Abstract Metastatic castration-resistant prostate cancer (CRPC) is a genetically heterogeneous disease. Effective translational research demands a diverse and representative set of preclinical models. A major obstacle in the field is the limited capacity to culture prostate cancer-derived cells in vitro. Few cell lines exist. Those that are available do not well represent the genetic diversity of the disease, nor do they accurately predict in vivo response. Recently, organoid culture techniques developed in the Sawyers and Clevers laboratories have increased our ability to grow metastatic tumor-derived prostate cancer cells in vitro. This represents a great step forward, but is restricted by limited availability of tissue and a success rate of <20% when establishing new lines. In the field of prostate cancer, patient-derived xenograft (PDX) lines are a model that broadly represents the clinical and genetic diversity of the disease. Like primary prostate tumors, PDXs have not been readily adaptable to in vitro culture (including the standard organoid method), making them poorly suited for mechanistic studies or high-throughput drug screens. Therefore, there is great interest in adapting the prostate cancer PDX model to an in vitro culture system. The LuCaP series of PDXs is a well-characterized set of ~40 diverse prostate cancer specimens. Using LuCaP PDXs, we systematically tested several modifications to the organoid culture method. This includes the addition of factors to modify metabolism, growth, and differentiation; removal of standard media components; and combinations thereof. We show that p38 is required for growth and survival of almost all the LuCaP organoids. Removal of the p38 inhibitor component from standard prostate organoid media improves initial growth and makes it possible to maintain many PDX-derived organoids over several generations. Additional modifications to individual LuCaP culture conditions improve growth. We successfully cultured 21 of 24 different LuCaPs attempted, for at least one generation, and 10 have been grown long-term. We show that PDX-derived organoids recapitulate the phenotype of the original PDX tumor, including AR-dependent growth. For example, we have shown that homologous recombination deficient LuCaP organoid responses to olaparib reflect the enhanced sensitivity seen for patients in clinical trials. Additionally, our method improves the success rate for culturing samples directly from patient biopsies. We show that 2 of 3 biopsy-derived organoid lines that we have established grow well in our conditions but not in standard published conditions. The third grows equally well in both. Further, our data suggest that p38 is a previously unappreciated target for CRPC treatment. Importantly, our work immediately and significantly increases the number of prostate cancer lines available to the community by making the LuCaP PDXs accessible to in vitro culture, and by improving the success rate for establishing organoids directly from patient samples. Citation Format: Michael L. Beshiri, Caitlin M. Tice, Crystal Tran, Holly M. Nguyen, Adam G. Sowalsky, Supreet Agarwal, Keith H. Jansson, Kerry McGowen, Aian Neil Alilin, JuanJuan Yin, Fatima H. Karzai, William L. Dahut, Eva Corey, Kathleen Kelly. PDX-derived and patient-derived prostate cancer organoids as a clinically relevant platform for translational research [abstract]. In: Proceedings of the AACR Special Conference: Prostate Cancer: Advances in Basic, Translational, and Clinical Research; 2017 Dec 2-5; Orlando, Florida. Philadelphia (PA): AACR; Cancer Res 2018;78(16 Suppl):Abstract nr B051.
Read full abstract