Introducing the Emerging Drug Safety Technology Program (EDSTP)
The EDSTP is specifically focused on the use of artificial intelligence (AI) and other emerging technologies in pharmacovigilance (PV) and is part of the Center for Drug Evaluation and Research’s (CDER) multifaceted approach to enhance mutual learning of where and how specific innovations, such as AI, can best be used across the drug product lifecycle.
Emerging Technology’s Potential in PV and Safety Surveillance
FDA has a longstanding commitment to ensure medicines marketed in the United States are safe through continued surveillance and research following approval. In the postmarket setting, regulated industry (per 21 CFR 314.80, 314.98, and 600.80) is obligated to review all adverse drug experience information received or otherwise obtained and submit reports to FDA. Both industry and regulatory authorities face challenges with timely and efficient collection, processing, and evaluation of single and aggregate patient safety data compounded by ever-increasing case volumes. Advances in emerging technology have the potential to address some of these challenges by creating more efficiencies within a PV system. For example, early adopters of AI are leveraging these emerging technologies to automate fundamental tasks (e.g., adverse event intake, data entry, and processing) with the intention to drive down associated administrative burden and costs. These technologies can also make safety surveillance more efficient and effective by capturing, aggregating, and analyzing larger and more diverse data sets.
Goals of the EDSTP
CDER recognizes industry’s interest in dialogue around AI capabilities that advance PV. Industry may have concerns that using such technologies could create uncertainties regarding satisfying regulatory obligations for PV. This is especially true while CDER familiarizes itself with new technologies (e.g., assessing their performance characteristics and efforts to validate and verify models) to determine how they may be evaluated within our regulatory framework.
In this context, CDER established the EDSTP with three goals:
- Serve as the central point of contact for discussion between industry and CDER on the use of AI and other emerging technologies in PV
- Enable knowledge management and knowledge transfer within FDA specific to the context of AI or other emerging technologies used in PV
- Understand the context of use of AI and other emerging technologies in PV to inform potential regulatory and policy approaches within PV
To help further these goals, CDER created the Emerging Drug Safety Technology Meeting (EDSTM) program, which will be administered by the EDSTP. FDA expects that increased communication with the broader pharmaceutical industry during EDSTMs will accelerate FDA’s understanding of how AI enabled tools and other emerging technologies are being used for PV, their performance characteristics, their associated risks and benefits, efforts to validate and verify relevant models, and barriers to implementation.
Questions and Answers
The EDSTM is a means by which eligible participants (see Eligibility Criteria) can meet with CDER to share information about their use of AI and other emerging technologies and its potential application in PV. The goal of the meeting is to facilitate mutual learning and discussion on the opportunities and challenges with using such technologies in PV. Those selected for a meeting will meet with CDER to discuss their research, development, and/or use of AI and other emerging technologies in PV.
EDSTMs will be attended by members of CDER’s Emerging Drug Safety Technology Program, which includes representatives from CDER staff with experience in emerging drug safety technologies, pharmacovigilance activities, policy, and relevant inspection programs (e.g., Postmarketing Adverse Drug Experience (PADE) Compliance Program). The relevant interdisciplinary experts attending an EDSTM will depend on the nature of the topic proposed by the meeting requester. Additional experts from other centers or offices may participate as resources and time permit.
CDER is interested in a deeper understanding of AI and other emerging technologies that industry is exploring or has applied to PV activities. CDER is also interested in additional safety-related use cases involving emerging technology in PV, including but not limited to, signal detection and evaluation. The FDA AI/ML for Drug Development Discussion Paper references common uses of AI in PV, particularly for postmarket safety surveillance such as for the processing, evaluation, and submission of individual case safety reports (ICSRs) (see pages 8 – 9).
When exploring use cases of AI and other emerging technologies in PV with industry in EDSTMs, CDER is also interested in understanding how industry establishes the credibility and trustworthiness of AI models, including the following areas of consideration:
- Human-led governance, accountability, transparency, and explainability
- Data quality, reliability, representativeness, and bias mitigation
- Model development, performance, monitoring, and validation.
These areas of consideration are noted in the AI/ML for Drug Development Discussion Paper (see pages 17 – 23).
CDER is now accepting requests for EDSTMs. Refer to the Submission Timeline and Process for quarterly submission deadlines. Requests will be reviewed on a quarterly basis for a total of up to nine participants in a 12-month period for the initial phase of the EDSTM.
Contact Us
For more information regarding CDER’s Emerging Drug Safety Technology Program (EDSTP), email AIMLforDrugDevelopment@fda.hhs.gov and include the subject line “EDSTP – General Inquiry”
