Effects of treatment with 5-azacytidine on the in vivo and in vitro hematopoiesis in patients with myelodysplastic syndromes

The myelodysplastic syndrome (MDS) comprises a group of clonal hematopoietic disorders derived from an abnormality affecting a multipotent hematopoietic stem cell. Despite trials testing numerous agents in patients with MDS, no single drug has yet emerged as the accepted standard of treatment. Observation and supportive care with blood products and antibiotics, when necessary, continue to be the mainstays of therapy. We administered 5-azacytidine, a cell-cycle specific ring analog of the pyrimidine nucleoside cytosine, as a continuous intravenous infusion, 75 mg/m² per day for 7 days every 4 weeks. Patients had refractory anemia with excess blasts (RAEB) or refractory anemia with excess blasts in transformation (RAEB-T). Responses were seen in 21 (49%) of 43 evaluable patients: five (12%) in complete remission (CR, complete normalization of bone marrow and peripheral blood counts); 11 (25%) in partial remission (PR, ≥ 50% restoration of the deficit from normal of all three peripheral blood cell lines, elimination of transfusion requirements, and a decrease in percentage bone marrow blasts by ≥ 50% from prestudy values); five (12%) improved (≥ 50% restoration in the deficit from normal of one or more peripheral blood cell lines and/or a ≥ 50% decrease in transfusion requirements). A trilineage improvement (CR and PR) occurred in 37% of the patients. The median survival for all patients was 13.3 months and the median duration of remission for those with CR and PR was 14.7 months. Mild to moderate nausea and/or vomiting was the most common side effect (63%). Myelosuppression, either bone marrow hypoplasia or drug related cytopenias requiring a reduction In the dose of azacitidine, occurred in only 33% of the patients. Prior to treatment, bone marrow erythroid progenitor cells were assayed in vitro. Colonies derived from erythroid burst-forming units (BFU-e) were undetectable in one patient and reduced in two. The number of colonies derived from erythroid colony-forming units (CFU-e)) were also reduced in two of the three patients. In the two patients with detectable colony growth prior to treatment, colony number decreased by day 8 of the first cycle, followed by a subsequent increase. Continued treatment with azacitidine led to normalization of the number of CFU-e derived colonies as well as an increase in the number of BFU-e derived colonies. This improvement in erythroid colony number correlated with the spontaneous rise in hemoglobin levels and red cell transfusion independence. Subsequent decrease in colony number (CFU-e, BFU-e) preceded a progressive fall in hemoglobin levels by 2 to 4 months. Azacitidine is an effective agent for the treatment of patients with RAEB and RAEB-T. The trilineage response rate is higher than those reported with the hematopoietic growth factors erythropoietin, interleukin-3, granulocyte-macrophage or granulocyte colony-stimulating factors (GM-CSF or G-CSF). Further trials testing azacitidine in a regimen administered in an ambulatory setting have been initiated.