Genome-wide analyses highlights gene interaction with processed meat and vegetable intake for colorectal cancer risk. J. Figueiredo1, L. Hsu2,15, E. White2,15, A. Chan3,16, B. Zanke4, J. Potter2,15, G. Casey1, C. Hutter5, H. Brenner6, B. Caan7, J. Chang-Claude8, S. Gallinger9, R. Hayes10, T. Hudson11, L. Le Marchand12, P. Newcomb2,15, R. Schoen13, M. Slattery14, U. Peters2,15 1) Preventive Medicine, USC, Los Angeles, CA; 2) Public Health Sciences, Fred Hutchinson Cancer Research Center, Seattle, WA; 3) Division of Gastroenterology, Massachusetts General Hospital and Harvard Medical School, Boston, MA; 4) Clinical Epidemiology Program, Ottawa Hospital Research Institute, Toronto, ON; 5) Division of Cancer Control and Population Sciences, National Cancer Institute, National Institutes of Health, Washington DC; 6) Division of Clinical Epidemiology and Aging Research, German Cancer Research Center, Heidelberg, Germany; 7) Division of Research, Kaiser Permanente Medical Care Program, Oakland, CA; 8) Division of Cancer Epidemiology, German Cancer Research Center, Heidelberg, Germany; 9) Department of Surgery, Mount Sinai Hospital, Toronto, ON; 10) Division of Epidemiology, Department of Environmental Medicine, New York University School of Medicine, New York, NY; 11) Department of Medical Biophysics, University of Toronto, Toronto, ON; 12) Epidemiology Program, University of Hawaii Cancer Center, Honolulu, HI; 13) Department of Medicine and Epidemiology, University of Pittsburgh Medical Center, Pittsburgh PA; 14) Department of Internal Medicine, University of Utah Health Sciences Center, Salt Lake City UT; 15) Department of Epidemiology, University of Washington School of Public Health, Seattle WA; 16) Channing Division of Network Medicine, Brigham and Women's Hospital and Harvard Medical School.

   Dietary factors including low consumption of red and processed meat and high intake of fruits, vegetables and fiber are associated with a lower risk of colorectal cancer; however, there is limited information as to whether these dietary factors interact with genetic variants to modify risk of colorectal cancer. To examine this question at a genome-wide level we tested interactions between these dietary factors and about 2.7 million genetic variants on colorectal cancer risk among up to 9,287 cases and 9,117 controls using individual-level data from 10 observational studies. We used logistic regression to detect potential multiplicative gene-diet (GxD) interactions for colorectal cancer as well as our recently developed Cocktail method that involves a screening step based on marginal associations and G-D correlations and a testing step for GxD interactions while correcting for multiple testing using weighted hypothesis testing. Evaluating risks per quartile increment in intake, red and processed meat were associated with statistically significant increased risks of colorectal cancer (OR=1.15, p=1.6E-18 and OR=1.11, p=4.2E-09, respectively) and vegetable, fruit and fiber intake with lower risks (OR=0.93, p=8.2E-05; OR=0.93, p=1.9E-05 and OR=0.91, p=5.6E-05, respectively). From the case-control analysis, we detected a statistically significant interaction between rs4143094 and processed meat consumption (p=8.7E-09). rs4143094 is located 1.42 kb upstream of GATA3 and DNase hypersensitivity marks and histone methylation patterns are consistent with open chromatin and poised/active promoters. Based on our cocktail method, we provided additional support for a statistically significant interaction between variants at the known locus 8q23.3/EIF3H, UTP23 and vegetable intake, which reached genome-wide significance by weighted hypothesis testing (for rs2450114: p-screen=1.4E-08, p-interaction=4.1E-03, p-alpha=5.0E-03). These genetic loci may have interesting biological significance given their location in the genome and further functional analyses are required. Our results identify a novel gene-diet interaction for colorectal cancer, highlighting that diet may modify the effect of genetic variants on diseases risk, which may have important implications for prevention.

You may contact the first author (during and after the meeting) at