FDA Regulatory Science Report Signals Top Priorities
A new FDA report identifies four regulatory science strategic initiatives for 2021 — unleashing the power of data; increasing choice and competition through innovation; empowering patients and consumers; and public health emergency preparedness and response. The Advancing Regulatory Science at FDA: Focus Areas of Regulatory Science report was developed to “identify and communicate priority areas where new or enhanced investments in regulatory science research capacity are essential to support FDA’s regulatory and public health mission,” the agency says. Increased research in these areas is intended to aid in innovative product development, provide data to inform regulatory decision-making, and improve guidance to sponsors, it says.
The report discusses the importance of real-world evidence (RWE) in regulatory decision-making, including its ability to provide “fit-for-purpose” clinically meaningful information about the safety and effectiveness of medical products (begins page 39). It outlines several projects that will advance the roles such data can play in supporting the evaluation of a product’s safety and effectiveness, it says. For example, FDA is funding a RWE demonstration project called “Randomized Controlled Trials Duplicated using Prospective Longitudinal Insurance Claims: Applying Techniques of Epidemiology.” The project will attempt to duplicate the results of recently completed randomized controlled clinical trials relevant to regulatory decision-making using RWE, based on health insurance claims data, FDA says.
The report also says FDA is participating in studies focused on understanding how real-world data (RWD) may be able to inform regulatory decisions with external RWD providers. “One objective is to facilitate the use of RWD to learn about the safety and efficacy of FDA-approved oncology drugs in populations generally under-represented in clinical trials,” it says. Additionally, the agency is supporting projects exploring analytic methods that inform RWE, such as machine learning, for drawing conclusions between the occurrence and causes of an adverse event (i.e., causal inference).