Hansa Biopharma partners with Cradle to design autoimmune drugs with AI protein engineering

Hansa Biopharma and Cradle said Thursday that Hansa will use Cradle’s AI-driven protein design and engineering platform across its discovery pipeline to design and optimize novel immunomodulatory candidates for severe autoimmune disease.

Listed drug companies are starting to buy protein-design software that learns from their own lab results, the way software teams buy coding agents. Hansa already sells an enzyme that cuts a common antibody, and it is putting Cradle across the next pipeline, a bet that AI sequence design will shorten the slog from an idea to a candidate for severe autoimmune disease.

On Thursday, 1 October 2026, Hansa Biopharma AB said it had formed a partnership with Cradle. Hansa lists on Nasdaq Stockholm under the ticker HNSA. The release calls Hansa a pioneer in treatments for rare immunologic conditions tied to IgG. IgG is immunoglobulin G, a common class of antibody. Cradle, the release says, builds an AI-driven protein design and engineering platform for biologics. A biologic is a drug made from a protein or another product of a living system, rather than a small chemical pill. The Cision page is datelined Lund, Sweden, and Amsterdam, the Netherlands, and it stamps Thu, Oct 01, 2026, 08:00 CET. Cradle’s blog carries the same announcement and dates it October 1, 2026. PR Newswire carries it as well, datelined Amsterdam, and stamps Oct 01, 2026, 07:29 ET. The pages do not print one shared clock. Those lines are the companies’.

What Hansa says it will do with the platform. Hansa will use Cradle’s end-to-end platform across its discovery pipeline to design, optimize, and advance novel therapeutic candidates. The line under the headline says Hansa is deploying that platform to develop novel immunomodulatory therapies in autoimmune disease. Immunomodulatory means the drug changes how the immune system behaves. Autoimmune disease is illness in which that system attacks the person’s own body. The release says the companies aim to speed new therapeutics for patients with severe autoimmune diseases. End-to-end, in this release, means the platform covers the discovery path, from an early hit through later optimization, rather than one model used for a single step. Those lines are Hansa’s and Cradle’s.

The lead product, kept separate from this partnership. Hansa’s lead product is imlifidase, a first-in-class enzyme. First-in-class means the first drug of its kind. It is commercially available in Europe under the brand name Idefirix. It is presently in review at the U.S. Food and Drug Administration, with a PDUFA date in December, for highly sensitized patients awaiting a kidney transplant. PDUFA is the law that sets the agency’s goal date for finishing a review. The release says December and does not print the year or the day. Highly sensitized means the patient’s blood already carries many antibodies that would attack a donor kidney. The release presents imlifidase as the product Hansa already has. It does not say imlifidase was designed on Cradle.

Why Hansa says it picked Cradle. Sofia Järnum, vice president and head of research and early development at Hansa, is quoted on the release. She said the company evaluated several AI models and vendors and selected Cradle because it combines three capabilities Hansa believes are essential for next-generation protein engineering. The first is end-to-end support across the discovery pipeline, from hit identification onward. A hit is an early molecule that shows a useful effect. The second is multi-parameter optimization across complex engineering objectives. That means improving several properties of the protein at once, rather than chasing a single score. The third is active learning from Hansa’s own data, so later results are more predictable. Active learning, here, means the model uses the company’s lab results to choose the next designs. She said each of the three accelerates Hansa’s lab-in-the-loop process. Lab-in-the-loop means the software and the experiments take turns: the model proposes, the lab tests, and the next round uses that result. She said Cradle’s platform lets Hansa scientists design and engineer better therapeutic candidates, faster and with a greater probability of success. That quotation is hers, in the release. The release does not print a measured success rate for a Hansa molecule designed on Cradle.

