5-Azacytidine (AZA) Compared to Intensive Chemotherapy in Elderly Acute Myeloid Leukemia (AML) Patients: Results of a Retrospective Single Centre Matched Analysis,

Abstract 3620

AML in elderly patients is characterized by a poor prognosis, especially in those patients aged >70, and/or in frail patients with comorbidities or poor performance status (PS). Moreover, several studies already suggested that elderly AML patients with unfavorable karyotype may not benefit from intensive chemotherapy. With this background, and using a matched analysis, this report aimed to assess the outcome of a single centre series of elderly AML patients who received either non intensive therapy by hypomethylating agents, or standard induction with intensive therapy.

All patients were aged over 60 and had de novo or secondary AML. For the purpose of this comparison, the cohort was divided in two distinct groups. Group A included 36 cases treated by intensive chemotherapy between 1995 and 2005 according to the GOELAMS AML-SA2002 or SA3&4 protocols (5+7 induction with idarabicine 5 mg/m²/d and cytarabine 100 mg/m²/d). In this group, patients who could achieve CR received either 3 or 6 consolidation courses delivered over 1 or 2 years (according the protocol AML-SA2002 versus SA3&4). Group B included another 36 patients who were treated between 2006 and 2010 with AZA according to the recommendations of the “compassionate use program” authorized by the French Health Agency (one cycle of AZA = 7 days of subcutaneous administration 75mg/m² every 28 days until progression).In this group, response was assessed after 3 cycles and qualified using IWG criteria. These two groups were matched based on cytogenetic features and age.

The median age for the total cohort was 72 years (range, 60–86). Groups were comparable for WBC, % marrow blasts infiltration, WHO subtypes, and cytogenetic features at diagnosis. A higher rate of secondary AML was observed in the AZA arm. CR and CR with incomplete hematological recovery (CRi) rates were significantly higher in the intensive vs. AZA arm (63% vs. 28%, p<0.0001). However, there was a trend for a higher rate of partial remission (PR) in the AZA Arm (25% vs. 5%, p=0.02). With a median follow-up of 13.3 months (range, 5–80) from diagnosis, median overall survival (OS) was comparable between the two arms: 10.4 vs. 10.3 months, p=0.3) In multivariate analysis for OS including treatment strategy, the strongest prognostic factors were an unfavorable karyotype (HR=2.05, 95%CI, 1.09–3.85; p=0.03), PS status (0 vs. 1–2; HR=2.04 95%CI, 1.16–3.58; p=0.01) and platelets number at diagnosis (analyzed as a continuous variable) (HR=1, 95%CI, 0.99–1.00; p=0.04). Of note, the treatment arm was not found to be a significant determinant for OS: (AZA vs intensive chemotherapy.; HR=1.86, 95%CI, 0.86–3.16; p=0.13).

This analysis suggests that the use of AZA as an alternative to intensive chemotherapy in elderly patients with de novo or secondary AML may lead to similar OS, despite a significant difference in terms of CR and CRi rate. The different mechanism of action of AZA in comparison to conventional chemotherapy, and the higher rate of PR that can be achieved after AZA therapy might contribute to improved OS through relatively long lasting disease control. These results set the frame for a prospective controlled trial to test AZA as an ambulatory alternative to standard intensive chemotherapy in elderly AML patients, especially those patients with unfavorable karyotype or poor PS and comorbidities.