Acute myeloid leukemia (AML) is a disease of older adults, with a median onset age at about 65-70 years. The treatment outcome of AML appears to be poorer with the age increasing, in part due to the poor performance status, concomitant end-organ dysfunction, higher incidence of unfavorable cytogenetic findings, frequent involvement of a more immature leukemic precursor clone, multidrug resistance mediated by MDR1/P-glycoprotein, and the presence of antecedent hematopoietic disorders. Treatment of elderly patients with AML remains highly challenging and controversial. The overall survival rates at 5-year of ≥ 60 years old AML patients are still less than 20% by now. At present, standard-dose induction chemotherapy using a cytarabine plus idarubicin(IA regimen) or daunorubicin (DA regimen) was considered by most to be the most effective upfront AML induction therapy. However, there are still quite a number of elderly patients could not tolerate because of poor performance status and complications. Therefore, low-intensity chemotherapy, including CAG regimen, which combine low-dose cytarabine, aclacinomycin and granulocyte colony-stimulating factor(G-CSF), was used for those were not appropriate for receiving standard-dose chemotherapy. Here we retrospectively analyzed the outcome and prognosis of elderly patients with AML treated with standard-dose or low-intensity induction therapy. 248 elderly patients with acute myeloid leukemia(AML) who received standard-dose or low-intensity induction therapy were enrolled in this retrospective clinical study, 186 patients in standard-dose group with 144 in IA and 42 in DA, 62 cases in low-intensityCAG group. The maininclusion criteria included age ≥ 60 years old, ECOG performance status ≤ 2, without severe complication of heart, liver, kidney or other important organ. The patients received standard-dose or low-intensity induction regimen according to their performance status and patient preference. 144 patients received IA regimen(idarubicin 10mg/m2/d ,d1-3; cytarabine 100mg/m2, q12h, d1-5 or 7), 42 patients received DA regimen(daunorubicin 60mg/m2/d, d1-3; cytarabine 100mg/m2 q12h, d1-5 or 7), and 62 patients received CAG regimen (cytarabine 10mg/m2 q12h, d1-14; aclacinomycin 20mg qd, d1-4; G-CSF 200ug/m2 qd, d0-14, or until bone marrow recover). The median survival time was 9.2 months. 1-year , 3-year and 5-year overall survival(OS) were 42.2%, 18.9% and 13.5%, respectively. After first induction cycle, complete remission(CR) rate was 49.3% in IA group, 35.7% in DA group and 32.3% in CAG group (P = 0.046). The median OS for IA, DA and CAG group were 10.0 months, 9.7months and 7.5 months, respectively. The early mortality of induction therapy and recurrence rate of three regimens showed no difference. IA could improve the long term survival compared to CAG and DA, with 3-year OS: 23.5%,15.9% and 8.3%, respectively; 5-year OS: 19.4%, 6.3%, and 0, respectively (P<0.01). The 67.0% patients relapsed within 24 months, with median relapse time of 8.4 months, 14.6 months and 8.3 months for IA, DA and CAG regimen, respectively. Moreover, Kaplan-Meier analysis showed that 7 parameters were adverse prognostic factors for OS, including age ≥ 70 years old, poor ECOG performance status, unfavorable cytogenetics, non-remission after first induction cycle, white blood cell (WBC) counts ≥ 50×10^9/L, percentage of bone marrow (BM) blast ≥ 80% and higher lactic dehydrogenase (LDH) . Multivariable analysis identified non-remission after first induction cycle (HR = 6.141, 95%CI: 3.585-10.52, P = 0.000) and LDH ≥ 490 IU/L(HR = 1.001, 95%CI: 1.000-1.001, P = 0.000) as independent significantly prognostic factors for OS. In conclusion, Our present data showed that standard-dose IA regimen could improve CR rate and prolong the survival time compared to low-intensity CAG regimen, and CAG regimen still has a certain therapeutic effect for those unfit for intensive chemotherapy. Recurrence is still a serious problem for those who do not receive Allo-HSCT for consolidation after CR. All prognostic factors should be considered before induction therapy to make sure the patients receive the best individualized treatment.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.