Hypometilating Agents (HMA) As Salvage Therapy in Relapsed or Refractory AML: A 5-Center Retrospective Study

Introduction: 5-azacytidine and decitabine have been widely studied as first line chemotherapy in acute myeloid leukemia (AML) patients not eligible for allogenic stem cell transplantation, but data on their use as salvage chemotherapy are limited. We tried to define efficacy and feasibility of hypometilating agents (HMA) as salvage chemotherapy in patients without previous allogenic stem cell transplantation.

Methods: We retrospectively reviewed clinical records of 35 patients treated with HMA as salvage therapy in our 5 institutions since their introduction in clinical practice for AML patients.

Results: Median age was 68 years. 22 patients were men and 13 women. One patient was AML with t(9;11), 18 were AML MRC, 4 were therapy related AML, 12 were AML NOS. Two patients were favorable risk sec ELN 2017, 20 were intermediate, 11 were adverse risk. Cytogenetic risk was not available for 2 patients. 66% of patients received HMA as second line therapy for their disease, 23% as third line and 11% were beyond the third line. 29 patients were treated with azacitidine and 6 with decitabine. Six patients reached CR or morphologic leukemia free state (MLFS) or PR after HMA. All patients underwent intensive chemotherapy (i.e. FLAI or 3+7 like) as first line induction, and we excluded patients who had an HMA as first line chemotherapy and another one as second line. Median number of hospitalization days during HMA therapy was 11(1-78); median number of HMA cycles was 3 (range 1-31). 7 episodes of sepsis demonstrated by positive blood cultures and 3 fungal infections were recorded during HMA therapy. Median OS was 197 days from the starting of HMA and median EFS was 125 days. Median EFS in patients with adverse cytogenetic risk was 68 days, while median EFS in patient with favorable and intermediate risk was 207 days (p=0,0124). Median OS in patients with adverse cytogenetic risk was 181 days, while median OS in patient with favorable and intermediate risk was 307 days (p=0,0401).We did not find significant difference between median OS ed EFS in patients with refractory disease compared to relapsed patients. 31% of patients with adverse cytogenetic risk reached a stable disease with HMA therapy while 64% of patients with favorable/intermediate cytogenetic risk reached a response (CR, PR; MLFS) or a stable disease. We did not find significant differences between transfusion needs before and after salvage therapy but this could be due to the small size of our sample.

Conclusion: HMA showed efficacy and a considerable OS in our patients. OS was significantly better in favorable/intermediate risk patients than adverse risk patients. Hospitalization was limited, and we can argue that this could lead to improved costs and quality of life, but further studies are needed. So HMA could be a good clinical option in a selected population of relapsed patients, especially in those not suitable for allogenic bone marrow transplantation, in whom the prognosis is generally extremely poor. Further studies are needed to define if there is a survival or quality of life advantage over other therapies.