Prostate cancer (PCa) is the most frequently diagnosed cancer (180,890 new cases in 2016) in men in the USA. Androgen deprivation therapy (ADT) is an effective first line therapy for locally advanced or metastatic disease. Unfortunately, once PCa recurs, the eventual development of castration-resistant prostate cancer (CRPC) remains an incurable disease and more effective therapies are needed. Currently, a limited number of cancer cell lines (LnCAP, PC3, DU-145, etc.) are available for research and many genetic mutations present in prostate cancer (e.g., SPOP mutation, FOXA1 mutation, TMPRSS2-ERG fusion, CHD1 loss) are not represented in such cells. New patient-derived cancer models are needed. However, patient-derived xenograft (PDX) models are successful at less than 2-5% efficiency with aggressive, high-grade metastatic tumors and organoid cultures only have an efficiency of 20%. However, our preliminary data demonstrate that conditional reprogramming (CR) has nearly a 100% success rate for establishing long-term cultures from either surgical prostate specimens or CT-guided biopsies. In this application, we propose the following specific aims to validate the potential of CR for translational use in human CRPC. We will first establish CR cultures from biopsies of 30 patients with CRPC and will characterize these culture genetically and phenotypically. Second, we compare the patients' drug response to those of corresponding tumor CR cells and their derivative CR- derived xenografts (CDXs). Lastly, we will use CR cultures in an unbiased high-throughput screen to identify new potential therapies for CRPC in collaboration with Dr. Craig Thomas at National Center for Advanced Translational Sciences (NCATS). New ?hits? from the screen will be validated by both in vitro cell assays and xenograft models.
Patient-derived cancer models are critical for both basic and translational studies of prostate cancer. We will establish conditionally reprogrammed (CR) cultures from castration-resistant prostate cancer (CRPC) patients and will validate the use of these models for translational applications, including the prediction of drug responses.