Front-Line FLAG-IDA and Nove-HiDAC Chemotherapy Improves Overall Survival (OS) and Complete Remission Rates (CR) for Patients with Secondary or Therapy-Related AML Compared to 3&7

Background: Therapy for patients (pts) with high risk AML remains unsatisfactory. Retrospective studies have demonstrated activity of fludarabine, cytarabine, GCSF and idarubicin (FLAG-IDA) and of mitoxantrone, etoposide and cytarabine (NOVE-HiDAC) as salvage therapy in pts with relapsed or refractory AML. A recent randomized trial indicated high complete remission (CR) rates with improved relapsed-free survival when FLAG-IDA is administered as frontline induction therapy (Burnett et al. J Clin Oncol 2013). Since 01/2011, we have used FLAG-IDA as a first line therapeutic option in pts with high risk AML (poor risk cytogenetics, antecedent myeloproliferative neoplasm or myelodysplastic syndrome, and/or therapy-related AML) in an attempt to improve CR rates and permit more patients with AML to advance to allogeneic hematopoietic cell transplantation (alloHCT). Prior to 2011, either 3&7 or NOVE-HiDAC was used as first line therapy in patients with AML.

Methods: We conducted a retrospective review of consecutive patients with high risk AML treated with front-line (a) FLAG-IDA between 01/2011 to 03/2015, (b) NOVE-HiDAC from 01/2006 to 12/2014, or (c) 3&7 from January 01/2011 to 12/2014 at the Princess Margaret Cancer Centre, to determine the CR rates and overall survival (OS) associated with the different regimens.

Results: Patients characteristics are in Table 1. Fifty-two, 32, and 30 pts received FLAG-IDA, NOVE-HiDAC or 3&7 as first induction, respectively. Patients receiving FLAG-IDA had more high-risk features (i.e. complex cytogenetics, more azacytidine failures) compared to those receiving 3&7. Overall CR rate (i.e. CR + [CRi] + [CRp]) with FLAG-IDA, NOVE-HiDAC, and 3&7 respectively was 86% (n=42/49), 84% (n=21/25) and 50% (n=13/26), respectively. Median CR duration, censored at time of transplant, for pts receiving FLAG-IDA, NOVE-HiDAC and 3&7 was 3 mos (0.5-15), 3.5 mos (1-9) and 5.5 mos (0.5-42), respectively. OS at 1 year with FLAG-IDA, NOVE-HiDAC and 3&7 was 61% (95% CI, 41% -75%), 55% (95% CI, 34%-72%) and 21.6% (95% CI, 7.4%-40.7%), respectively (log-rank test p-value=0.0076). On subgroup analysis, there was no statistical difference in OS for pts ≥70 years. Of those with a donor identified, 35% (n=13/37), 73% (n=11/15) and 29% (n=5/17) of pts who were treated with FLAG-IDA, NOVE-HiDAC and 3&7 underwent an alloSCT, respectively. Pts with sAML may have had a higher transplant rate due to donor searches initiated earlier. Probable and possible invasive aspergillosis infections in pts receiving FLAG-IDA, NOVE-HiDAC and 3&7 were 50%, 34% and 33% respectively. Institution of earlier bronchoscopies led to increased fungal detection in the FLAG-IDA group. Median length of stay and ICU transfers were similar between groups. Induction deaths were secondary to sepsis, respiratory failure, invasive aspergillosis, and hemorrhage; these were similar across groups. Two pts receiving NOVE-HiDAC, with prior MPN, died of progressive splenomegaly and liver failure.

Conclusions: Toxicities associated with frontline FLAG-IDA and NOVE-HIDAC induction are acceptable. FLAG-IDA and NOVE-HiDAC induction can result in durable CR, permitting patients with high risk AML to proceed to alloSCT and providing more favourable survival rates than frontline 3&7. Randomized studies are needed to confirm these findings for pts with poor-risk sAML and tAML.