Randomized trial of treatment with cisplatin and interleukin-2 either alone or in combination with interferon-alpha-2a in patients with metastatic melanoma

BACKGROUND, The objective of the current study was to evaluate the response rate, survival, and toxicity of treatment with cisplatin and high dose int ravenous continuous infusion interleukin-2 (IL-2) with or without interfero n-alpha-2a (IFN) in patients with metastatic melanoma. METHODS. One hundred and seventeen patients with metastatic melanoma random ly were assigned to receive cisplatin, 100 mg/m(2), followed after a 3-day rest period by IL-2, 18 x 10(6) IU/m(2), on Days 3-6 and Days 17-21 (Arm 1) or cisplatin and IL-2 using an identical schedule plus subcutaneous IFN, 9 x 10(6) U, 3 times a week during IL-2 administration (Arm 2). In the absen ce of disease progression or undue toxicity, the cycle could be repeated on Day 29, Patients who responded after two cycles eventually could receive a third cycle. One hundred and one patients were evaluable for toxicity and efficacy. RESULTS. On treatment Arm 1, 3 patients (6%) achieved a complete response ( CR) and 5 patients (10%) achieved a partial response (PR) with a median res ponse duration of 3.8 months for the CRs and 8.7 months for the PRs. On tre atment Arm 2, 2 patients (3%) achieved a CR (durations of 5.9 and 33.1 mont hs, respectively) and 11 patients (21%) a PR with a median response duratio n of 8.3 months. The median durations of overall, survival were 10.4 months (range, 1.1-39.7+ months) and 10.9 months (range, 0.5-38.1+ months) far tr eatment Firms I and 2, respectively. The toxicity profile was consistent wi th the known side effects of this IL-2 intravenous regimen combined with ci splatin chemotherapy and IFN. Toxicity was more pronounced in treatment Arm 2 compared with treatment Arm 1. There were 2 and 4 patients, respectively , in treatment Arms 1 and 2 who died within 28 days after completion of tre atment. CONCLUSIONS. The observed overall response rates of 16% and 25% in treatmen t Arms 1 and 2, respectively, is lower than that expected with biochemother apy; despite the fact that the objective of the trial was not to show any d ifference between the 2 treatment arms, our results indicate that the addit ion of IFN, at the dose and schedule used in this trial, fails to improve t he activity of a cisplatin/IL-2 regimen significantly in patients with meta static melanoma. Although response rates were relatively law, the median ov erall survival was nearly 1 year in both groups. Cancer 1999;85:1060-6. (C) 1999 American Cancer Society.