Effects of emixustat hydrochloride in patients with proliferative diabetic retinopathy: a randomized, placebo-controlled phase 2 study

Abstract

Abstract Purpose To evaluate the effects of oral emixustat hydrochloride on pro-angiogenic and inflammatory cytokines in the aqueous humor, as well as other ophthalmic parameters, in subjects with proliferative diabetic retinopathy (PDR). Methods Twenty-three patients with PDR, with or without diabetic macular edema (DME), were assigned to emixustat or placebo in daily oral doses ranging from 5 to 40 mg over a step-up titration period, for 84 days. The main outcome measures included levels of IL-1β, IL-6, IL-8, TGFβ-1, and VEGF in the aqueous humor. Results Seven of 12 subjects (58%) who were randomized to emixustat and 11 of 12 subjects (92%) who were randomized to placebo completed the study. No statistically significant differences between treatment groups were observed for changes in any of the aqueous humor cytokines tested. However, median VEGF levels were slightly reduced in the emixustat but not the placebo group (− 70.0 pg/mL versus + 42.7 pg/mL, or − 11.8% versus + 6.7%). In a post hoc analysis of all subjects (with or without DME), statistically significant differences between treatment arms in mean changes from baseline in central subfield thickness (CST; emixustat − 11.9 μm, placebo + 36.2 μm; P = 0.076) and total macular volume (TMV; emixustat − 0.13 mm3, placebo + 0.23 mm3; P = 0.026) were observed, both favoring emixustat. Emixustat’s safety profile was consistent with prior studies (i.e., the adverse events of delayed dark adaptation and visual impairment were more common in subjects treated with emixustat). Conclusion Although this pilot study did not demonstrate statistically significant differences in changes in aqueous humor cytokine levels between the emixustat and placebo groups, VEGF levels were slightly reduced in the emixustat but not in the placebo group. In addition, statistically significant differences favoring the emixustat group were observed in CST and TMV among all subjects. These data warrant further investigation of emixustat’s potential therapeutic effects in diabetic retinopathy. Trial registration ClinicalTrials.gov identifier: NCT02753400 (April 2016)