Open Access Open Badges Research article

Evaluation of the rheumatoid arthritis susceptibility loci HLA-DRB1, PTPN22, OLIG3/TNFAIP3, STAT4 and TRAF1/C5 in an inception cohort

Ann W Morgan1*, James I Robinson1, Philip G Conaghan1, Stephen G Martin1, Elizabeth MA Hensor1, Michael D Morgan1, Lori Steiner2, Henry A Erlich2, Hock-Chye Gooi3, Anne Barton4, Jane Worthington4, Paul Emery1 and UKRAG Consortium and the YEAR Consortium

Author Affiliations

1 NIHR - Leeds Musculoskeletal Biomedical Research Unit, Leeds Institute of Molecular Medicine, University of Leeds, Chapel Allerton Hospital, Chapeltown Road, Leeds LS7 4SA, UK

2 Roche Molecular Systems, 4300 Hacienda Drive, Pleasanton, CA 94588-8566, USA

3 Clinical Immunology, St James's University Hospital, Beckett Street, Leeds LS97TF, UK

4 arc-Epidemiology Unit, Stopford Building, The University of Manchester, Oxford Road, Manchester M13 9PT, UK

For all author emails, please log on.

Arthritis Research & Therapy 2010, 12:R57  doi:10.1186/ar2969

Published: 30 March 2010



This study investigated five confirmed rheumatoid arthritis (RA) susceptibility genes/loci (HLA-DRB1, PTPN22, STAT4, OLIG3/TNFAIP3 and TRAF1/C5) for association with susceptibility and severity in an inception cohort.


The magnitude of association for each genotype was assessed in 1,046 RA subjects from the Yorkshire Early RA cohort and in 5,968 healthy UK controls. Additional exploratory subanalyses were undertaken in subgroups defined by autoantibody status (rheumatoid factor and anti-cyclic citrullinated peptide) or disease severity (baseline articular erosions, Health Assessment Questionnaire (HAQ) score and swollen joint count (SJC)).


In the total RA inception cohort, the HLA-DRB1 shared epitope (per-allele odds ratio (OR) = 2.1, trend P < 0.0001), PTPN22 (per-allele OR = 1.5, trend P < 0.0001), OLIG3/TNFAIP3 locus (per-allele OR = 1.2, trend P = 0.009) and TRAF1/C5 locus (per-allele OR = 1.1, trend P = 0.04) were associated with RA. The magnitude of association for these loci was increased in those patients who were autoantibody-positive. PTPN22 was associated with autoantibody-negative RA (per-allele OR = 1.3, trend P = 0.04). There was no evidence of association between these five genetic loci and baseline erosions or SJC in the total RA cohort, after adjustment for symptom duration. TRAF1/C5 was significantly associated with baseline HAQ, however, following adjustment for symptom duration (P trend = 0.03).


These findings support the mounting evidence that different genetic loci are associated with autoantibody-positive and autoantibody-negative RA, possibly suggesting that many of the genes identified to date are associated with autoantibody production. Additional studies with a specific focus on autoantibody-negative RA will be needed to identify the genes predisposing to this RA subgroup. The TRAF1/C5 locus in particular warrants further investigation in RA as a potential disease severity locus.