Background. Current methods to predict survival in patients with advanced, metastatic melanoma are limited. To determine clinical prognostic factors that accurately predict survival in patients with metastatic melanoma, a retrospective analysis was performed. Methods. Clinical, hematologic, and biochemical pretreatment parameters from 284 patients entered on 18 Phase I and II clinical trials were analyzed to determine their prognostic effect on survival. A multivariate parametric regression based on the Weibull distribution was derived to estimate survival. Results. Multivariate Weibull survival regression analysis showed that a simple model using the logarithm of the pretreatment values of lactate dehydrogenase and serum albumin, dichotomized as high and low, significantly and adequately predicted survival. A more complex multivariate model was also derived that involved the pretreatment platelet count, visceral organ involvement, and gender as additional factors. However, a larger study is needed to statistically validate such a model. Conclusions. The pretreatment values of serum lactate dehydrogenase and albumin are independent prognostic factors for survival in patients with metastatic melanoma. These two factors can be used to estimate survival of patients with advanced, metastatic melanoma and should be considered when designing melanoma trials in which survival is an endpoint.

Twenty-one patients were entered into a phase I trial to evaluate toxicity, antitumor effects, and biological responses at tumor sites during treatment of R24, an immunoglobulin G3 (IgG3) mouse monoclonal antibody (mAb) against GD3 ganglioside. Toxicity was related to dose of R24. Urticaria and pruritus were the most prominent side effects, with nausea, vomiting, and diarrhea occurring at the highest dose levels. Partial responses were observed in four patients lasting from 6 to 46 weeks, and mixed responses were seen in two patients. Responses occurred as early as 4 weeks and as late as 10 weeks after beginning treatment. Twenty of the 21 patients developed human IgG antibodies against R24. Antimouse Ig antibodies were first detected at a median of 14 days after starting treatment, but three of the four patients who had a partial response developed the antimouse Ig responses later than 20 days. Peak serum levels of R24 were related to dose and ranged from a mean of 0.9 micrograms/mL at the lowest dose level (1 mg/m2/d) to 44 micrograms/mL at the highest dose (50 mg/m2/d). The amount of R24 reaching tumor sites corresponded to the dose administered, and R24 could be detected in tumors as late as 30 days after finishing treatment. Inflammation at tumor sites was observed during treatment. Biopsies of tumors taken before, during, and after treatment revealed that R24 induced deposition of complement components, increased numbers of mast cells with mast cell degranulation, and infiltration of T lymphocytes. These results suggest that treatment with R24 can produce a localized inflammatory response at tumor sites that is capable of producing tumor regression.

No abstract

No abstract