Impact of Post-Remission Therapy on Survival of Older Patients with AML Treated with Intensive Chemotherapy: Results of the Multicenter ALFA-9803 Trial.

Background: More than 50% of newly-diagnosed patients with AML are older than 60 years of age. In this age group, the benefits of intensive chemotherapy are debated, due to excessive toxicity and short duration of response. Negative impacts of unfavorable cytogenetics and associated comorbidities are assumed to explain these results. Prolonged survival may, however, be observed in some patients even there is no well-established post-remission therapy at the present time.

Methods: In the randomized ALFA-9803 trial in AML patients aged 65 years or more, we randomly compared idarubicin or daunorubicin throughout the study (R1) and two different post-remission strategies (R2): one single intensive consolidation course similar to induction versus six ambulatory cycles with one dose of idarubicin/daunorubicin (day 1) and 2x60 mg/m²/d cytarabine SC (day 1 to 5) delivered in out-patients on a monthly basis. Primary endpoint was 2-year overall survival (OS). From Nov 1999 to Apr 2006, 416 patients with de novo (#353) or post-MDS AML (#63) were included (median actuarial follow-up, 34 months). Median age was 72 years. High-risk cytogenetics was defined as complex (5 abns or more), monosomy 7, chromosome 5 and 7 abns, or 3q abn. Associated comorbidities were scored prospectively.

Results: Complete remission (CR) rate was 57%, with 4% CRi and 10% induction deaths (ID), after a 4+7 induction with either 45 mg/m²/d daunorubicin or 9 mg/m²/d idarubicin (day 1 to 4) and cytarabine 200 mg/m²/d CIV (day 1 to 7), followed by lenograstim G-CSF until myeloid recovery. Salvage with intermediate-dose cytarabine and mitoxantrone was used in 54 patients. By multivariate analyses, only high comorbidity score (p=0.001) and advanced age (p=0.02) predicted ID. High-risk cytogenetics (p<0.001 and <0.001), high comorbidity score (p=0.001 and <0.001), high WBC (p=0.05 and <0.001), and advanced age (p=0.13 and =0.003) predicted lower CR rate and shorter OS, respectively. Median OS was only 4 months in a subset of about 20% of patients (high-risk cytogenetics and/or high comorbidity score and/or age 80+) who could thus be considered as “unfit” for intensive chemotherapy, even if meeting all usual selection criteria for an intensive approach. Median OS of the remaining 80% patients was 15 months. Long-term outcome, CR and ID rates did not differ according to R1. In the 164/236 CR patients randomized for R2 (69%), estimated 2-year OS from CR was 55% (95% CI, 43 to 66) in the ambulatory arm as compared to 36% (95% CI, 18 to 43) in the intensive arm (p=0.045). Remission duration was similar among both consolidation arms, while incidence of death in CR was lower in the ambulatory arm (1% versus 9% at 2 years).

Conclusion: These results indicate that an ambulatory post-remission strategy has an anti-leukemic efficacy similar to intensive consolidation in older patients with AML and could become a reference arm to which alternative strategies incorporating new agents should be compared. In addition, such an “out-patient” strategy is easier to apply in most patients who have reached CR after intensive induction.