AAO 2026: DAYBREAK shows tarcocimab and tabirafusp-ted noninferior to aflibercept in wet AMD

At the American Academy of Ophthalmology (AAO) 2026 Retina Subspecialty Day in New Orleans, Charles C. Wykoff, MD, PhD, presented 1-year primary end point results from the phase 3 DAYBREAK trial. The trial evaluated 2 investigational intravitreal therapies against a shared aflibercept 2 mg (Eylea; Regeneron) control arm in eyes with neovascular age-related macular degeneration (nAMD).

Study design

DAYBREAK included 2 separate pivotal analyses within a single trial, each with its own statistical analysis plan and alpha. The control arm received aflibercept as 3 monthly doses followed by dosing every 8 weeks. The primary end point was change in best-corrected visual acuity (BCVA) from baseline at weeks 40, 44, and 48, and follow-up continues through 2 years. Baseline characteristics were balanced, with mean BCVA of approximately 20/60 and central subfield thickness (CST) of approximately 350 µm.

Tarcocimab tedromer (Kodiak Sciences) is an anti-VEGF biologic given as a 5-mg injection. Of that dose, 1 mg is unconjugated protein and 4 mg is conjugated to an optically clear biopolymer intended to extend ocular residence time. Tabirafusp alfa tedromer (tabirafusp-ted; Kodiak Sciences) is an anti-VEGF/anti–interleukin 6 (IL-6) bispecific. It is also a 5-mg injection, with 1.5 mg free protein and 3.5 mg conjugated.

AI-guided retreatment

Wykoff described the retreatment approach as the trial's most interesting element. At monthly visits, sites uploaded optical coherence tomography (OCT) volume scans to an artificial intelligence algorithm that quantified intraretinal and subretinal fluid. Treatment was given when disease activity exceeded a low threshold. "Critically, we did not wait for a specific CST increase, and we did not wait for visual acuity loss," Wykoff said.

Tarcocimab results

After 4 monthly loading doses, tarcocimab was given as needed at monthly visits, up to a maximum interval of 6 months. The primary end point was met. At week 48, BCVA improved by a mean of 7.2 letters with tarcocimab (n = 220) and 7.6 letters with aflibercept (n = 224). CST trajectories overlapped between arms. At the primary end point, 54% of tarcocimab-treated patients had reached the 6-month interval.

No cases of intraocular inflammation (IOI), occlusive retinal vasculitis, or endophthalmitis occurred in either arm. Cataract was reported in 1 patient receiving tarcocimab and 2 patients receiving aflibercept.

Tabirafusp-ted results

Tabirafusp-ted (n = 225) was given as 4 monthly doses, then every 8 weeks, with additional as-needed doses at monthly visits in between. Both BCVA and CST were alpha-protected end points. Tabirafusp-ted met noninferiority for BCVA, with gains of 6.3 letters vs 7.6 letters with aflibercept (n = 225). It also met noninferiority for CST, with reductions of 131 µm vs 138 µm.

One case of IOI occurred with tabirafusp-ted and none with aflibercept. Cataract occurred in no patients receiving tabirafusp-ted and 2 receiving aflibercept.

Next steps

Wykoff said Kodiak anticipates submitting a multi-indication biologics license application for tarcocimab in 2026. That application would be supported by pivotal data from DAYBREAK, DAYLIGHT, GLOW1, GLOW2, and BEACON. For tabirafusp-ted, he said additional subanalyses are needed to identify nAMD subpopulations that may benefit from IL-6 inhibition. The molecule is also being studied in diabetic macular edema.

Reference

Wykoff CC, et al. [The DAYBREAK Phase 3, Randomized, Double-masked, 3-Arm, Clinical Trial: First-time Results of Tarcocimab vs Tabirafusp vs Aflibercept for Neovascular AMD]. Presented at: Retina Subspecialty Day, American Academy of Ophthalmology 2026 Annual Meeting; October 9-10, 2026; New Orleans, LA.