To investigate the relationship between radiographic disease progression in the presence or absence of rheumatoid arthritis (RA) linked HLA-DRB1 alleles after early introduction of disease modifying antirheumatic drug therapy in patients with RA over a study period of 6 years.One hundred nine patients of a trial comparing intramuscular (im) gold sodium thiomalate (GSTM) and im methotrexate (MTX) in early erosive RA were followed for 6 years with regular assessments of clinical and laboratory data and yearly radiographs of hands and feet, and they were typed for HLA-DRB1 genes. Radiographic progression was analyzed for an influence of rheumatoid factor (RF) status and HLA-DRB1 genes.Twenty-seven patients (25%) were positive for two, 46 (42%) for one, and 36 (33%) for none of the disease linked alleles. A decrease of the rate of radiographic disease progression with treatment in this group of patients was reflected by the decline in the slope of the radiographic score. Seropositive patients (n = 71, 68%) had a significantly more destructive disease course than RF negative patients. In seropositive disease, patients with a "double dose" of RA linked alleles showed a tendency to greater progression during the first 12-24 mo of treatment, but no significant difference in the longterm radiographic outcome could be detected between subgroups defined by the presence or absence of HLA-DRB1 genes. There was no significant difference throughout the study period with respect to the clinical disease course as assessed by joint swelling, C-reactive protein, and erythrocyte sedimentation rate. The majority of the seronegative population (n = 38, 32%) had a benign disease course with the exception of patients (n = 6) with the double allele; they had radiographic disease progression comparable with the seropositive patients.Our data do not provide evidence for a more aggressive disease course in patients bearing double RA linked HLA-DRB1 alleles.