Analysis of 39 Crohnʼs Disease Risk loci in Swedish Inflammatory Bowel Disease Patients

To the Editor: Whether Crohn's disease (CD) and ulcerative colitis (UC) represent distinct disorders or comprise a continuum is a key question in inflammatory bowel disease (IBD). Recently, genome-wide association studies (GWAS) have identified several susceptibility loci, particularly for CD. This progress has enabled attempts to define the molecular similarity of IBD subtypes by specifically testing CD loci in UC.1,2 Here we extend this approach by genotyping a Swedish sample of 736 CD patients (age 49.4 ± 15.9 SD, 50.9% males), 935 UC patients (age 51.6 ± 13.1 SD, 54.2% males) and 1460 common controls (age 49.8 ± 15.6 SD, 35.6% males) for 39 of 40 loci nominally associated with CD in a recent GWAS meta-analysis.3 IBD patients were recruited at the Karolinska University Hospital, Stockholm, and in other participating centers across Sweden. Diagnosis of IBD (CD or UC) was based on standard clinical, endoscopic, radiologic, and histologic criteria. Control individuals were Swedish healthy blood donors, and subjects free of inflammatory disease enrolled in the Swedish EIRA study of rheumatoid arthritis.4 Informed consent was obtained from all participants and local ethics committees approved the study. One single nucleotide polymorphism (SNP) was selected at each CD locus based on previous associations and genotyped with iPlex chemistry (www.sequenom.com). The average genotyping success rate was 96.3%, and no marker deviated significantly (P < 0.001) from Hardy–Weinberg equilibrium in controls. SNPs were tested for association with CD and UC by trend tests implemented in PLINK (pngu.mgh.harvard.edu/≈purcell/plink), with >80% power to detect odds ratios >1.3 for 90% of the markers (risk allele frequency >0.10). A Bonferroni correction for the number of tested loci (39) was applied, which set P < 0.0013 as the level of significance in our experiment (Table 1). Association of CD-loci with Crohn's Disease and Ulcerative Colitis in Sweden Association of CD-loci with Crohn's Disease and Ulcerative Colitis in Sweden With the exception of IL23R and NOD2,5,6 few of the CD loci studied here have been investigated in Swedish CD patients. We replicate associations withstanding correction for multiple testing for SNPs in or near IL23R, IRGM, ZNF365, LRRK2, and C13orf31. Loci showing best nominal significance (P < 0.05) include regions harboring TNFSF18, IL18RAP, JAK2, CUL2, and NKX2-3, whereas other loci consistently replicated across different populations appeared only of marginal (ATG16L1, NOD2, and TNFSF15) or no relevance (C5orf56 and PTPN2) in our sample. In particular, the minor role of NOD2 in Swedish CD has been reported previously, and stems from the low frequency of its risk variants in Scandinavia.5 Three CD loci were associated with UC in our study after correction for multiple testing: those containing IL23R, MST1, and GSDMB. Interestingly, the latter 2 appeared to be relevant primarily to UC in our sample, an observation that diverges from previous reports in UK patients, where MST1 was found to be associated with both CD and UC,7 while the region harboring GSDMB did not show evidence of association with UC.2 One locus, CUL2, showed P-values nearing the Bonferroni significance level, and had similar effects on both CD and UC susceptibility. Finally, although type 2 errors cannot be excluded, another 8 CD loci also recently implicated in UC in different studies2,7,–9 only showed nominal significance (C1orf81, BTNL2, JAK2) or were not associated in our sample (ILRAP18, IL12B, LYRM4, CDKAL1, STAT3), despite Swedish control frequencies similar to those of other European populations.1,–3,7,–9 Our findings are only partially concordant with previous results, and highlight the importance of assessing the relevance of genetic risk variants in different populations, even within Europe. Several GWA screens have been performed, and others are imminent. These rapidly increasing data promise to more conclusively define the genetic architecture of IBD, and the magnitude of the effects that specific risk alleles have in its various forms, and in different populations. Supported by funds from the IRIS Center, the Swedish Research Council, and the Söderbergs Foundation to S.P., and the Swedish Combine program to S.P. and L.P.