e14003 Background: Brain metastases (BrMet) remain a clinically challenge. There is an increase interest in evaluating the efficiacy of systemic therapy for BrMet. Patient-derived organoids (PDO) and xenografts (PDX) are thought to capture the tumor heterogeneity and molecular alterations of the source tumor, and may be used as ‘avatars’ for therapeutic response assessment of the source patient. PDO has an advantage over PDX with a shorter establishment time, and thus, may allow a real-time drug sensitivity testing. The objective of this study was to determine the feasibility of establishing PDO from resected breast cancer BrMet and to evaluate the drug sensitivities in a time period amenable to eventual implemention into clinical utiliity. Methods: Under an IRB-approved protocol, resected BrMet tissues were prospectively collected at the time of clinically indicated neurosurgical procedure as part of a banking program. Tumors were directly cultured as 3-D organoids. DNA and RNA were collected. Drug testing panels included standard clinical care drugs or drug combinations as well as those selected based on the molecular profiling. When clinically conducted next-generation sequencing testing reports were available, we also tested drugs (or class of drugs) that were suggested as a potential therapeutic consideration in these reports. When possible, Nanostring PanCancer Panel evaluation was conducted at the time of PDO establishment to guide the selection of drugs. Results: To date, 11 breast cancer BrMet samples have been collected. 8/11 were successfully established as ‘direct from operating room’ PDO and underwent drug panel testing. The reasons for 3 cases not drug tested include: no viable tumor (confirmed by clinical pathology report as necrosis only) and primary brain tumor (not metastasis). Of the 8 PDO, the clinical subtypes were 5HER2+, 2HR-HER2-, and 1HR+HER2- cases.The median time from collection to the drug testing results availability was 11 days (range: 7-19 days). The median number of drugs or drug combinations tested was 23 (range: 7-32). The drug testing revealed various patterns of sensitivities to chemotherapies and targeted therapies. Nanostring PanCancer testing of PDO identified potential targetable pathways for which PDO demonstrated sensitivity when the drugs were matched to the associated molecular pathway. Conclusions: While the expected number of cases has been much lower due to COVID-19 pandemic-related change in clinical practice patterns and research operations, we have successfully demonstrated that the real-time establishment of BrMet PDO is feasible and may be used as a platform for further investigations. Real-time PDO could be potentially employed to predict drug sensitives for prioritizing drug therapy options in a clinically meaningful time-frame. PDO platform may be used to investigating biomarkers and mechanisms of response and resistance to therapy.