LAUSR

Background Genetic factors play a prominent role in AS pathogenesis. So far over 40 non-MHC Ankylosing Spondylitis(AS) susceptibility loci with genome-wide or suggestive significance have been initially reported in Caucasians, however, lack of association evidence of most loci was seen in Chinese Han and some results seemed controversial. Objectives Here, we present a systematic evaluation of 47 non-MHC AS susceptibility loci using GWAS datasets in Chinese Han. Methods Totally 1853 AS cases and 4048 newly matched controls in 4 cohorts were obtained, after imputation meta-analysis results of 93 589 variants within 47 reported loci were extracted. Best-guess genotype data were used for interaction analysis and weighted genetic risk score model construction which was then assessed by receiver operator characteristic analysis. Functional annotation was conducted using HaploReg, RegulomeDB and rVarBase Database. Results We revealed 14 AS-associated variants with nominal evidence in Chinese Han, including rs10865331(p=2.96E-9), rs10050860 (p=1.84E-4) and rs8070463(p=2.81E-4) and found potential associated variants within these loci. We then extracted variants in ERAP1 as well as HLA-B27 tag snp rs13202464 for HLA-B27-ERAP1 interaction analysis (figure 1). Epistatic association between ERAP1 (rs30187, rs10045403) and HLA-B27 (rs13202464) was confirmed. Among those 14 variants, rs30187 showed weaker risk effect in Chinese while rs10050860 and rs12504282 seemed to attribute more risk (Table 1). Genetic prediction model combining 14 variants in 11 loci with HLA-B27 achieved better discrimination ability(AUC=0.884, 95%CI=0.873~0.895) than HLA-B27 alone(p=2.17E-6). We also identified some likely functional variants at these loci. Abstract THU0022 – Figure 1Abstract THU0022 – Table 1 Comparison of 14 shared associated non-MHC SNPs across European and Chinese Locus SNP Nearby Gene(s) Risk allele RAF% (CEU/CHN) OR(CEU/CHN) PARP%(CEU/CHN) 1 p36 rs6600247 RUNX3 C 0.50/0.69 1.16/1.15 7.41%/9.42% 2 p15 rs10865331 Intergenic A 0.38/0.49 1.34/1.27 11.44%/11.60% 3 p24 rs10510607 CMC1 C 0.83/0.54 1.15/1.14 11.07%/7.03% 4q21 rs12504282 ANTXR2 T 0.54/0.91 1.14/1.20 7.03%/15.38% 5 p13 rs11742270 IL7R G 0.73/0.84 1.11/1.14 7.43%/10.53% 5q15 rs30187 ERAP1 T 0.34/0.53 1.32/1.11 9.81%/5.47% 5q15 rs10045403 ERAP1 A 0.73/0.82 1.20/1.18 12.74%/12.92% 5q15 rs10050860 ERAP1 C 0.78/0.95 1.18/1.45 12.31%/30.01% 6q15 rs639575 BACH2 T 0.61/0.49 1.08/1.10 4.65%/4.69% 14q13 rs8006884 PPP2R3C C 0.35/0.40 1.11/1.09 3.71%/3.44% 17q11 rs2297518 NOS2 A 0.19/0.16 1.13/1.11 2.41%/1.72% 17q21 rs9901869 NPEPPS A 0.52/0.62 1.15/1.14 7.24%/7.95% 17q21 rs8070463 TBKBP1 C 0.51/0.44 1.14/1.16 6.66%/6.64% 21q22 rs2836883 Intergenic G 0.74/0.83 1.19/1.16 12.33%/11.71% Conclusions Our results provided a detailed spectrum of non-MHC AS susceptibility loci in Chinese Han and highlighted 2 p15, ERAP1 and TBKBP1 may play a critical role in AS pathogenesis. References [1] IGAS, Cortes A, Hadler J, et al. Nature genetics. 2013Jul; 45(7):730–738. [2] Ellinghaus D, Jostins L, Spain SL, et al. Nature genetics. 2016May; 48(5):510–518. Disclosure of Interest None declared