How the release frames the bet, and Cradle’s speed claim. The collaboration, the release says, reflects Hansa’s continued investment in AI-powered drug discovery as it builds the next generation of its pipeline. It says Cradle’s record with leading biopharmaceutical companies made Cradle a natural partner as Hansa expands the use of AI across new discovery programs. The release does not name those other companies. It describes a shift toward one platform that combines candidate generation, optimization, and continuous learning from experimental data. According to Cradle, customers report development timelines up to 18 months faster. Up to, in that sentence, is a ceiling on what customers have reported. It is not a clock on this partnership. The same sentence suggests that a faster timeline would turn into significant research-and-development savings per clinical candidate. The release does not print a dollar figure for those savings, and it does not print a price for the Hansa partnership.

Cradle’s chief executive is quoted on the same release. Stef van Grieken, chief executive and co-founder of Cradle, said: “The future of multimodality therapeutics engineering isn’t about spending more AI tokens or navigating a stable of models. It’s about empowering scientists with a platform that learns alongside them, while enabling them to apply their unique human expertise to solve complex molecular challenges at scale, with speed and precision.” A modality, in drug discovery, is the kind of molecule, such as an antibody or an enzyme. The release does not define his use of the word further. He said Cradle is excited to partner with Hansa “to accelerate the discovery and development of novel, high quality therapeutics for patients suffering from rare and severe autoimmune disorders.” Those quotations are his, in the release.

Who the about boxes say the companies are. Hansa calls itself a commercial-stage biopharmaceutical company developing immunomodulatory therapies for acute or complex immune disorders. Its IgG-cleaving enzyme technology, the box says, addresses needs in transplantation, gene therapy, and autoimmune diseases. Cleaving, here, means cutting the antibody. The portfolio named on the page is imlifidase, which the box says has been shown to enable kidney transplantation in highly sensitized patients, and HNSA-5487, a next-generation IgG-cleaving molecule that will be developed for Guillain-Barré syndrome. Guillain-Barré syndrome, shortened to GBS, is a rare condition in which the immune system attacks nerves. The box says HNSA-5487 will be developed for that use. It does not say the molecule was designed on Cradle, and it does not say it is approved. Hansa is based in Lund, Sweden, with operations in Europe and the United States. Cradle’s mission line says it wants every scientist to be able to turn today’s impossible cures, foods, and materials into tomorrow’s ordinary. Its platform, the box says, helps scientists engineer biologics from design through a candidate, bringing AI-driven molecular engineering, active learning from experimental data, and multi-property optimization into one system. Multi-property optimization is the about box’s wording for the idea Järnum calls multi-parameter optimization. The box says the platform is used across more than 80 active research-and-development programs, by companies in pharmaceuticals, industrial bio, agriculture, and consumer goods. It does not print those names. The 80 is Cradle’s count of programs. It is not a count of Hansa programs. Cradle is based in Amsterdam, Zürich, and the United States. The box says it is backed by IVP, Index Ventures, and Kindred Capital.

The picture is Cradle’s product screen. A navy field shows a Round 3 workflow for HER2 antibodies, on the step that generates candidates. HER2 is a protein some cancer drugs are aimed at. An antibody is a protein the immune system uses to recognize a target. The screen is Cradle’s example of that design workflow. The Hansa release does not say this screen is a Hansa project, and it does not say Hansa is designing an anti-HER2 antibody. The screen does not print a calendar date.

In plain terms, Hansa said on Thursday that it will run Cradle’s AI protein-design platform across its discovery pipeline, to design and optimize new immunomodulatory candidates for severe autoimmune disease. Järnum’s account of the choice is end-to-end discovery support, optimizing several properties at once, and learning from Hansa’s own lab data. Cradle says customers have reported timelines up to 18 months faster, and that its platform is in use across more than 80 research programs. Those speed and program figures are Cradle’s. Imlifidase, sold in Europe as Idefirix, is the enzyme Hansa already has in FDA review for highly sensitized kidney-transplant patients. The pages do not print a dollar value for the partnership, an equity stake, a clinical trial started because of this deal, or an FDA approval of a Hansa candidate designed on Cradle. They do not say imlifidase was designed on Cradle.