DOP011 Genotype–phenotype association analysis of inflammatory bowel disease in 6395 East Asian subjects: results from the international IBD Genetics Consortium. (16th January 2018)

• Record Type: Journal Article
• Authors: Abedian, S; Wong, S H; Van Sommeren, S; Takahashi, A; Cheon, J H; Kim, W H; Yamazaki, K; Yang, S -K; Kubo, M; Weersma, R K; Alizadeh, B Z; Ng, S C
• Is Part Of: Journal of Crohn's and colitis. Volume 12:Number 1(2018:Jan.)Supplement 1
• Page Start: S037
• Page End: S037
• Publication Date: 2018-01-16
• Subjects: Inflammatory bowel diseases -- Periodicals; 616.344005
• DOI: 10.1093/ecco-jcc/jjx180.048
• Languages: English
• ISSNs: 1873-9946

Abstract

Abstract: Background: Inflammatory bowel disease (IBD), including Crohn's disease (CD) and ulcerative colitis (UC), is a debilitating disease with rising incidence in Asia. Recent European genome-wide and Immunochip genotype data showed strong association of disease sub phenotypes with genetic risk score (GRS) from known IBD risk alleles. Here, we undertook a cross-population genotype-phenotype association study to further understand whether GRS of IBD, CD, and UC could explain the proportion of the variance, and if they are predictive of the corresponding phenotypes in three East Asian (EA) populations (Japan, Korea, and China). Methods: This study included 6395 subjects, consisting of 2764 IBD patients (1668 CD; 1096 UC), and 3631 country-, age- and gender-matched controls, genotyped on the Immunochip array. We shuffled the three populations in three settings: first we combined two populations to calculate odds ratio (OR) for 176 shared known genome-wide significant IBD-associated variants, then we built IBD GRS for the third target population. We generated a multi-locus GRS for each studied population to summarise the total load of genetic risk for the phenotype of interest. Risk scores were analysed to compare IBD, CD or UC vs. controls. Subsequently, we estimated explained variance and tested predictability of generated GRS for IBD, CD and UC independently for each population. Results: Demographic features of the study population showed that CD patients were more likely to be younger at diagnosis, male, smokers and had an affected family member with IBD compared with UC patients. GRS could significantly explain up to 4.4%, 1.51%, and 1.34% variance of IBD, but given a prevalence of 0.08%, 0.04%, and 0.009% for IBD in Japan, Korea, and China, respectively, it yields to a negligible predictive (posterior) probability of 8.8 × 10 –4, 4.76 × 10 –4, and 1.05 × 10 –4 in three populations for IBD. CD-GRS and UC-GRS could significantly explain CD and UC to a lesser extent compared with IBD and again given a lower prevalence of CD (up to 0.02%) and UC (up to 0.06%) in three countries, they yield to a negligible predictive probability (Table). Conclusions: This study shows that genetic findings based on transethnic analyses are applicable across the EA populations. Our findings suggest that the association of GRS which was built upon combining the effect of genome-wide associated risk alleles, is unlikely to provide a strong predictive probability of IBD, CD, and UC in EA. Using only GRS alone may not predict the clinical characteristics in EA populations.