Comparison of DNA Methyltransferase Inhibitors in Patients with Myelodysplastic Syndrome

The aim of this retrospective study was to compare the response rates (RR), overall survival (OS), time–to–acute myeloid leukemia–transformation (TTAMLT), hematologic toxicity and transfusion requirement of azacitidine and decitabine in patients with Myelodysplastic Syndrome Refractory Anemia with Excess Blasts (MDS RAEB) types 1 and 2. A total of 51 patients (35 male, 16 female, median age (years): 67) diagnosed with MDS RAEB-1 and RAEB-2 treated with azacitidine (n=31) or decitabine (n=20) in three Turkish institutes were evaluated . Response to treatment was classified according to International Working Group response criterias. Competing risk analysis was used to compare TTAMLT. There was no significant difference between the two groups in terms of age, gender, performance score, diagnosis, etiology, treatment cycles, bone marrow cellularity and blast percentage. Most patients received at least 4 cycles of azacitidine or decitabine. Patients who received azacitidine or decitabine had comparable response rates (58.1% vs 66.7%, respectively, p= 0.55). There was no difference in terms of red blood cell and platelet transfusion requirement and febrile neutropenia episodes (p= 0.46, p= 0.27, p= 0.19, respectively). Median follow–up was 31 months for azacitidine and 20 months for decitabine. Median OS was 18.4 months for azacitidine and 13.4 months for decitabine (p= 0.16). Only predictor of OS was at least 4-cycle treatment with any DNA methyltransferase inhibitors. 1–year risk of AML transformation was comparable across the groups (26% vs 33%, p= 0.55). In conclusion, both azacitidine and decitabine have similar efficacy and toxicity profiles in the treatment of MDS RAEB-1 and RAEB-2. Until a head–to–head comparison in prospective, randomized studies are conducted, the current source of available data will derive from meta–analyses that consist of the indirect comparison of treatment arms, or retrospective analyses with inevitable biases.