A strategy for evaluation of new treatments in untreated patients: application to a clinical trial of AMSA for acute leukemia.

This clinical trial (DT7995) was designed to evaluate amsacrine (AMSA) plus cytosine arabinoside (ara-C), vincristine, and prednisone (OAP) therapy in previously untreated patients with adult acute leukemia and to investigate a new strategy for assignment of patients to treatment using estimated probabilities of complete remission (PPR) based on six prognostic factors. In the first stage of the trial, patients with unfavorable prognosis (PPR less than .40) received AMSA-OAP for remission induction and patients with favorable prognosis (PPR greater than or equal to .40) received Adriamycin [Adria Laboratories, Columbus, OH] plus OAP (Ad-OAP). As AMSA-OAP was found to be promising in patients with unfavorable prognosis, it was administered to relatively more favorable patients (PPR less than .60) in the second stage of the trial and to all patients in the third stage. There were 242 patients entered into study; 134 received AMSA-OAP and 108 received Ad-OAP. Outcomes were compared with 242 paired patients who received Ad-OAP therapy from 1973 to 1977. The estimated complete remission rate in previously untreated adults with acute leukemia is 61% for patients receiving Ad-OAP (95% confidence interval, 59% to 64%). Overall, the survival experience for the 242 patients on DT7995 was significantly better than that in the control series (P = .03), but there was no strong statistical evidence (P = .10) that the 134 patients receiving AMSA-OAP had better survival than control patients receiving Ad-OAP, with a median of 32 v 21 weeks, respectively. It is concluded that AMSA-OAP is equivalent to Ad-OAP in the induction of complete remissions (estimated complete remission rate, 61%) and that assignment of patients to treatment based on predicted prognosis is an ethical and efficient strategy for the evaluation of new therapies in previously untreated patients with acute leukemia.