Effect of recombinant human granulocyte-macrophage colony-stimulating factor in patients with Hodgkin's disease: a phase I/II study.

PURPOSE As bone marrow toxicity is the major limitation of the optimal administration of chemotherapy, we investigated whether recombinant human granulocyte-macrophage colony-stimulating factor (rhGM-CSF) could prevent myelotoxicity or accelerate hematopoietic recovery after mechlorethamine, vincristine, procarbazine, and prednisone (MOPP) chemotherapy. PATIENTS AND METHODS Twenty-four previously untreated patients with Hodgkin's disease were included in a phase I/II study in which standard MOPP chemotherapy was followed by 5 days of GM-CSF at every other cycle. Patients were entered sequentially to receive one of four dosc levels (2, 4, 8, and 16 micrograms/kg of glycoprotein; 1.4, 2.8, 5.5, and 11.0 micrograms/kg of protein) and were randomly allocated to either 24-hour continuous intravenous (IV) infusion or twice daily subcutaneous (SC) injection of rhGM-CSF. RESULTS WBC counts (mainly neutrophils, eosinophils, and monocytes) were significantly higher in cycles with rhGM-CSF than in cycles with MOPP alone. The total number of days of leukopenia (WBC count less than or equal to 2.0 x 10(9)/L) and neutropenia (absolute neutrophil count [ANC] less than or equal to 1.0 x 10(9)/L) was reduced in cycles with rhGM-CSF from 6.3 to 0.8 days and from 5.4 to 1.0 days, respectively. All dose levels of rhGM-CSF were effective in increasing the ANC, but only at the dose levels of 8 and 16 micrograms/kg did this significantly affect the scheduling of chemotherapy. Mild and reversible adverse reactions included low-grade fever, chest/bone pain, myalgias, erythemia, headache, fatigue, and periorbital edema. CONCLUSIONS rhGM-CSF can be administered safely to patients with Hodgkin's disease and results in improved hematologic recovery after MOPP. Full-dose chemotherapy can be administered on time, resulting in an increase in the overall tolerated dose of myelosuppressive drugs when compared with historical controls. SC administration proved to be at least as effective as continuous IV infusion and should be preferred.