RATIONALE: Erythropoietin and colony-stimulating factors such as filgrastim stimulate the production of blood cells. It is not yet known whether erythropoietin with or without filgrastim is more effective than standard blood transfusions in reducing the need for transfusions in patients who have anemia associated with myelodysplastic syndrome.

PURPOSE: Randomized phase III trial to compare the effectiveness of erythropoietin with or without filgrastim with that of standard blood transfusions in reducing the need for transfusions in patients who have anemia associated with myelodysplastic syndrome.

OBJECTIVES:

• Compare the benefit of erythropoietin vs standard transfusion support in reducing transfusion requirements in patients with myelodysplastic syndromes.
• Compare the clinical response, disease progression, and survival in patients treated with these regimens.
• Compare the toxicity of these regimens in these patients.
• Evaluate whether adding filgrastim (G-CSF) or increasing the erythropoietin dose will reduce the transfusion requirement in patients who do not respond to erythropoietin alone.
• To compare the benefit of erythropoietin versus supportive care alone on quality of life (QOL) in persons with myelodysplastic syndromes.

OUTLINE: This is a randomized, controlled, multicenter, cross-over study. Patients are stratified according to morphologic subtype (refractory anemia [RA] vs RA with ringed sideroblasts vs RA with excess blasts), transfusion requirement (yes vs no), prior erythropoietin treatment (yes vs no), and erythropoietin level (at least 200 mU/mL vs less than 200 mU/mL). Patients are randomized to one of two treatment arms.

• Arm I (standard transfusion support): Patients receive red cell and platelet transfusions for symptoms or to maintain hematocrit level of 25% or above. Patients undergo bone marrow aspirate and biopsy at 4 months and then every year until development of acute leukemia or completion of study. Patients with progressive disease may cross over to arm II after at least 4 months on study and up to 1 year from the time of randomization. Patients who cross over receive erythropoietin alone.
• Arm II (Erythropoietin): Patients receive erythropoietin subcutaneously (SC) or intravenously (IV) daily. Patients undergo bone marrow aspirate and biopsy as in arm I. Treatment continues daily for a maximum of 1 year.

Patients with stable or progressive disease at day 120 receive filgrastim (G-CSF) SC daily or 3 days a week and erythropoietin SC daily for up to 6 months. Patients with no response to G-CSF and lower-dose erythropoietin may proceed to a higher dose of erythropoietin.

Quality of life is assessed at baseline, every 4 months during study, and at study completion.

Patients are followed every 4 months for 2 years, every 6 months for 3 years, and then annually for 5 years.

ACTUAL ACCRUAL: A total of 118 patients were accrued for this study.

Inclusion Criteria:

• At least 18 years of age
• Diagnosis of a myelodysplastic syndrome
• Refractory anemia (RA)
• RA with ringed sideroblasts
• RA with excess blasts (RAEB). RAEB patients must have a bone marrow blast count of less than 20% and less than 5% blast forms on peripheral blood
• Eastern Cooperative Oncology Group (ECOG) Performance Status of 0 to 3
• Platelet count greater than 30,000/mm^3 (without platelet transfusions)
• Hematocrit less than 30% (pretransfusion)
• Bilirubin less than 3 mg/dL
• Blood urea nitrogen (BUN) less than 40 mg/dL or Creatinine less than 2.0 mg/dL
• Prior epoetin alfa allowed provided dosage was less than 30,000 units per week for less than 1 month duration
• At least 1 month since prior erythropoietin
• At least 2 months since prior recombinant growth factor
• At least 2 months since prior chemotherapy for other malignancy or autoimmune disease
• At least 2 weeks since prior androgen or steroids for treatment of myelodysplastic syndromes

Exclusion Criteria:

• RAEB in transformation
• Chronic myelomonocytic leukemia
• Splenomegaly greater than 6 cm below the left costal margin or greater than 3 times normal size
• Uncontrolled hypertension
• Sensitivity to E. coli-derived proteins
• Sensitivity to epoetin alfa or any of its components (e.g., human albumin)
• Documented iron deficiency. If marrow iron stain is not available, the transferrin saturation must be greater than 20% or ferritin greater than 100 ng/dL
• Active infection or bleeding
• Other uncontrolled malignancy
• Pregnant or nursing. Fertile patients must use effective contraception.