The aim of the study is to investigate whether patients with moderate-to-severe VKC and keratitis will benefit from the addition of Diquafosol ophthalmic solution in terms of symptoms and signs of VKC, tear film metrics, ocular surface inflammation and quality of life. Potential beneficial effect of Diquafosol treatment on top of current standard use of topical immunosuppressant therapy in VKC would be studied.

Subject at least aged 6 and above with clinical diagnosis of vernal keratoconjunctivitis would be invited to join this study and perform the following assessments:

1. best corrected visual acuity,
2. slit lamp biomicroscope examination,
3. tear meniscus height and non-invasive keratographic tear breakup time(NIKBUT) measurement,
4. cornea fluorescein staining (CFS) and
5. Schirmer's I test. Doctor will prescribe eyedrop, 3% Diquafosol ophthalmic solution and 0.1% Cyclosporin A cationic ophthalmic emulsion or 0.1% Cyclosporin A cationic ophthalmic emulsion at the end of examination. Subjects will be followed up at 4 weeks, 8 weeks, 16 weeks after treatment.

All participants will be required to answer a standardised questionnaire relating to severity of VKC and dry eye disease.

Aim To compare the four-month clinical efficacy of topical treatment with 3% Diquafosol ophthalmic solution and 0.1% Cyclosporin A cationic ophthalmic emulsion versus 0.1% Cyclosporin A cationic ophthalmic emulsion in reducing symptoms and signs of active moderate-to-severe vernal keratoconjunctivitis (grade of 2 to 4 on the Bonini severity scale) with keratitis (corneal fluorescein staining score of 1-5 on the modified Oxford scale).

Study design A single-centre prospective double-blinded randomized controlled trial

Hypothesis The addition of topical treatment with 3% Diquafosol ophthalmic solution is superior to 0.1% Cyclosporin cationic ophthalmic emulsion alone in improving symptoms and signs of moderate-to-severe VKC with keratitis at four months of use.

Methods This is a single-centre, prospective, 16-week, randomized, double-blinded controlled clinical trial investigating the clinical efficacy and safety of combined topical treatment with 3% Diquafosol ophthalmic solution and 0.1% Cyclosporin A cationic ophthalmic emulsion and versus topical treatment with 0.1% Cyclosporin A cationic ophthalmic emulsion in patients with moderate-to-severe vernal keratoconjunctivitis and keratitis. The study will also investigate the longitudinal changes in tear cytokines and correlate them with changes in VKC signs and tear film metrics. The study will be performed in accordance with the principles of the Declaration of Helsinki, the International Conference on Harmonization, Good Clinical Practice guidelines, and all applicable laws and regulations. Written informed consent will be obtained from each patient before enrolment in the study. A clinical trials certificate will be obtained from the Department of Health, Hong Kong SAR.

Sample size The study will recruit 94 subjects with moderate-to-severe VKC at eye clinics in Hong Kong West cluster (QMH and GH).

Sample size Calculation Sample size calculation is based on the findings from a prospective study comparing the combined use of 3% Diquafosol and 0.1% Cyclosporin versus 0.1% Cyclosporin alone in the treatment of dry eye disease in 279 subjects [36]. It is anticipated that a similar, if not larger, effective size in VKC patients compared to dry eye patients would be seen. In this study the mean change in cornea staining score between baseline and week 12 was significantly higher in Cyclosporin A and Diquafosol combination therapy (2.91 ± 3.93) than in Cyclosporin A monotherapy (0.93± 2.22). The estimated Cohen's d for this study is approximately 0.620. Based on calculations, approximately 41 participants per group to achieve a study power of 80% with a significance level of 0.05 and the estimated effect size of 0.620 is needed. Considering a 10% dropout rate, the adjusted sample size per group would be approximately 47 participants.

Randomization method Randomisation is conducted by computer-generated random number allocation and will be applied to sequentially enrolled participants. The randomisation schedule is pre-determined, prior to commencing participant recruitment, such that the investigator involved in baseline participant assessment will have no involvement in treatment allocation. Block randomization is performed to ensure 1:1 distribution into treatment groups.

Double-blinding To keep the investigators and subjects blind to the study treatment, all investigational drugs are removed from their commercial packaging, wrapped in aluminium foil, double-wrapped in plastic bags, and distributed to participants by an independent clinical research coordinator.

Rescue Therapy During the study period rescue medication (Loteprednolol 0.5% one drop 4x/day for up to 5 days) is permitted in the event of worsening of ocular surface symptoms. A maximum of 2 courses were allowed between study visits. Any use of rescue therapy is documented in the eyedrop diary and serves as one of the outcome measures.

Patient Workflow The patient workflow is aimed at obtaining clear and comprehensive documentation of symptoms and signs baseline and each subsequent visit after initiation of intervention. Furthermore, documentation of adverse events is achieved via assessment of best corrected visual acuity, intraocular pressure measurement, general periorbital area assessment and slit lamp biomicroscope assessment. See Figure 1. Study Workflow

In particular, the sequence of clinical assessments must be done in order of sequence from least invasive to most invasive, as any touching of the ocular surface or manipulation of the eyelids will stimulate tear secretion. Least invasive tests include imaging-based assessments without manipulation or touching of the ocular surface, include best corrected visual acuity, slit lamp biomicroscope assessment, NIKBUT and tear meniscus height. Mildly invasive tests include those that require gentle touching of the ocular surface, such as fluorescein cornea staining. Moderately invasive tests include those with prolonged ocular surface contact, such as the Schirmer's I test for tear collection

Inclusion Criteria:

• Subject at least aged 6 and above
• Clinical diagnosis of vernal keratoconjunctivitis
• Evidence of active moderate-to-severe disease: At least Grade 2 on the Bonini scale of VKC clinical presentation (moderate-to-severe) [38].
• Evidence of keratitis: At least Grade 1 on the Modified Oxford scale for cornea fluorescein staining as reported by Bron et al [39].
• Experienced 1 or more recurrences of VKC during the previous year

Exclusion Criteria:

• Nasolacrimal duct obstruction
• Impaired blinking function
• Active ocular infection or history of ocular herpes, varicella zoster or vaccinia virus infection
• Any ocular disease that would require topical ocular treatment during the study
• Use of cyclosporin A or tacrolimus eyedrops, or use of systemic immunosuppressants within 3 months before enrolment
• Any ocular surgery within 6 months before enrolment

It is a double-masked randomized study. Assigned treatment of a specific patients would be kept confidential in both the recruited participants and investigators. To keep the investigators and subjects blind to the study treatment, all investigational drugs are removed from their commercial packaging, wrapped in aluminium foil, double-wrapped in plastic bags, and distributed to participants by an independent clinical research coordinator.