Johnson & Johnson is paying $2.2 billion upfront for a retina drug whose two completed trials enrolled 188 patients. OLN324 blocks the same two targets as Roche’s Vabysmo with a protein just over a quarter of its size. This brief explains how it works, what its one head-to-head trial showed, and what four phase 3 trials are built to measure.
The OLN324 mechanism of action is a double blockade. One small protein, about 41 kDa, binds VEGF-A and angiopoietin-2 (Ang2). Those two signals make retinal blood vessels leak and overgrow in wet age-related macular degeneration (wet AMD) and diabetic macular edema (DME). Roche’s Vabysmo (faricimab) blocks the same pair with a full-length antibody of about 149 kDa.
OLN324 is built differently. It is a single chain that Johnson & Johnson calls a minibody: an antibody fragment against VEGF-A fused to a single-domain antibody against Ang2. At 4 mg it puts 2.4 times as many molecules into the eye as a faricimab dose (computed). Ollin Biosciences, its developer, reports up to 60 times the potency of faricimab against Ang2 in laboratory tests.
On 16 November 2026 Johnson & Johnson agreed to buy Ollin for $2.2 billion upfront and up to about $1.125 billion in milestones. Blackstone Life Sciences will pay $750 million towards phase 3. The molecule’s two completed trials enrolled 188 patients between them (24 + 164).
01OLN324 mechanism of action: two targets, one 41 kDa protein
Two signals do most of the damage in wet AMD and DME. VEGF-A makes vessels leak and pushes new, fragile ones to grow. Ang2 loosens the vessel wall and leaves it more sensitive to VEGF-A. Faricimab was the first approved drug to block both. Its US label puts it this way: blocking VEGF-A cuts permeability and new vessel growth, and blocking Ang2 is thought to promote vascular stability.
OLN324 goes after the same two signals with one chain. At one end, the N-terminus, sits a single-chain variable fragment (scFv) that binds VEGF-A. A single-domain antibody that binds Ang2 sits at the other, where it keeps Ang2 off its receptor, Tie-2. The two are joined by a flexible glycine-serine linker. The description comes from the phase 1 paper (Wang et al., via PubMed), a 2025 ARVO abstract and an Innovent release from November 2023.
No full-length antibody is involved, JADE investigator Margaret Chang told Retinal Physician in a sponsored interview. The chain weighs about 41 kDa against about 149 kDa for faricimab. Ollin gives three reasons to expect more from it: higher potency against Ang2, a smaller format, and more molecules per injection.
How OLN324 differs from faricimab
OLN324 molecular weight and molar dose
Target for target, the OLN324 mechanism of action matches faricimab’s. The differences are potency and the number of molecules in each injection. Ollin says OLN324 delivers a higher molar dose than faricimab and than aflibercept, Eylea HD included. With the molecular weight in print, that claim can be checked.
Aflibercept, Eylea HD 8 mg / 115 kDa = 69.6 nmol
OLN324 2 mg 2 mg / 41 kDa = 48.8 nmol
OLN324 4 mg 4 mg / 41 kDa = 97.6 nmol
The claim holds in full only at 4 mg. The 2 mg dose carries 21% more molecules than faricimab and 30% fewer than Eylea HD (computed). And OLN324 is not the smallest drug to make this argument. Brolucizumab (Beovu), a 26 kDa antibody fragment, delivers about 231 nmol in its 6 mg dose (computed).
The formulation is still changing. In phase 1 and in JADE the 2 mg dose came in the standard 0.05 mL and the 4 mg dose in 0.1 mL, twice the usual volume. Both work out to 40 mg/mL (computed). Ollin is reformulating so that 4 mg fits in 0.05 mL, which means 80 mg/mL (computed). Faricimab and brolucizumab are both sold at 120 mg/mL.
02OLN324 vs faricimab: what the JADE trial showed in 164 patients
JADE is the only trial that has compared the two drugs. It randomised 164 patients at 30 sites, 29 in the United States and one in Puerto Rico, to OLN324 2 mg, OLN324 4 mg or faricimab 6 mg. Everyone had three monthly injections and was followed to week 20. The registry lists adverse events as its primary outcome. Ollin assessed efficacy at week 12.
DME: the anatomy result
The clearest gap was in retinal thickness in DME. One week after the first injection, central subfield thickness had fallen by 79.0 µm on OLN324 4 mg and by 44.8 µm on faricimab. At week 12 the figures were 180 µm and 121 µm.
