Matching-adjusted indirect comparison of phase 3 clinical trial outcomes of OC-01 (varenicline solution) nasal spray and lifitegrast 5% ophthalmic solution for the treatment of dry eye disease

BACKGROUND: Matching-adjusted indirect comparison (MAIC) is a methodology for cross-study comparisons after adjusting for baseline characteristic imbalances. It is a comparative analytical approach used across therapeutic areas absent head-to-head trial outcomes. OBJECTIVES: To compare the efficacy of OC-01 (varenicline solution) 0.03 mg nasal spray (OC-01 VNS) to lifitegrast 5% ophthalmic solution on tear production and patient-reported eye dryness in patients with dry eye disease (DED) using data from phase 3 clinical trials via MAIC analysis. METHODS: Individual patient data (IPD) from the phase 3 registrational trial of OC-01 VNS and aggregate data from 2 phase 3 trials of lifitegrast in the publicly available XIIDRA New Drug Application were used. Using unanchored MAIC methods, IPD were weighted on clinically relevant baseline variables (age, race, sex, baseline Schirmer’s test score [STS], and Eye Dryness Score [EDS]) to produce weighted OC-01 VNS datasets matched to the same lifitegrast datasets’ variables. Least-squares (LS) mean change from baseline (CFB) in STS for OC-01 VNS was calculated using the identical analysis of covariance model and covariates used to calculate the same values for lifitegrast in the XIIDRA New Drug Application and was then compared. LS mean EDS (based on a 100- point Visual Analogue Scale) was compared via analysis of covariance in the weighted OC-01 VNS and lifitegrast datasets. OC-01 VNS at 2 and 4 weeks compared to lifitegrast data at 2 and 6 weeks. RESULTS: Data from 511 subjects (n = 260 treated; 251 vehicle control [VC]) in the OC-01 VNS phase 3 trial, 588 (n = 293 treated, 295 VC) in the lifitegrast phase 3 OPUS-1 trial, and 718 (n = 358 treated, 360 VC) in the lifitegrast phase 3 OPUS-2 trial were analyzed. The LS mean STS CFB for OC-01 VNS at 2 and 4 weeks was significantly greater than that for lifitegrast at 2 and 6 weeks in OPUS-1 and OPUS-2 (P < 0.0001 for all comparisons). The LS mean EDS CFB for OC-01 VNS at 2 and 4 weeks was significantly greater than that for lifitegrast at 2 and 6 weeks in OPUS-1 (P < 0.0001 for both comparisons) and at 4 weeks vs lifitegrast at 6 weeks in OPUS-2 (P < 0.0001). CONCLUSIONS: This MAIC analysis demonstrates OC-01 VNS produced significantly greater improvement in mean STS and comparable or greater improvement in EDS compared with lifitegrast in phase 3 trials. These findings suggest a potentially greater magnitude of improvement achieved with OC-01 VNS compared with lifitegrast for the treatment of DED within the conditions of the analysis methodology.