Prognosis of Patients with Standard- or Intermediate-Risk Acute Myeloid Leukemia Presenting with Residual Disease after One Induction Cycle Is Improved If Remission Is Achieved Following Re-Induction Given By Day 14 Compared to Second-Line Therapy Given 24-40 Days after Induction

Introduction: Failure to achieve remission following standard induction is a poor prognostic sign in patients with acute myeloid leukemia (AML). However, as there is no consensus on the optimal number of induction cycles and/or the best indication and timing for re-induction, the definition of refractory AML is still debatable. A clinically relevant definition of refractoriness should be associated with poor long-term prognosis. This is of particular importance for patients who eventually achieve remission (CR1) and in whom referral for allogeneic stem cell transplantation (alloSCT) in CR1 may be controversial. It was previously reported that AML patients who achieved CR1 following re-induction given on day 14 of therapy, had a long-term survival identical to that of patients who achieved CR1 after one induction cycle. To the best of our knowledge, no data are available regarding the prognosis of patients who achieved CR1 after a second cycle of intensive chemotherapy, administered at the end of neutropenic period.

The current study compared the prognosis of AML patients who achieved CR1 after one induction cycle to that of patients who achieved CR1 following a second cycle of chemotherapy given either as re-induction by day 14 or as a second-line therapy administered on days 24-40 of the first induction.

Patients and Methods: Data of all leukemia patients admitted to the Rambam Medical Center between the years 1993 and 2015 were retrospectively collected. Adults diagnosed with de-novo AML who achieved CR1 following one or two cycles of therapy were included in the study. Patients diagnosed with APL or high-risk AML, as defined based on a poor-risk karyotype or presence of FLT3-ITD mutation, were excluded from the analysis. It is common practice at our center to assign patients for re-induction according to day 14 bone marrow (BM) results. Yet, for various reasons, this paradigm is not followed in some patients. Overall survival (OS), leukemia-free-survival (LFS), and death rates were compared in the study participants. Multivariate analysis for prediction of death, including the use and timing of second-line therapy, was performed.

Results: Six hundred and forty three de novo AML patients were treated at Rambam during the study period. After exclusion of patients with high-risk AML, 492 individuals were left. Of them, 242 patients achieved CR1 after one cycle of induction (control group), 53 after induction + re-induction administered by day 14 (re-induction group) and 38 achieved CR1 after induction followed by second-line therapy prescribed on days 24-40 (second-line group).

Median OS was 25.9, 23.5 and 15.1 months for patients in control, re-induction and second-line groups, respectively. Median LFS was 14.7, 10 and 8.16 months for the corresponding groups, respectively. Differences in OS and LFS were statistically significant between the control and second-line groups (p<0.0001 for both comparisons), and between the re-induction and second-line groups (p=0.007 for OS; p=0.001 for LFS). Differences in OS and LFS between the control and re-induction groups were insignificant (p=0.46, p=0.49). In multivariate analysis including all patients, age and cytogenetics were the only significant factors to predict survival. When the 40 patients presenting with standard-risk (favorable) cytogenetics were excluded, age and treatment were the only significant factors to predict survival in multivariate analysis.

Conclusions: Despite previous reports showing similar outcomes in patients who achieved CR1 post re-induction and those achieving CR1 with one induction cycle, some physicians are reluctant to start re-induction as early as day 14. We herein report a significantly better prognosis in patients achieving CR1 after re-induction given by day 14 compared to those achieving CR1 after a second cycle of therapy prescribed on days 24-40. These data may have 2 practical implications. First, in AML patients with residual disease, an effort should be made to start re-induction as early as day 14. Second, for intermediate-risk patients, a failure to achieve CR1 after one induction cycle should not alter post-remission decisions, if CR1 was achieved with re-induction by day 14. This may not be true for patients who achieved CR1 with second-line therapy prescribed on days 24-40. Our study was not powered to examine whether alloSCT could improve the outcome of these refractory AML patients.