insitro, the South San Francisco AI drug-discovery company founded by Daphne Koller, has signed a worldwide exclusive research collaboration with AstraZeneca to find new targets and medicines for metabolic dysfunction-associated steatohepatitis (MASH) and cardiomyopathies, the diseases of heart muscle.
Under the deal announced on 15 October 2026, insitro will receive $40 million upfront and is eligible for up to approximately $1.46 billion in potential discovery, development and commercial milestones, plus single-digit royalties on potential global commercial sales. Those larger sums are potential only: they depend on programmes succeeding at later stages, and the companies have not said any medicine from this collaboration is in the clinic yet.
What each side will do
insitro will apply its Virtual Human platform to identify causal disease biology in both indications and will advance AstraZeneca-nominated targets from discovery through to a development candidate. AstraZeneca will then lead clinical development, regulatory work and commercialisation.
MASH is a serious form of fatty liver disease tied to metabolic dysfunction. Cardiomyopathies are conditions in which the heart muscle is abnormal, which can impair how the heart pumps. Both, insitro said, are defined by patient heterogeneity: the same diagnostic label can cover people whose underlying biology differs, which has made them hard to treat with a one-size-fits-all approach.
What the companies said
Daphne Koller, founder and chief executive of insitro, said: “MASH and cardiomyopathies are prevalent, underdiagnosed diseases where high-content data holds immense potential for uncovering disease drivers. By applying AI to these complex datasets, we can unlock greater discovery power and reveal novel genetic insights.”
Sharon Barr, executive vice president and head of BioPharmaceuticals R&D at AstraZeneca, said: “We know that disease biology is incredibly complex and one of the most critical decisions in R&D is selecting the right target. This strategic collaboration underscores our commitment to maximizing the potential of cutting-edge AI to tackle the root cause of disease.”
How big are the numbers, really?
The $40 million upfront is cash tied to signing. The “up to approximately $1.46 billion” is a ceiling on potential milestones across discovery, development and commercial stages. Milestone language in biotech deals usually means payments that only become due if specific goals are hit — for example advancing a candidate, starting a trial, or reaching sales thresholds. TSN has not seen a schedule that breaks those payments down by stage, so readers should treat the headline total as an upper bound, not money already earned.
Royalties are described as single-digit and based on potential global commercial sales. There is no approved product from this collaboration to generate those sales today.
Company background (as claimed)
insitro describes itself as unlocking “causal human biology” with AI and says it has built what it calls the world’s largest integrated multi-modal corpus of human and cellular data, plus the Virtual Human engine and a TherML platform for designing medicines. Those platform claims are the company’s own characterisation in its release; TSN has not independently audited the corpus size or ranked it against peers.
The company also says it is advancing a pipeline of nine programmes, including liver and ALS efforts, and that it is backed by more than $750 million from investors including a16z, ARCH, BlackRock and others, with prior collaborations with Bristol Myers Squibb, Lilly and Gilead. Those are company statements in the same release.
Why it matters
Big pharma continues to buy access to AI discovery platforms rather than building every capability in-house. Pairing AstraZeneca’s development and commercial scale with insitro’s stated strength in heterogeneous human biology puts two hard disease areas — fatty liver and heart muscle disease — under a partnership that is explicitly about target choice, not about announcing a finished medicine.
For patients, nothing changes yet: no trial start date was given for assets emerging from this specific collaboration. For the AI-for-science sector, the deal is another data point that large milestone packages are still being written around platform collaborations, even when the near-term cash is a much smaller upfront.
What’s still unknown
- Which exact AstraZeneca targets will be nominated, and how many programmes the collaboration covers in parallel.
- Whether any development candidate will reach the clinic, and on what timeline.
- How the approximately $1.46 billion potential milestones split across discovery, development and commercial events.
- Independent validation of insitro’s “largest corpus” claim.
What to watch
Watch for a first development-candidate nomination under the deal, any clinical-trial registry entries that name both parties, and AstraZeneca pipeline updates that list MASH or cardiomyopathy assets linked to the collaboration.
What this means
- Deal signed; drugs not yet. The collaboration and $40 million upfront are announced; clinical medicines from it are not.
- Big number is potential. Up to about $1.46 billion in milestones is an eligibility ceiling, not cash in hand.
- AstraZeneca runs the clinic. insitro works through development candidate; AstraZeneca leads thereafter.
Sources
Related stories
- Deep Origin and BioAge Labs Team Up on AI Drug Discovery, With First Hits Confirmed in the Lab
- Generate Biomedicines and Zealand Pharma Team Up to Design Obesity Protein Drugs With AI
- Multiply Labs’ $75M Series B: robots between AI discovery and patient doses
Follow @tsnmediahub on X for breaking AI news as it happens.
Follow on X
