PWE-031 No association between HLA-DQA1*05 and -DQB1*02 gene dosage and clinical phenotype of coeliac disease

Introduction Coeliac disease is a common gluten-sensitive enteropathy. Disease susceptibility is strongly associated with specific Human Leukocyte Antigen (HLA)-DQA1 and -DQB1 loci. Individuals with the HLA-DQA1*05:HLA-DQB1*02 heterodimer (referred to as HLA-DQ2.5) have the highest risk of developing coeliac disease, particularly if two copies of HLA-DQB1*02 are present. Understanding whether differences in the frequency of DQA1*05 and DQB1*02 accounts for differences in the clinical phenotype of coeliac disease is important for the development of personalised medicine for patients and was addressed in the following study. Methods Demographic, clinical and laboratory data was retrospectively collected from adult patients attending the specialist coeliac disease clinic, Royal Hallamshire Hospital between 2008 to 2016 and correlated with the number of DQA1*05:DQB1*02 combinations forming the DQ2.5 heterodimer (DQ2.5 gene dose), as well as the frequency of DQB1*02. Results Four hundred and ninety patients had biopsy-proven coeliac disease and HLA genotype information. Individuals who were positive for the DQ2.5 heterodimer were more likely to be EMA positive than those who were DQ2.5 negative (p=0.0132). There were no linear associations between the DQ2.5 gene dosage and clinical or laboratory parameters assessed. 190/490 (39%) patients carried two copies of DQB1*02, 278/490 (57%) patients had a single copy and DQB1*02 was absent in 22/490 (4%) patients. The prevalence of folate deficiency was higher and mean haemoglobin levels lower, in individuals carrying two copies of DQB1*02 than those with a single copy of DQB1*02 (p 0.05). The carriage of two copies of DQB1*02 did not correlate with any other parameters. Conclusions The presence of DQA1*05 and DQB1*02 is well documented to correlate with risk of developing coeliac disease. Our results suggest that there is no association between homozygosity/heterozygosity for DQA1*05 and DQB1*02 and the clinical phenotype of active disease. These results provide an important insight into the interpretation of HLA-DQ data in the setting of coeliac disease.