Induction chemotherapy versus palliative treatment for acute myeloid leukemia in a consecutive cohort of elderly patients

A retrospective survey of 210 consecutive patients aged ≥65 years (median age 69 years, range 65–88 years) with acute myeloid leukemia (AML) diagnosed at a single center over a 6-year period (January 2001 to December 2006) is presented. De novo AML was diagnosed in 179 (85.2 %) patients and 31 (14.7 %) patients had a secondary AML. Twenty-three patients had M0 (11 %), 36 M1 (17.15 %), 57 M2 (27.1 %), eight M3 (3.8 %), 45 M4 (21.4 %), 31 M5 (14.8 %), one M6 (0.5 %), one M7 (0.5 %), and eight patients had unclassified myeloid leukemia (3.8 %) according to French-American-British (FAB) Study Group Classification. Eight patients with M3 (acute promyelocytic leukemia) were excluded from the study. Cytogenetic analysis was performed in 172/202 (85 %) patients. The normal karyotype was found in 81/172 (47 %), high risk aberrations in 32/172 (18.6 %), and favorable karyotype in 13/172 (7.5 %) patients. Supportive and palliative therapies were applied in 115 (56.9 %) patients, a no induction chemotherapy (NIC) group, and 87 (43.1 %) patients received induction chemotherapy (IC group). Complete remission (CR) was achieved in 45/87 (51.7 %) in the IC group and in 5/115 (4.3 %) in the NIC group of patients. After a median follow up of 4 years, 194 (96 %) patients died. The variables significantly associated with a longer overall survival (OS) by univariate analysis were an age of <75 years, a better ECOG performance status (PS) (p = 0.000, CI 95.0 %, 1.358–2.049), a serum LDH activity <600 U/l (p = 0.000, CI 95.0 %, 1.465–2.946), lower white blood cell (WBC) count at diagnosis (p = 0.011, CI 95.0 %, 1.102–2.100), lower comorbidity HCT-CI index (p = 0.000, CI 95 % 2.209–3.458), absence of splenomegaly (p = 0.015, CI 95.0 %, 1.082–2.102) and hepatomegaly (p = 0.008, CI 95.0 %, 1.125–2.171), and no preceding nonhematological malignancy. Multivariate analysis showed that significant factors affecting OS in the IC group were achievement of CR (p = 0.000), the ECOG PS (p = 0.045) and the ECOG PS (p = 0.000), and HCT-CI (p = 0.000) in the NIC group of elderly patients. The present study suggests that a subgroup of elderly patients with both ECOG PS and HCT-CI ≤ 2 at presentation may be eligible for intensive induction chemotherapy.