Abstract 230: Modeling tumor progression and identifying genotype-drug interactions by the sequential introduction of cancer mutations into the genome of human normal cells

Abstract

Abstract The mutational landscape of cancer genomes displays a complex combination of genetic lesions affecting oncogenes and tumor suppressor genes. The construction of cellular models closely recapitulating cancer-linked genetic alterations is a prerequisite to understand their role in tumor progression and to identify genotype- specific pharmacological responses. To recapitulate point mutations affecting oncogenes we used homologous recombination to ‘knock-in’ specific nucleotide changes in the genome of human cells. We focused on the following alleles: EGFR (delE746-A750), KRAS (G13D), BRAF (V600E) and PIK3CA (E545K and H1047R), that are found in multiple human cancers. As recipient cells, a human non transformed epithelial cell line of breast origin (hTERT-HME1) has been employed. This cell line has been immortalized by hTERT and harbors a mutation on TP53 gene, a genetic event frequently observed in naturally occurring human tumors. To recapitulate the inactivation of additional tumor suppressor genes in our cellular models, we exploited shRNAs to permanently knock-down the expression of PTEN and RB1 genes, as they are frequently down regulated in human tumors. Knock-down of PTEN and RB1 was performed both in the parental hTERT-HME1 cells and in the ‘knock-in’ models, thus allowing us to combine tumor suppressor genes inactivation with mutational activation of specific oncogenes. The resulting cellular models are hereafter refereed to as combinatorial ‘matrix’. To assess how specific combinations of genetic lesions affect the tumorigenic properties of not transformed cells, we measured the ability of the ‘matrix’ to grow in soft agar. At the same time we injected our cellular models in the cleared, humanized mammary fat pad of immunocompromised mice. For the latter approach, we focused on genotypes recapitulating mutations found in breast tumors, such as PTEN or RB1 inactivation and PIK3CA or KRAS mutations, alone or in combination. In vitro and in vivo results showed that those combinations of genetic events are not sufficient to transform the normal hTERT-HME1 cells. However, preliminary data indicates that few genotypes of the ‘matrix’ acquired the ability to colonize the murine fat pads. With the rational of unveiling new pharmaco-genetic interactions involving one or multiple cancer mutations, the “matrix” was used to screen a panel of compounds used for cancer therapy, or for which an anti-proliferative activity was previously reported. This approach led to the identification of a subset of genotype-drug interactions thus underscoring how the multifactorial genomic milieu of cancer cells impact the response to therapies. Grant support: This work is part of the EU FP7 grant COLTHERES Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 230. doi:10.1158/1538-7445.AM2011-230