Abstract
Introduction
The standard first line treatment for acute myeloid leukemia (AML) for patients eligible for intensive chemotherapy is a combination of cytarabine and anthracyclines called "3+7" regimen. However, some patients are ineligible because of cardiac comorbidities or prior exposure to anthracyclines in the case of therapy-related AMLs. To date, there is no consensual regimen for these patients. In this study, we retrospectively analyzed the results of 3 cytarabine-based, anthracycline-free, regimens in which topotecan, clofarabine or fludarabine were used.
Methods
From 2008 to 2016, 52 AML patients aged ≥18 years with comorbidities were treated intensively in 3 French centers (CHU of Nice, Antoine Lacassagne Center and Paoli-Calmettes Institute). All patients had new diagnosed or relapsed/refractory (R/R) AML (APL were excluded) and at least one contra-indication to receive "3+7" regimen. 3 types of regimen were used for induction. FLAG regimen combined fludarabine (days 1-4, 30 mg/m²/d), cytarabine (days 1-4, 2000mg/ m²/d) and G-CSF (day 0 up to ANC > 1 x 109/l, 400 µg/m²/d). CLARA regimen combined clofarabine (days 1-3(or 5), 30mg/m²/d) and cytarabine (days 1-7 (or 10), 20mg/m²/d). TA regimen combined topotecan (days 1-4, 1.25 mg/m²/d) and cytarabine (days 1-4, 1000 mg/m²/12 h days 1-4). Responses were scored according to ELN 2010. Overall (OS), Event free survival (EFS) were censored to allo stem cell transplantation (ASCT). All statistical analyses were performed using SPSS v.22 software (IBM SPSS Statistics).
Results
Among 52 AML patients, sex ratio M/F 27/25, median age was 66 years (20-79, range). 50/52 patients had performans status £ 2. 30 (58%) and 22 (42%) patients had de novo or relapsed/refractory AML, respectively. ELN risk categories were favorable, intermediate and adverse in 4 (8%), 27 (52%) and 20 (39%) patients, respectively. 7/28 patients were FLT3-ITD and 4/28 were NPM1 mutated. 24 and 28 patients were treated by TA or FLAG/CLARA regimens, respectively. Main comorbidities leading to use of TA/FLAG/CLARA regimens were cardiac in 50% and prior anthracyclin exposure above MTD in 50% of patients. 56% achieved complete remission after induction (74% in TA and 40% in FLAG/CLARA, p=0.11 in newly diagnosed AML; and 25% in TA and 53% in FLAG/CLARA, p=0.59 in R/R AML). Although duration of neutropenia was similar in both groups, (25 vs 26 days in TA and FLAG/CLARA subgroups, respectively, p=0.62), we observed a significant difference for duration of thrombopenia less than 50 Giga/l (24 vs 34 days in TA and FLAG/CLARA subgroups, respectively, p=0.03). Median OS was no significant different between the 2 groups (10.2 months [1.1-19.2] vs 9.8 months [1.1-22.1] in TA and FLAG/CLARA subgroups, respectively, p=0.59). There was a trend for a better median EFS in TA subgroup (8.8 months [5.6-11.9] vs 1.6 months [3.6-9.2] in TA and FLAG/CLARA subgroups, respectively, p=0.09) which reached statistical significance after censoring for ASCT (16.1 months [8.4-23.7] vs 4.7 months [4.3-5.1] in TA and FLAG/CLARA subgroups, respectively, p=0.001). This effect was marked only in new diagnosed AML group (16.1 months [11.2-20.9] vs 5.9 months [4.3-7.6], in TA and FLAG/CLARA subgroups, respectively, p=0.03). This effect was confirmed in the unfavorable ELN-risk group (17.8 months [1.1-32.5] vs 4.1 months [2.1-6.1] in TA and FLAG/CLARA subgroups, respectively, p=0.006). Finally, we observed no significant difference for cardiac toxicity between TA and FLAG/CLARA subgroups (29% vs 23%, respectively, p=0.78).
Conclusion
TA, FLAG and CLARA regimens are efficient strategies in AML patients with comorbidities ineligible for classical "3+7" regimen. Safety was acceptable for all regimens. TA showed a better EFS in newly diagnosed AML and adverse ELN-risk and need to be confirmed in a larger study, especially in this subgroup of patients with cardiac comorbidities.
Charbonnier: Novartis: Consultancy, Speakers Bureau; Incyte: Consultancy, Speakers Bureau; Pfizer: Consultancy, Speakers Bureau; BMS: Consultancy, Speakers Bureau.
Author notes
Asterisk with author names denotes non-ASH members.