Background Advances in imputation methods have enabled association studies of HLA protein amino acid polymorphisms and rheumatoid arthritis (RA) susceptibility. A recent meta-analysis of this region has attributed most of its association with seropositive RA to 5 amino acid positions . Radiological progression in RA has a significant genetic component. Previous studies have linked shared-epitope alleles to radiological damage in RA; the association between HLA amino acid polymorphisms and radiological progression is unexplored.
Objectives To evaluate the relationship between HLA amino acid polymorphisms and radiological progression (defined as the development of new erosions over 2 years) in early, active RA patients enrolled to a clinical trial.
Methods The Combination Anti-Rheumatic Drugs in Early RA (CARDERA) trial randomised 467 patients to receive methotrexate with or without ciclosporin, corticosteroids or both treatments in a factorial-design. We evaluated 421 European ancestry patients with archived DNA (passing QC procedures) and radiological data available. Radiographs were evaluated 6-monthly (for 2 years) for the development of new erosions. Genotyping was performed on the ImmunoChip; HLA alleles and amino acid polymorphisms were imputed using SNP2HLA . We evaluated 237 amino acid positions in 3 MHC class I and 5 MHC class II proteins and 255 HLA alleles for their association with new erosions. Association testing was performed using UNPHASED version 3.0.7, which utilises a standard retrospective likelihood model for a binary trait and allows the use of polymorphic markers . A P-value threshold of 0.0002 was applied based on permutation testing. Ethical approval was obtained; all patients provided consent.
Results One amino acid position, which was position 10 in HLA-DRβ1 (P=0.0001) passed our pre-defined significance threshold for association with new erosions. Using glutamine as the reference amino acid, the OR for new erosions with tyrosine was 0.443 (95% CI 0.278-0.706). A similar trend was observed for HLA alleles; the strongest association was seen with HLA-DRB1*04 (P-value=0.0002), which encodes glutamine at position 10. The absence of *04 alleles provided an OR for new erosions of 0.513 (95% CI 0.360-0.730).
Conclusions We demonstrate an association between amino acid polymorphisms at position 10 in the HLA-DRβ1 protein and the development of radiological erosions in early active RA. Unlike amino acid polymorphisms associating with RA susceptibility (located in binding grooves) this polymorphism influences β- and α-chain binding; it could therefore influence radiological damage through attenuating antigen presentation within the joint. Replication of our finding in other cohorts is required to confirm its validity.
Raychaudhuri S et al. Five amino acids in three HLA proteins explain most of the association between MHC and seropositive rheumatoid arthritis. Nat Genet 2012 29; 44: 291-6.
Jia X et al. Imputing amino acid polymorphisms in human leukocyte antigens. PLoS One 2013; 8: e64683.
Dudbridge F. Likelihood-based association analysis for nuclear families and unrelated subjects with missing genotype data. Hum Hered 2008; 66: 87-98.
Acknowledgements Arthritis Research UK (Grant Reference Number 19739 to ICS); National Institute for Health Research (NIHR) Biomedical Research Centre at Guy's and St Thomas' NHS Foundation Trust in partnership with King's College London.
Disclosure of Interest : None declared