The influence of induction chemotherapy dose and dose intensity on the duration of remission in acute myeloid leukemia. Australian Leukemia Study Group.

The aim of this study was to assess the influence of dose and dose intensity (DI) of induction and consolidation chemotherapy on relapse rates in 264 de novo patients with acute nonlymphocytic leukemia (ANLL). Patients were randomised to receive cytosine arabinoside (ARAC) 100 mg/m2 continuous infusion for 7 days and daunorubicin (DNR) 50 mg/m2 IV day 1-3 (7-3) or the same drugs with the addition of etoposide 75 mg/m2 IV days 1-7 (7-3-7). Cox proportional hazards regression models were used throughout to identify prognostic factors, including dose delivery parameters, influencing the rate of relapse. Of 152 patients who achieved a complete remission (CR), 104 have relapsed with a median duration of CR of 15.8 months. Actual dose delivered was prospectively documented. Cox regression analysis identified the most significant prognostic factors jointly influencing duration of CR as performance status groups (p < 0.0001), percentage peripheral blasts (p = 0.0015), 7-3-7 arm (p = 0.0075), age < 40 years (p = 0.022) and induction dose ARA-C plus DNR (p = 0.029). In this analysis patients randomized to the 7-3-7 arm had an estimated 43% reduction in the relapse rate and each 10% reduction of doses ARA-C and DNR was associated with an estimated 45% increase in the relapse rate. The number of induction courses, delays in treatment and induction dose intensity did not significantly influence the duration of CR nor did any of the consolidation treatment parameters.(ABSTRACT TRUNCATED AT 250 WORDS)