The validity of serum sICAM-I levels to assess Multiple Sclerosis (MS) activity was evaluated in 49 untreated definite relapsing-remitting (RR) patients. sICAM-I levels were significantly (P = 0.0009) higher in the 'clinically active' group (No 22) than in the 'clinically inactive' (No 27), whereas no different values were found between patients with Gd-enhancing lesions at MRI (Gd-positive) (No 32) and patients without such lesions (Gd-negative) (No 17) independently of their clinical activity. Among the 'clinically active' MS, the Gd-positive (No 16) subgroup showed significant (P < 0.05) lower sICAM-I levels when compared to the Gd-negative (No 6) subgroup, but higher (P = 0.009) than those of the 'clinically inactive Gd-positive' (No 16) patients. The sICAM-I levels did not differ between the two 'clinically inactive' subgroups Gd-positive (No 16) and Gd-negative (No 11). Finally the clinically active Gd-negative (No 6) showed sICAM-I levels higher (P = 0.002) than the clinically inactive Gd-negative (No 11). The specificity of high serum sICAM-I levels (above M +/- 2 s.d. of control values) to assess the disease activity in MS resulted higher (100%) using clinical than Gd-MRI activity (76%) as gold standard. The changes induced by 1 year recombinant Interferon-beta-Ib (rIFN beta-Ib) treatment on sICAM-I serum levels were also longitudinally investigated in 36 of the 49 RR MS. sICAM-I levels at baseline significantly increased in the first 2 months (baseline vs 1st month P < 0.0001 and 1st vs 2nd month P = 0.02), persisted at high levels without any significant change after 3 months, showed a temporary decrease at 6 months, then significantly increased again at 9 and 12 months. Fourteen patients experienced relapses, with a total of 20 relapses, during the whole treatment duration. The mean relapse/rate and the frequency of patients with Gd-positive MRI scans resulted significantly higher in the first semester compared to the second semester of treatment. This study adds further insights into the validity of serum sICAM-I to assess disease activity in MS and on the immunomodulatory properties of rIFN beta-Ib.