A comparative clinical trial of two regimens of combination chemotherapy has been accomplished in acute leukemia by four separate medical and pediatric services. Sixty-five patients were allocated at random to one of two treatment programs. Daily administration of methotrexate with daily 6-mercaptopurine has been compared to methotrexate every third day in the same total dose with daily 6-mercaptopurine.

No difference in frequency of remission, extent of remission or toxicity was observed between the two groups. Among those patients who attained remission status, however, duration of remission (P = .05-.10) and of survival (P = <.05) was longer for the continuous group.

All remissions in children occurred in acute lymphocytic leukemia, whereas all remission in adults were observed in acute myelocytic leukemia. The duration of remissions was somewhat shorter for children with acute lymphocytic leukemia than for adults with acute myelocytic leukemia. The frequency of remission, either partial or complete, was higher in children, however (36 per cent), than in adults (19 per cent), although the confidence limits for each figure overlap.

The duration of acute leukemia in previously untreated patients did not influence response to therapy from the two antimetabolite regimens in this study. In patients who had had prior antimetabolite therapy, however, complete remissions were attained less often than in previously untreated patients.

The toxic manifestations encountered during the administration of these antimetabolites are described. Seventeen deaths occurred during this study, of which 8 occurred in the first 10 days, presumably from leukemia and not drug toxicity. Five patients died with hypoplastic marrows ascribed to drug toxicity.

The toxic manifestations were qualitatively and proportionately the same in patients who attained remission status, and in those patients who failed to remit, but who lived long enough to recognize the onset of remission if it were going to occur. No indication was obtained, therefore, that patients who attained remission were subjected to a greater toxic hazard, in order to achieve the therapeutic benefits observed, than those who did not remit.

The median survival of patients who achieved remission was longer (p <.05) than for patients who did not remit. Since the survival time of remitters from relapse to death was almost identical with the survival time of nonremitters from onset of treatment to death, this difference can be accounted for by the time spent in remission and getting to remission.

The median survival time from symptomatic onset for all children in this study was 12 months, and for adults, 7 months. The median in children is similar to that reported from other clinics. This is evidence that a comparative therapeutic trial in acute leukemia can be accomplished without recognizable compromise of patient welfare.