Background: An intensive in-patient Phase II study of continuous infusion recombinant human interleukin-2 (rhIL-2) 3 X 10(6) Roche units (9 MIU/m2)/day plus intramuscular injection of interferon-alpha-2a (rIFN-alpha-2a) 5 x 10(6) U/m2/day, four consecutive days per week, was performed to determine whether four consecutive weeks of treatment every 6 to 8 weeks might yield a response rate of 30 to 40%, supporting an additive or synergistic effect of these agents in patients with metastatic melanoma.
Methods: Eligibility was limited to otherwise healthy metastatic melanoma patients with no brain metastases. Tumor sites were measured at six-week intervals. Toxicity was recorded using the National Cancer Institute's common toxicity criteria.
Results: From 1988 through 1989, 23 patients received 47 courses of rhIL-2 plus rIFN-alpha-2a. Fifteen patients (65%) had visceral metastases. Eight patients (35%) had prior systemic chemotherapy. There were two partial responses, both after one course, for a response rate of 8% (95% confidence interval: 1% to 28%). A previously untreated patient had more than 90% diminution in extensive liver, adrenal, duodenal, and soft tissue metastases lasting 9 months. A second patient, who had failed prior chemotherapy, had a 50% reduction in lymph node metastases lasting 8 months. Median overall survival was 10.4 months (21 and 70+ months for the two responders). Cumulative reversible toxicities of debilitating fatigue, malaise, depression, poor appetite, and acute toxicities of hypotension, oliguria, infection, and cortical dysfunction were responsible for dose modifications in 16% of 188 anticipated weeks of treatment, usually near the end of the second or subsequent treatment courses.
Conclusions: This study demonstrates a low response rate and lack of additive benefit for rhIL-2 and rIFN-alpha-2a administered intensively in metastatic melanoma patients. Since this biotherapy regimen, in the same dose and weekly schedule, has yielded substantial clinical activity with chemotherapeutic agents, this activity cannot be significantly attributed to the independent activity of rhIL-2 plus rIFN-alpha-alone.