The objectives of this phase I-II trial were to assess the toxicity, immunological and clinical responses induced in stage III/IV melanoma patients by the subcutaneous administration of xenogenic polyantigenic vaccine (XPV) prepared from disrupted murine melanoma (B16) and carcinoma (LLC) cells. An inducing course of vaccinotherapy consisted of ten immunizations (five at weekly and five at fortnight intervals). Twenty-four hours following each of the first five vaccinations, the patient was subcutaneously given a low dose of the recombinant interleukin-2 (IL-2). A consolidating course of the vaccinotherapy consisted of monthly vaccinations. Grade 3 or 4 toxicities, as well as laboratory and clinical signs of developing autoimmune disorders, were recorded in none of the 40 XPV-treated evaluable patients. A significant increase in delayed-type hypersensitivity (DTH) skin reaction to vaccinal B16, but not to LLC antigens (Ags), occurred in patients after inducing vaccinations. At the same time, those patients demonstrated a marked augmentation of blood lymphocyte proliferation responses not only to B16 but also to LLC Ags. Vaccinations also led to increased cell-mediated reactivity to murine non-tumor, spleen cell (SC)-associated Ags, which, however, was not as significant as that to tumor-associated antigens (TAAs). Of great importance was the fact that XPV administration resulted in increased blood lymphocyte proliferative reactivity of patients to human melanoma-associated Ags, while not affecting their reactivity to the control alloantigens. With immunotherapy, concentrations of both interferon-gamma (IFN-gamma) and interleukin-4 (IL-4) were elevated in patients' sera, suggesting an intensification of the T helper1/ T helper 2-mediated responses in the XPV-treated patients. The average survival of the 32 stage IV melanoma XPV-treated evaluable patients was noticeably higher than that of the 32 clinically comparable control patients (13 vs. 5 months). The overall 3 year-survival rate in the XPV-treated group and the control group was 25% (8 patients) and 3% (1 patient), respectively. In general, the results suggest that xenogenic tumor cells may provide a novel feasible approach to constructing clinically effective vaccines.