Official: FDA modernizing pharmacovigilance oversight with AI tools
BALTIMORE – To modernize its pharmacovigilance efforts, the US Food and Drug Administration (FDA) Office of Surveillance and Epidemiology (OSE) is using artificial intelligence (AI) to support the review of adverse event reports and is also piloting the standardization of Risk Evaluation and Mitigation Strategies (REMS) data to make it easier to use, said OSE Deputy Director Robert Ball.Speaking at DIA's Global Pharmacovigilance and Risk Management Strategies Conference in Baltimore, MD, Ball said that OSE launched its Information Visualization Platform (InfoViP) an AI-based decision tool in 2022 to help the agency respond to the growing number of adverse event reports. The FDA receives more than 2 million reports a year, and this number is increasing every year. With these reports, FDA must decide whether to generate safety signals for marketed drugs.
Ball explained that “it is a very manual process and [involves] multiple steps with human interactions, and it is challenging for us in generating all the safety information that has to be processed.”
Reviewing these reports involves four activities: intake, evaluation, follow up, and distribution, activities that an AI model can emulate.
InfoViP has three main functions: a temporal data visualization feature, which applies natural language processing (NLP) to extract clinical concepts from each individual case safety reports (ICSRs) to visualize the timeline of adverse events, a duplicate algorithm where NLP is applied to extract key features in structured data, and an assessability algorithm where NLP and machine learning models classify repots according to the quality of information. Assessable ICSRs that contain sufficient information are then used for causality assessments.
REMS reporting
Other efforts are in the works to incorporate and standardize REMS data into the health care system for prescribers. FDA awarded a 13-month $2 million grant to MITRE Corpration to integrate REMS data in August 2023; the pilot uses the HL7 international standard to support exchanging health data across the health care system. By using this standard, prescribers can complete their REMS requirements without having to go outside their workflows.
FDA plans to launch real world pilots this year and to develop an initial HL7 implementation guide.
International colaboration
There is also heightened interest among regulators worldwide in exploring the use of AI in pharmacovigilance.
Last fall, FDA and Health Canada convened a workshop to discuss their internal experiences and some general principles to frame the discussion of AI going forwards.
Other initiatives include FDA and European Medicines Agency’s (EMA) launch of a working group exploring AI in the lifecycle of medicines in the fall of 2023 with Health Canda and Japan’s PMDA joining as observers in in 2024.
In addition, the Council for International Organizations of Medical Sciences (CIOMS) XIV recently convened a working group to develop a scientific consensus document on AI in pharmacovigilance. The group consists of regulators, industry, and members of academia.
Lastly, the Pharmaceutical Inspecting Cooperation’s Scheme’s (PIC/S) is Good Pharmacovigilance Practices (GVP) group is exploring creating an expert circle focused on incorporating AI and machine learning in the pharmacovigilance area.
×
Welcome to the new RAPS Digital Experience
We have completed our migration to a new platform and are pleased to introduce the updated site.
What to expect: If you have an existing login, please RESET YOUR PASSWORD before signing in. After you log in for the first time, you will be prompted to confirm your profile preferences, which will be used to personalize content.
We encourage you to explore the new website and visit your updated My RAPS page. If you need assistance, please review our FAQ page.
We welcome your feedback. Please let us know how we can continue to improve your experience.