Results of Therapy for Acute Myeloid Leukemia in First Relapse

Abstract

Fifty-four consecutive patients with acute myeloid leukemia (AML) in first relapse presenting at a single institution were studied to determine factors affecting response to re-induction therapy. For purposes of analysis, re-treatment protocols were grouped into two categories, one with high dose and the other with standard dose cytosine arabinoside. Most regimens also included an anthracycline, mitoxantrone or amsacrine. Thirty-one of the 51 patients (61%) who received therapy achieved a second complete remission (CR-2). Median duration of CR-2 was 6 months (range 1–24+ months). Five patients remain in CR-2, three of whom received bone marrow transplants (median follow-up 24 months). The variables, age, gender, FAB subtype, leukocyte and platelet count, duration of CR-1, the initial and re-induction regimens were analyzed for prognostic value in attaining and maintaining CR-2. Only younger age (p < 0.001) and longer CR-1 duration (p < 0.05) were significantly correlated with greater likelihood of attaining CR-2 with univariate analysis, and only age was correlated with CR-2 rate using multivariate analysis (p = 0.018). Younger age was associated with longer CR-2 duration (p = 0.003) using multivariate analysis, a correlation that persisted when transplanted patients were excluded. There was no advantage to the use of high dose versus standard dose cytosine arabinoside in the reinduction regimen with respect to the ability to either achieve or sustain CR-2. Our data indicate that although the remission induction rate for AML in first relapse is high, remissions are brief and other strategies are required to improve outcome of patients in second remission.

Key Words:

• Acute myeloid leukemia
• first relapse
• remission duration
• chemotherapy
• prognostic factors