October 2026 retina recap: Duravyu stumbles, DAYBREAK delivers, and optogenetics takes the Nobel

October was a conference month for retina, with AAO 2026 in New Orleans and late EURETINA and Retina Society coverage driving most of the news. There were no FDA approvals in retina, but 2 pivotal readouts pointed in opposite directions for durable nAMD therapy: the vorolanib insert missed noninferiority a second time, while tarcocimab met it in DAYBREAK. Optogenetics had its moment as the Nobel committee honored the field's founders the same week 4-year RP data and a planned GA trial were presented for son-vec. Below is an overview of 9 key developments from October, plus honorable mentions worth a look.

Late-phase readouts in nAMD

Duravyu misses wet AMD noninferiority in LUCIA and LUGANO as EyePoint pivots to DME

EyePoint reported that LUCIA, the second of 2 pivotal phase 3 trials of the vorolanib intravitreal insert (Duravyu; EyePoint), missed its primary end point in neovascular age-related macular degeneration (nAMD). The sustained-release tyrosine kinase inhibitor (TKI), dosed every 6 months, did not achieve noninferiority to aflibercept 2 mg every 8 weeks for best-corrected visual acuity (BCVA). The same day, EyePoint said it would cut approximately 53% of its workforce, end both wet AMD trials early, and focus on diabetic macular edema (DME).

In LUCIA (NCT06683742), insert-treated eyes gained a least squares mean of 2.5 ETDRS letters at blended weeks 52 and 56 vs 6.1 letters with aflibercept, a difference of −3.7 letters against a −4.5-letter noninferiority margin. Treatment burden fell 38% (3.3 vs 5.3 injections after loading; nominal P < .001), and central subfield thickness (CST) decreased 66 µm vs 82 µm.

LUGANO (NCT06668064), presented from the podium by Carl D. Regillo, MD, at the Retina Society meeting, showed a gain of approximately 2 letters vs approximately 6 letters despite near-identical anatomic control. Regillo attributed part of the gap to 9 insert-treated patients who lost 3 or more lines for reasons other than nAMD; a post hoc analysis excluding them narrowed the difference to 2.4 letters.

Because the primary end point was missed in both trials, secondary results carry nominal P values only. Pivotal DME data from COMO and CAPRI (551 patients combined) are expected in the fourth quarter of 2027.

Click here to read the full coverage on Modern Retina.

Tarcocimab and tabirafusp-ted meet noninferiority to aflibercept in phase 3 DAYBREAK

Charles C. Wykoff, MD, PhD, presented 1-year primary end point results from the phase 3 DAYBREAK trial at AAO 2026 Retina Subspecialty Day. DAYBREAK compared 2 Kodiak Sciences agents, tarcocimab tedromer, an anti-VEGF antibody biopolymer conjugate, and tabirafusp alfa tedromer (tabirafusp-ted), an anti-VEGF/anti–interleukin 6 bispecific, with aflibercept 2 mg every 8 weeks in nAMD. Retreatment was guided by an AI algorithm that quantified fluid on monthly OCT scans.

At week 48, BCVA improved by a mean of 7.2 letters with tarcocimab (n = 220) vs 7.6 letters with aflibercept (n = 224), meeting the noninferiority end point. At the primary end point, 54% of tarcocimab-treated patients had reached the 6-month maximum interval. Tabirafusp-ted (n = 225) gained 6.3 letters vs 7.6 letters and met noninferiority for both BCVA and CST (−131 µm vs −138 µm).

No intraocular inflammation, occlusive retinal vasculitis, or endophthalmitis occurred with tarcocimab; 1 case of intraocular inflammation occurred with tabirafusp-ted. Kodiak anticipates a multi-indication BLA for tarcocimab in 2026, supported by DAYBREAK, DAYLIGHT, GLOW1, GLOW2, and BEACON.

Click here to read the full coverage on Modern Retina.

Uveitis and uveitic macular edema

FDA accepts vamikibart BLA for uveitic macular edema as 1-year data hold

At AAO 2026, Roche presented 52-week data from the identical phase 3 MEERKAT (NCT05642312) and SANDCAT (NCT05642325) trials of vamikibart (Roche), an intravitreal anti–interleukin 6 monoclonal antibody, in adults with uveitic macular edema (UME). Roche also announced that the FDA has accepted its BLA, with a decision expected by July 2027. Coverage published October 13, 2026.

