explained why the between-group difference in dryness score did not reach statistical significance.

This study and the ESSENCE-2 study showed similar findings in the primary and key secondary outcomes. The between-group difference in the mean change in tCFS score was more prominent in this study than that in the ESSENCE-2 study (−1.8 vs −0.4). Combining the results of this bridging study and the original ESSENCE-2 study, pooled analysis of the entire population showed greater improvement in tCFS with cyclosporine treatment compared with vehicle (Δ = −0.7) and improvement in dryness score in both the treatment and control groups. Considering that NEI grading is nonlinear, the improvement in tCFS score was especially relevant for patients with high staining scores. It was reported in the ESSENCE-1 and ESSENCE-2 studies that tCFS responders showed clinically important improvements in reading speed and various ocular symptoms compared with nonresponders. Consistent with these findings, in this study, tCFS responders, regardless of treatment, were associated with improvements in a variety of symptoms of DED. This further demonstrates that a grade-3 or higher improvement in tCFS (a meaningful improvement for caregivers) is also relevant for the patients with regard to their symptomatology. In this study, cyclosporine demonstrated higher tCFS response rate compared with vehicle, indicating that cyclosporine could bring clinically meaningful improvements in tCFS score. In addition, other end points measured using objective staining showed greater improvement with cyclosporine vs vehicle at day 29. These results were consistent with those reported in the ESSENCE-2 study and demonstrated restoration of the ocular surface integrity. DED is associated with a disorder of tear quality or quantity on the ocular surface, and symptom control of DED is important to improve patients’ quality of life. In this study, the blurred vision VAS score and unanesthetized Schirmer I test score of all participants improved at day 29 with superiority in cyclosporine over vehicle. Moreover, all other symptoms reported by participants, as measured by VAS and the Ocular Surface Disease Index, improved during study treatment in both the cyclosporine and vehicle groups.

The incidence and distribution of TEAEs were similar between groups. No serious TEAEs were reported. These results were consistent with those reported in previous studies using the water-free cyclosporine, 0.1% (CyclASol [Novaliq]). For other safety evaluations, the presenting visual acuity and intraocular pressure were stable, and no significant changes were observed in slitlamp biomicroscopy and dilated ophthalmoscopy. Cyclosporine treatment for another 12 weeks did not result in an increased risk of TEAEs, and no new safety signals were observed. Cyclosporine and vehicle were considered comfortable, the instillation-site reactions were low (2.9% with cyclosporine vs 0 with vehicle), and the comfort assessment yielded satisfactory results.

Limitations

One limitation of this study is that participants were classified as having predominantly aqueous deficient disease; therefore, the clinical outcome of cyclosporine in other forms of DED remains inconclusive. A second limitation is the duration of treatment and follow-up. A 29-day treatment period and an additional 12-week follow-up were relatively short compared with the life-long course of DED. The long-term efficacy and safety are currently being assessed in a 52-week open-label study (CYS-005). Lastly, the study results may not generalize to populations of other ethnicities as this study was conducted among the Chinese population.

Conclusions

In this randomized clinical trial, water-free cyclosporine, 0.1%, effectively reduced tCFS score in Chinese participants with DED compared with vehicle. Assessment of central CFS, total conjunctival staining, and blurred vision also favored cyclosporine over vehicle on day 29. The eye drop was well tolerated with high treatment satisfaction.

Supplement 1.

Trial Protocol.

Supplement 2.

eAppendix. Secondary End Points

eTable 1. Central CFS Score at Day 15 and Change From Baseline (NEI Scale)

eTable 2. Total Conjunctival Lissamine Green Staining Score at Day 15 and Change From Baseline (Oxford Scale)

eTable 3. Symptom Assessments Using VAS and Change From Baseline at Each Visit

eTable 4. OSDI Total Score and Change From Baseline at Each Visit

eTable 5. Unanesthetized Schirmer I Test Results at Day 29 and Change From Baseline

eTable 6. Unanesthetized Schirmer I Test Response Rate at Day 29

eTable 7. Improvement in Symptoms per VAS in tCFS Responders vs Nonresponders Irrespective of Treatment at Day 29

eTable 8. Eye Drop Comfort Questionnaire

eTable 9. Eye Drops Questionnaire (on a scale of 1-10)

eTable 10. Summary of TEAEs Following Treatment With SHR8028 Over the 12-Week Follow-Up Period

Supplement 3.

Data Sharing Statement.