HLA-DRB1 allele distribution in polymyalgia rheumatica and giant cell arteritis:: Influence on clinical subgroups and prognosis

Objective: To evaluate HLA-DRB1 associations in patients with polymyalgia rheumatica (PMR) and giant cell arteritis (GCA) in the Spanish population, especially those alleles that include the disease-linked sequence motif DRYF (positions 28 to 31 of the HVR2). Methods: We performed a PCR based HLA-DRB1 genotyping in 89 PMR patients, 44 GCA patients, and 99 unrelated healthy controls from the same geographic area. Results: We did not find any significant difference between the whole group of PMR/GCA patients (n = 133) compared with the healthy controls with the exception of a lower frequency of HLA-DRB1*0405 in the patient group (odds ratio [OR], 0.1 [CI0.02 to 1.2]; P = .04). The distribution of DRB1 alleles was very similar between PMR patients and controls. However, DRB1*0401 (OR, 3.1 [1.1 to 8.6]; P = .02) and DRB1*0404 (OR, 3.5 [0.97 to 12.9]; P = .04) were overrepresented in patients with GCA compared with the control group. DRB1*04 (OR, 1.9 [0.96 to 3.8]; P = .06), especially *0401 (OR, 2.8 [1 to 7.7]; P = .04), and DRB1*07 (OR, 2.3 [1.2 to 4.6]; P = .01) were more frequent in GCA than in PMR. Frequency of the DRYF 28-31 motif was similar among GCA (79.5%), PMR (89.9%), and controls (87.9%) and did not confer any significant risk of the development of systemic vasculitis. We also compared the DRB1 allele distribution in patients with classic PMR (n = 58) and those with an erythrocyte sedimentation rate (ESR) <40 mm/hour (n = 31). Patients with classic PMR expressed DRB1*07 less frequently (OR, 0.4 [0.1 to 1]; P = .04) and had a higher frequency of the DRYF 28-31 motif (94.8% vs 80.6%; P = .03) than patients with ESR <40. Within the GCA group, DRB1 alleles were not predictive for the development of severe ischemic complications. However, the development of relapses in patients with PMR was associated with a higher frequency of DRB1*09 (5.6% vs 0%; P = .04). Conclusions: Our data suggest that the HLA-DRB1 alleles associated with susceptibility for developing PMR and GCA are different. Whether PIVIR with low ESR represents a different clinical subset of the disease should be clarified in a larger sample of patients. HLA-DRB1 genes might predict the presence of relapses in PIVIR, but they do not seem to be indicators of severe disease in GCA patients.