
Merck and Isomorphic Labs announced a research collaboration on 19 November 2026 to find small-molecule drug candidates against one undisclosed cancer target, pairing Isomorphic’s AI drug design engine with Merck’s proprietary data. It follows Isomorphic’s $5 billion Series C announced two days earlier and adds a Merck programme to the pharmaceutical partnerships Isomorphic has announced with Lilly, Novartis and Johnson & Johnson.
What happened
Merck, known as MSD outside the United States and Canada, and Isomorphic Labs published the agreement through a Business Wire release datelined Rahway, New Jersey, at 6:45 a.m. Eastern time on Thursday 19 November 2026, according to Merck’s news release. The two companies will combine Isomorphic’s AI approach to drug discovery with Merck’s proprietary data resources and discovery experience to identify therapeutic small-molecule candidates against what the release calls an oncology target of strategic importance. The target is not named.
STAT, which reported the deal the same day, quotes George Addona, Merck’s global head of discovery, preclinical development and translational medicine: “It is a target of high interest to us, a target where we have deployed internal efforts.” STAT describes the pairing as Isomorphic’s drug design engine with Merck’s existing wet-lab data and calls the agreement “only the latest such deal” between a large pharmaceutical company and Isomorphic.
Key details of the agreement
The release and trade coverage describe a division of labour in which the AI company designs and the pharmaceutical company develops. The reported terms are summarised below.
| Element | What the companies say |
|---|---|
| Scope | Small-molecule candidates against a single oncology target that is not named |
| Isomorphic Labs’ role | In silico predictions and molecular designs from its drug design engine, IsoDDE, using select proprietary Merck data sets |
| Merck’s role | Preclinical and clinical development, manufacturing and commercialisation of any resulting products |
| Payments to Isomorphic | An upfront payment, milestone payments tied to research, development, regulatory and commercial achievements, and tiered royalties on net sales |
| Co-funding | An option for Isomorphic to co-fund development activities in exchange for elevated royalties |
| Amounts | Financial terms were not published |
Merck’s release describes IsoDDE as Isomorphic’s proprietary, modality-agnostic drug design engine, built from foundational AI models. The release says the engine supplies the predictive fidelity needed to work with novel biological systems and that Isomorphic continually refines its model architecture. Contract Pharma and Investing.com report the same roles and payment structure.
Background and context
Isomorphic Labs was founded in 2021 to apply AI to drug discovery. Its chief executive is Sir Demis Hassabis and its president is Max Jaderberg, and the company is headquartered in London with offices in Cambridge, Massachusetts, and Lausanne, Switzerland, according to the boilerplate in Merck’s release. STAT describes the company as built on a Nobel-winning technology that can model how molecules bind to proteins; our guide to the AlphaFold protein structure database explains the public resource from the same research lineage.
The Merck agreement is narrower than Isomorphic’s earlier partnerships. On 7 January 2024 the company announced multi-target small-molecule collaborations with Eli Lilly and Novartis. Isomorphic’s announcement lists upfront payments of $45 million from Lilly and $37.5 million from Novartis and says the two deals could be worth nearly $3 billion to the company, excluding royalties. On 20 January 2026 it announced a cross-modality, multi-target collaboration with Johnson & Johnson covering small molecules, antibodies, peptides and molecular glues, with no terms published. The Merck agreement is limited to one target and one modality.
Contract Pharma places the deal in a wider pattern of pharmaceutical interest in generative AI and machine learning for discovery and cites other recent AI drug discovery partnerships involving Roche, AstraZeneca and Quotient Sciences.
Reaction
Addona said in the release that the key to using AI and machine learning for molecular invention is “close coordination of wet-lab and dry-lab functions”, and that combining Merck’s internal data and discovery expertise with Isomorphic’s modelling and compute capabilities is intended to produce better candidates faster. Sarah Skerratt, Isomorphic Labs’ chief research officer, said in the same release that the company will apply its drug design engine to the program to unlock significant discovery progress, and described the collaboration as reflecting both companies’ commitment to patients with unmet needs.
What it means for researchers, universities and research offices
The agreement is between two companies, and neither the release nor the coverage describes a university role. Its value to research offices is as a worked example of how an AI developer and a pharmaceutical sponsor split the pieces of a discovery programme, which is the structure many institutions meet when an industry partner brings models or proprietary data to a joint project.
Three features stand out. First, the data flows one way into the models: Merck supplies selected proprietary data sets, and Isomorphic returns molecular designs. Institutions negotiating similar arrangements need the data-rights and background-IP terms settled before work starts; our definitions of a research collaboration agreement and a sponsored research agreement set out the two common vehicles, and the guide on publication delay and review clauses covers a term that commonly matters to academic partners.
Second, the economics follow a familiar licensing shape: an upfront payment, milestones across research, development, regulatory and commercial stages, tiered royalties and a co-funding option that raises the royalty rate. Technology transfer offices can compare that structure with their own term sheets, bearing in mind that the amounts here have not been published.
Third, the release places preclinical and clinical development with Merck. Any candidate that emerges would therefore be developed under the sponsor’s programme and, as with other drug development work, regulators’ expectations for the use of AI apply; the guide to FDA AI guidance for drug development sets out that framework.
What happens next
The release gives no timeline, target name or milestone schedule. Candidates produced under the collaboration would move into Merck’s preclinical and clinical development if they meet its criteria, and Isomorphic is eligible for payments as research, development, regulatory and commercial milestones are reached. Further details will depend on what the two companies choose to publish.
Sources
- Merck: Merck and Isomorphic Labs Enter Strategic Research Collaboration, 19 November 2026
- STAT: Merck, Isomorphic Labs partner to develop oncology drug candidates, 19 November 2026
- Contract Pharma: Merck, Isomorphic Labs Team Up on AI-Driven Cancer Drug Discovery, 19 November 2026
- Investing.com: Merck partners with Isomorphic Labs on AI drug discovery, 19 November 2026
- Isomorphic Labs: Isomorphic Labs kicks off 2024 with two pharmaceutical collaborations, 7 January 2024
- Isomorphic Labs: Isomorphic Labs enters into a research collaboration with Johnson & Johnson, 20 January 2026







