P687. Genetic associations with high-grade dysplasia and colorectal cancer in patients with colonic inflammatory bowel disease: preliminary results from ImmunoChip using a targeted analytic approach
M. Waterman1, J. Knight2, J. Stempak3, K. Krishnaprasad4, I. Cleynen5, S. Vermeire6, P. Schumm7, D. McGovern8, S. Brant9, G. Radford-Smith10, M. Silverberg11, 1Rambam Health Care Campus, Department of Gastroenterology, Haifa, Israel, 2Neuroscience Research, Toronto, Canada, 3Mount Sinai Hospital, Toronto, Canada, 4Australia and New Zealand Inflammatory Bowel Disease Consortium, Australia, 5Katholieke Universiteit Leuven, Department of Clinical and Experimental Medicine, Leuven, Belgium, 6Department of Gastroenterology, University hospitals Leuven, Belgium, 7University of Chicago, United States, 8Cedars Sinai Hospital, University of California, Los Angeles, United States, 9The Johns Hopkins Hospital Division of Gastroenterology, the Meyerhoff Inflammatory Bowel Disease Center, Baltimore, United States, 10Inflammatory Bowel Diseases Research Group, Queensland Institute of Medical Research, Brisbane, Australia, 11Mount Sinai Hospital, University of Toronto, Toronto, Canada
Colonic IBD is a significant risk factor for colorectal cancer (CRC) and high-grade dysplasia (HGD). Extensive involvement of the colon, the co-existence of primary sclerosing cholangitis (PSC), disease activity, family history of CRC and disease duration have been shown to increase the risk for CRC. However, no specific genetic association has been repeatedly shown to be associated with CRC in patients with colonic IBD. We aimed to identify genetic associations with CRC/HGD in patients with colonic IBD using data from the ImmunoChip in a large multi-national cohort of patients with colonic IBD.
Members of the International IBD Genetic Consortium (IIBDGC) were asked to identify patients with colonic IBD who developed CRC or HGD, verified by pathology. Demographic and clinical data were also collected. For each HGD/CRC case 1–2 controls were matched (by IBD subtype, disease duration, endoscopic extent (Montreal Classification), ethnicity, co-existence of PSC and gender). Phenotypic variables were compared between CRC/HGD cases and controls using chi square for categorical variables and student's t-test for continuous variables. Preliminary analysis of genotypes generated from available iChip data was undertaken using logistic regression in PLINK looking specifically at recently-reported IBD associations, PSC associations and SNPs in CRC-associated genes. p-value ≤5×10−4 was considered significant.
Overall, 585 colonic IBD cases (390 UC, 171 CD, 19 IBDU) were reported, 274 had either HGD or CRC and 311 were matched controls. There were 58% males, 95% Caucasian, and 12% current smokers. The mean age at diagnosis was 33. Disease duration to CRC/HGD diagnosis was significantly longer than length of F/U in controls (230 months vs. 180 months, respectively, p = 0.005). There was no significant difference in proportions of patients with extensive disease between the groups. However, PSC was more common in cases vs. controls (11.3% vs. 2.9%, respectively, p = 0.044). Genetic data from iChip were available on 90 cases, 222 controls (53.3% of the total cohort; 50% males). Only Caucasians were selected for analysis. Six covariates were included to account for population stratification and one to control for disease duration. There were no significant genetic associations.
In this initial analysis, no genetic associations could be found related to risk of CRC/HGD in IBD. The IIBDGC will continue to gather cases to increase the sample size.