Article Text

OP0190 Personalized Genetic Medicine: Amino Acid Positions 11, 71 and 74 in HLA-DRB1 PREDICT Disease Severity, Treatment Response and Mortality in Rheumatoid Arthritis; Multi-Centre Prospective Cohort Studies
  1. S. Viatte1,
  2. D. Plant1,
  3. B. Han2,
  4. B. Fu1,
  5. A. Yarwood1,
  6. W. Thomson1,
  7. D.P. Symmons3,
  8. J. Worthington1,
  9. A. Young4,
  10. K.L. Hyrich3,
  11. A.W. Morgan5,
  12. A.G. Wilson6,
  13. J.D. Isaacs7,
  14. S. Raychaudhuri2,
  15. A. Barton1
  16. on behalf of Biologics in Rheumatoid Arthritis Genetics and Genomics Study Syndicate BRAGGSS
  1. 1Arthritis Research UK Centre for Genetics and Genomics, Manchester, United Kingdom
  2. 2Broad Institute of MIT and Harvard, Cambridge, United States
  3. 3Arthritis Research Uk Epidemiology Unit, Manchester
  4. 4St. Albans City Hospital, St. Albans
  5. 5Leeds Institute of Rheumatic and Musculoskeletal Medicine, Leeds, United Kingdom
  6. 6Conway Institute of Biomolecular & Biomedical Research, Dublin, Ireland
  7. 7Musculoskeletal Research Group, Newcastle upon Tyne, United Kingdom


Background The shared epitope (SE) is a group of alleles of the HLA DRB1 gene and was thought to have the strongest effect on rheumatoid arthritis (RA) susceptibility. However, recently, HLA-DRB1 position 11,outside of the classical SE, has been shown to be a stronger predictor of RA susceptibility. Positions 11,71 and 74 define 16 haplotypes, the effect of which ranges from risk to protective on RA susceptibility. Their effect on RA severity, treatment response or mortality in patients has not previously been studied.

Objectives To assess whether HLA-DRB1 positions 11,71,74 can also be used to predict radiological outcome, anti-TNF response and mortality in patients with RA.

Methods We used 3 independent prospective cohort studies: the Norfolk Arthritis Register-NOAR (1691 patients with 2811 x-rays); the Early Rheumatoid Arthritis Study-ERAS (421 patients with 3758 x-rays); a cohort from 57 UK centres-BRAGGSS (1846 patients with treatment response). HLA typing was determined using a reverse dot-blot method or dense genotyping of the HLA region by the ImmunoChip array, followed by imputation.Longitudinal modelling of the presence of erosions was performed with generalized estimating equation (GEE) models whilst the Larsen score was modelled with Generalized Linear Latent and Mixed Modelling (GLLAMM). Change in Disease Activity Score28 was modelled with linear regression and EULAR response with ordinal logistic regression. Cox proportional hazard models were used for all cause and cardiovascular mortality studies.

Results Valine at position 11 of HLA-DRB1 (Val11) is a new and the strongest independent genetic determinant of radiological damage in RA and this finding has been replicated in separate cohorts (OR in NOAR: 1.75, 95%CI 1.52 to 2.01, p=8.7E-15). Positions 71 and 74 represent independent predictors and the 3 positions define 16 haplotypes strongly associated with disease outcome (multivariate p=2.83E-12), superseding the SE. The hierarchy, ranging from risk to protective effects, is perfectly correlated with that observed for disease susceptibility. HLA-DRB1 haplotypes associated with RA susceptibility and severe outcome are also predictors of good treatment response with anti-TNF therapy. For example, the Val11Lys71Ala74-haplotype, carried by 52% of patients, is associated with good EULAR response (OR: 1.24 95%CI 1.07 to 1.44 p=5.31E-03). On average, 17 patients need to be treated with anti-TNF to see one more patient responding better, based solely on the carriage of this haplotype. Both all-cause and cardiovascular mortality are also predicted by the 16 haplotypes.

Conclusions Combinations of amino-acids at positions 11, 71 and 74 of the HLA-DRB1 gene predict severe disease, treatment response and mortality in RA, superseding the classical SE. At disease onset, this allows stratification of patients to identify those at risk of joint damage and early death but who are more likely to respond to anti-TNF biologic therapy.

Disclosure of Interest None declared

DOI 10.1136/annrheumdis-2014-eular.3155

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.