Arthritis & Rheumatism, Volume 63,
November 2011 Abstract Supplement

Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Chicago, Illinois November 4-9, 2011.

The Hierarchy of HLA-DRB1 Shared Epitope Risk Genotypes Associated with Rheumatoid Arthritis in African-Americans Differs From That in Europeans.

Ahmed1,  Altan F., Hughes1,  Laura B., Conn2,  Doyt L., Jonas3,  Beth L., Callahan4,  Leigh F., Smith5,  Edwin A., Brasington6,  Richard D.

University of Alabama at Birmingham, Birmingham, AL
Emory Univ School of Medicine, Atlanta, GA
University of North Carolina at Chapel Hill, Chapel Hill, NC
University of North Carolina, Chapel Hill, NC
Med Univ of South Carolina, Charleston, SC
Washington Univ School of Med, St. Louis, MO
University of Pittsburgh, Pittsburgh, PA


HLA-DRB1 alleles encoding the shared epitope (SE) are associated with RA in persons of European ancestry (~60–70% of RA vs ~40% of controls have >=1 SE allele). SE-containing alleles are also associated with RA in African-Americans (Af-Amer) but have a lower allele frequency than in Europeans (~43% of RA vs ~25% of controls). Several SE classification systems have been developed to define the relative contributions of individual SE alleles/combinations on susceptibility to RA, but these have not been tested in Af-Amer because of insufficient numbers of subjects. Here we evaluate the du Montcel classification system in 920 Af-Amer RA pts and 1,006 Af-Amer controls.


High-resolution HLA-DRB1 genotyping was performed in patients and controls from the CLEAR Registry and from local Birmingham controls. HLA-DRB1 susceptibility (S) alleles are those with the amino acid sequence RAA at positions 72–74 (Table). S alleles are subdivided according to the amino acid at position 71 (S1: ARAA [S1A] or ERAA [S1E], S2: KRAA, S3: RRAA). The S3 alleles are subdivided further according to the amino acid at position 70 (S3D: DRRAA, S3P: QRRAA or RRRAA). S1A, S1E, and S3D are classified as low-risk (L) alleles. All other alleles are non-susceptibility alleles. To estimate the odds ratio of association of each genotype with RA, we fit log-linear models with case/control as the response, also stratified by autoantibody positivity, with levels of genotype as predictors.

HLA-DRB1Amino Acid Residue 
0401, 1303 KRAAS2
0101, 0102, 0404, 0405, 0408, 10, 1402, 1406Q/RRRAAS3P
1501, 1502, 1503 ARAAL (S1A)
0103, 0402, 1301, 1302 ERAAL (S1E)
1102, 1103, 1202, 1305, 16DRRAAL (S3D)


The genotype combination with the largest effect for Af-Amer RA was S2/S3P (OR 9.1, CI 3.4–24, p=5.7 × 10-6) followed by S3P/S3P (OR 3.2, CI 1.5–6.7, p=1.9 × 10-3), S2/L (OR 2.6, CI 1.9–3.6, p=2.7 × 10-9), S2/S2 (OR 2.3, CI 0.62–8.3, p=.20), S3P/L (OR 1.9, CI 1.5–2.4, p=1.8 × 10-8). When analyzing autoantibody positive patients only, the risk hierarchy was nearly identical, and the overall effect was slightly larger. In contrast, when stratifying by autoantibody negative patients, there was no association of genotype with RA (c2=1.9, DF=5, p=0.86).


Among British RA the S2/S2 genotype provided the strongest association with RA, but was not associated with RA in Af-Amer. Therefore, the HLA-DRB1 risk hierarchy differs between Af-Amer and Caucasians. The much lower risk of the S2/S2 genotype is likely due to its low frequency (0.56%, both cases and controls) in Af-Amer. The lack of association among autoantibody negative patients further substantiates the role of the HLA in the pathogenesis of autoantibody positive RA. The results illustrate the importance of considering ethnic differences, which may include heterogeneous genetic admixture and/or allele frequency differences, when defining genetic susceptibility loci for RA and other complex diseases.

To cite this abstract, please use the following information:
Ahmed, Altan F., Hughes, Laura B., Conn, Doyt L., Jonas, Beth L., Callahan, Leigh F., Smith, Edwin A., et al; The Hierarchy of HLA-DRB1 Shared Epitope Risk Genotypes Associated with Rheumatoid Arthritis in African-Americans Differs From That in Europeans. [abstract]. Arthritis Rheum 2011;63 Suppl 10 :2157

Abstract Supplement

Meeting Menu