Economic evaluation of prophylactic granulocyte colony stimulating factor during chemotherapy in elderly patients with aggressive non-Hodgkin's lymphoma.

BACKGROUND AND OBJECTIVES: Treatment with CHOP chemotherapy in elderly patients with aggressive non-Hodgkin's lymphoma (NHL) is less effective and accompanied by more adverse effects than in younger patients. The prophylactic use of granulocyte colony-stimulating factor (G-CSF) might improve the results, but increases the costs of treatment. We analyzed the costs of therapy and follow-up of patients with NHL treated with CHOP with or without G-CSF prophylaxis.

DESIGN AND METHODS: Four hundred and eleven patients were randomized for treatment with CHOP or CHOP+G-CSF. A detailed study of treatment costs from randomization until 3 years of follow-up or death was performed in a subset of 100 out of 389 eligible patients. Because costs during follow-up were independent of the use of G-CSF during treatment, costs of follow-up and second-line treatment were calculated irrespective of the treatment arm.

RESULTS: Total hospital costs for first-line treatment were 12178 [95% CI 10297 - 14059] for CHOP alone and 18356 [95% CI 15807 - 20906] for CHOP + G-CSF. Costs during follow-up showed a wide difference (range 74 - 53925) depending on disease status and choice of treatment in the case of relapse or progression.

INTERPRETATION AND CONCLUSIONS: The clinical study showed no difference between the treatment arms in response, overall survival or event-free survival, while the costs were significantly higher in the G-CSF arm. We conclude that the addition of prophylactic G-CSF to CHOP chemotherapy is not cost-effective in these patients.