Abstract B178: A functional mutational profile of a compendium of 350 patient-derived tumor xenografts (PDXs).

Abstract

Abstract Introduction: PDXs are an increasingly popular tool for drug discovery. One reason is that well-characterized tumor models enable streamlining of testing procedures, as molecular markers and drug efficacy can be assessed simultaneously. At Oncotest we have >350 well-characterized PDXs representing many tumor types including niche tumors. The aim of this study was to define PDXs by functionally distinct mutational profiles. Methods: Mutational analyses were conducted for well-studied oncogenes and tumor suppressor genes by Sequenom (OncoCarta panels 1, 2 and 3), the results of which were validated and extended by Sanger sequencing (targeted sequencing of exons containing hotspots of mutations). Results: We categorized mutational profiles of tumor types well-represented in the PDX collection by identifying common driver mutations and then dividing the mutation carrying tumors into sub-groups based on additional mutations. For example, we identified melanoma PDXs (MEXFs) with activating BRAF mutations and MEXFs with activating NRAS mutations. We then analyzed whether BRAF or NRAS mutant MEXFs also carried a PTEN mutation. We found that of 13 BRAF-mutant MEXFs, 8 had a PTEN mutation. Of the 7 NRAS mutant MEXFs, 2 carried a PTEN mutation. When testing a drug in melanoma, it will be of interest to include representative models from each of the sub-groups with distinct mutational profiles, as they can be expected to respond differently to treatment. We were also interested in screening all tumor types for mutations primarily present in distinct histologies and identifying mutations in new histologies. This could lead to a new histology being of interest for therapeutic intervention with a targeted agent, or the definition of tumors by a specific mutation or mutational profile, rather than by histology, as a potential concept for patient selection for future trials. For example, we analyzed IDH1 and IDH2 mutations, which are mostly studied in glioma and myeloproliferative disease in PDXs from >25 histologies. We found five PDXs with mutations in IDH1 and IDH2 known to produce the oncometabolite R(-)-2-hydroxyglutarate: Two melanoma PDXs, one liver PDX, two colon cell-line derived xenografts (CDXs) and one NSCLC PDX. The mutations are rare events in each of the histologies, but, if broad mutational screening is done, patients carrying druggable mutations regardless of the tumor histology may benefit from treatment targeting these mutations. Conclusions: Compiling mutation data of several genes leads to the identification of genetic sub-groups of a tumor histology, which is a useful basis for the efficacy testing of novel drugs, as potential biomarker information is simultaneously collected. Screening different histologies for well-characterized mutations can potentially identify additional histologies or individual tumors for the further use of drugs targeting such mutations. Citation Information: Mol Cancer Ther 2013;12(11 Suppl):B178. Citation Format: Rebekka Krumbach, Jagatheswari Virajah, Thomas Metcalfe, Heiner Fiebig, Vincent Vuaroqueaux. A functional mutational profile of a compendium of 350 patient-derived tumor xenografts (PDXs). [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2013 Oct 19-23; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2013;12(11 Suppl):Abstract nr B178.