Week 12 180 / 121 = 1.49 times
No edema at week 12 88.5% − 57.1% = 31.4 points
Those shares match 23 of 26 and 16 of 28 patients
At week 12, 88.5% of patients on the 4 mg dose and 84.6% on 2 mg had no edema, defined as thickness below 325 µm. On faricimab the share was 57.1%. The P value of .003 was nominal, Ophthalmology Times reported from the ASRS meeting in July. The patient counts in the box are Witfire’s, worked back from the percentages.
Vision: the smaller gap
Vision moved less. In DME the mean gain at week 20 was 8.1 letters on OLN324 4 mg, 7.4 on 2 mg and 7.7 on faricimab. Those are the figures Chang gave Retinal Physician. So the higher dose led faricimab by 0.4 letters and the lower dose trailed it by 0.3 (computed). In wet AMD the gains were 8.6, 8.4 and 6.4 letters, a lead of 2.2 and 2.0.
Durability and safety
Longer gaps between injections are what this drug class promises, and on that JADE says little. Twelve weeks after the last injection the retreatment-free rates were four points apart in DME and one point apart in wet AMD.
What JADE cannot show
Ollin’s chief executive said in January that OLN324 “is the first and only therapy to demonstrate superior anatomic efficacy” against faricimab. The trial behind that sentence was “not powered for definitive efficacy comparisons”, as Ophthalmology Times put it in its report on the 20-week data. It ran for 20 weeks. In the wet AMD cohort the faricimab group started with thicker retinas, investigator Arshad Khanani told Healio. And the 4 mg dose went in at twice the standard volume.
Rare safety events are also beyond a trial this size. Brolucizumab (Beovu), the 26 kDa fragment, won US approval in October 2019. Eight months later the FDA added retinal vasculitis and retinal vascular occlusion to its label, Novartis said in June 2020. No intraocular inflammation or retinal vasculitis was seen on OLN324 in JADE or in the phase 1 study. Those two trials enrolled 188 patients, and not all of them received OLN324.
03OLN324 phase 3 trials: GOBI, ATACAMA, MOJAVE and SAHARA
Four phase 3 trials are registered, each named after a desert. Together they plan 1,800 patients.
None of the four uses faricimab as the comparator. Three test OLN324 against aflibercept, covering 1,400 of the 1,800 patients, and one tests it against ranibizumab (computed). Any claim over Vabysmo will go on resting on JADE.
Faricimab took the same route. Its four registration trials enrolled 3,220 patients and compared it with aflibercept 2 mg. A margin of four letters defined non-inferiority (Wykoff et al. and Heier et al., Lancet 2022, via PubMed). The observed differences ran from 0.2 letters behind to 1.5 ahead. OLN324’s programme is 56% of that size (computed).
Vision is the primary endpoint in all four OLN324 trials, and vision is where JADE’s gaps were 0.4 and 2.2 letters. Both sit inside the four-letter margin faricimab’s trials used. Retinal thickness and absence of edema, where OLN324 pulled ahead, are secondary endpoints in both DME trials, measured at weeks 20 and 48. The wet AMD trials time their anatomy endpoints earlier, at week 16 in MOJAVE and week 12 in SAHARA.
Entry rules differ by disease. The DME trials admit a limited number of previously treated patients, and the wet AMD trials take only untreated eyes.
Two of faricimab’s lead investigators now work with its challenger. Charles Wykoff, first author of the DME paper, sits on Ollin’s scientific advisory board. Arshad Khanani, second author of the wet AMD paper, was a JADE investigator and is on the same board.
For all four trials the registry gives 15 October 2026 as the start date, and it still lists them as not yet recruiting. Innovent said on 16 November that recruitment had begun. Primary completion is set for 15 December 2028, which is 760 days after the deal (computed).
04What J&J is paying for OLN324, and who funds phase 3
Upfront against equity raised $2,200M / $430M = 5.1 times
Blackstone per planned phase 3 patient $750M / 1,800 = about $417,000
Neither company has said what the $750 million covers beyond the four registered studies.
Blackstone appears twice. It was among the new investors in Ollin’s $330 million Series B in June, pharmaphorum reported. Its life sciences arm will now finance the trials for the buyer. If OLN324 reaches the market, Blackstone collects milestones and a single-digit royalty from Johnson & Johnson.
Trung Huynh of RBC Capital Markets called the deal a logical fit and said the price “appears high” next to other companies in the field, BioPharma Dive reported. One earlier retina buyout gives a yardstick.
J&J’s upfront is 1.7 times Merck’s, and far more of the price is fixed on day one (computed). Witfire took the same kind of deal apart in GSK’s $10.6 billion purchase of Nuvalent.
