How AI simplifies data management for drug discovery
Calithera is running registered clinical trials of its products to study their safety, whether they are effective in patients with specific gene mutations, and how they work in conjunction with other therapy. The company has to collect detailed data on hundreds of patients. While some of its trials are in the early stages and only involve a small number of patients, others cover more than 100 research centers around the world.
“In the world of life sciences, one of the biggest challenges we have is the amount of data we have, more than any business,” said Behrooz Najafi, Calithera’s leading information technology strategist. (Najafi is also chief information and technology officer for healthcare tech company Innovio.) Calithera must keep and manage the data while making sure it is easily accessible when needed, even years from now. . It must also follow specific FDA requirements on how data is produced, stored, and used.
Even something as simple as upgrading a file server must follow a strictly defined FDA protocol with several steps of testing and review. Najafi said all these changes in compliance-related data could add 30% to 40% to the overhead of a company like his, at the same direct cost and staff time. These are resources that can be put into a lot of research or other activities that add to the cost.
Calithera has left most of the additional cost and further improved its ability to track its data by placing it in what Najafi calls a safe “container container,” a protected area for regulated access. , which is part of a much larger cloud document management application, mostly driven by artificial intelligence. AI can’t sleep, doesn’t get bored, and learns to recognize among hundreds of different types of documents and data forms.
Here’s how it works: clinical or patient data is put into the system and scanned by AI, which identifies specific areas involved in accuracy, completeness, compliance with regulations, and other aspects of the data. The AI can be flagged if there is a missing test result, or if a patient does not submit a required diary entry. It knows who is allowed to access specific types of data and what they are and is not allowed to do so. It will detect ransomware attacks and turn them off. And it can automatically document everything to the satisfaction of the FDA or any other regulatory body.
“This approach takes away the burdens of following us,” Najafi said. If data from multiple research sites is on the platform, Calithera knows that AI will make sure it is safe, complete, and compliant with all regulations, and take advantage of any problems.
Managing drug discovery data to meet research requirements and vendor requirements can be, as Najafi observed, cumbersome and expensive. The life sciences industry may borrow data management methods and platforms designed for other industries, but these will need to be modified to control the level of security and verification, and the detailed procedures for -audit, that’s a way of life for drug makers. AI can speed up these tasks, improving data security, consistency, and accuracy-without paying the overhead for drug companies and research organizations to apply to their core mission.
A complex data management environment
Compliance with the regulation will help ensure that new drugs and devices are safe and function as intended. It also protects the privacy and personal information of thousands of patients who participate in clinical trials and post-market testing. No matter their size-very large worldwide or small startups looking to get a product on the market-drug manufacturers must follow the same standard procedures to document, audit, verify, and protect every bit of information connected to a clinical trial.
If researchers run a double-blind study, the gold standard for verifying a drug’s effectiveness, they should keep patients ’information anonymous. But they need to be able to quickly de-anonymize the data later, make it identifiable, so that patients in the control group can receive medication, and for the company to track — sometimes for years — how the product of real world use.
The burden of data management falls on hard-to-come and midsize biosciences companies, said Ramin Farassat, chief strategy and product officer at Egnyte, a Silicon Valley software company that creates and supports the platform. to manage AI-driven data used by Calithera and hundreds of other life- science companies.
“This approach takes away the burdens of following us,” Najafi said. Once data from multiple research sites is on the platform, Calithera knows that AI will make sure it’s safe, complete, and compliant with all regulations, and flag any problems.
Download full report.
Its content was created by Insights, the standard MIT Technology Review content. It was not written by the editorial staff of the MIT Technology Review.