|Grant Number:||5R03CA171014-02 Interpret this number|
|Primary Investigator:||Newcomb, Polly|
|Organization:||Fred Hutchinson Cancer Research Center|
|Project Title:||Lipid Genotypes, Phenotypes, and Colorectal Adenomas: Elucidating Mechanisms|
DESCRIPTION (provided by applicant): Evidence from epidemiologic studies shows an association between blood lipid concentrations and colorectal adenomas. It has been difficult, however, to determine if this link is indicative of a causal mechanism or is the consequence of shared risk factors for cancer and cardiovascular disease such as diet, obesity, smoking, and physical inactivity. Dyslipidemia is common, affecting up to one-third of the adult US population, and has a strong hereditary component. Some studies have identified associations between genetic loci known to influence lipid levels and the risk of colorectal neoplasia, including PCSK9, ABCG8, and APOE, but these studies have been limited to only a few candidate genes. Many recent genome-wide association studies (GWASs) have been conducted for lipid phenotypes, and nearly 100 loci have been linked to blood lipid levels. Teslovich at al. (2010) conducted a combined analysis of 46 GWASs involving plasma lipid measurements from over 100,000 individuals. Their analysis resulted in a set of 102 single nucleotide polymorphisms (SNPs) related to blood levels of total cholesterol (TC), high- and low-density lipoprotein (HDL, LDL) cholesterol, and triglycerides (TG). Notably, these newly recognized variants affect pathways known to be important for carcinogenesis: inflammation, apoptosis, angiogenesis, cellular signaling, adhesion, migration, hormone synthesis, and growth factor regulation. In the proposed post-GWAS study, we will evaluate the association between adenoma occurrence, lipid phenotypes, and lipid genotypes. Our Mendelian randomization approach, which exploits genetic proxies for observed phenotypes, will help avoid problems with confounding and reverse causation that have hampered previous observational studies of this question. This method has proved to be especially effective for studies of cholesterol and cardiovascular disease. Specifically, our proposed study will: 1) test the association between adenomas and blood concentrations of LDL, HDL, TG, and TC in our study population; 2) determine the association between adenomas and 102 GWAS-identified SNPs related to blood lipids; and 3) evaluate whether these genetic associations with adenoma risk vary according to measured blood lipid concentrations. We will utilize extant biospecimens and questionnaire data from a completed colonoscopy study among enrollees of Group Health (GH), a large healthcare provider in Washington State. Participants underwent colonoscopy for any indication at GH between 1998 and 2007. In total, there were 889 pathologist-confirmed adenoma cases and 1,037 with a polyp-free colonoscopy that serve as controls. New data collection includes: 1) genotyping extracted DNA; 2) measuring lipids from stored plasma or from medical records; and 3) measuring use of cholesterol medication from pharmacy records. Odds ratios for lipid measures will be estimated from logistic regression. Variants will be considered in both single-SNP analyses and as part of genetic risk scores according to mechanistic pathway. Understanding how lipid genes influence adenoma risk may inform mechanisms of carcinogenesis, ultimately helping to reduce the burden of CRC.