Abstract 635: Establishment of a large panel of patient-derived tumor xenograft models of prostate, bladder and kidney cancers

Abstract

Abstract Prostate, bladder and kidney cancers represent 1 900 000 new cases and 620 000 deaths per year worldwide with an incidence increasing by 1-10% each year. Surgery is curative at localized stages; however, current therapies are inefficient at advanced stages. One of the major needs in new drugs development is the availability of clinically pertinent models faithfully reproducing the heterogeneity of patient tumors. Patient-derived tumor xenografts (PDX) are thus developed as essential tools for drug testing and identification of predictive biomarkers for a better clinical response, the first step for personalized medicine. Prostate, bladder and kidney tumors were obtained from patients undergoing surgery, subcutaneously (and some orthopically) xenografted in nude mice and serially passaged up to passage (P) 12. Normal corresponding tissues were also harvested. Tissues were conserved at all passages for characterization and grafting. For all patients, informed consent and clinical history are available. For the 3 cancers, primary tumor and tumors grown in mice were characterized for growth behavior, histopathology, genetic stability (short tandem repeat fingerprinting), mRNA expression profiling and response to current therapies. For each cancer type, we also investigated more specific features: expression of the androgen receptor, PSA and pan-cytokeratin for prostate PDXs, expression and status of hotspot mutations including FGFR3, PIK3CA, HRAS, RXRa and p53 for bladder PDXs, and von Hippel-Lindau gene mutation status for kidney PDXs. Metastatic models were followed by infrared imagery (IR780 dye). So far and since 9 years, we have collected 230 prostate tumors, 130 bladder tumors and 336 kidney tumors at all stages, and established 5, 25 and 31 models (> P3 in mice) respectively (8.8% success rate). Tumor take rate was positively correlated to advanced stage and high grade and for kidney cancer, to sarcomatoid component. Tumor growth evaluation reveals that PDXs were stable from mouse to mouse and throughout passages. Histopathologic and genetic characteristics were preserved between original tumors and case-matched PDXs. Molecular characteristics were also stable with less than 5% of genes differentially expressed between the primary tumors and the PDXs. In bladder PDXs, this analysis and mutation status of the selected genes allowed to define molecular subtypes. The comparison with patient therapeutic response, when available, showed the clinical predictivity of the models. Orthotopic models developed metastases at classical sites. In conclusion, we developed here a unique platform of preclinical PDXs models for urologic cancers with stable biological characteristics and clinically predictive. This is an invaluable tool for the clinical design of efficient therapies, the identification of predictive biomarkers and translational research. Citation Format: Hervé Lang, Claire Béraud, Audrey Bethry, Sabrina Danilin, Véronique Lindner, Catherine Coquard, Sylvie Rothhut, Thierry Massfelder. Establishment of a large panel of patient-derived tumor xenograft models of prostate, bladder and kidney cancers. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 635.