This prospective, interventional, randomized, controlled, double-blind clinical study evaluates the effect of Low-Level Light Therapy (LLLT) delivered by a polychromatic periocular mask (Eye-light®️) versus a sham device in adults with dry eye disease (DED) associated with Graves' ophthalmopathy, refractory to conventional topical treatment. The LLLT protocol applies photobiomodulation to eyelid and periocular tissues to support meibomian gland function and reduce tear inflammatory biomarkers. The primary endpoint is the change in tear film stability measured by Non-Invasive Tear Break-Up Time (NIBUT) and Fluorescein Tear Break-Up Time (FBUT). Secondary outcomes include symptom scores (OSDI), Schirmer I test, tear osmolarity, ocular surface staining scales, automated conjunctival hyperemia, blink analysis, meibomian gland morphology (meibography/meiboscale), tear inflammatory biomarkers, and safety/tolerability.

This randomized, single-masked, parallel-group clinical trial was conducted to evaluate the efficacy and safety of low-level light therapy (LLLT) in the management of dry eye disease (DED) in patients diagnosed with Graves Ophthalmopathy. The study was carried out at the Department of Ophthalmology of Clinical Hospital CF Iaşi, Romania, University of Medicine and Pharmacy "Grigore T. Popa" (Iași, Romania), in accordance with the Declaration of Helsinki and ICH-GCP guidelines.

Dry eye disease is a multifactorial disorder characterized by tear film instability and ocular surface inflammation. Patients with Graves' ophthalmopathy often experience a severe form of dry eye that is resistant to conventional topical treatments, resulting in a need of finding new therapies to stabilize the tear-film.

LLLT is a non-invasive photobiomodulation technique that delivers low-intensity, polychromatic red and near-infrared light to target mitochondrial cytochromes and promote cellular activity. Its application in ophthalmology has shown potential to improve meibomian gland function, enhance tear film stability, and downregulate inflammatory cytokines on the ocular surface. However, its utility in patient with diagnosed DED associated with Graves Ophthalmopathy has not been formally evaluated.

Eligible participants included consecutive adults (≥18 years) diagnosed with DED associated with Graves Ophthalmopathy that are refractory to conventional dry eye treatments. Key exclusion criteria were pregnant or breastfeeding patients, uncontrolled systemic disease (other than Graves Disease), use of general corticosteroids in the last 6 months, ocular and orbital surgery in the last 6 months, use of contact lenses in the last month and any active ocular infection or allergy.

Eligible participants were randomly assigned to each the active low-level light treatment group or the sham-control group with the Eye-light unit (Espansione Group Ltd, Funo, Italy). Randomization was performed using a computerized allocation sequence (http://www.sealedenvelope.com). Allocation concealment was ensured through sequentially numbered, opaque, sealed envelopes. Both participants and investigators responsible for clinical evaluations remained unaware of group assignment throughout the study period.

Participants were randomized into two groups: the intervention group received LLLT using the Eye-light® device (Espansione Group, Italy), and the control group received sham treatment with an identical, inactive device. Each treatment session lasted a total of 15 minutes. The LLLT masks consisted of light emitting diodes (LEDs) with red light, 633 nm wavelengths. Participants attended a total of seven visits. The first 4 visits comprised of treatment sessions, each session separated by 1 week. The next 3 visits comprised of a follow-up assessment, 1 month following the final treatment session, 60±7 days after the last treatment and 90±7 days after the last treatment, targeting the periocular region and meibomian glands.

Clinical assessments were performed at baseline (T0), before first mask session, 7 days after T0 (T1), 7 days after T1 (T2), 7 days after T2 (T3), 30±4 days after T3 (T4), 60±7 days after T4 (T5) and 90±7 days after T5 (T6), and included:

1. Tear Collection Procedure: Tear samples were collected using Schirmer type I strips without topical anesthesia to assess changes in tear inflammatory biomarkers using Luminex Human Discovery Assay 9 plex- CD 27, CD 40, ICAM 1, IL 1 beta, IL 10 , MMP8, TNF alfa, TNF RI, TRAIL R2 in patients with Graves' ophthalmopathy . The strips were placed in the lower conjunctival fornix for up to 5 minutes, then individually stored in sterile microtubes at -80 °C.
2. Tear Protein Extraction: For protein extraction, the strips were thawed and incubated in 400 μL of 1.5 M Tris-HCl buffer (pH 8.8) containing protease inhibitors for 3 hours, followed by centrifugation at 16,000× g for 15 minutes (4 °C). Total protein content was determined by Nanodrop (ThermoFischer Scientific), and tear cytokines and chemokines were quantified using a multiplex bead-based immunoassay (R&D Systems) on a Luminex FlexMap3D platform.
3. Analysis Method: Tear proteins will be analyzed using the Luminex® multiplex immunoassay, as recommended by the manufacturer.
4. Clinical Measurements: Ocular Surface Disease Index (OSDI) scores, Schirmer 1 test, number of total blinks in 30s and number of partial blinks within that timeframe, tear meniscus height, subjective lipid layer pattern grading, automated bulbar conjunctival and limbal hyperemia (DEA 520), fluorescein corneal staining and fluorescein staining using the Oxford scale, tear osmolarity and meibography drop-out graded with the meiboscale for upper and lower eyelids, will be measured at each session (T0-T6).

The primary outcome was the change in tear film stability (NIBUT and FBUT) from baseline to 3 months post treatment. Secondary outcomes included all objective ocular surface parameters. Safety was evaluated by monitoring treatment-emergent adverse events and assessing patient tolerability of the device.

Statistical analysis was performed using SPSS (version 30.0.0.0, SPSS Inc., Chicago, IL, USA). The normality of data was evaluated using the Shapiro-Wilk test. For normally distributed variables, repeated measures ANOVA followed by Dunnett's post-hoc tests were used. For non-normally distributed data, the Friedman test with Dunn's correction was applied. A p-value < 0.05 was considered statistically significant.

This is the first prospective study to investigate LLLT in the diagnosed DED associated with Graves Ophthalmopathy patients. The findings aim to provide evidence for integrating non-pharmacological interventions in ocular surface optimization for this category of patients.

Inclusion Criteria:

• Diagnosed with dry eye disease, particularly associated with Graves' ophthalmopathy.
• Refractory to conventional dry eye treatments (e.g., artificial tears, cyclosporine, corticosteroid eye drops).
• Adult patients (over 18 years old).

Exclusion Criteria:

• Pregnant or breastfeeding.
• Uncontrolled systemic diseases (other than Graves' disease).
• Use of general corticosteroids in the last 6 months.
• Ocular or orbital surgery in the last 6 months.
• Use of contact lenses within the last week.
• Active ocular infection or allergy.

De-identified individual participant data (IPD) will be shared, including demographics (age, sex), baseline and follow-up ocular surface measures, including BUT, Schirmer I test, tear osmolarity, ocular surface staining scales, automated conjunctival hyperemia, blink analysis, meibomian gland morphology, tear inflammatory biomarkers, and thyroid-related clinical scores and biomarkers. No personally identifiable information will be shared. Data will be provided upon reasonable request for ethics/IRB-approved research purposes only.