A company focused on discovering drugs using artificial intelligence and machine learning is moving one of its technology-discovered compounds into the clinic later this year.
Salt Lake City-based Recursion said Tuesday that the Food and Drug Administration had approved its Investigational New Drug application to start a Phase I study of REC-994 in cerebral cavernous malformation, a genetic disease that affects up to 1.5 million people in the United States.
When Investment Rhymes with Canada
Canada has a proud history of achievement in the areas of science and technology, and the field of biomanufacturing and life sciences is no exception.
Chris Gibson, the company’s CEO and co-founder, said in a phone interview that the clinical trial would likely start in the third or fourth quarter of this year. At the same time, he clarified that while it would be accurate to say REC-994 – a superoxide dismutase mimetic – was discovered using machine learning, its discovery was not based on artificial intelligence. The difference between AI and ML is subtle and even itself the subject of debate among experts, he said, and AI itself is a form of ML. However, the discovery platform used to discover REC-994 did not include iterative feedback in which algorithms learn and get better and better over time, meaning it was not truly AI, though the company’s current discovery platform is AI-based.
The company’s goal is to discover 100 drugs using AI/ML by 2025, and its pipeline includes about 30 programs in various disease states. The next program will probably be IND-ready in about 18 months, Gibson said. What sets Recursion’s approach apart from competitors, he said, is that the company has developed its own data set internally, whereas other companies tend to rely on public data sets, in addition to being able to reduce what he called the “Achilles’ heel” of phenotypic screening, whereby researchers find interesting drugs but are unable to determine how they work.
Other companies using AI/ML in drug discovery include BioXcel, NuMedii and Qrativ.
AI/ML has grown in importance in drug discovery and development, with large drugmakers and technology companies forming partnerships to use the processes to find and develop new compounds. Swiss pharmaceutical giant Novartis formed a partnership with chip maker Intel to use deep neural networks to cut the time for analyzing microscopic images from 11 hours to 31 minutes. Recursion has also formed partnerships with large drug companies, particularly French firm Sanofi and Japan’s Takeda Pharmaceutical Co., Gibson noted.
Other such deals include a drug-discovery partnership announced last year between British drugmaker GlaxoSmithKline and startup Exscientia to discover small-molecule drugs, and an immuno-oncology partnership between Pfizer and IBM using the latter’s Watson platform. BioPharm Insight reported in September 2017 that it would likely be two to five years before AI/ML became mainstream in drug development.
Photo: MF3d, Getty Images
