Identification of gene expression differences between primary pediatric tumors and their PDX models

Abstract Molecularly targeted therapies inhibit specific cancer pathways and have fewer harmful side effects than cytotoxic chemotherapies; however, the development of targeted therapies for childhood cancers has lagged behind that of adult cancers. Although adult and childhood cancers have different molecular mechanisms, similar druggable cancer pathways may be activated. We propose using gene expression analysis to identify opportunities to reposition FDA-approved targeted therapies for childhood cancer patients. We developed an N-of-1 gene expression approach that leverages a compendium of ~11,000 cancer gene expression profiles. To evaluate the utility of this approach for identifying cancer driver pathways, we analyzed gene expression data for 9 patient-derived xenograft (PDX) tumors. PDXs were tested for sensitivity to selected targeted therapies. Our analysis correctly predicted drug sensitivity in 8 of the 9 PDXs based on overexpression of specific cancer pathways. We also compared gene expression between PDX and matched primary tumor samples (n=7). Using TumorMap analysis, we found that PDX tumors cluster with their matched primary tumor and other samples with a similar cancer diagnosis. We also ran differential expression analysis and found 260 overexpressed genes and 630 underexpressed genes in PDX tumors. Underexpressed genes were associated with immune functions, extracellular matrix proteins, and growth factor signaling. Overexpressed genes were involved in TNF-α; signaling. This analysis confirmed expected changes in gene expression while also identifying TNF-α; signaling as a potential biologic artifact of PDX tumors. These features may influence the results of PDX experiments and should be accounted for when using this model. Overall, the PDX model recapitulates gene expression features of pediatric cancer and is a powerful tool for evaluating novel precision medicine approaches. Citation Format: Jacob J. Pfeil, Alex G. Lee, Leanne C. Sayles, Ellen T. Kephart, Holly C. Beale, Lauren M. Sanders, Olena Morozova, Sofie R. Salama, Alejandro Sweet-Cordero, David Haussler. Identification of gene expression differences between primary pediatric tumors and their PDX models [abstract]. In: Proceedings of the AACR Special Conference: Pediatric Cancer Research: From Basic Science to the Clinic; 2017 Dec 3-6; Atlanta, Georgia. Philadelphia (PA): AACR; Cancer Res 2018;78(19 Suppl):Abstract nr A22.