J&J has tried retina medicines before. It bought a gene therapy for X-linked retinitis pigmentosa outright from MeiraGTx in December 2023 for $130 million upfront. Results shared on 2 May 2025 showed the therapy had missed its primary endpoint in phase 3. That was 563 days before the Ollin deal (computed). The rest of its eye business sells contact lenses and surgical products.
05Who gets paid: Ollin’s investors, Innovent and Blackstone
Ollin’s backers have had a short wait. The company raised $100 million when it left stealth in September 2025 and $330 million in June 2026, a disclosed total of $430 million (computed). The upfront is 5.1 times that sum. Investors in the second round reached an agreed sale in 145 days (computed).
Innovent Biologics discovered the molecule in Suzhou and ran the first human study in 24 patients in China. Twelve had a single dose of 0.5, 2 or 4 mg and twelve had three monthly doses of 2 or 4 mg, with no dose-limiting toxicity. The results were published in January (Wang et al., BMJ Open Ophthalmology, via PubMed). What it earns from the sale is not public. Its Hong Kong filing on 16 November named no payment, stake, milestone or royalty. The route is the one Witfire mapped in the China biotech pipeline grab: discovered in China, developed by a young US company, bought by a multinational.
Ollin’s second programme goes with it. OLN102, a bispecific for thyroid eye disease, came from VelaVigo of Shanghai in April 2025 in a licence worth up to $440 million. It is still preclinical.
06Where OLN324 would compete: Vabysmo’s $5 billion and a slower US market
The drug OLN324 is chasing is large, and in the United States it has slowed. Vabysmo’s US sales grew 2% at constant exchange rates in the first nine months of 2026. Sales outside the US, Europe and Japan grew 65%, according to Roche’s nine-month report. J&J’s upfront equals 41.5% of one year of Vabysmo sales (computed in dollars).
The US branded market for eye injections shrank 15% in 2025, Roche’s pharmaceuticals chief said, Fierce Pharma reported. OLN324’s phase 3 comparators, aflibercept and ranibizumab, are the older drugs in that market. A result that matches them would put OLN324 where faricimab stood in 2022, and faricimab has had a head start since then.
07OLN324 and India: no sites yet, and a regulator that asks for local data
The need is large. Among Indians aged 40 or older who have diabetes, an estimated 12.5% have diabetic retinopathy. The vision-threatening form affects 4.0%, about 3 million people (Raman et al., 2022, via PubMed).
India has had no part in OLN324 so far. The two completed trials list 31 sites on the registry: 29 in the United States, one in Puerto Rico and one in China. The phase 1 paper counts six Chinese sites. None is listed yet for the four phase 3 trials.
There is a recent local precedent. Roche launched Vabysmo in India in March 2024. Roche later sought a new use, macular edema after retinal vein occlusion. The regulator’s ophthalmology committee met on 16 July 2025 and told the company to run a phase 3 trial in India first, Medical Dialogues reported. Roche had offered a phase 4 study in its place.
So a waiver under Rule 101 cannot be assumed, even after a US approval that is at least two years away. Indian sites in phase 3 would be the first sign that J&J plans an early filing there. Whether they appear will show on the registry, the test Witfire applied in India’s next CRO boom.
08Witfire Risk Score: 5.1 / 10
This score covers Johnson & Johnson’s OLN324 bet as it stands at signing.
09What to watch
- By 31 December 2026. Closing of the acquisition (JNJ). Johnson & Johnson has said it will state the accounting treatment on or before closing.
- The registry. A change from not yet recruiting, the first site lists, and the aflibercept dose. Indian sites would show there first.
- 28 January 2027. Roche’s full-year results (ROG). Vabysmo’s US sales grew 2% in the first nine months of 2026.
- March 2027. Innovent’s annual results (1801.HK), on last year’s timing. They are the first accounts in which a payment linked to the sale could appear.
- 15 December 2028. Primary completion for GOBI, ATACAMA, MOJAVE and SAHARA.
Nothing in the OLN324 mechanism of action is new. Faricimab has already shown that blocking VEGF-A and Ang2 together works, and it sells more than $5 billion a year.
What J&J bought is a different build of the same idea. It weighs 41 kDa against 149, is more potent against Ang2 in the laboratory, and is ahead of faricimab on retinal drying in 83 DME patients. The price was set on anatomy. Approval will be decided on vision, where the gaps in JADE were 0.4 and 2.2 letters. None of the 1,800 phase 3 patients will be randomised against faricimab.
