Publication Only: Acute myeloid leukemia - Clinical
“3+7” represents the current standard of care for acute myeloid leukemia (AML). Fludarabine-based regimens have limited application due to the absence of survival benefit and prolonged hematological toxicity after 2nd cycle. Infections in neutropenic patients represent an important cause of mortality and morbidity.
To analyze the infective risk associated with intensive chemotherapy based on double-FLAI induction.
We retrospectively collected data of 40 consecutive adult patients with newly diagnosed AML treated with FLAI-5 induction therapy (Fludarabine IV 25 mg/sqm/die days 1-5, Cytarabine IV 2000 mg/sqm/die days 1-5 and Idarubicin 10 mg/sqm/die on days 1, 3 and 5). Patients in complete remission (CR) proceeded to second FLAI-5 followed by allogeneic transplant or chemotherapy consolidation. The incidence and type of infective events were collected in 36/40 patients. Patients were compared with a historical court of 39 patients treated with non-fludarabine based induction. MRD was assessed by qRT-PCR or WT1 gene expression. Data were harmonized using RedCap custom eCRFs.
Forty patients received FLAI-5 from August 2015 to December 2018. Median age at diagnosis was 53 years (range 19-66). According to NCCN-2016 stratification criteria 12/40 (30%) patients were favorable, 10/40 (25%) intermediate and 17/40 (42.5%) high risk; one patient wasn't evaluable. Thirty-five patients (87.5%) achieved a CR after induction, 4 (10%) patients were refractory and 1 patient (2.5%) died during induction. Twenty-nine of 40 patients were evaluable for MRD analysis and 15/29 (52%) patients achieved MRD negativity. Twenty-two of the 35 patients in CR proceeded to 2nd FLAI-5 chemotherapy regimen.
At induction, patients recovered from neutropenia (ANC> 500 x 109/L) after 21.5 days (range 13-32) from the end of chemotherapy and reached platelet > 20 x 109/L after a median of 16 days (range 10-33). Patients spent a median of 33 nights in hospital (range 14-61; historical control 35 nights, range 31-42; p = ns). After 2nd cycle, median days for ANC and platelet recovery were 29 and 26 (range 16-45 and 13-60 respectively, historical control for ANC recovery 20 days, range 18-25; p = 0.001). Median patients' hospitalization was 39 nights (range 26-55; historical control 32.5 nights, range 24-47; p = 0.003).
Patients and historical controls received same antimicrobial prophylaxis in induction and experienced respectively 33/36 (91.7%) and 36/39 (92.3%) infective events. Patients had a higher number of infections of unknown origin (15/33 vs 6/36, p .002), and lower incidence of Gram-positive isolates (4/12 vs 15/22, p = 0.002). Any-mould infection rate was comparable in the two groups.
Twenty-seven of 36 patients and 22/39 controls proceeded to consolidation chemotherapy. Antimould prophylaxis was different, with patients receiving mainly posaconazole 17/27 (63 %), fluconazole 5/27 (18.5%), itraconazole 4/27 (14.8%) and voriconazole 1/27 (3.7%); while controls received posaconazole 4/22 (18.2%) fluconazole 5/22 (22.7%), itraconazole 12/22 (54.5%) and voriconazole 1/22 (4.5%). Twenty-two of 27 patients (81.5%) and 17/22 (77.3%) in controls had an infection. Three patients of 27 had a probable mould infection and 1 possible, compared to 1 proven mould infection and 1 possible in the 22 controls (any-mould infection 4/27 vs 2/22).
In our experience FLAI-5 is associated with high CR rates, but prolonged neutropenia and hospitalization after 1st consolidation. However, this doesn't translate in a higher infection risk in this set of patients.