Real-world data and artificial intelligence will soon become part of Bristol-Myers Squibb’s oncology arsenal, under a partnership with a firm focused on real-world evidence generation in cancers.
Boston-based Concerto HealthAI and the New York-based drugmaker said Thursday that they had formed a partnership to integrate various data sources and apply AI and machine learning in clinical trials, protocol design and the generation of insights for precision treatment and improved patient outcomes. Under the deal, BMS will use Concerto HealthAI’s eurekaHealth real-world data and AI insights system. The idea, they said, is to accelerate insights through novel health economic outcomes and also synthetic control arm studies in clinical development.
Physicians often use the term “real world” to refer to the use of interventions outside the confines of a clinical trial. Real-world data means evidence gathered through routine patient care from sources like electronic health records, medical claims, lab tests and health technology connected to mobile devices, from which real-world evidence, or RWE, can be developed. In December, the Food and Drug Administration issued guidelines for use of real-world evidence in the context of regulatory decisions. The agency, outgoing Commissioner Scott Gottlieb wrote, has already begun incorporating real-world evidence into regulatory decisions, and it released the December guidance under requirements in the 21st Century Cures Act.
That doesn’t mean applications of RWE and synthetic control arms are without their critics, such as Dr. Vinay Prasad, an associate professor of medicine at Oregon Health and Science University. Prasad, a prominent critic of the use of non-randomized clinical trial data in regulatory decisions, tweeted last month, “It’s painful to listen to calls for ‘synthetic’ control arms or RWE for drug approval when the real is obvious: Make [randomized, controlled trials] easier to do (and this has been done!).”
Other uses of AI/ML in drug development have cropped up as well, such as in drug discovery and development. In a newsletter Monday, drugmaker Agenus said it would begin using its own AI platform – called Adaptive Learning Platform Systems, or ALPS – to provide insights into cancer biology and identify new targets for development by integrating large amounts of laboratory data. That, in turn, will help the company find new ways to combine immuno-oncology therapies like checkpoint inhibitors and cell therapies. “For us, the machine learning comes into play by integrating different data sets,” Agenus senior scientist Thomas Horn said in a phone interview.
Photo: MF3d, Getty Images
