Effects of a formulary change from granulocyte colony-stimulating factor to granulocyte-macrophage colony-stimulating factor on outcomes in patients treated with myelosuppressive chemotherapy

Study Objective. To evaluate the effects on efficacy and safety of a formulary change from granulocyte colony-stimulating factor (G-CSF) to granulocyte-macrophage CSF (GM-CSF). Design. Retrospective chart review. Setting. Single-center academic institution. Patients. Fifty-six patients aged IS years or older with breast cancer, lung cancer, melanoma, Hodgkin's lymphoma, or non-Hodgkin's lymphoma who developed neutropenia within 4 weeks after treatment with myelosuppressive chemotherapy and who had been given five or more doses of CSF as primary or secondary prophylaxis from January 1995-March 2002. Twenty-nine patients treated before January 2000 were given G-CSF-1 after the formulary change in January 2000, 27 patients were primarily given GM-CSF Measurements and Main Results. The primary efficacy end point was time to an absolute neutrophil count of 1.5 x 10(3)/mm(3) or greater after treatment with CSF Second and third efficacy end points, respectively, were frequency of febrile neutropenia and effect of CSF treatment on schedule and dose intensity of subsequent chemotherapy cycles. Primary and secondary safety end points, respectively, were frequency of adverse events and use of resources used to manage these events. The time to neutrophil recovery was similar with G-CSF and GM-CSE Febrile neutropenia was more common in the patients given GM-CSF Chemotherapy dose delays also were more common in patients treated with GM-CSF, as was the frequency of fever. Use of resources (platelet and red blood cell transfusions, intravenous antibiotics, and hospitalizations) was greater in the patients treated with GM-CSF. Conclusion. The formulary change to GM-CSF was associated with a higher frequency of febrile neutropenia, resultant chemotherapy dose delays, more adverse events, and greater use of resources to manage the adverse events. These results suggest that G-CSF and GM-CSF are not therapeutically equivalent, with G-CSF having a superior safety and efficacy profile for the prevention of chemotherapy-induced neutropenic events.