Merck and Isomorphic Labs partner on AI cancer drug discovery | Magica
Merck taps Isomorphic Labs’ AI for an undisclosed cancer target
Editorial Team
••📖6 min read
Merck will pair its experimental data with Isomorphic Labs’ AI to seek small-molecule cancer drug candidates. The deal follows Isomorphic’s $5 billion financing, but its payments, target and development timetable remain undisclosed.
Merck will supply proprietary data for AI designs against an undisclosed cancer target, then handle development and commercialization.
Isomorphic’s latest pharmaceutical partnership follows a $5 billion financing round; the deal payments are separate and undisclosed.
The test is whether model predictions produce experimentally validated candidates worth taking into development.
U.S. drugmaker Merck has agreed to combine its experimental data with London-based AI drug designer Isomorphic Labs’ computational engine to seek small-molecule candidates against one oncology target. The November 19 collaboration adds another major pharmaceutical partner to Isomorphic’s roster, while keeping the work of developing any resulting medicines with Merck.
Under the companies’ announcement, Isomorphic will make computer-based predictions and deliver molecular designs using its proprietary Isomorphic Labs Drug Design Engine, or IsoDDE, incorporating selected Merck datasets. Merck takes responsibility for preclinical and clinical development, manufacturing and commercialization.
Merck has already worked internally on the target, its discovery leadership said in an interview about the agreement. The collaboration therefore tests whether Isomorphic can improve molecular invention using data from an established research effort. The target and specific cancer indications remain undisclosed, preventing an assessment of its biological rationale or competing programs.
Merck, the U.S. pharmaceutical company known as MSD outside the United States and Canada, brings a large oncology business. Its 2025 financial results recorded $31.7 billion in worldwide sales of the cancer treatments Keytruda and Keytruda Qlex against $65.0 billion in total company sales. Dividing those figures puts the products at about 49% of annual revenue. That establishes the scale of Merck’s oncology interests, without implying a sales forecast for this discovery program.
OpenAI reports sharply lower reasoning-extraction rates in internal GPT-6.2 tests and backs sanctions against illicit foreign distillation. Its own research and rival Anthropic’s experience show why protecting hidden reasoning leaves capability transfer and customer costs unresolved.
Editorial Team
Isomorphic was founded in 2021 to build on Google DeepMind’s AlphaFold protein structure model, as a report on its financing explains. The Alphabet-backed company develops AI systems for drug design and pursues both partnered and wholly owned programs. Its technical overview describes jointly releasing AlphaFold 3 with Google DeepMind in 2024; IsoDDE extends that work into predicting additional properties needed for molecular design.
Isomorphic raised $2.1 billion in May and $5 billion in November. These investments are separate from Merck’s undisclosed collaboration payments. Source: Fierce Biotech.
A capital-rich partner, with deal economics undisclosed
Isomorphic will receive an upfront payment, potential payments tied to research, development, regulatory and commercial milestones, and tiered royalties on net sales. It can also choose to co-fund development in exchange for higher royalties. No payment amounts, royalty rates or co-funding requirements were disclosed.
That option could give Isomorphic a larger stake in eventual sales at the cost of committing more capital. The missing terms prevent a public assessment of either the agreement’s value or the risk Isomorphic would take by exercising it.
There is an earlier disclosure baseline. Isomorphic’s January 2024 announcement put upfront payments from drugmakers Eli Lilly and Novartis at $45 million and $37.5 million, respectively. It said the two collaborations could be worth nearly $3 billion excluding royalties, including contingent milestone payments. Both covered multiple small-molecule targets, so their potential totals cannot establish the value of Merck’s single-target agreement.
Johnson & Johnson is also a partner, as an account of the Merck deal noted. The agreement came two days after Isomorphic announced a $5 billion Series C round, following a $2.1 billion round in May. These are investments in Isomorphic, separate from Merck’s undisclosed payments.
Greenoaks and Sequoia Capital co-led the new financing, with continued investment from Alphabet. Isomorphic said in its funding announcement that proceeds would also support translational and clinical development infrastructure as it advances candidates toward human trials. The spending plans extend beyond improving models to building the capabilities needed to develop its own medicines.
Isomorphic chief executive Demis Hassabis, who led Google DeepMind until August and is now its chair and Alphabet’s chief scientist, described more than 20 programs and an ambition to produce hundreds of drugs a year in a Bloomberg interview relayed by the financing report. He also described plans to expand beyond small molecules into biologics. Those statements describe the wider platform’s scale and ambitions, rather than the scope of Merck’s agreement.
Hassabis said Isomorphic had secure computing supply, while the report noted that Google designs its own AI chips. That is management’s assurance about infrastructure access; it provides no program-level computing costs or capacity figures.
Benchmarks support a narrower claim than drug success
In its February technical overview, Isomorphic said IsoDDE more than doubled AlphaFold 3’s accuracy on the hardest category of the Runs N’ Poses protein–ligand structure benchmark, testing systems most dissimilar to the training data. That comparison concerns a specified prediction task, rather than an overall doubling of drug discovery productivity.
