Abstract 34: A Drosophila approach to personalized cancer therapeutics

Personalized cancer genomics is providing unprecedented access into the genetic complexity and diversity of human tumors. The next challenge is to utilize this information to establish effective therapeutics. Functional interrogation of cancer genomes using genetic model systems provides a powerful step towards realizing this goal. To capture the genetic complexity and diversity of human tumors, we identified the most frequently observed double, triple and quadruple combinations of mutations in human colon cancer genomes generated by TCGA. We used this information to generate the corresponding multigenic models in Drosophila and investigated the tumorigenic and metastatic potential of these 32 models by activating transgenes specifically in the adult Drosophila colon. These models recapitulate key features of human cancer, many of which arise as emergent properties of multigenic combinations. Importantly, we found that multigenic models were more resistant to targeted drugs and compounds: 12/16 of the targeted agents we tested were effective against ras G12V alone while 0/16 were effective against the four-hit model ras G12V p53 RNAi pten RNAi apc RNAi . We identified a potential mechanism for resistance to PI3K pathway inhibitors as well as biomarkers for single agent response and resistance. With this in hand, we developed a combination therapy that overcame resistance of our four-hit model to single agent PI3K pathway inhibitors. We validated our findings in human colorectal cancer cell lines including xenografts as well as 3D colospheres and allografts derived from genetically engineered mouse colorectal cancer models. Overall, our models provide an excellent opportunity to study tumorigenesis and metastasis in the context of the whole animal and explore compound effects on a genetically diverse set of models. Through these efforts we have developed a platform designed to screen large numbers of personalized fly models in a rapid and cost effective manner. We are now leveraging these technologies—using personalized fly models to identify personalized drug cocktails—to treat individual patients in a clinical study focusing on medullary thyroid cancer and colorectal cancer. Briefly, we first generate high quality genomic profiles for our patients and use this information to build a personalized fly model for each patient. These models are then screened against a large library of FDA approved drugs in an iterative manner to identify drug combinations specifically tailored to each patient. Our approach to personalized cancer therapeutics leverages sophisticated genetic tools and high throughput drug screening methods in Drosophila to address tumor and whole body complexities and provides a unique opportunity to identify novel treatment options for individual patients based on functional exploration of their tumor genomes. Citation Format: Erdem Bangi, Claudio Murgia, Alexander Teague, Peter Smibert, Jessica Esernio, Nelson Gruszczynski, Caitlyn Yeykal, Owen Sansom, Ross Cagan. A Drosophila approach to personalized cancer therapeutics. [abstract]. In: Proceedings of the AACR Precision Medicine Series: Integrating Clinical Genomics and Cancer Therapy; Jun 13-16, 2015; Salt Lake City, UT. Philadelphia (PA): AACR; Clin Cancer Res 2016;22(1_Suppl):Abstract nr 34.