Abstract 4917: Aligning xenograft models to glioblastoma (GB) patient tumors to assess chemovulnerability of patients

Abstract

Primary human tumor xenografts are able to capture the heterogeneity of Glioblastoma (GB) and can be exploited to identify patterns within genomic profiles that are indicative of vulnerability to specific treatments. The Ivy Genomics-Based Medicine (IGBM) project accessed 40 GB human xenografts in mice, and evaluated responses to 4 different treatments. Additionally, gene expression data for 301 GB patient samples from TCGA were investigated to find context-specific gene interactions within patient data. We discovered 12 sets of genes with corresponding subsets of samples with coherent expression profiles, termed contexts. Each context was tested for enrichment of known GB subtype and patient survival characteristics. Our goal is to align 39 xenografts to the 12 contexts, by establishing the similarity between the context samples and the mapped xenografts, and characterize the contexts for their potential therapeutic responses. We first eliminated confounding probe hybridization from mouse tissue. Removal of such “confounders” was inadequate to resolve gene expression disparity between xenografts and patients. We then used Kolmogorov-Smirnov test to determine a set of genes that retain similar distributions between the expression data sets, resulting in 1,650 genes. Initial clustering analysis indicated that xenografts could be aligned across GB patient samples, without any apparent segregation. Xenografts were mapped to individual contexts using various similarity metrics such as correlation or Mahalanobis distance based on expression data, and the quality of alignment was measured using Silhouette scores. We have also conducted preliminary survival analysis for the xenograft subgroups (untreated) and compared it to the patient data. Finally, we compared survival data of 12 treated xenografts to determine if certain contexts are associated with improved or decreased sensitivity to four different drug regimens (temozolomide (TMZ), TMZ PARP inhibitor, BEZ235 and Sorafenib). The results indicated that xenografts aligned to context enriched with classical subtype showed sensitivity to the TMZ drug regimens, whereas contexts enriched with proneural subtype were sensitive to TMZ PARP inhibitor and resistance to TMZ; xenografts not associated with a subtype showed resistance to Sorafenib. While this preliminary analysis needs to be strengthened with more xenografts, aligning xenografts to GB patients with specific gene expression patterns appears to be promising in finding drug targets for these patient groups. This project has been funded in whole or in part with Federal funds from the National Cancer Institute, National Institutes of Health, under Contract No. HHSN261200800001E. The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. government. It is also supported by the Ben&Catherine Ivy Foundation. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 4917. doi:1538-7445.AM2012-4917

Contemporary murine models in preclinical astrocytoma drug development

“…Therefore, systematic drug efficacy screening in multiple genomically characterized PDX models might be useful for prospectively identifying sets of tumors that are likely to respond. 18,97 In order to evaluate the predictive accuracy of these models and maximize their utility for biomarker discovery and development, preclinical studies in PDX models should be performed earlier in the drug development process, ideally prior to initiating clinical trials or concurrently as "co-clinical trials." 98 Rather than relying on data from preclinical models, predictive biomarker discovery for TMZ and EGFR TKIs was largely conducted by retrospective molecular analyses of clinical trial specimens.…”

“…The multi-institutional Ivy Genomicsbased Medicine Project is currently utilizing this approach to investigate both novel and conventional cytotoxic agents and develop predictive biomarkers in a genomically diverse panel of PDX models. 18,97 Genetically Engineered Human Cell Models Models using genetically engineered normal human brain cells (geHCs) have been recently developed to overcome some of the limitations of human astrocytoma ECL and PDX models (Fig. 3C).…”

“…184 Moreover, more systematic use of multiple genomically diverse models in preclinical drug efficacy screens promises to aid development of predictive genomic biomarkers. 18,97…”