Patients received vamikibart 0.25 mg, 1 mg, or sham every 4 weeks through week 16, then as needed through week 52. At the week 16 primary end point, previously reported at ARVO, the proportion gaining 15 or more letters was 26.1% (0.25 mg) and 43.2% (1 mg) vs 6.3% with sham in MEERKAT, and 34.0% and 24.2% vs 13.3% in SANDCAT. UME resolution (CST below 325 µm) ranged from 56.7% to 63.3% across vamikibart arms vs 16.3% and 18.0% with sham.

At 52 weeks, Roche reported that gains on the primary end point and key secondary BCVA and CST end points were maintained, and approximately two-thirds of eligible patients needed no retreatment after week 16. Specific 52-week values were not reported in the coverage.

Click here to read the full coverage on Modern Retina.

Brepocitinib meets all phase 3 end points in noninfectious uveitis

Roivant Sciences and Priovant Therapeutics reported positive topline results from the phase 3 CLARITY program of oral brepocitinib (Lisraya; Priovant Therapeutics), a tyrosine kinase 2/Janus kinase 1 inhibitor, in noninfectious uveitis. The 2 identical trials enrolled 371 patients randomized to brepocitinib 45 mg daily or placebo for 48 weeks, with all patients completing a mandatory 8-week oral corticosteroid taper to zero.

The hazard ratio for treatment failure vs placebo was 0.27 (95% CI, 0.17-0.43; P < .0001) in CLARITY-1 (n = 166) and 0.22 (95% CI, 0.14-0.34; P < .0001) in CLARITY-2 (n = 204). Pooled across both trials, 62% of brepocitinib patients achieved steroid-free quiescence at week 48 vs 20% with placebo. Of particular relevance to retina specialists, 42% of patients with macular edema at baseline achieved resolution at week 48 vs 4.8% with placebo (P < .0001).

Three percent of participants discontinued for an adverse event, and no new safety signals were identified. Priovant plans to file a supplemental NDA in the first half of 2027.

Click here to read the full coverage on Modern Retina.

Inherited retinal disease and optogenetics

Tinlarebant slows Stargardt lesion growth 36% in phase 3 DRAGON ahead of a February 2027 FDA decision

Michel Michaelides, MD, FRCOphth, of Moorfields Eye Hospital discussed phase 3 DRAGON (NCT05244304) results for tinlarebant (Belite Bio), an oral, once-daily retinol-binding protein 4 antagonist, in adolescents with Stargardt disease type 1. Three days later, Jacque Duncan, MD, reviewed the data at AAO 2026.

DRAGON randomized 104 patients aged 12 to 20 years 2:1 to tinlarebant or placebo for 24 months. Tinlarebant reduced growth of definitely decreased autofluorescence lesions by about 36% vs placebo (P = .0033). Visual acuity changed little in either arm over 2 years, consistent with natural history, which leaves open how well the anatomic end point reflects functional benefit.

The most frequent ocular adverse events were xanthopsia, delayed dark adaptation, and night vision impairment, consistent with the drug's mechanism; 4 such events led to drug discontinuation among 69 treated patients. The FDA has granted the NDA priority review with a target action date of February 12, 2027. If approved, tinlarebant would be the first treatment for Stargardt disease.

Click here to read the full coverage on Modern Retina.

Optogenetics wins the Nobel as son-vec posts 4-year RP data and heads to a GA trial

On October 5, 2026, the 2026 Nobel Prize in Physiology or Medicine was awarded to Karl Deisseroth, MD, PhD; Peter Hegemann, PhD; and Georg Nagel, PhD, for discoveries concerning light-gated ion channels and optogenetics. The award landed as the most advanced ophthalmic optogenetic therapy, sonpiretigene isteparvovec (son-vec; MCO-010; proposed brand name Mogenry; Nanoscope Therapeutics), moves through FDA review for retinitis pigmentosa (RP) after BLA acceptance in September.

At AAO 2026, Kenneth C. Fan, MD, MBA, presented 4-year REMAIN follow-up from the phase 2b/3 RESTORE trial (NCT04945772). A single intravitreal injection improved BCVA by 0.337 logMAR (high dose; P = .02) and 0.382 logMAR (low dose; P = .03) at week 52; at week 204, gains were 0.261 and 0.244 logMAR, roughly 3 ETDRS lines. No serious adverse events were reported.

