BACKGROUND: Anthracycline based treatment for acute myeloid leukemia (AML) can be associated with significant morbidity and mortality amongst older patients or those with significant co-morbidities. Furthermore, for patients with previous anthracycline exposure or preexisting cardiac disease anthracyclines pose an increased risk of cardiotoxicity. For such patients intensive treatment options are limited.

The FLAG (fludarabine, cytarabine and filgrastim) regimen is a non-anthracycline based chemotherapy useful for relapsed/refractory AML and as initial therapy for “fit” eligible patients. We present our experience using FLAG as first line therapy in a cohort of newly diagnosed AML patients with advanced age, significant co-morbidities or preexisting cardiac disease.

METHODS: A single institution retrospective chart review was undertaken of patients treated with FLAG as frontline therapy from 2004 – 2013 with follow-up until June 2014. All patients were considered ineligible for standard ‘3+7’ treatment by the attending physician.

RESULTS: Over the study period 56 patients received FLAG. Due to patient refusal or induction complications bone marrow assessments to ascertain remission status were not performed in 10 patients – to minimize bias these patients were evaluated for toxicity and overall survival (OS) but excluded from complete remission (CR) and relapse free survival (RFS) analysis.

Of the 56 patients, 68% were males. The median age was 69 (21 – 80) with 79% aged ≥ 60 and 43% aged ≥ 70. Fifty-five percent (31) of the patients had primary AML and cytogenetics were favorable in 5% (3), intermediate in 66% (37), poor-risk in 21% (12) and failed in 7% (4). Forty-six percent (26) were treated with FLAG due to preexisting cardiac disease and others due to advanced age (20%), poor performance status (16%), co-morbidities (16%) or previous anthracycline exposure (2%). The ages of patients treated with FLAG due to cardiac disease or other reasons was similar (63.5 years vs. 66.9 years, p=0.27). Amongst patients with cardiac disease a pre-chemotherapy cardiac evaluation revealed wall motion abnormalities in 39% and an ejection fraction (EF) ranging from 33% - 81% with a borderline (≤ 50-51%) EF in 46%.

Febrile neutropenia was observed in 82% (46) of patients, 27% (15) required intensive care support and the induction mortality was 16% (9). Amongst patients evaluated for a CR (46/56), 48% (22/46) obtained a CR. Although a CR was seen in 100% (3/3), 46% (15/33) and 37% (3/8) of patients with favorable, intermediate and poor-risk cytogenetics respectively there was no statistically significant difference (p=0.22) between groups. A CR was obtained in two additional patients following a second course of FLAG for an overall CR rate of 52% (24/46). Of these 24 patients, 8 (33.3%) received no additional treatment while 16 (66.7%) received consolidative chemotherapy. Of 10 eligible patients, 4 proceeded to an allogeneic stem cell transplant (allo-SCT) including two with preexisting cardiac disease.

Of patients not transplanted 50% (10/20) eventually relapsed. There was no difference in relapse rates for those receiving FLAG for age/co-morbidities vs. cardiac disease (60% vs. 40%, p= 0.66) or for patients age <60 vs. ≥ 60 (50% vs. 50%, p=1.0). The median RFS was 441 days with no difference between patients treated with FLAG for cardiac reasons vs. other reasons (p=0.33) or for patient’s age < 60 vs. ≥ 60 (p=0.66).

The median OS for the 56 patients was 278 days with 1 and 2 year survivals of 44% and 22% respectively. The median OS for patient’s age < 60 was higher than those age ≥ 60 (1102 days vs. 218 days, p=0.009) but there was no difference in the OS between patients treated with FLAG for cardiac disease and those treated for other reasons (305 days vs. 254 days, p=0.202). The median survival of patients with favorable, intermediate and unfavorable risk cytogenetics was 523 days, 311 days and 87 days respectively but this was not statistically significant (p=0.25) likely due to limited patient numbers.

Conclusions: The FLAG regimen is a non-anthracycline based regimen that may serve as an alternative to the standard ‘3+7’ induction for AML in older adults or those with significant co-morbidities including preexisting cardiac disease. It is associated with comparable remission rates and overall survival in older patients. In addition, it may allow some patients with preexisting cardiac disease to proceed to allo-SCT.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.