Objective To analyze the effect of 3% diquafosol sodium ophthalmic solution on tear film proinflammatory cytokine levels, ocular symptoms, and tear film stability in patients with dry eye disease. Methods This was an open-label, noncomparative, single-cohort, interventional, proof-of-concept study among patients with dry eye disease who were administered diquafosol eye drops six times a day for 4 weeks. Pretreatment and posttreatment tear samples were collected, and ocular surface disease index scores and fluorescein tear break-up time were measured. Interferon-gamma, interleukin-1 beta, interleukin-6, and tumor necrosis factor-alpha levels were measured in the tear samples using a cytokine magnetic bead panel. Independent associations of cytokine responses with sex, age, tear break-up time, and ocular surface disease index were determined. Results Thirty-one eyes (n = 31) in 31 patients with dry eye disease were included. At baseline, interleukin-6, tumor necrosis factor-alpha, and interferon-gamma were detected in all samples, whereas interleukin-1 beta was detected in 3% (1/31) of the samples. Baseline concentrations of interleukin-6, tumor necrosis factor-alpha, and interferon-gamma decreased by ≥25% in 58%, 51.6%, and 45.1% of the eyes, respectively, reaching significance for tumor necrosis factor-alpha (W = 101, z = 2.694, p = 0.006, two-tailed) and interleukin-6 (W = 134, z = 2.224, p = 0.026, two-tailed). The tear break-up time (from 3.95 to 4.65 s; p = 0.013) and ocular surface disease index scores (from 20.57 to 13.85; p = 0.005) significantly improved following treatment. Individual responses were found to be independent of age, sex, eye laterality, pretreatment tear break-up time, and ocular surface disease index scores based on the multivariate logistic regression. Conclusion Diquafosol reduced the levels of proinflammatory cytokines in the tears of patients with dry eye disease after 4 weeks of treatment, suggesting their possible role in modulating ocular surface inflammation. Larger prospective studies are recommended to validate these results.
Tanchuling et al. (Sat,) studied this question.