This prospective and randomized trial reports the effects of three chemotherapeutic regimens on three different clinical parameters in adults with supratentorial malignant brain gliomas: 1) duration of the interval between surgery and relapse (the "free interval"); 2) total survival time; and 3) rate and length of objective remissions. All patients were irradiated with 5500 to 6000 rads after neurosurgery aimed at an optimum removal. The administration of VM-26, 60 mg/sq m on Day 1, plus CCNU, 130 mg/sq m on Day 2, repeated every 6 weeks, had no significant effect on the length of the free interval between surgery and relapse. Thus, the median duration of this period was 39 weeks in the group of 61 patients who received adjuvant chemotherapy and 30 weeks in 55 controls without treatment. Neither was the total survival time prolonged by the administration of early chemotherapy. The best prognostic factor for the duration of the free interval and survival was age: patients under 50 years old survived statistically significantly longer. The rate of the objective remission, defined as a clear-cut clinical improvement persisting 6 weeks after complete discontinuation of steroids, was measured after tumor relapse. In patients who were not previously treated with chemotherapy, CCNU alone or VM-26 plus CCNU produced objective remissions in only 15% of treated patients. Out of 17 cases treated previously by VM-26 plus CCNU, none responded to procarbazine after relapse.