e14130 Background: There are few concrete examples of real world evidence (RWE) used to support clinical development in regulatory filings despite growing interest in this field. This study systematically reviewed FDA oncology approvals to identify use cases of RWE. Methods: FDA’s new drug application (NDA) and biologics license application (BLA) approvals for oncology products from 2015-2019 were systematically reviewed. Among cases with RWE, data characterizing the submission and RWE details (data source, study design, FDA comments) were synthesized. Results: 93 approved NDAs and BLAs were identified; 6 included RWE in support of efficacy (see Table), approved on or after 2017, and were largely retrospective studies that contextualized results to pivotal trial, with primary endpoints overall survival (OS), overall response rate (ORR), and time to treatment discontinuation (TTD). Flatiron data were used in 3 of these as database analyses, 1 was an expanded access program (EAP), 1 was a meta-analysis, and 1 was a retrospective chart review. Conclusions: In the past 5 years, few FDA decisions incorporated RWE in oncology drug approvals. When used, RWE has been a complement rather than a supplement for clinical trial data. Early engagement, a priori protocol development, and robust research design (adjusting for bias, comparability to clinical trial population) remain key determinants for successful use of RWE in FDA decision making. [Table: see text]