Efficacy and toxicity of Decitabine in patients with acute myeloid leukemia (AML): A multicenter real-world experience

BACKGROUND: The hypomethylating agent Decitabine (DAC) is a valuable treatment option in acute myeloid leukemia (AML), particularly in elderly patients (pts) not suitable for intensive chemotherapy (CHT). However, limited data are available about efficacy and safety of DAC in clinical practice. PATIENTS AND METHODS: We retrospectively reviewed data of 104 AML pts treated with DAC in eight Italian Hematological Centers from 2015 to 2017. The objective of this study was to evaluate the efficacy and safety of DAC in older AML pts outside of clinical trial. Seventy-five (75%) pts received DAC as first line treatment (Cohort 1) and 29 pts as salvage therapy (Cohort 2). All pts received a DAC schedule of 20 mg/sqm IV for 5-days, every 28 days. The median age was 72.5 years (74 in cohort 1 and 66 in cohort 2) and 16% of pts had an ECOG performance status >2 at the start of DAC treatment (with non-significant difference in the two cohorts). The cumulative illness rating scale (CIRS) was > 6 in 27% of pts. Forty-five pts (43%) had secondary AML. Bone marrow blast count was > 30% in 64% of patients (67/104). In the relapsed cohort 17/29 (59%) patients were treated with DAC after conventional CHT, 5/29 (17%) after allo-SCT and 7/29 (24%) after azacitidine therapy. RESULTS: A total of 469 DAC cycles were given to the 104 pts with a median of 3 cycles (range 1-21) and 45/104 (43%) pts received > 4 cycles. The Overall Response Rate (ORR = Complete Remission-CR plus Partial Remission-PR) was 33%, significantly higher in Cohort 1 (42%) compared to Cohort 2 (14%) (p = 0.009). The median duration of response was 6 months (range 1-20). In Cohort 1 the best response (CR or PR) was obtained between 3th and 6th cycle. In multivariate Cox regression analysis, achievement of CR or PR (HR = 0.78; p = 0.0004), CIRS < 6 (HR = 0.9; p = 0.04) and complex karyotype (HR = 0.8; p = 0.03) were significant predictors of better overall survival (OS). Median OS from the start of DAC therapy was 11 months for the whole population with a significant OS advantage in Cohort 1 (median OS 12.7 mths vs 6.3 mths; p = 0.003); median OS was significantly longer in responders compared to non-responders (22.6 mths vs 5.7 mths; p < 0.0001). At the last follow-up, 56 patients (54%) are still alive and 48 (46%) are dead (71% due to disease progression). The most common toxicities were myelosuppression and documented infectious complications that occurred mainly during the first 4 cycles. CONCLUSION: These data confirm the efficacy (ORR 33%) and the acceptable safety profile of DAC in the real life management of AML in elderly pts unsuitable for intensive CHT, with a significant better performance in first line therapy (ORR 42%, median OS 12.7 mths). The efficacy of DAC, both in first line and as salvage therapy, may probably be improved with combined treatment strategies and/or with different DAC schedules that could increase its anti-leukemic effect.