Also at AAO, investigators reported that son-vec was associated with a mean 12-letter gain in Stargardt patients whose atrophy was limited to the macula, and they plan to move directly to a randomized phase 2 trial in geographic atrophy (GA). That subgroup figure comes from conference data without a reported subgroup size or follow-up duration, and the GA rationale rests on a nonhuman primate model. Jordi Monés, MD, PhD, laid out the case for testing vision restoration in advanced GA on October 19.

Click here to read the full coverage on Modern Retina.

Geographic atrophy and DME pipeline

Kamuvudine K8 implant slows GA lesion growth 54% in phase 2

On October 9, 2026, Liangbo Shen, MD, of Duke University presented phase 2 results for kamuvudine K8 (Inflammasome Therapeutics), an intraocular inflammasome inhibitor, in geographic atrophy (GA) at AAO 2026 Retina Subspecialty Day. K8 is a bioerodible implant delivered in the office through a 24-gauge preloaded injector and dosed quarterly.

The multicenter US trial enrolled 30 patients with bilateral GA. The worse-seeing eye received 1 of 3 K8 doses at baseline and month 3, and the untreated fellow eye served as control. At the 0.7-mg dose, K8 reduced GA growth by 54% over 6 months (43% over months 0 to 3 and 70% over months 3 to 6). In a prespecified stratum of non-subfoveal lesions, BCVA improved by 4 letters vs controls (P = .004), although the study was not powered for visual acuity.

No drug-related serious adverse events, dose-limiting toxicities, or increased rates of neovascular conversion or intraocular inflammation occurred through 6 months. A larger phase 3 trial with longer follow-up is needed.

Click here to read the full coverage on Modern Retina.

Gene therapy 4D-150 enters phase 3 in DME with 4SIGHT

On October 2, 2026, Modern Retina reported that the first patients have been enrolled in 4SIGHT, a global phase 3 trial comparing a single intravitreal injection of 4D-150 (4D Molecular Therapeutics) with aflibercept 2 mg every 8 weeks in treatment-naive diabetic macular edema (DME). 4D-150 is an intravitreal AAV gene therapy encoding aflibercept plus a microRNA that suppresses VEGF-C. DME is its second indication in phase 3, after wet AMD.

4SIGHT plans to enroll 514 patients; only those who respond to aflibercept loading are randomized. The primary end point is noninferiority in mean BCVA change at week 52. In the open-label phase 2 SPECTRA trial, 9 patients at the phase 3 dose gained a mean of 10.8 letters with a 176-µm CST reduction at 2 years and needed a mean of 5.2 supplemental aflibercept injections. No intraocular inflammation was seen in any of the 22 SPECTRA patients.

The efficacy signal comes from 9 patients without a concurrent control, and the company's 61% burden reduction is calculated against projected, not observed, on-label dosing. Durability across the second year is the key open question.

Click here to read the full coverage on Modern Retina.

Biosimilars

CHMP backs ranibizumab biosimilar Fovlussea as Alvotech files the first aflibercept 8 mg biosimilar

The European Medicines Agency's Committee for Medicinal Products for Human Use adopted a positive opinion for Fovlussea (ranibizumab; Accord Healthcare), a biosimilar to Lucentis (ranibizumab; Novartis). The recommended indications match the adult Lucentis label: nAMD, visual impairment due to DME, proliferative diabetic retinopathy, macular edema secondary to RVO, and choroidal neovascularization.

The EMA concluded that Fovlussea has comparable quality, safety, and efficacy to Lucentis, but neither the meeting highlights nor the summary of opinion identifies the pivotal trial or reports its results; those will appear in the European public assessment report. The European Commission typically decides within 67 days of a CHMP opinion. Fovlussea would join ranibizumab biosimilars available in Europe since 2023, including Byooviz (Samsung Bioepis) and Ximluci (STADA/Xbrane).

The same day, Alvotech and Advanz Pharma reported submitting AVT29, a proposed biosimilar to aflibercept 8 mg (Eylea 8 mg; Bayer/Regeneron), to the EMA as a line extension to Mynzepli, Alvotech's aflibercept 2 mg biosimilar. The companies describe it as the first regulatory submission worldwide for an aflibercept 8 mg biosimilar. Its phase 3 comparative trial in DME is expected to read out in mid-2027, after the filing.

Click here to read the full coverage on Modern Retina.