BCCMA: VA Colorectal Cancer Clinical & Computational Collaborative (VA-5C)

About This Grant

Objectives: This BLRD-CSRD-Collaborative MERIT Award (BCCMA) application is comprised of three projects unified by their focus on CRC precursors in Veterans. We call our team-science consortium the “VA5C,” for VA Colorectal Cancer Clinical & Computational Collaborative. Our multidisciplinary collaborative is linked by a common focus on CRC precursors as targets for intervention. With a lifetime development risk of 5%, colorectal cancer (CRC) is the third-most common cancer and the second major cause of cancer-related deaths in the US and within the VA Healthcare System. Specifically, this proposal will be part of a cluster of merit review projects that include the San Diego VA and Boston VA that have a track record on ongoing research collaborations aimed at advancing our understanding of colorectal neoplasia. Of particular interest are two distinct phenotypes involving serrated adenomas and Veterans with multiple adenomatous polyps. The role of the gastrointestinal microbiome has been identified as a potential area of significant interest. Furthermore, the role of diet and microbiome in colon neoplasia have been supported by epidemiological studies. It is hypothesized that genotoxins from ingested food are linked to the development of malignant transformation as well as specific microorganisms such as Fusobacterium. In other studies, cultured human colon cancer has been shown in the APC/min model to promote colorectal cancer. In this model there is an increase in mucosal E. coli that may promote colon cancer by regulation of intestinal cytokines. A colonic association of both types of bacteria is linked to the development of more advanced stages of colon cancer. We therefore hypothesize that the intestinal microbiome, through its effects on the colonic mucosa and the release of bioactive microbial products, will be associated with clinical/histologic phenotypes and molecular subclasses defined by transcriptional profiles in Veterans with colorectal adenomas compared to normal colonic mucosa. We hypothesize that multiple, lower penetrance genes interacting with non-genetic, environmental factors contribute to colonic adenomas in Veterans. To study this in aim 1 we will utilize our existing clinical infrastructure to develop a biobank of colorectal polyps and fecal samples that will enable to investigate the association of tumor-associated fecal microbiota with colon polyp phenotype using transcriptomics. In the second aim, we will examine through genome wide association studies the relationship between the observed transcriptomics data in Aim 1 and genetic variants in over 400,00 Veterans with colon polyps that are part of the Million Veteran Program (MVP). Of particular interest are the examination of patients with multiple adenomatous polyps. In the third aim¸ we will explore the colonization of germ-free APC/min mice with human donor stool to evaluate whether microbiota associated with more advanced colon CA accelerate tumor development. These studies will be performed in Veterans who have developed colon polyps using biorepositories established from tissue and donor stool specimens from the collaborating research site. Results from these studies will better inform investigators about the role of the host-microbiome interactions that may predispose to the development of colon adenomas.