Blackstone carries the trial cost and J&J carries the $2.2 billion. Whether size matters will first be tested after 15 December 2028.
FAQOLN324: questions and answers
What is the OLN324 mechanism of action?
OLN324 is a bispecific protein given by injection into the eye. One end, a single-chain antibody fragment, binds VEGF-A. The other end, a single-domain antibody, binds angiopoietin-2 (Ang2). Those two signals drive leak and abnormal vessel growth in wet AMD and diabetic macular edema. The OLN324 mechanism of action is the same double blockade used by faricimab (Vabysmo), delivered in a molecule of about 41 kDa. Ollin reports up to 60 times the potency against Ang2 in laboratory tests.
How is OLN324 different from faricimab (Vabysmo)?
Both block VEGF-A and Ang2. Faricimab is a full-length antibody of about 149 kDa, dosed at 6 mg. OLN324 is a single-chain fusion protein of about 41 kDa that Johnson & Johnson calls a minibody, tested at 2 mg and 4 mg. At 4 mg it delivers about 2.4 times as many molecules per injection as faricimab (computed).
What is the molecular weight of OLN324?
About 41 kDa, according to the published phase 1 paper and a 2025 ARVO abstract. Faricimab is about 149 kDa, aflibercept about 115 kDa, ranibizumab about 48 kDa and brolucizumab about 26 kDa. Because OLN324 is small, a 4 mg dose carries about 97.6 nanomoles of drug, against 40.3 for faricimab 6 mg and 69.6 for aflibercept 8 mg (computed).
What did the JADE trial of OLN324 show?
JADE was a 164-patient phase 1b trial comparing OLN324 with faricimab. In diabetic macular edema, 88.5% of patients on OLN324 4 mg had no edema at week 12 against 57.1% on faricimab (nominal P = .003). Vision gains at week 20 were 8.1 letters against 7.7. In wet AMD the gains were 8.6 letters against 6.4. The trial was not powered for efficacy comparisons.
What are the OLN324 phase 3 trials?
Four are registered. GOBI and ATACAMA each plan 450 patients with diabetic macular edema and compare OLN324 with aflibercept. MOJAVE plans 500 patients with wet AMD against aflibercept, and SAHARA plans 400 against ranibizumab. The primary endpoint in all four is change in best-corrected visual acuity. The registry lists primary completion on 15 December 2028.
How much is Johnson & Johnson paying for Ollin Biosciences?
$2.2 billion upfront, plus up to about $1.125 billion in regulatory and commercial milestones, for a maximum of about $3.325 billion. Blackstone Life Sciences will separately pay $750 million over several years to fund phase 3, in return for milestones and single-digit royalties. Johnson & Johnson expects the acquisition to close before the end of 2026.
Who discovered OLN324?
Innovent Biologics of Suzhou, China, where the molecule is coded IBI324. Innovent ran the first trial, a 24-patient dose-escalation study in China, and licensed the drug to Ollin Biosciences of Austin, Texas. Innovent’s filing on the Johnson & Johnson deal disclosed no payment, stake, milestone or royalty terms.
When could OLN324 be approved?
No filing date has been announced. All four phase 3 trials list primary completion on 15 December 2028 on ClinicalTrials.gov, and their primary endpoints are measured at week 36 or week 48. OLN324 is investigational and is not approved in any country.
Is OLN324 available in India?
No. OLN324 is not approved anywhere. Neither completed trial had an Indian site, and the phase 3 records list no sites yet. India’s regulator can consider waiving a local trial for drugs approved in the United States, the UK, Japan, Australia, Canada or the EU. In July 2025 its expert committee asked Roche for an Indian phase 3 trial before adding a new use for faricimab.
Author. Dr. Akhilesh Vats is Editor-in-Chief of Witfire Elite and founder of ACME Research Solutions, a pharmaceutical contract research organisation.
Disclosure: Editorial analysis, not investment or medical advice. The Witfire Risk Score is our own framework, weighting Regulatory Exposure (30%), Competitive Displacement (25%), Capital Position (20%), Evidence Integrity (15%) and Execution & Credibility (10%) on a 1–10 scale. Scores are judgments, not forecasts.
Put this framework on your own file.
The same method used in these briefs, applied privately to a decision you are actually facing — a target, a partner, a facility, a filing.
- CMC & manufacturing risk assessment
- Regulatory / CRL readiness review
- Due diligence on a target or partner
- Investor brief or market analysis
- Commissioned editorial
- Something else
The signal before the market catches up.
A low-noise briefing on the pharma events that change strategy, risk and value. No more than one a week.
