Soluble HLA-DR antigen levels in serum correlate with rheumatoid arthritisdisease activity and the presence of disease-associated epitopes

We investigated correlations between soluble HLA-DR (sHLA-DR) molecules and several clinical, biological and genetic parameters associated with rheuma toid arthritis (RA) disease activity. Serum sHLA-DR concentrations were det ermined in 146 samples from 89 RA patients by an ELISA format, using an ant ibody combination of mouse and rat monoclonal anti-human HLA-DR antibodies. The mean sHLA-DR serum level in RA patients was significantly increased wi th 277+/-19 ng ml compared to 142+/-13 ng ml of 80 healthy controls (P<0.00 1). In ascending order of significance correlations were found between seru m sHLA-DR and EULAR swelling and pain scores. Waaler-Rose RA factor, ESR an d CRP (P=0.025 to P<0.001). High sHLA-DR levels were defined above 374 ng m l that was the 95% confidence interval of the controls. Thirty-seven blood samples (25%) in 31 RA patients were above this level. The EULAR pain and s welling scores erythrocyte sedimentation rate (ESR). C-reactive protein (CR P) and RA factor were higher (P=0.044 to P<0.001) at the moment of high sHL A-DR concentrations compared to the lower concentrations. Higher disease ac tivity was further found in groups of RA patients respectively heterozygous or homozygous for the disease-associated epitope (Q)R KRAA within the HLA- DRB1 chain, compared to the group without this epitope (P<0.017 for part of the results). Likewise sHLA-DR was respectively 169+/-17 (no disease assoc iated epitope). 324+/-34 (heterozygous) and 442+/-69 ng ml (homozygous for the disease associated epitope on HLA-DRB1 alleles) (P<0.017). In conclusio n, this study shows significant correlations between serum sHLA-DR levels a nd RA disease activity parameters as well as increased sHLA-DR in patients with disease-associated epitope on HLA-DRB1 alleles.