Several groups have described the efficacy of interleukin 2 (IL-2) plus lymphokine-activated killer (LAK) cells in the treatment of cancer patients with significant response rates noted in patients with renal cell cancer and malignant melanoma; however, the optimum regimen remains undefined. The Biological Response Modifiers Program of the National Cancer Institute conducted two consecutive Phase I/II studies evaluating the toxicity and clinical efficacy of different methods of IL-2 and LAK cell therapy. In the first trial, we modified the standard Rosenberg regimen by decreasing the duration of priming in an attempt to reduce the toxicity related to this phase of the therapy and thereby administer more IL-2 doses with the LAK cells. In the second trial, we used a continuous i.v. infusion IL-2 regimen and altered both the leukapheresis procedure and the LAK cell culture techniques based on our in vitro and preclinical studies suggesting that 2-day LAK cells were superior. Thirty cancer patients received i.v. bolus IL-2 at 100,000 units/kg every 8 h for 3 days during priming and for 5 days during LAK cell administration. A second group of 22 cancer patients received IL-2 by continuous i.v. infusion at 3 x 10(6) units/m2 for 5 days during priming and an additional 5 days of IL-2 with the LAK cell phase of the treatment. The timing of the start of the leukapheresis procedures, their duration and number, and the LAK cell culture techniques differed in the two trials. Overall, 52 patients with various cancers were treated. The toxicities associated with each regimen were similar to those seen in other IL-2 plus LAK cell trials. Four patients (one each with melanoma and diffuse large cell lymphoma and two with renal cell cancer) exhibited partial responses lasting 2, 4, 10, and 15+ mo. Serial tumor biopsies from treated patients demonstrated that therapy can produce a marked mononuclear cell infiltrate and an increase in HLA-DR expression on tumor cells. There was no difference in the overall response rate between the two regimens, but toxicity was less with continuous i.v. infusion IL-2. The 5-day continuous i.v. infusion regimen resulted in significantly higher rebound lymphocytosis, cell yield from leukapheresis, and number of LAK cells harvested from culture.