Abstract Introduction: Patient-derived models such as organoids are a powerful translational tool for preclinical research with several advantages over current conventional models: 1) they are derived from stem cells with self-organizing and renewal properties reflecting the 3D structure and morphology of original tissue; 2) they preserve the original genomic, physiological, and multicellular identity thereby representing patient diversity and heterogeneity in a dish; 3) they provide in vitro convenience for greater throughput for screening large numbers of compounds and/or patients; 4) they are amenable to genetic engineering. Methods: We have used Hubrecht Organoid Technology (HUB) protocols to systematically create a biobank of organoids derived from a well-annotated patient-derived xenograft (PDX) library (the world's largest commercially available PDX library with over 2,500 models, covering a variety of carcinomas, with extensive pathology, genomic, and treatment information), referred to as PDX-derived organoids or PDXOs. Results: We have currently developed more than 200 PDXOs covering many cancer types, including colorectal, gastric, liver, lung, bladder, breast, ovarian, pancreatic cancer, etc. Genomic analyses have demonstrated that PDXOs largely preserve the histo- and molecular pathology of the original PDX. In vitro maintenance of PDX models as organoids enables genetic manipulation to be performed, which is not possible for in vivo PDX models, thereby creating new models to facilitate drug development. We have employed both lentivirus-based and CRISPR/Cas9-based approaches to modify various tumor organoid models with luciferase, tumor associated antigens (TAA), and some EGFR mutants unavailable in current preclinical model panels. These engineered organoid models, alongside corresponding in vivo matched models, have been incorporated into a new drug efficacy platform. Preliminary results show that these engineered organoid models greatly expand the capacity and application of our current organoid bank. Conclusions: In summary, engineered cancer organoids are a great tool to enrich and diversify current tumor organoid biobanks, particularly for unavailable and clinically relevant models to be used for drug testing, as well as for studies on mechanisms of action and drug resistance. Citation Format: Fei Wang, Rongyun Zhou, Xinlong Dong, Yuqing Ma, Jun Zhou, Xiaoxi Xu, Yujun Huang, Henry Li, Phillip S-Z Wang. The emergence of tumor organoid engineering as a powerful tool for drug efficacy and mechanism of action profiling [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr LB-007.