Key points are not available for this paper at this time.
Introduction: Dry eye disease (DED), also known as keratoconjunctivitis sicca (KCS), imposes a substantial global burden, yet the clinical trial landscape guiding therapeutic development remains incompletely characterized. Methods: We conducted a cross-sectional study mapping interventional dry eye disease trials (2000-2024) indexed in Trialtrove. We analyzed 1,171 prospectively registered trials with dry eye disease as the primary indication, characterizing design, phase, geography, sponsorship, endpoints, and therapeutics. Results: Trial activity rose steadily, peaking in 2023-2024, and was concentrated in the United States (n=773), followed by India (n=111), China (n=71), and Japan (n=58). Sponsorship was balanced between industry (45.5%) and academia (44.4%). Most trials were completed and in later phases (Phase 4: 40.3%; Phase 2: 25.0%; Phase 3: 17.0%). Adult cohorts predominated, yet 59.9% did not report severity strata, and where strata were prespecified, thresholds based on symptom questionnaires were heterogeneous. Monotherapy designs were most common (71.6%), with head-to-head comparisons in 22.4%. Frequently studied interventions included other ocular medications, ocular anti-inflammatory agents, and artificial tears. Primary endpoints centered on ocular surface staining, with the Ocular Surface Disease Index, tear film break-up time, and Schirmer's test commonly used. Discussion: These commonly used endpoints vary in psychometric validity and standardization, which can undermine data quality, complicate analysis and interpretation of individual trial results, and limit cross-trial comparability. Overall, the ecosystem appears mature but conservative, emphasizing later-phase, single-agent studies, traditional endpoints, and limited severity stratification and geographic diversity. Future trials should prespecify severity and subtype strata, use validated mechanism-aligned endpoints with standardized assessment windows and longer follow-up, and broaden geographic representation to improve generalizability and cross-trial comparability.
Jiang et al. (Wed,) studied this question.