Novartis and Microsoft conclude a five-year AI alliance2. October 2019
Novartis and Microsoft conclude a five-year AI alliance
Microsoft’s mission is to help Novartis better understand and utilize the wealth of data generated from its laboratory experiments, clinical trials, and manufacturing facilities.
For Microsoft, the partnership with Novartis provides the opportunity to test its AI technology for precise applications such as CAR-T therapy delivery and on a global enterprise scale. Their agreement is an alliance that benefits both the technology company and the pharmaceutical company. “It really is an equal partnership,” said Bertrand Bodson, Chief Digital Officer at Novartis, in an interview with BioPharma Dive. “It’s not a supplier relationship at all.”
The two companies’ employees will work together at Novartis headquarters in Basel, Switzerland, at the Novartis Service Center in Dublin, Ireland, and at Microsoft’s research laboratory in the UK.
Initially, they will focus on treating age-related macular degeneration and improving the production of CAR-T cancer therapies. Through a newly created AI innovation laboratory, they hope to make AI-based applications accessible and usable throughout the Novartis organization.
For Novartis CEO Vas Narasimhan, digital data is an important part of the transformation of the Swiss pharmaceutical company into a “leading pharmaceutical company”.
Novartis estimates that over the past two decades, clinical trial data worth 2 million patient years have been collected, while laboratories store data on 1.5 million chemical compounds.
The problem is not just the overwhelming volume of data, wrote Peter Lee, head of Microsoft Healthcare, in a blog post announcing the partnership.
“Much of the information is in the form of unstructured data, such as research lab notes, medical journal articles, and clinical trial results, which are typically stored in separate systems. Novartis and Microsoft will seek to overcome these hurdles in three areas – two specific and one broad. For example, in age-related macular degeneration, the companies will use AI to scan images of patient eyes and look for patterns that could pave the way for more personalized dosing.
“We have thousands and thousands of OCT scans,” says Bodson, referring to an imaging technique called optical coherence tomography. “This is a nice input for AI because one of the simplest areas for pattern recognition and patient stratification is imaging.
On the agenda is also the use of CI to improve the production of CAR-T cells, a time-consuming and resource-intensive endeavor that is causing Novartis headaches in the commercial launch of its drug Kymriah (Tisagenlecleucel).
Novartis and Microsoft also plan to apply AI to generative chemistry to develop better medicines through advanced computing.
“Typically, we have to design many molecules one after the other to do the right thing,” says Bodson. “We hope to accelerate and improve this process so that we can create molecules that have the highest chemical probability of binding in the right way to have the right structure.
The collaboration with Microsoft fits well with Narasimhan’s vision for Novartis. Another pillar of the CEO’s plan to build a leading pharmaceutical company is the development of complex drugs such as Kymriah or its gene therapy Zolgensma (onasemnogene abeparvovec).
However, Narasimhan and Novartis are involved in a regulatory scandal involving the transmission of manipulated data to the Food and Drug Administration. An investigation is underway and could lead to penalties for the company.