Hormones, such as glucocorticoids and androgens, are generally of little benefit to patients with MDS.
Recombinant myeloid growth factors such as granulocyte-macrophage colony-stimulating factor (GM-CSF) have been studied in myelodysplasia. Circulating granulocytes usually increase in a dose-dependent manner during therapy with GM-CSF but usually return to pretreatment levels when the agent is discontinued. Platelet and reticulocyte counts usually do not respond. The effect of GM-CSF treatment on infection rate, morbidity, mortality, and disease progression is not yet known.[14,15,16,17] Some patients respond to GM-CSF with increased circulating blasts. A randomized trial in which granulocytopenic MDS patients were assigned to GM-CSF or observation showed no advantage to the prophylactic use of that cytokine.
The nucleoside 5-azacitidine is an inhibitor of DNA methyltransferase. Following a series of phase II studies suggesting significant activity of 5-azacitidine in patients with MDS, a randomized trial was conducted by the Cancer and Leukemia Group B. In the trial, 191 patients were randomized to receive 5-azacitidine (75 mg/m2 /day subcutaneously daily for 7 days every 28 days) or observation. The antineoplastic nucleoside 5-azacitidine was continued for a minimum of four cycles. Patients who did not improve on the observation arm crossed over to receive 5-azacitidine. Hematologic responses occurred in 60% of patients on the 5-azacitidine arm (7% complete response, 16% partial response, and 37% improved response) compared with 5% in the observation arm (responses in the latter group consisted of hematologic improvement associated with an increased white cell count caused by the progression to acute leukemia). The response data of the original Cancer and Leukemia Group B trial have been reanalyzed in the CLB-8421, CLB-8921, and CLB-9221 studies, respectively, using the International Working Group MDS Response Criteria; total hematologic response rate was 47% and included 10% complete responses. Median time to leukemic transformation or death was 21 months for azacitidine versus 13 months for supportive care (P = .007). Median duration of response was 15 months, and fewer than 1% of treated patients died on study. Quality-of-life assessment found significant major advantages in physical function, symptoms, and psychological state for patients initially randomized to azacitidine.[20,21,22][Level of evidence: 1iiDii]
Results have been reported from a phase III randomized controlled trial (AZA PH GL 2003 CL 001 [NCT00071799]) of 5-azacitidine versus other regimens. The other regimens included low-dose cytarabine, acute myeloid leukemia-type remission induction chemotherapy, or best supportive care, and the trial was limited to patients with higher-risk MDS subtypes. The median and 2-year overall survival (OS) favored the 5-azacitidine arm, at 24 months versus 16 months (P = .0001) and 51% versus 26% (P < .0001), respectively.[Level of evidence: 1iiA]
The azacitidine congener decitabine demonstrated similar activity in phase II trials with an overall response rate of 49% and a median response duration of 39 weeks. A randomized trial of decitabine versus supportive care in patients with International Prognostic Scoring System (IPSS) Int-1 or greater led to an overall response rate of 30% in the decitabine arm versus 7% in the observation arm (P < .001). Median time to acute myeloid leukemia (AML) or death was 12.1 months in the treated arm versus 7.8 months in the supportive care arm (P = not significant). Fourteen percent of patients treated with decitabine died on the study. Quality-of-life assessment found advantages to decitabine similar to those of 5-azacitidine. Although considerably fewer than originally planned, the median number of cycles administered was three, the decrease possibly attributable to the toxicity of the dose schedule studied[Level of evidence: 1iiDii]