High-Dose Cytarabine (HiDAC) Improves the Cure Rate of Patients with Newly Diagnosed Acute Myeloid Leukemia (AML): Is It Better to be Given As Induction Therapy or As Consolidation Therapy?

Introduction: The optimal treatment approach for newly diagnosed patients with AML remains uncertain. HiDAC is widely considered to increase the proportion of patients cured compared to standard-dose cytarabine. However, it remains uncertain whether HiDAC is best given during induction or consolidation, and how many cycles of HiDAC are optimal. Many centres in Australia treat younger patients (age ≤60 yrs) with newly diagnosed AML with one of two approaches: either 7+3 induction followed by HiDAC-2 consolidation for 2 cycles; or a single course of HiDAC-3±7 induction followed by 2 cycles of lower dose cytarabine-based therapy (eg 5+2±5). Our retrospective study compared the outcomes of these 2 approaches in a large cohort of Australian patients treated at 5 centres.

Methods: Consecutive patients aged ≤60 yrs with a new diagnosis of AML (de novo or secondary) were included in the study if they were planned for treatment with either: 1) cytarabine 100 mg/m² for 7 days plus idarubicin 12 mg/m² for 3 days (7+3) induction followed by 2 cycles of HiDAC 3 g/m² days 1,3,5,7 plus idarubicin 12 mg/m² for 2 days (HiDAC consolidation cohort); or 2) HiDAC 3 g/m² days 1,3,5,7 plus idarubicin 9-12 mg/m² for 3 days ± etoposide 75-100 mg/m² for 7 days as induction followed mostly by cytarabine 100 mg/m² for 5 days plus idarubicin 9-12 mg/m² for 2 days ± etoposide 75-100 mg/m² for 5 days as consolidation (HiDAC induction cohort). Patients were diagnosed from 1999 to June 2013, and were followed for at least 12 months with data cut off June 2014.

Results: 486 patients were included: HiDAC consolidation cohort n=251; HiDAC induction cohort n=235. The HiDAC consolidation cohort had a greater median age (49 vs 47 yrs, p=0.02) and more patients with good risk cytogenetics (16% vs 8%, p=<0.005). Other baseline demographics were well matched. For the HiDAC consolidation cohort and the HiDAC induction cohort, respectively, CR1 rate was 80% vs 91% (p=0.001); TRM 8% vs 5% (p=0.14); OS (5 yrs) 49% vs 50% (p=0.7); DFS (5 yrs) 47% vs 41% (p=0.24) and the cumulative incidence of relapse (CIR) 41% vs 50% (p=0.1). The CIR was greater in the HiDAC induction cohort despite a higher allogeneic hematopoietic stem cell transplantation (alloHSCT) in CR1 rate (18% vs 29%, p=0.002) in this cohort. For the 301 patients who achieved CR1 and did not undergoing alloHSCT in CR1, CIR was greater in the HiDAC induction cohort (49% vs 60%, p=0.059) leading to a reduced DFS (58% vs 46%, p=0.058), and OS (59% vs 49%, p=0.13) in that subset of patients. Excluding patients with good risk cytogenetics from the analyses did not change the results significantly.

Conclusions: OS and PFS using HiDAC as induction or consolidation therapy were similar, and compared favourably to published data. Interestingly, the better CR rate and a greater use of alloHSCT in CR1 in the HiDAC induction cohort did not lead to a better PFS or OS - because of a greater relapse rate in this cohort - primarily seen in those patients not undergoing alloHSCT in CR1. In the absence of mutational prognostic information, these data may suggest that HiDAC as induction therapy can achieve CR in patients with biologically higher risk disease who have a higher relapse rate, and that including 2 cycles of HiDAC in consolidation in the absence of alloHSCT in CR1 is a more effective therapy than a single cycle of HiDAC administered during induction therapy.