The company also described predicting binding affinity—how strongly a molecule binds to its target—and locating protein pockets where molecules might bind. It claimed binding-affinity accuracy exceeding physics-based methods on public benchmarks at lower time and cost. Those methods, along with earlier structure models, are relevant comparison points: Isomorphic is seeking improvements on existing computational approaches.
The overview itself acknowledged that understanding molecular structures alone was insufficient for computer-based drug discovery. Merck’s announcement likewise emphasized coordinating physical laboratory work with computational work. Neither document provides candidate-level results, a comparison against Merck’s existing methods, or measured time or cost savings for this target.
The next decision depends on experimental evidence
Isomorphic says its engine is producing positive preclinical results in internal oncology and immunology programs and with partners. That broader portfolio claim, relayed in the financing report, does not establish results from the newly announced Merck collaboration.
Clinical timing remains unresolved for the wider platform too. The report said Hassabis had expected Isomorphic’s first clinical trials by the end of 2026, but gave no updated target in the interview. He said regulatory submissions were on track and more announcements would follow in 2027. None of those statements supplies a timetable for Merck’s program.
For this partnership, the consequential next step is whether the designs yield experimentally validated candidates that Merck elects to advance. Disclosing a candidate and the evidence behind that decision would begin to show what the pairing of Merck’s data and Isomorphic’s models has achieved. The agreement alone cannot answer that question.
Alberta proposes 600-metre setbacks for large data centres and expanded utility oversight, but exemptions and rules still to be written will determine how much protection communities receive.
President Donald Trump backed Commerce Secretary Howard Lutnick as lawmakers scrutinize alleged delays in restricting a suspected route for AI chips to China. A new congressional inquiry also questions Nvidia’s influence on an existing investigation.
Amazon’s month-old Echo Time is $50 off for Black Friday. Its touchscreen, home hub and optional sleep-aware alarm cost $90 more than the reported Echo Spot price, while Alexa+ access and evidence of better waking remain key buying questions.
Thinking Machines Lab released a roughly 20GB open-weights model that reuses its middle layers for reasoning. It brings the Inkling family to local hardware, with competitive coding results but workload-dependent costs and preliminary evidence on oversight.
Apple’s first HomePod mini update since 2020 adds Mac audio and four-speaker surround, but Siri AI remains unavailable, core home-hub features carry over, and reviews disagree on playback reliability.
Treasury Secretary Scott Bessent says an unnamed Anthropic founder will announce a large gift to children's investment accounts within weeks, but the amount, recipients and connection to existing philanthropic pledges remain undisclosed.
Google says air cooling will limit water use at its Visakhapatnam AI hub, but environmental proceedings and residents’ accounts raise broader questions about land, electricity, public incentives and the evidence behind promised benefits.
Nvidia slipped behind Alphabet to fourth in U.S. market value during November 20 trading. The reversal follows a steep selloff despite strong reported revenue growth, with a permanent CEO still to be chosen.
Google will close the Nest app on January 25, 2027 and retire Assistant on home speakers and displays in August. Device transfers are automatic, but continued voice access requires accepting Gemini; basic assistance remains free.
Amazon says Cairn agents exploited a MONAI Label zero-day to take over a university server. Other attacks failed, and maintainers list no patched package release.
Federal Reserve Governor Lisa D. Cook sees early AI pressure on coding, translation and young workers, but says a solid overall labor market and persistent inflation still call for tighter monetary policy.
President Donald Trump threatened a tax on U.S. companies building data centers abroad without specifying how it would work. The proposal raises questions for an AI export program that has received 78 applications and for domestic construction constrained by power and community opposition.
Donald Trump said only older chips should be sold to China while discussing a broader deal with Beijing. His remarks leave unresolved which products would qualify as Washington tightens access to American AI hardware and overseas manufacturing.
Treasury Secretary Scott Bessent expects no change to the US policy permitting limited H200 sales to China despite sparse orders. Nvidia’s earlier filing shows Chinese restrictions and tariff costs also obstruct sales, while the company warns that exclusion helps competitors.
AI infrastructure stocks fell on November 20 amid cloud financing concerns and a reported NAND demand warning. Customer contracts support the buildout, but IREN’s filing shows the conditions and cash demands behind its expansion.
Jensen and Lori Huang’s foundation has committed $350 million to CoreWeave computing for research recipients. The purchase links charitable access to Nvidia’s commercial ecosystem, while pricing, allocation and supplier-selection terms remain unclear.
The mathematicians claim finite-time blow-up without external forcing, a significant advance if verified. Their repository omits the numerical certificate and its checking computation, and says the patched Lean package was not rebuilt.
The Pentagon treated Anthropic’s military-use safeguards as a supply chain threat. Later court rulings diverged under different statutes, while the dispute exposed how difficult it was to replace Claude in classified systems.
A new investigation describes decades of alleged humiliation under Nvidia co-founder and former CEO Jensen Huang. He denies a former HR chief’s claims, while employee support and stock rewards complicate the account of his management.