GIANT-CELL ARTERITIS AND POLYMYALGIA-RHEUMATICA CAN BE DIFFERENTIATEDBY DISTINCT PATTERNS OF HLA CLASS-II ASSOCIATION

Objective. To determine whether patients with polymyalgia rheumatica ( PMR) and giant cell arteritis (GCA) exhibit identical HLA class II ass ociations. Methods. A case-control association study was performed on a population sample from Lugo, in Northwestern Spain. DNA samples were available for 128 patients and 145 ethnically matched controls. Withi n the patient group 26 exhibited both PMR and GCA, 75 PMR alone, and 2 7 GCA alone. HLA-DRB1, DQA1, and DQB1 phenotypes were defined by molec ular based techniques. Results. KLA-DRB10401 was associated with GCA regardless of PMR status, although this only reached statistical signi ficance in the total GCA group. This was also seen for DRB10101, *010 2, although the association was less strong. Patients with PMR without GCA were not associated with DRB10401 or *0101, *0102, but exhibited a significant association with DRB113, *14. Nonsignificant increases in DQA1 and DQB1 phenotype frequencies appeared to reflect known patt erns of linkage disequilibrium with the HLA-DRB1 alleles associated wi th GCA and PMR groups. An association was observed between the presenc e of the RA DRB1 shared epitope (SE) and GCA but not with PMR in the a bsence of GCA. This association was primarily accounted,for by the pre sence of a single copy of the SE, and homozygosity for the SE did not confer additional risk. A high frequency of SE-bearing DRB1 alleles wa s observed in patients with GCA with jaw claudication or visual manife stations, although the sample size of these subgroups was small. Concl usion. PMR and GCA in a Northwestern Spanish population have distinct HLA class II associations. HLA is unlikely to account for the observed high level of overlap in these patients, and other etiological factor s may be involved.