Back to: FDA Regulatory Education for Industry (REdI) 2024 Conference – Plenary and Drugs Track
Presenter
Qi Liu, PhD, MStat, FCP
Associate Director for Innovation & Partnership
Office of Clinical Pharmacology (OCP)
Office of Translational Sciences (OTS)
Center for Drug Evaluation and Research (CDER)
US Food and Drug Administration (FDA)
Qi Liu, PhD, MStat, FCP, is the Associate Director for Innovation & Partnership in OCP, OTS/CDER/FDA. She leads OCP’s innovative initiatives through strategic partnership. She has helped develop OCP’s portfolio on machine learning/artificial intelligence, real-world evidence, and digital health technologies, collaborating with internal and external experts. She led OCP’s Physiologically Based Pharmacokinetic Modeling and Simulation Oversight Board and co-led the Biologics Oversight Board. She was also a co-lead initiating the Real-Time Oncology Review and Assessment Aid Pilot Programs. During her career at the FDA, she also contributed to over 200 NDA/sNDA reviews, 20 BLA/sBLA reviews, and numerous IND reviews to support drug development. She worked on working groups for FDA guidance documents and Manual of Policies & Procedures. She is an Associate Editor of Clinical Translational Science and on the editorial board of five scientific journals. Before joining FDA, Dr. Liu was a senior pharmacokineticist at Merck & Co. Inc. She obtained her PhD degree in Pharmaceutics and a concurrent Master’s degree in Statistics from the University of Florida in 2004. In addition, she has a Master’s degree in Pharmaceutics and a Bachelors’ degree in Clinical Pharmacy from West China University of Medical Sciences.
Abstract
This presentation by Dr. Qi Liu from the FDA’s Office of Clinical Pharmacology discusses the rapidly expanding application of Artificial Intelligence (AI) and Machine Learning (ML) in drug development and regulation. AI/ML offers significant potential to improve the efficiency and probability of success of drug development and to advance precision medicine. The talk covers fundamental definitions, highlights important ML algorithms like neural networks and transformers, and outlines their diverse applications across the entire drug development lifecycle, from drug discovery to post-marketing surveillance. Regulatory submissions involving AI/ML are increasing sharply at CDER. Key regulatory considerations include defining the context of use, ensuring fitness for purpose, and applying a risk-based credibility assessment framework that weighs model influence and decision consequences. The FDA encourages model transparency and the development of best practices. Examples of sponsor proposals, such as using ML for patient risk stratification or generative AI for digital twins and covariate adjustment, are reviewed, alongside an example of FDA’s own use of ML to identify a suitable patient population for a COVID-19 treatment under Emergency Use Authorization. Challenges discussed include bias, generalizability, explainability, and data diversity. The FDA is actively engaged through initiatives like the Innovative Data Analytics Program, a dedicated ML review team, the Quantitative Medicine Center of Excellence, and plans for upcoming guidance documents, emphasizing the need for continued education, research, and collaboration in this evolving field.