Abstract 812: Multiplatform Modeling of Pancreatic Cancer Using Patient-Derived Cells: A New Approach for Defining Drug Resistance Mechanisms

Abstract Background: Pancreatic adenocarcinoma (PA) is the fourth leading cause of cancer-related death in the USA. Current treatments utilize Gemcitabine alone or in combination with other drugs, such as nab-paclitaxel, however, a growing number of patients have shown resistance to these regimes. In order to enhance clinical response, it is important to better understand the mechanisms of drug resistance. To date, one of the main limitations in PA research has been the reliance on a few commercially available PA cell lines or a limited number of genetically engineered or PDX mouse models for in vitro and in vivo studies. It was therefore imperative to develop more accurate cellular models that better represent a given patient’s tumor. In our current study, we present the use of patient-derived primary PA cells as a model system for basic and translational research, and for personalized medicine. Methods: Patients’ biopsies were collected after surgery, and long-term cultures of PA cells were established using the conditional reprogramming of cells (CRC) approach we developed. KRAS and p53 sequencing verified the PA origin of both the patient samples and the matched CRC lines. The IC50s for Gemcitabine and Abraxane were determined and used to establish drug resistant clones. Our two dimensional (2D) CRC cultures have shown to be adaptable to different platforms, such as 3D spheroid cultures as well as zebrafish- and mouse- PDX models. Results: We established KRAS-mutant primary cell lines derived from patients’ PA samples. From two different parent cell lines we created five nab-paclitaxel and two Gemcitabine resistant clones. The clones were 3-1000 times less sensitive to the drugs as compared with the parent lines. Drug resistance ratios between parent and resistant clones were confirmed both in 3D cultures and our zebrafish PDX platform. The injection of both parent and resistant clones into NCI/nude mice induced tumor masses in 100% of the cases. Unlike the parent cell-derived tumors, the nab-paclitaxel clone-derived tumors exhibited a metastatic phenotype. Histological analyses of these PDX models recognized the presence of morphological features of human PA, including ductal acinar structures comprised of pleomorphic epithelial cells with extensive stromal desmoplasia surrounding the tumors, and thereby represent a fundamentally important advancement in the PDX modeling field. Conclusion: Overall, the ease of culture, the genetic stability and the high-throughput ability to identify differences between patients sensitivity to FDA approved drugs confirm the power of this technology for on-demand in vitro use in PA research. The in-vivo approach demonstrated the accuracy of our model system in recapitulating the patient tumor. Our method now enables the high-resolution experiments necessary to better understand the resistance mechanisms that directly affect clinical outcomes Citation Format: Erika Maria Parasido, Praathibha Sripadhan, George Avetian, Richard Schlegel, Jonathan Brody, Jordan Winter, Charles J. Yeo, Michael J. Pishvaian, Erik Glasgow, Stephen Byers, Christopher Albanese. Multiplatform modeling of pancreatic cancer using patient-derived cells: A new approach for defining drug resistance mechanisms [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 812. doi:10.1158/1538-7445.AM2017-812