Abstract B025: Inducible mouse models of cancer driver mutations as a powerful tool for in vivo and epigenetic studies of colorectal cancer

Abstract Human colorectal cancer (CRC) most often arises as a result of spontaneous genetic driver mutations such as APC, BRAF, KRAS, SMAD4, and TP53. These are among the most common drivers of CRC and are thought to contribute directly to cancer initiation and growth. CRC tumors harboring BRAFV600E mutation exhibit a CpG Island Methylator Phenotype (CIMP). Other drivers are also more weakly associated with distinct DNA methylation profiles. It is not clear whether these DNA methylation profiles are a result of differences in the cells-of-origin for these cancers, with pre-existing methylation patterns or whether there is selective pressure for various DNA methylation alterations that cooperate with each driver. To address this question, we have modeled this in mice by inducing colon tumors in a controlled cell population at the same age, with either BRAFF-V600E, KRASLSLG12D, Smad4, or Trp53 drivers in combination with Apc heterozygosity. We have profiled DNA methylation in tumors arising in these mice to determine the interaction between different driver mutations and the tumor methylome. Using a multiple linear regression model, we have identified DNA methylation alterations that are shared across all tumor samples, as well as changes that are specific to each of the different genetic drivers. The DNA methylation profiles accurately recapitulate well-established findings, including high levels of hypermethylation in KRASLSLG12D and BRAF F-V600E tumors reminiscent of CpG Island Methylator Phenotype (CIMP), universal tumor hypermethylation at binding sites for transcription factors involved in intestinal epithelial differentiation, and hypomethylation at AP1 complex component binding sites. Driver-specific histology findings broadly align with those observed in human cancers, such as sessile serrated morphology in BRAFF-V600E mutated tumors. The tumor models with different driver mutations also show variation in hypermethylation of polycomb repressive complex target sites and replicative history. We are using the differences in the DNA methylation profiles between the driver tumor models in conjunction with RNA-seq analysis and conservation of the alterations in human CRC to identify epigenetic events cooperative with the drivers and functionally relevant to tumorigenesis. Citation Format: Manpreet Kalkat, Toshinori Hinoue, Liang Kang, Peter W. Laird, Mary Olesnavich. Inducible mouse models of cancer driver mutations as a powerful tool for in vivo and epigenetic studies of colorectal cancer [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: DNA Methylation, Clonal Hematopoiesis, and Cancer; 2025 Feb 1-4; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2025;85(3 Suppl):Abstract nr B025.