Maintenance Therapy with Low-Dose Subcutaneous 5-Azacitidine in Older Patients with AML in 1st Remission.

Elderly patients with AML have a poor outcome, with low complete response (CR) rates and durations of CR that are typically less than 1 year, highlighting the need for more effective post-remission therapy. 5-azacitidine (AZA) is a nucleoside analogue/DNA methyltransferase inhibitor approved for use in all FAB subtypes and risk categories of myelodysplastic syndromes (MDS). In higher-risk patients with MDS, including those with AML (former RAEB-T subtype), AZA improves overall survival and delays the time to leukemia transformation. We undertook a phase 2 pilot study of low-dose subcutaneous (SC) AZA in older adult patients with AML in 1st CR or CR with incomplete platelet recovery (CR_p) following standard induction therapy.

Study objectives included the following: 1) determine the one year disease-free survival in elderly patients with acute myeloid leukemia (AML) in first CR/ CR_p treated with low-dose SC AZA as post-remission therapy. 2) determine safety and tolerability of SC AZA administered in the post-remission setting. 3) investigate the relationship between bone marrow genomic promoter methylation with 1-year disease-free survival. Eligibility included age ≥ 60 with AML in 1st CR/ CR_p following 1-2 cycles of induction chemotherapy and 1-2 cycles of consolidation therapy, ECOG PS 0-2, adequate end-organ function. AZA was administered subcutaneously on 1 of 2 different dosing schedules: A) 50 mg/m²/d x 5d (d 1-5), or B) 50 mg/m²/d x 7d (d 1-5, 8-9). Cycles were repeated every 4 weeks, and up to 12 cycles.

As of August 2009, 16 patients have been enrolled on the study and 15 are currently evaluable. Nine patients received dosing schedule A and 6 received schedule B. Median age was 69 years (range 62-81); M/F was 13/2; baseline cytogenetic risk categories at initial diagnoses: poor (6), intermediate (8), and unknown/not done (1). Two of 15 patients had a history of antecedent MDS. Nine of 15 patients (60%) required 2 cycles of induction chemotherapy to achieve CR/CR_p. The median time from achievement of CR/CR_p to AZA initiation was 13.6 weeks (range 7 – 21.9 weeks). To date, the median number of AZA cycles received was 4, ;5 patients have received ≥ 6 cycles, 2 of whom have received the 12 planned cycles and another who remains on-study after 11 cycles. Median duration of CR/CR_p was 54.8 weeks (95% CI: 28.1-96.4 weeks) estimated using the Kaplan-Meier method. Only 2 of 15 (13%) patients developed grade 3-4 non-hematologic adverse events (colitis, headache). Six of 15 (40%) patients developed reversible grade 3-4 neutropenia or thrombocytopenia, but only 3 required dose reduction. No patients discontinued AZA due to toxicity, and there were no deaths that occurred on-study. Criteria for early stoppage, based upon toxicity, have not been reached. Methylation array analyses are ongoing.

SC AZA administered as maintenance therapy in older patients with AML in 1^st CR/CR_p appears feasible and safe. Extended treatment was possible in a high proportion of patients, with encouraging early signs of durable remissions. Accrual to this trial is ongoing and updated results will be presented.

Lancet:Celgene: Research Funding. Komrokji:Celgene: Research Funding, Speakers Bureau. de Castro:Celgene: Speakers Bureau. Rizzieri:Celgene: Research Funding, Speakers Bureau. List:Celgene: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding, Speakers Bureau.