Celgene taps Exscientia’s AI drug discovery tech for 3 new programs

Exscientia is adding another Big Pharma to its list of partners. This time, the drug discovery specialist is teaming up with Celgene on three programs in oncology and autoimmunity. Celgene is handing over a $25 million upfront payment, as well as promising the usual milestones and royalties, though those remain under wraps for now.

Under the three-year pact, the partners will use Exscientia’s artificial intelligence-based drug discovery platform, dubbed Centaur Chemist, to speed up the discovery of small molecules for three targets. Celgene has already selected the targets, said Alex Snow, chairman of Exscientia. The platform consists of two main parts: a data integration engine and an automated modeling and active learning engine, he said. The engine designs new molecules and assesses them for their potency, selectivity and ability to bind to specific targets and uses a rapid “design-make-test” cycle to quickly alter candidates toward the desired criteria.

The technology can crunch Big Data—data libraries, large patient databases, historical data and the like—as well as what Snow calls “small data.” He means what drug developers might use when they foray into discovery areas where there is no data, such as when trying to find a target compound that has no comparable compounds. In that case, Exscientia might use its platform on data from a competitor compound that’s off-patent or a simple protein structure.

RELATED: Sanofi inks €250M bispecific small molecule diabetes deal

“The platform is reasonably agnostic; it handles both large or very, very small data well. We’ve seen great results doing it both ways,” he said.

The Oxford, U.K.-based company is working with GlaxoSmithKline, Sanofi, Roche and Evotec in a similar capacity, but the Celgene partnership is bigger in terms of size and scale, as well as the amount of control Exscientia will have over the project, Snow said.

“In our other collaborations, we have tended to do the algorithm-driven and AI-based work, and the pharma partner will use their in-house chemistry, or third-party chemistry to keep the data flowing,” Snow said. In the Celgene deal, Exscientia will take care of the entire project, "from gene to the drug candidate," which means using its own chemistry and CRO partners. It could make a “significant difference to the speed and outcomes of this collaboration,” he said.

RELATED: GlaxoSmithKline, Exscientia ink AI-based drug discovery deal worth up to $42M

Exscientia is looking to sign more partners over the course of the year, but it's also working on its own in-house pipeline using its technology. At the moment, the company is working on about 20 discovery targets and wants that number to double in 2020. That will come from the expansion of existing partnerships, one or two new ones to come in the first half of this year, and the building out of Exscientia’s internal programs, Snow said.