Expression of QK/QR/RRRAA or DERAA motifs at the third hypervariable region of HLA-DRB1 and disease severity in rheumatoid arthritis.

OBJECTIVE: To examine the relationship between disease severity in patients with confirmed rheumatoid arthritis (RA) and the carriage of alleles expressing the high risk epitope (HRE) QK/QR/RRRAA or the low risk epitope (LRE) DERAA at positions 70-74 of the third hypervariable region of HLA-DRB1.
METHODS: A case-control design to compare allele carriage rates in 204 Caucasian subjects with severe RA and mild RA and healthy controls. Patients had a mean disease duration of 12-18 years and severity of RA was defined using clinical and therapeutic criteria. Molecular typing at the HLA-DRB1 locus was performed using a polymerase chain reaction method.
RESULTS: Eighty-seven percent of patients (52/60) with severe RA had one or more of the alleles bearing the QK/QR/RRRAA motif or HRE, compared with 54% (21/39) with mild RA (OR 5.57, p = 0.0007) and 39% (41/105) of controls (OR 10.15, p < 0.0001). Twenty-five percent of patients (15/60) with severe disease expressed 2 disease associated HRE DRB1 alleles, compared with 13% of patients (5/39) with mild disease (OR 2.3, p = NS) and 5% (5/105) of controls (OR 6.67, p = 0.0003). In contrast, only 5% of patients (3/60) with severe RA expressed one of the LRE alleles that carry the DERAA motif at positions 70-74, compared with 31% of patients (12/39) with mild RA (OR 0.12, p = 0.0013) and 22% of controls (23/105) (OR 0.19, p = 0.0082). No patient or control was homozygous for LRE alleles. Eighty-three percent (50/60) of patients with severe RA expressed the HRE without the LRE, compared with 44% (17/39) of those with mild disease (OR 6.47, p < 0.0001) and 35% (37/105) of controls (OR 9.19, p < 0.0001). In contrast, only one patient (2%) with severe disease expressed the LRE without the HRE, compared with 20% (8/39) of those with mild disease (OR 0.07, p = 0.0047) and 16% (17/105) of controls (OR 0.09, p = 0.009). There was no significant difference between the 3 groups in the frequency of patients who expressed both or neither epitope. Logistic regression showed that age at disease onset (p = 0.0009), duration of disease (p = 0.007), positive rheumatoid factor status (p = 0.003), and presence of the HRE or LRE (p = 0.00005) were significantly associated with the presence of severe disease.
CONCLUSION: HLA-DRB1 alleles appear to confer an important bidirectional influence on the risk of disease severity in RA, with 20-fold difference in OR between those associated with the highest (HLA-DRB1*0401) and lowest (HLA-DRB1*1301/02) risk. The HRE and LRE exhibit diametrically opposed effects, which may be mutually antagonistic. These data support a multistep pathogenesis in which MHC class II genes are one component of a coordinate genetic and environmental interaction leading to immunological injury and joint destruction.