Arthritis & Rheumatism, Volume 62,
November 2010 Abstract Supplement

Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Atlanta, Georgia November 6-11, 2010.

The Associations between RA Genetic Risk Alleles and Seropositive but Non-Erosive Rheumatoid Arthritis.

P. Liao,  Katherine, Cui,  Jing, E. Weinblatt,  Michael, Iannaccone,  Christine, Shadick,  Nancy, M. Plenge,  Robert, H. Solomon,  Daniel


Presently, little is known about the subgroup of subjects with seropositive rheumatoid arthritis (RA) who never develop bone erosions in contrast to those with erosions that remain stable or progress despite treatment. Our objective was to determine whether specific RA genetic risk alleles were associated with remaining erosion-free in a longitudinal RA cohort.


Our study was conducted in a prospective observational cohort of >1100 RA patients recruited from the outpatient practice of an academic medical center. We included patients with bilateral hand radiographs obtained at baseline and at 2 yr follow-up with formally assessed Sharp scores. The primary outcome was erosion-free status at recruitment and at 2 yr follow-up. We genotyped 31 validated RA risk alleles in a seropositive (RF+ or anti-CCP+) subset of this cohort.

To assess whether individual risk alleles for seropositive RA were also associated with erosion-free status, we tested each polymorphism using a log additive model (0, 1, 2) in a logistic regression adjusted for significant clinical predictors of erosion-free status. The significant clinical predictors were determined in a previous study: younger age at RA onset, male gender, and shorter RA duration. We quantified the association between risk alleles and erosion-free status using odds ratios (OR), 95% confidence intervals (CI) and p-values from the logistic regression models.


Summary Three hundred-three RA subjects in the cohort had Sharp scores and genotype information; 18% were erosion-free at recruitment and 2 year follow-up: 75% female, mean age at RA onset of 44.3 yrs (SD 10.1) and mean RA duration of 8 yrs (SD 8); 94.6% were anti-CCP+, 83.6% were RF+, and 76.6% had >= 1 copy of the HLA-shared epitope (HLA-SE). In the erosive group, 84% were female, with mean age at RA onset of 41.8 yrs (SD 14) and mean RA duration of 18 yrs (SD 12.3). There was no significant difference between anti-CCP status, methotrexate and anti-TNF use at baseline between the erosion-free group and those with erosions.

Two alleles were independently associated with erosion-free status after adjusting for clinical factors (age at RA onset, gender, RA duration): REL (rs13031237, T allele) with erosion-free status and KIF5A (rs775322, T allele) with not remaining erosion-free (Table 1). Neither allele reached statistical significance after Bonferroni correction. The HLA-SE was not associated with erosion-free status in this seropositive cohort.

Table 1. Association between REL and KIF5A with erosion-free status in seropositive RA adjusted for clinical predictors of erosion-free status.

VariablesClinical model*+RELClinical model*+KIF5A
 OR95% CIp-valueOR95% CIp-value
Age at RA onset (every 5 yrs)0.770.66, 0.900.00010.780.67, 0.910.0013
Male gender2.871.25, 6.580.0132.811.23, 6.420.015
RA duration (yrs)0.860.82, 0.91<0.00010.870.83, 0.91<0.0001
Risk allele1.951.19, 3.170.00760.540.33, 0.880.013
*Clinical model includes age at onset, gender, and RA duration.


REL which encodes a component of the osteoclastogenesis pathway may predict erosion-free status in seropositive RA beyond significant clinical predictors. KIF5A, whose role in RA is unclear, was associated with erosive disease in this small study. Larger studies in independent cohorts are needed to confirm these associations.

To cite this abstract, please use the following information:
P. Liao, Katherine, Cui, Jing, E. Weinblatt, Michael, Iannaccone, Christine, Shadick, Nancy, M. Plenge, Robert, et al; The Associations between RA Genetic Risk Alleles and Seropositive but Non-Erosive Rheumatoid Arthritis. [abstract]. Arthritis Rheum 2010;62 Suppl 10 :1610
DOI: 10.1002/art.29376

Abstract Supplement

Meeting Menu