280 Background: Since the 21st Century Cures Act, there has been growing interest in using real-world evidence (RWE) to support regulatory filings for new drugs and indications. The goal of this study was to provide a comprehensive source of RWE use cases of U.S. Food and Drug Administration (FDA) approvals for oncology products. Methods: A systematic review of FDA new drug applications (NDAs) and biologics license applications (BLAs) from 1/2015-2/2020 for approved oncology products was performed. Data on the RWE study were extracted (data source, study design, statistical methods, results), and corresponding FDA comments were synthesized to identify patterns of the FDA’s review. Results: We identified 102 NDAs and BLAs, 8 (8%) of which included RWE, all post-Cures Act (see Table). RWE supporting avelumab, axicabtagene ciloleucel, and avapritinib were received positively and used by the FDA in their approval decision. RWE results were used to provide contextualization to the pivotal trial, rather than statistical comparison. Common data sources included Flatiron (38%) and chart reviews (38%). FDA critiques included lack of a priori study protocol, incomparability with the pivotal trial population and endpoints, and uncontrolled confounding. Conclusions: There have been few examples of RWE in oncology submissions, and most served to complement clinical trial results. To meet FDA standards, RWE studies should be clearly designed and discussed with the FDA and include robust methods to minimize bias. [Table: see text]
A Pragmatic Approach to Data Source Selection for use In Real-World Evidence (Rwe) Generation
