Circulating forms of L-selectin were found to be elevated in several autoimmune diseases. ICAM-1 has been suggested a predictor of the onset of GO. The aim of the study was an estimation of serum L-selectin and ICAM-1 in patients with Graves' disease with ophthalmopathy during treatment with corticosteroids to assess their potential as a guideline of immunosuppressive therapy. We detected serum L-selectin and ICAM-1 in three groups of subjects: 20 patients with Graves' disease without ophthalmopathy (Gd), 17 patients with clinical symptoms of ophthalmopathy (CAS> or =3, anamnesis of GO> or =1 year) and 24 healthy volunteers. Corticosteroid therapy consisted of intravenous infusions of methyloprednisolone (MP) and subsequent treatment with oral prednisone (P). The serum samples were collected 24 h before MP, 24 h after MP, 12+/-2 days of treatment with prednisone and after the end of the corticosteroid therapy. The levels of soluble L-selectin and ICAM-1 in the serum were determined by the ELISA method. The statistical significance was estimated by the Mann-Whitney U-test. Serum L-selectin and ICAM-1 were significantly increased in patients with GO (respectively 1182+/-222 and 438+/-68 ng/ml) and in patients with Graves' disease without ophthalmopathy (respectively: 1168+/-130 and 343+/-109) in relation to the controls. After MP treatment in corticosteroid-responsive patients (improvement in CAS < or =1) serum concentration of L-selectin and ICAM-1 decreased significantly and gradually increased during subsequent treatment with prednisone. In corticosteroid-responsive patients serum L-selectin was significantly higher before MP administration and after P treatment in relation to corticosteroid-resistant subjects.
Conclusions: 1. Serum L-selectin and ICAM-1 were elevated in patients with active GO 2. Methyloprednisolone decreased levels of the studied adhesion molecules in corticosteroid-responsive patients with GO 3. Lack of clinical results in corticosteroid therapy in patients with a low starting L-selectin level would suggest the possibility of serum L-selectin estimation as a prognostic for immunotherapy efficacy.