Novo’s $1.4 Billion Bet on Oral Macrocycles: What a Deal Validates and What It Doesn’t
Novo signed with Orbis for up to $1.4bn on oral macrocycles. Against two 2026 approvals, what the deal establishes and what it does not.
On 17 September 2026 Orbis Medicines announced a discovery and licensing agreement with Novo worth up to USD 1.4 billion in upfront and milestone payments, plus tiered royalties and a strategic equity investment, to design orally bioavailable macrocycles against undisclosed cardiometabolic targets. No molecule from the collaboration has entered the clinic, and the headline figure is the sum of milestones that mostly will not be earned, because most discovery programmes fail. What the deal establishes is that a company of Novo's size will pay platform prices for discovery-stage access to the modality, six months after the FDA approved icotrokinra and two months after it approved enlicitide.
On 17 September 2026 Orbis Medicines announced a discovery and licensing agreement with Novo worth up to USD 1.4 billion in upfront and milestone payments, plus tiered royalties and a strategic equity investment, to design orally bioavailable macrocycles against cardiometabolic targets.1 Neither company disclosed the number of programmes, the targets, or how the 1.4 billion divides between signature and milestones. What makes the deal worth reporting is not its size. It is that it is the third event this year in which the oral macrocyclic peptide stopped being a chemistry problem and started being a commercial position. Condor Research supplies peptides as reference materials for laboratory research use only; this article is commentary on drug development and contains no human use guidance.
What was signed
Orbis Medicines is a Copenhagen and Lausanne company founded in January 2025 with EUR 116 million raised to date, including a EUR 90 million Series A led by NEA with Lilly Ventures, Cormorant, EIFO, Forbion and Novo Holdings.1 Its platform, nGen, pairs generative design with automated synthesis: hit-finding libraries on the order of 100 billion compounds, target-specific libraries designed by generative models, and a chemistry system the company says can synthesise and analyse up to 100,000 distinct synthetic macrocycles in weeks, feeding results back as training data. The molecules are termed nCycles.
Under the agreement Orbis is eligible for up to USD 1.4 billion in upfront and potential development and commercial milestones, plus tiered royalties on future sales, and Novo will take a strategic stake.1 Novo’s stated rationale, from Brian Vandahl, senior vice president for Therapeutics Discovery, is access to a differentiated platform combining macrocycle chemistry with AI-enabled molecular design.
Context on the buyer’s side is relevant. Novo has lost its lead in obesity and diabetes to Eli Lilly, has changed chief executives, cut staff and reshuffled its board over the past year, and announced a corporate rebranding the same week that drops “Nordisk” from day-to-day use.2 This is a company buying into a modality it does not own, during a reset.
Why macrocycles, and why orally
The gap the modality is aimed at is easy to state. Small molecules are absorbed well and can enter cells, but struggle to bind the large, flat protein-protein interfaces that many biological targets present. Antibodies and peptides bind those surfaces precisely, but are digested or poorly absorbed if swallowed, which is why they are injected. A macrocycle is a peptide whose backbone has been closed into a ring, often with further modifications, which makes it rigid, harder for proteases to cut, and in the right cases permeable enough to survive the gut.
The size of the problem being solved is best shown by what happens without it. As BioPharma Dive notes, the highest dose of Novo’s oral semaglutide contains twenty times the highest injectable dose, because most of the active ingredient is destroyed before it reaches the bloodstream. The same arithmetic has been run in public on a single molecule: zenagamtide was trialled both injected and swallowed, and the comparison is unusually explicit about what the oral route costs.2 That is the tax an oral peptide pays. A macrocycle that avoids most of it changes the economics of every peptide target at once: cost of goods, manufacturing scale, cold chain, and who is willing to start treatment.
The interesting claim is not that peptides can be swallowed. It is that they can be swallowed without paying twenty times the dose for the privilege.
The 2026 sequence
A single licensing deal proves little. Three events in seven months, two of them regulatory approvals rather than promises, is a different kind of evidence.
| Date | Event | What it establishes |
|---|---|---|
| 18 March 2026 | FDA approves ICOTYDE (icotrokinra), a 13-amino-acid oral cyclic peptide blocking the IL-23 receptor, for moderate-to-severe plaque psoriasis3 | An oral peptide can clear Phase 3 against both placebo and an oral small molecule, at a fraction of one percent oral bioavailability and with no absorption enhancer |
| July 2026 | FDA approves LIPFENDRA (enlicitide), a macrocyclic peptide and the first once-daily oral PCSK9 inhibitor, at 20 mg, on the CORALreef Phase 3 programme4 | A macrocyclic peptide can take a target previously reachable only by injected antibody or siRNA and deliver it as a tablet |
| 17 September 2026 | Novo signs with Orbis for up to USD 1.4 billion on undisclosed cardiometabolic targets1 | Nothing clinical. It establishes that a company of Novo’s size is willing to pay platform prices for discovery-stage access |
The third row is the one to read carefully, and it is where most coverage of this deal goes wrong. An approval is evidence: a molecule was tested in patients and a regulator agreed it worked. A discovery deal is a forecast. No nCycle has entered the clinic. The headline figure is the sum of every milestone that could be earned across an undisclosed number of programmes, and in this structure the large majority of it is payable only on events that mostly will not happen, because most discovery programmes fail. What the deal genuinely de-risks is Orbis’s balance sheet.
Orbis is also not alone. Unnatural Products and Dayra Therapeutics have each signed with larger drugmakers, and macrocycles are not the only modality being pushed at hard targets: peptide-drug conjugates are being built on the same logic from the other direction.2 A modality with three startups taking partner money is a modality where the buyers have decided the risk is now worth pricing.
