Polyunsaturated fatty acids and colorectal tumor risk: a molecular and genetic epidemiology study

PROJECT SUMMARY/ABSTRACT Arachidonic acid, a long-chain ω-6 polyunsaturated fatty acid (PUFA), has been demonstrated to affect carcinogenesis in animal and in vitro studies. The effect of arachidonic acid is believed to be largely due to overproduction of the eicosanoid, prostaglandin E2 (PGE2). The other class of PUFAs, ω-3, also bind to the

same enzymes involved in arachidonic acid metabolism; however, the resulting set of eicosanoids are anti- inflammatory. Thus, ω-3 PUFA metabolism could indirectly inhibit PGE2 production and reduce cancer risk. Over the past few years, multiple genetic variants have been identified to be associated with PUFAs. The goal

of the proposed K99/R00 award is to elucidate the potential causal association between long-chain PUFAs and colorectal tumor risk using Mendelian randomization (MR), an approach that may avoid potential pitfalls of conventional observational epidemiologic research. Using fine-mapping, the proposed study will identify

additional variants in key loci (11q12.2 and 6p24.2) that are involved in the conversion from short- to long- chain PUFAs to improve the PUFA genetic instruments. The proposed study will utilize individual-level data from the ColoRectal Transdisciplinary Study (CORECT) consortium; blood and tissue samples from the

Tennessee Colorectal Polyp Study (TCPS); and genotype and phenotype information from Vanderbilt University's de-identified electronic medical record DNA bio-repository (BioVU). Specifically, we propose the following aims: (1) to conduct a MR study for the association between long-chain PUFAs and colorectal cancer

risk; (2) to conduct fine-mapping to identify additional variants in key PUFA metabolism loci to improve the genetic instruments for MR; (3) to evaluate potential interactions of genetically predicted long-chain PUFAs (using the improved genetic instrument) and use of aspirin and non-steroidal anti-inflammatory drugs (NSAID)

with the risk of colorectal tumors; (4) to investigate associations between genetically predicted long-chain PUFAs and selected tumor biomarkers; and (5) to conduct a mediation analysis to determine whether PGE2 is a mediator on the causal pathway between long-chain PUFAs and colorectal adenoma risk. This innovative

study will be the first to develop an instrumental variable using a polygenic risk score in order to identify potential causal association between long-chain PUFAs and colorectal cancer tumor risk, and will be cost- efficient. The proposed study will help elucidate associations between long-chain PUFAs and colorectal

tumors, which could lead to potential risk reduction strategies. Lastly, the proposed career development award will equip the candidate with the additional didactic and research training necessary for building an independent research program in the areas of nutrition, genetics and molecular epidemiology, and cancer.