Abstract 3225: Identification of predictive drug resistance markers in preclinical xenograft models (PDX) of soft tissue sarcomas

Abstract

Abstract Soft tissue sarcomas are a rare and heterogeneous type of cancer. Although targeted therapies have extended and improved treatment options for many cancers, treatment of sarcomas is still mainly based on standard chemotherapy with marginal improvements over the last years. As cell cultures can hardly model the complex environment of the disease we established a large panel of patient derived sarcoma xenografts (PDX). A set of 11 new sarcoma PDX models was established from samples of 29 primary human soft tissue sarcomas patients which were obtained directly from the clinic. The tumour samples were taken from representative peripheral areas of the original tumour. Fragments from each tumour were transplanted subcutaneously into immunodeficient mice. These models were together with another panel of sarcoma PDX models (Hoffmann et al. 1999) screened for sensitivity to a set of standard of care drugs (e.g. ifosfamide, docetaxel, gemcitabine, doxorubicin, dacarbazin) and some new investigational compounds (everolimus, imatinib). The main objective of the study was to establish and characterise the soft tissue sarcomas PDX, the evaluation of therapeutic options and the identification of predictive markers. Conservation of the typical sarcoma morphology of the PDX (pleomorphic-, ewing-, clear cell-, myxofibrosarcoma) was compared with the original tumour by histological and immunohistochemical staining. Heterogeneity of tumour cells in size and shape and nuclear irregularities were observed. The models were further analyzed for common oncogenic mutations using the Illumina TrueSeq Cancer Panel. Frequent mutations were detected in KDR, MET, RB1 and TP53. However no typical mutational profile was found within the panel as many of the sarcomas showed individual profiles. Most of the PDX responded to docetaxel and gemcitabine followed by doxorubicin and ifosfamide whereas only some were sensitive to dacarbazine. Sensitivity to some targeted drug has been evaluated in a subset of the PDX models. All PDX tested so fare did not respond to everolimus or imatinib. Comparison of the chemosensitivity profile of the PDX and the patient outcome was possible for 9 patients. Analysis revealed that the prediction for a therapy resistance was correct in 10 out of 13 cases. Further analysis for correlations between mutation and gene expression profiles and drug response will be performed. We have shown, that our set of established sarcoma PDX models closely resemble the patient malignancy and particularly predict treatment response in the clinic. Available pharmacological data on drug response together with the molecular characterization will allow comprehensive investigations of therapy-related response markers any may help to introduce new targeted drugs in sarcoma therapy. Citation Format: Jana Rolff, Frank Traub, Per-Ulf Tunn, Jens Hoffmann, Iduna Fichtner. Identification of predictive drug resistance markers in preclinical xenograft models (PDX) of soft tissue sarcomas. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 3225. doi:10.1158/1538-7445.AM2015-3225