Where the constraint moves next
If discovery becomes tractable, the binding constraint moves downstream, and this is the part that touches laboratory supply rather than pharma strategy. Macrocyclic peptides are harder to make than linear ones: cyclisation adds steps, non-natural residues are common, and purification is less forgiving. Solid-phase synthesis at commercial scale is already under pressure from solvent regulation, and peptide manufacturing capacity is being built out worldwide on the assumption that demand keeps rising.
A generative platform that can design 100,000 candidates in weeks does not relieve that pressure. It increases it, by producing more molecules that need to be made, characterised and made again at scale. Capacity is being bought as well as built, as Samsung Biologics’ bid for PolyPeptide showed. The analytical burden rises with structural complexity: a cyclic peptide with non-natural residues is a harder identity problem than a fifteen-residue linear chain, and the methods that establish what is in the vial have to keep up with what the chemists are designing. On what those methods can and cannot establish, see our note on the difference between purity and identity, and on why proving two peptide preparations are the same thing is harder than proving it for a small molecule.
We covered the chemistry of oral peptide design in how drug designers are breaking the bioavailability barrier, which is the companion to this piece and explains how icotrokinra achieves oral activity. The supply side is in the global peptide manufacturing arms race and the solvent problem, and the market framing in why Wall Street is suddenly paying attention.
How this was checked. Deal terms, platform description and quotations are from Orbis Medicines’ own announcement of 17 September 2026, read on the company site rather than from aggregated coverage. The oral semaglutide dose comparison, the position of Novo’s business and the names of competing startups are from BioPharma Dive of the same date. Approval dates and indications for icotrokinra and enlicitide are from FDA-facing company announcements and were cross-checked against our own earlier article on icotrokinra. A figure circulating this week attributing a specific preclinical oral bioavailability percentage to Orbis’s macrocycles is not in the company’s announcement and we could not verify it in a primary source, so it is not reported here. Version 1.0, first published 21 September 2026.
Condor Research supplies peptides as characterised reference materials for laboratory research use only: not for human or veterinary use, not for diagnostic or therapeutic application, and not for any food or cosmetic purpose. None of the compounds discussed in this article are supplied by Condor Research, and nothing here describes, recommends or implies human use of any compound, approved or investigational.
Condor Research · Scientific desk
Atrio Sciences s.r.o., IČO 57 669 651, Nitra (SK) · info@condorresearch.com
- Orbis Medicines is eligible for up to USD 1.4 billion in upfront and milestone payments plus tiered royalties; Novo will also take a strategic equity stake.
- Neither company disclosed the number of programmes, the targets, or how the headline figure divides between signature and milestones.
- The nGen platform pairs generative design with automated synthesis: hit-finding libraries of 100 billion compounds and up to 100,000 macrocycles synthesised and analysed in weeks.
- Macrocycles aim at the gap between small molecules and biologics: large enough to bind flat protein surfaces, rigid enough to resist proteases, sometimes permeable enough to survive the gut.
- The scale of the problem: the highest dose of Novo's oral semaglutide contains twenty times the highest injectable dose, because most of the drug is destroyed before absorption.
- This is the third event of 2026 in the modality: icotrokinra approved 18 March, enlicitide approved in July as the first oral PCSK9 inhibitor, and this deal on 17 September.
- The first two are approvals and therefore evidence; the third is a forecast. A discovery deal de-risks the balance sheet of the smaller company, not the science.
- Novo signed during a reset: loss of the obesity lead to Eli Lilly, a change of chief executive, layoffs, a board reshuffle and a rebranding that drops Nordisk from daily use.
- Orbis is not alone: Unnatural Products and Dayra Therapeutics have each signed with larger drugmakers, which is what a modality looks like once buyers decide the risk is priceable.
- If discovery becomes tractable the constraint moves to manufacturing and analysis: cyclisation adds steps, non-natural residues are common, and identity is harder to establish.
What did Novo actually agree to pay?
Orbis Medicines is eligible for up to USD 1.4 billion in upfront and potential development and commercial milestone payments, plus tiered royalties on future sales, and Novo will make a strategic investment in the company. Neither party disclosed the upfront amount, the size of the equity stake, the number of programmes or the targets.
Does this deal mean oral macrocyclic peptides work?
No. It means a large buyer is willing to pay for discovery-stage access. No molecule from this collaboration has entered the clinic. The evidence that the modality can work comes from two separate 2026 approvals: icotrokinra for plaque psoriasis in March, and enlicitide, the first once-daily oral PCSK9 inhibitor, in July.
Why are macrocyclic peptides interesting?
They occupy the space between small molecules and biologics. Small molecules absorb well but struggle to bind large flat protein-protein interfaces; antibodies and peptides bind those surfaces precisely but are digested or poorly absorbed when swallowed. Closing a peptide backbone into a ring makes it rigid, harder for proteases to cut, and in the right cases permeable enough to survive the gut.
How hard is oral delivery for a peptide?
Hard enough that the highest dose of Novo's oral semaglutide contains twenty times the highest injectable dose, because most of the active ingredient is destroyed before reaching the bloodstream. A macrocycle that avoids most of that penalty changes cost of goods, manufacturing scale, cold chain and who is willing to start treatment.
What does this mean for peptide manufacturing?
It increases pressure rather than relieving it. A platform that can design 100,000 candidates in weeks produces more molecules that must be made, characterised and remade at scale, while solid-phase synthesis is already under regulatory pressure over solvents. Analytical burden rises with structural complexity: a cyclic peptide with non-natural residues is a harder identity problem than a short linear chain.