Association of giant cell arteritis and polymyalgia rheumatica with different tumor necrosis factor microsatellite polymorphisms

Objective To determine whether giant cell arteritis (GCA) and polymyalgia rheumatica (PMR) are associated with different tumor necrosis factor (TNF) microsatellite polymorphisms. Methods Typing of TNF microsatellite polymorphisms was carried out by molecular‐based techniques on DNA obtained from a population sample of residents from Lugo, northwestern Spain. A case–control approach was used to compare 136 patients with GCA and/or PMR with 147 ethnically matched controls. The association of disease with TNF microsatellite polymorphisms was investigated using chi‐square tests and multivariate logistic regression analyses. Results Different TNF microsatellite associations were found with GCA and PMR. In patients with isolated GCA, the primary association was with TNFa2, which was independent of the GCA associations with HLA–DRB1*0401 and *0101. A negative association was found with TNFa10. In patients with isolated PMR, there was a positive association with TNFb3. This was found to be independent of the HLA–DRB1*13/*14 association in isolated PMR. TNFd4 was negatively associated with isolated PMR. Forward stepwise logistic regression analyses indicated that the strongest association with GCA was provided by the TNFa2 allele, although DRB1*0401 and *0101 were still associated. PMR was primarily associated with TNFb3. A direct comparison of TNF allele frequencies between isolated GCA and isolated PMR indicated that the main difference between these conditions occurred in the frequency of TNFa10. Conclusion GCA and PMR in individuals from northwestern Spain are associated with different TNF microsatellite polymorphisms. The primary TNF associations (TNFa2 and TNFb3) appear to influence susceptibility to these conditions independent of any HLA–DRB1 association.