Aims and Background: A soluble form of intercellular adhesion molecule-1 (sICAM-1) has been recently identified in patients with malignant melanoma. It has been demonstrated that inflammatory cytokines can modulate the cellular expression of ICAM-1 and the shedding of this molecule by cells. To our knowledge, few data exist on serum sICAM-1 levels in cancer patients treated with immunomodulators. Liposomes containing muramyl tripeptide (MLV MTP-PE) can activate monocytes from cancer patients in vitro and in vivo, making them cytotoxic such as tumor necrosis factor-alpha (TNF-alpha) and Interleukin-6 (IL-6). The purpose of the present study was to evaluate the levels of sICAM-1 and their possible correlation with serum inflammatory cytokine levels in melanoma patients treated with MLV MTP-PE., Methods: The sera from 9 patients with metastatic melanoma treated with MLV MTP-PE, 4 mg i.v. twice a week for 12 weeks, were tested in ELISA system to detect sICAM-1, TNF-alpha, IL-6, Interleukin-1 beta (IL-1 beta) and Interferon-gamma (IFN-gamma) before, and 2 and 24 h after the 1st, 12th and 24th infusion of MLV MTP-PE., Results: Baseline levels of sICAM-1 were elevated in all patients (median 540 ng/ml: range 400-1030 ng/ml). Twenty-four h after the 1st infusion of MLV MTP-PE, we observed 6 increases in sICAM-1 levels, 1 decrease and 2 stable values (median 720 ng/ml: range 410-1820; P = 0.060). Twenty-four h after the 12th infusion, sICAM-1 increased in 3 patients and did not change in 4 (median 790 ng/ml: range 495-1650 ng/ml; P = 0.069). At the 24th infusion, sICAM-1 increased in 4 of 6 evaluable patients and remained stable in 2 (median 802 ng/ml: range 510-1450 ng/ml; P = 0.045). To better analyze the variations in sICAM-1, the patients were arbitrarily divided into two groups according to their clinical behavior: 4 presented stabilization (all lesions, n = 2; some lesions, n = 2) (Group A); 5 presented progressive disease (Group B). In Group A, sICAM-1 levels remained stable or showed a modest increase during treatment (except in 1 patient, who exhibited a substantial variation after the 12th infusion). In contrast, in Group B very high levels of sICAM-1 were observed at the beginning of the study therapy in 1 patient and after the 1st infusion in 3 patients; these values remained high until the 24th infusion. In most of the patients, TNF-alpha and IL-6 increased after the 1st infusion, but not thereafter. IFN-gamma was never detected; IL-1 beta was detectable in a few cases, but only before the infusions., Conclusions: baseline levels of sICAM-1 were elevated in all patients and further increased during treatment only in patients with more aggressive disease. No correlation was found between sICAM-1 and inflammatory cytokines. It would therefore seem that in patients with advanced disease, higher levels and a progressive increase in sICAM-1 may be unfavorable prognostic factors.