AstraZeneca pays insitro $40 million upfront in AI drug-target deal worth up to $1.5 billion

AstraZeneca announced a drug discovery collaboration with insitro on Thursday, Oct. 15, aimed at two diseases where the biology has resisted researchers: metabolic dysfunction-associated steatohepatitis (MASH), an advanced form of fatty liver disease, and cardiomyopathies, a group of heart-muscle disorders.

The headline number is up to about $1.5 billion. The guaranteed number is $40 million.

How the money is structured

Insitro gets $40 million upfront and is eligible for up to roughly $1.46 billion in discovery, development and commercial milestones. It would also collect single-digit royalties on global sales of any resulting product.

AstraZeneca takes exclusive worldwide rights to advance programs that come out of the work. It leads all clinical development, regulatory work and commercialization.

Bloomberg described the deal as worth up to $1.86 billion. The figures the companies disclosed add up to about $1.5 billion, and the companies have not explained the gap.

The announcement does not say how much of the milestone money could be paid before a candidate reaches the clinic.

What insitro is supposed to do

Insitro will apply its Virtual Human platform, which combines human and cellular data with machine learning, to identify causal disease biology in both conditions. It will then advance AstraZeneca-nominated targets from discovery through to a development candidate.

Both diseases carry a single diagnostic label over patients whose underlying biology differs. That heterogeneity is why they have been hard to treat, and it is the problem the collaboration targets.

"By applying AI to these complex datasets, we can unlock greater discovery power and reveal novel genetic insights," said Daphne Koller, insitro's founder and CEO.

Sharon Barr, AstraZeneca's executive vice president and head of BioPharmaceuticals R&D, said: "One of the most critical decisions in R&D is selecting the right target."

Koller told Bloomberg that "the vast majority of AI drug-discovery companies focus on what happened after a target has been identified." Insitro, she said, works on the earlier step of understanding disease biology.

She said the company has made considerable progress on MASH and expects to identify targets on a relatively near-term timeline. The cardiovascular program will take longer, she said.

AstraZeneca's recent record in these areas

According to Bloomberg, AstraZeneca has stumbled in both fields. Its drug Wainua failed to show benefit in a heart-muscle disease trial earlier this year, and the company dropped a mid-stage MASH candidate last year.

AstraZeneca is also chasing a large goal: roughly $80 billion in revenue and 20 new medicines launched by 2030. It has added AI partnerships beyond insitro, including Algen Biotechnologies for immunology targets, a Modella AI acquisition and an agentic AI licensing deal with Owkin.

Small upfront, big ceiling

The analysts at Tikr put the $40 million in context. AstraZeneca spent $13.85 billion on R&D in 2025, up from $8.05 billion in 2021. The upfront equals about 0.3% of that, roughly one day of research spending. The full $1.5 billion maximum would be about 11% of one year's R&D budget.

Tikr called the structure "an option, not a bet." Its read is that big pharma sees AI discovery as "worth a seat at the table, not yet worth paying for in advance."

The numbers fit that view. Two days earlier, Zealand Pharma agreed to pay Generate Biomedicines $30 million upfront against nearly $1.2 billion in milestones for AI-designed obesity drugs. In both deals the upfront is under 3% of the headline figure.

CEO Pascal Soriot said in July that AstraZeneca does not need more business development to deliver its targets, according to Tikr. This deal is a bounded target-discovery arrangement, not an acquisition.

Investors responded mildly. AstraZeneca shares edged up in premarket trading after the announcement.

A crowded field

Insitro is not new to this. Gilead signed with the company in 2019 and Bristol Myers Squibb in 2020, and BMS has extended and expanded that deal over the years. Eli Lilly teamed up with insitro in 2024, which led to a platform being offered to select biotechs in 2025.

Other drugmakers are moving the same way. Novo Nordisk said last month it would use Anthropic's Claude Science platform for drug research. Roche said earlier this year it would build a supercomputer with Nvidia.

None of these arrangements has yet produced an approved drug from an AI-identified target. The test for this one comes first in MASH: Koller expects targets there relatively soon, and AstraZeneca has not said which targets it will nominate or when.