Use of granulocyte colony-stimulating factor (G-CSF) in patients receiving myelosuppressive chemotherapy for the treatment of cancer. Provincial Systemic Treatment Disease Site Group.

Guideline questions: 1) Does G-CSF reduce the incidence of important adverse clinical outcomes due to infections in patients with cancer treated with myelosuppressive therapy? 2) Does G-CSF allow maintenance of the chemotherapy dose with the goal of improving survival?

Objective: To evaluate the evidence for the role of G-CSF in patients receiving myelosuppressive chemotherapy for the treatment of cancer.

Outcomes: Clinical outcomes reflecting events that may affect quality of life and/or resource utilization (e.g., rates and duration of hospitalization, antibiotic use); outcomes reflecting the effect of treatment on infection rates, tumour response and survival and those related to the biological effect of G-CSF.

Perspective (values): Evidence was selected, reviewed and synthesized by members of the Provincial Systemic Treatment Disease Site Group (DSG) of the Cancer Care Ontario Practice Guidelines Initiative. Drafts of this document have been circulated and reviewed by members of the Systemic Treatment DSG. The DSG comprises medical oncologists, pharmacists, supportive care personnel and administrators. Evaluation by clinicians was considered in the final practice guideline. Community representatives did not participate in the development of this report but will in future reports. Guidelines approval does require participation by community representatives.

Quality of evidence: Two published guidelines and an update of one guideline were identified. Ten eligible randomized controlled trials published in English were included.

Benefits: A meta-analysis of data from 8 trials showed that the odds of experiencing febrile neutropenia with G-CSF were significantly reduced (odds ratio 0.38; 95% confidence interval [CI] 0.27 to 0.52; p < 0.00001). G-CSF reduced the risk of febrile neutropenia by 34% (risk ratio 0.66; 95% CI 0.51 to 0.86; p = 0.0015). The use of G-CSF was associated with a significant reduction in antibiotic usage and days spent in hospital in 2 trials and had no effect in the other 4 in which it was measured. Five trials reported no difference in overall median survival, with 2 small trials detecting a significant increase related to G-CSF. However, further research is necessary to confirm these results.

Harms: The toxic effects of G-CSF are relatively mild. The most consistent clinical symptom attributed to G-CSF is bone pain, reported in incidence rates ranging from 20% to 50% in 3 trials. Except for one case, reported bone pain was mild.

Practice guideline: In cancer patients receiving myelosuppressive chemotherapy, granulocyte colony-stimulating factor (G-CSF) may be beneficial for some patients. If a reduction in the number of febrile neutropenic episodes, or in the duration of such episodes, is expected to improve quality of life, then G-CSF is a reasonable treatment option for selected patients. A clear justification for the use of G-CSF should be stated. If the objective of using G-CSF is to maintain dose intensity of antitumour agents, then G-CSF can be recommended where reduction in dose intensity has been shown in randomized controlled trials to reduce survival or disease-free survival. Although the evidence is weaker, the Systemic Treatment DSG would support the practice endorsed by other guidelines (American Society of Clinical Oncology, Ontario Drug Benefit Plan) and would recommend G-CSF for patients receiving potentially curative chemotherapy: i) as primary prophylaxis; that is, where dose reductions below a specified level are required because of a known high risk of febrile neutropenia, or ii) as secondary prophylaxis in patients receiving chemotherapy of established efficacy who have suffered a prior serious episode of febrile neutropenia due to the same chemotherapy regimen. The exact cut-off for dose reductions is unknown at this time and ought to be left to the judgement of the clinician. In general, the use of G-CSF for dose reductions lower than 20% is not recommended. (ABSTRACT TRUNCATED)