Kodiak's DAYBREAK Win Beat an Outdated Eylea Dose

Kodiak's Zenkuda met its primary endpoint in the DAYBREAK trial for wet macular degeneration, and the stock more than doubled, adding over $3 billion in value while Regeneron fell more than 4%. The trial matched aflibercept dosed every eight weeks; the newer Eylea HD and Vabysmo, the drugs Zenkuda must actually out-dose commercially, have yet to be tested against it. Eli Lilly, flush and shopping, has pointed its deal hunt elsewhere.

Kodiak Sciences shares more than doubled on Monday, adding more than $3 billion in market value, after its drug Zenkuda hit the primary endpoint in the Phase 3 DAYBREAK trial for wet age-related macular degeneration. Regeneron, whose Eylea franchise Zenkuda is built to challenge, fell more than 4%, a decline that arrives on top of the biosimilar pressure Eylea already faces as its own patents age. Both stock moves treat the result as settled. It is real, but it answers a narrower question than the market is pricing: Zenkuda proved non-inferior to old-dose Eylea, not superior to the two drugs it will actually have to beat once it reaches the market.

The choice of comparator shapes what the result can establish. DAYBREAK dosed aflibercept at its original label strength, every eight weeks, the regimen doctors have been moving away from since Regeneron launched the higher-dose Eylea HD, which runs every 16 to 20 weeks, and since Roche brought Vabysmo to market on a similar 16-week schedule. Zenkuda matched that older, more frequent regimen on a non-inferiority basis, with a p-value of 0.0007, while gaining 7.2 letters of vision against aflibercept's 7.6, a small numerical shortfall the trial's statistics treat as immaterial. Kodiak has not yet run the head-to-head against the newer, longer-acting drugs that actually define the category it is entering.

DAYBREAK also tested durability under a strict protocol. Fifty-four percent of patients reached six-month, 24-week dosing intervals at one year, after four monthly loading doses and an AI-guided algorithm that retreated whenever OCT imaging detected any fluid, on no fixed schedule. Kodiak's molecule has a mean ocular half-life of about 20 days, roughly three times that of approved agents, which is the pharmacokinetic reason it can stretch dosing that far. Durability also cuts both ways commercially. Existing retina drugs cost roughly $1,800 to $2,500 per injection, according to Forbes, so a drug given twice a year instead of six times has to command a much higher price per shot to earn the same revenue per patient. Safety looked clean: no intraocular inflammation, and cataracts in 0.5% of patients against 0.9% on aflibercept. A companion asset from the same readout, the anti-IL-6/VEGF-trap bispecific tabirafusp-ted, met its own vision primary endpoint (p=0.0036) and anatomical secondary endpoint (p<0.0001), with a 6.3-letter vision gain, giving Kodiak two paths to a first approval. Kodiak is also running tabirafusp in a roughly 910-patient Phase 3 called ALTO, testing superiority over aflibercept in diabetic macular edema, and it owns full rights to both molecules, with no partner entitled to a share of any deal economics.

Kodiak argues that trial-defined dosing intervals for Eylea HD and Vabysmo may overstate what those drugs deliver when patients skip real-world visits. A durability win over old-dose Eylea could still give Zenkuda an edge over newer drugs in practice. That may hold up, but Kodiak will file for approval before any head-to-head or real-world comparison exists, and retina specialists will want one before switching patients off Vabysmo. The filing is targeted for the fourth quarter of 2026, built on five Phase 3 trials: DAYBREAK, the companion DAYLIGHT study in wet AMD, GLOW1 and GLOW2 in diabetic retinopathy, and BEACON in retinal vein occlusion, Kodiak's release shows. Two approvals are possible by the end of 2027 or early 2028.

The same Monday, Eli Lilly's chief executive David Ricks told Bloomberg that investors should expect "slightly larger" deals, along the lines of the $7.8 billion Centessa acquisition, aimed at what he calls the "white spaces" of science: infectious disease, women's health and psychiatry. Lilly has done about 40 deals this year worth more than $20 billion, its biggest year on record, including three vaccine makers for as much as $3.8 billion and the psychedelics developer AtaiBeckley for about the same. Ricks also said Lilly feels no urgency, calling deal-making under pressure "shopping when you're hungry, and you can make mistakes."

Lilly's deals look exploratory rather than defensive. Tirzepatide's patents run to 2036, which gives Lilly a decade of GLP-1 cash to spend on options before a patent cliff forces its hand. Ricks said Lilly now likes more mature assets when they are solving brand new problems, and has historically bet early, before companies are worth $5 billion. The common thread is novelty: new diseases for Lilly, new mechanisms, new markets.

Lilly's stated priorities give it little reason to pursue Kodiak. Yet Kodiak, now worth close to $5 billion with a filing weeks away, is the sort of late-stage asset investors assume a cash-rich giant will want. It fails Lilly's own test. Zenkuda is a longer-lasting version of an established drug class in a market Roche and Regeneron already dominate, retina is absent from Ricks's list, and Lilly has no eye-care sales force. A Zenkuda launch needs retina reps in clinics that Roche, Regeneron and Novartis already staff. Those companies have the reason to pay a premium for a validated retina drug; Lilly's shopping list points elsewhere.

Monday's result came four years and seven months after Zenkuda's first Phase 3 failed to match Eylea, BioPharma Dive noted, and Kodiak redesigned the trial around fluid-triggered retreatment to fix what went wrong. Kodiak ended June with $125.9 million in cash, enough to reach 2027 but not to launch a drug, so the desk expects it to raise equity on the back of Monday's rally before year end, ahead of two catalysts already on the calendar: Zenkuda's BLA filing in the fourth quarter of 2026, and topline data from the KSI-101 PEAK trial in December 2026.