Abstract
OBJECTIVE: Refractory celiac disease-type II (RCDII) is the more severe and adverse form of celiac disease; however, its association with other autoimmune diseases remains unclear. We conducted a phenome-wide association study (PheWAS) to examine the association between the polygenic risk score (PRS) for RCDII and autoimmune diseases. METHODS: To construct the PRS-RCDII, we extracted summary statistics for three non-human leukocyte antigen genetic variants, which were independently associated with RCDII ( r2 < 0.001; P < 5 × 10 -5 ) in a genome-wide association study. We then conducted a PRS-PheWAS in the UK Biobank to investigate the associations of PRS-RCDII with 27 autoimmune diseases, adjusting for age, sex, genetic batch, and genetic ancestry. False discovery rate (FDR < 0.05) correction was applied to account for multiple comparisons. RESULTS: Our study population comprised 373 022 UK Biobank participants (mean age: 57.2 years), of whom 202 865 (54.4%) were females. We constructed the PRS-RCDII, using three genetic variants, namely rs2041570 on chromosome 7p14.3 ( FAM188B ), rs7324708 on chromosome 13q22.1 ( KLF12 ), and rs205047 on chromosome 17p12 ( SHISA6 ). In the PRS-PheWAS, two phenotypes were initially associated with RCDII at a nominal P value threshold, ankylosing spondylitis and systemic sclerosis; however, after adjusting for multiple comparisons, only the association with ankylosing spondylitis remained statistically significant (odds ratio per 1 SD increase = 1.13; 95% confidence interval: 1.04-1.22; PFDR = 0.023). Sex-stratified and single-nucleotide polymorphism (SNP)-by-SNP analyses revealed no significant heterogeneity. CONCLUSION: Our study identified an association between the genetic risk score for RCDII and ankylosing spondylitis, but not with other autoimmune diseases. This finding may have clinical importance for people with RCDII, although replication in future studies is needed.