AAO 2026: BRUNELLO 1-year results for remigromig in DME

Remigromig (MK-3000, formerly EYE103), a tri-specific agonist of the Wingless-related Integration Site (Wnt) pathway, met its primary non-inferiority endpoint sagainst ranibizumab at 1 year in the pivotal phase 2b/3 BRUNELLO study of adults with diabetic macular edema (DME), according to Donald J. D’Amico, MD. D’Amico presented the results at the American Academy of Ophthalmology (AAO) 2026 Annual Meeting, held October 9-12 in New Orleans.

BRUNELLO is a randomized, double-masked, non-inferiority trial that compared 2 doses of remigromig, 0.5 mg and 0.8 mg, with ranibizumab 0.5 mg, each given monthly by intravitreal injection. The trial enrolled 984 adults with diabetes, visual acuity of 25 to approximately 74 letters, and central subfield thickness of 325 microns or greater. D’Amico described this as a typical DME population and said baseline characteristics were well balanced across groups.

The primary end point was letters gained from baseline at year 1. The 0.5-mg dose gained 9.1 letters and the 0.8-mg dose gained 8.7 letters, compared with 11.8 letters with ranibizumab. D’Amico said these differences are not considered clinically meaningful.

Safety profile and mechanism

D’Amico described the safety profile as excellent, with one exception. There were no episodes of retinal vasculitis, and intraocular inflammation occurred at low rates that were balanced across arms. Indicators of proliferative diabetic retinopathy (PDR) were increased in the remigromig groups, at approximately 6.5% overall compared with about 1% with ranibizumab.

D’Amico said this finding is mechanistically expected. Remigromig works by activating the Wnt pathway, which is important for vascular health in the retina and for restoring and maintaining the blood-retinal barrier. It is orthogonal to anti-VEGF, he said, operating by a completely different molecular mechanism, so it is not designed to be an anti-neovascular agent. He added that the diabetic retinopathy rates in the remigromig arms were better than those seen in earlier control arms, including the sham arm in RISE and RIDE and the laser arm in VISTA and VIVID. Among patients who reached PDR endpoints, 85% had improved vision at year 1.

D’Amico noted that 40% of patients do not achieve satisfactory visual results with anti-VEGF alone, whether from persistent fluid, a relapsed response, or an impermanent response. He said the BRUNELLO results point to a new pathway for treating DME and potentially other retinal vascular diseases. He added that the agents might be used in combination in the future, and that much remains unknown.

A companion trial of similar size and design, BAROLO, is expected to report in the next several months. D’Amico said it will allow further validation of the findings and analyses of subgroups.