Abstract 1925: Establishment and characterization of 3D cancer organoids as clinically relevant ex vivo drug screening tools for cancer translational research and drug discovery

Patient derived cancer organoids (PDO), a form of comprehensive 3D spheroid cultures that harbor cancer multicellular components and mimic cancer lesion structures and tumor heterogeneity, have been proven to be innovative preclinical model system for use in both academic research and industrial cancer drug development. It has been demonstrated by a large amount of independent literature reports that 3D PDOs can better predict drug response than 2D cell line settings due to their clinical relevant tumor micro-environment. We have the world’s largest patient derived xenograft (PDX) tumor model collection (~2500 models), which were annotated and validated genomically and phenotypically. However, there appears to be limitations when using PDX models to evaluate compounds in large scale fashion with narrow time windows and limited budget allocation. To address these limitations and to fully leverage preclinical values of these PDX model assets, we have established and characterized a series of PDX-derived cancer organoid (PDXO) models so that they could be used as scalable and high throughput compatible drug screening platforms. As first phase of development, we started with low passage PDX primary tumors from 5 major cancers types, including colorectal, lung, breast cancer, gastric and pancreatic carcinoma. The PDX tumor tissues were mechanically and enzymatically dissociated and the digested tumor fragments were cultured and passaged in Matrigel containing culture medium supplemented with stem cell niche factors. Technical protocols for 3D cultures including culture propagation procedures and medium optimization have been established. Characterization and QC of these PDXO models were done by gross morphology, histopathology, and genomic profiling. Histopathological analysis showed cellular/structural similarities (ductal, mucous or carcinoid) between PDXO models and original PDX tumor tissues, suggesting that tissue specific structural features were maintained in the 3D organoids. Moreover, ~10 PDXO models were tested for ex vivo cytotoxicity with SOC drugs. The preliminary data indicated that majority of (73%, 24/33) the PDXO ex vivo drug sensitivity IC50 datasets correlated with the in vivo PDX drug sensitivity TGI data. The negative and positive predictive values for PDXO are 94% and 53%, respectively. Genomic characterization including RNAseq and WES, is in progress. In summary, we successfully established PDXO models that were stably passaged from PDX primary tumors and resembled original PDX tumors both morphologically and histopathologically. In addition, the PDXOs apparently had encouraging predictive values for SOC drug response. Further in-depth model validation/characterization is warranted. Citation Format: Xiaoxi Xu, Satya Pathi, Limei Shang, Yan Liu, Peng Han, Likun Zhang, Binchen Mao, Davy Ouyang, Henry Li, Wenqing Yang. Establishment and characterization of 3D cancer organoids as clinically relevant ex vivo drug screening tools for cancer translational research and drug discovery [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 1925.