We investigated the utility of serum S100 determined by means of immunoradiometric assay in a cohort of 438 patients affected by cutaneous melanoma (126 untreated and 312 previously treated). Using 0.2 microg/l cut-off value, determined in 134 healthy blood donors, the sensitivity was 4.2% in stage I patients (4/94), 5.3% in stage II patients (1/19), and 38.5% in stage III patients (5/13). Even though the sensitivity increased progressively from stage I to stage II and III, these differences were not statistically significant. The prognostic significance of S100 evaluation at diagnosis was investigated in terms of survival but no statistical correlation between S100 basal levels and survival was found. In the 312 previously treated patients serum S100 levels were correlated to disease extent, high levels of the marker were observed in 42.8% (9/21) of patients with local recurrence, in 32% (16/50) of patients with lymph node and/or in-transit metastases, in 77.3% (17/22) of patients with distant metastases, and in patients with NED, the specificity of the marker was 96.8% (212/219). The difference between these groups were statistically significant. In conclusion, S100 protein was abnormally high in patients with metastatic malignant melanoma. Serial S100 measurements in a follow-up study are necessary to test the importance of the protein in the management of patients with metastatic malignant melanoma.