Abstract B110: The use of patient-derived xenograft models to explore new treatment strategies in uterine sarcomas

Abstract Background: Uterine leiomyosarcomas (uLMS) are rare, aggressive tumors. The limited benefit of current treatments emphasizes the need for new treatment modalities. Our aim is to gain insight into the molecular determinants of response to targeted and systemic therapies in uterine sarcomas, using pre-clinical models. Methods: Nine uLMS (uLMS1-9) patient-derived tumor xenograft (PDTX) models were established in collaboration with the PDTX platform. In the LMS1 model, we selected 6 treatment groups (+placebo group; 4 mice/group; 28 days) based on molecular findings and literature. First, SNP-array technique was used to detect copy number alterations. Heterozygous deletions were detected of phosphatase and tensin homolog (PTEN) and PH domain and leucine rich repeat protein phosphatase 1 (PHLPP1), both negative regulators of the PI3K/mTOR pathway. Immunohistochemistry confirmed mTOR pathway activity through phospho-S6 ribosomal protein detection. Therefore, we selected the PI3K/mTOR inhibitor BEZ235. Second, pazopanib (inhibits PDGFR, VEGFR, c-KIT and FGFR) was recently approved for treatment of LMS. ULMS1 showed homogeneous expression of PDGFR. Third, histone deacetylase inhibitors have recently been shown to inhibit cell growth in many cancers, including LMS. Therefore, valproate was selected in single therapy, as well as in combination therapy with pazopanib and BEZ235. Last, trabectedin was included as state-of-the-art chemotherapy. Based on this pilot experiment, the uLMS1 model was again treated with BEZ235, trabectedin and the combination of both compounds in a larger cohort for 22 days (7-9 mice/group). Additionally, a second uLMS model (uLMS2) was treated with the same compounds for 22 days (5-6 mice/group). Results: In the uLMS1 pilot experiment, treatment with trabectedin and BEZ235 resulted in stable tumor volume, while other compounds showed no effect. No significant differences were noted, possibly due to the low number of mice used. Therefore, the experiment was repeated using only BEZ235, trabectedin and the combination treatment in a higher number of mice. Although BEZ235 and trabectedin showed minor reductions in tumor growth, the effects were not significant. However, in the uLMS2 model, BEZ235, trabectedin and the combination treatment all led to a highly significant reduction in tumor volume, while placebo-treated tumors grew as large as 4 times the starting volume. Placebo vs trabectedin, BEZ235 and the combination all showed p-values lower than 0.0001. The combination treatment showed a stronger effect than single-agent trabectedin (p = 0.0001) and single-agent BEZ235 (p = 0.0003). Conclusions: BEZ235, trabectedin and the combination of both compounds showed effect in a uterine sarcoma PDTX model. Next, their effects will be assessed in different uLMS models and responses will be linked to selected pathway activities, determined by gene/protein expression, protein activation and copy number alterations. Citation Format: Tine Cuppens, Els Hermans, Debby Thomas, Ellen Gommé, Matthieu Moisse, Diether Lambrechts, Frederic Amant. The use of patient-derived xenograft models to explore new treatment strategies in uterine sarcomas. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2015 Nov 5-9; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2015;14(12 Suppl 2):Abstract nr B110.