The aim of this study was to assess the correlations of the serum soluble interleukin 2 receptor (sIL-2R) concentrations with disease activity parameters and response to treatment with second line drugs in patients with rheumatoid arthritis (RA). Sixty-seven patients with active disease completed a 24-week, open, randomized study of methotrexate (MTX) versus sulphasalazine (SSZ) or hydroxychloroquine (HCQ). Serum sIL-2R levels were evaluated before entry and after 24 weeks by ELISA. Serum sIL-2R were significantly higher in RA patients than in controls (P = 0.0001) and correlated significantly only with erythrocyte sedimentation rate (P = 0.03) and with Chronic Arthritis Systemic Index (P = 0.01) at study entry. No correlation was found between serum sIL-2R and other laboratory and clinical indices of disease activity. After 24 weeks of treatment no differences in serum sIL-2R in comparison with basal levels were found in either responding or in non-responding patients, although the mean reduction of sIL-2R was more marked in the MTX-treated cohort than in the HCQ and SSZ-treated groups. These data suggest that in RA the measurement of sIL-2R should be used with caution as an isolated index of disease activity and that it is not a useful marker of response to treatment with second line drugs.