1. Home>
2. Specialties>
3. Oncology and Hematology>
4. Summary and Comment

Effects of G-CSF and GM-CSF on Chemotherapy-Related Mortality and Infection

A meta-analysis revealed that mortality rates were similar with or without CSFs during chemotherapy, although rates of infection and infection-associated morbidity were lower with CSFs.

Cytokine supportive care with the colony-stimulating factors G-CSF (granulocyte colony–stimulating factor) or GM-CSF (granulocyte-macrophage colony–stimulating factor) is used widely for patients who are undergoing chemotherapy or stem-cell transplantation. Shorter duration of neutropenia and lower infection rates have been reported with CSF use, but the effect of CSFs on rates of serious infection and overall mortality has been unclear because of the small size of individual studies.

Researchers evaluated all-cause and infection-related mortality rates in a comprehensive meta-analysis of randomized controlled trials in which G-CSF or GM-CSF was compared with placebo or no CSF treatment in chemotherapy or stem-cell–transplantation patients. Studies were included in the meta-analysis if CSF was given concurrently with or subsequent to chemotherapy but before onset of any neutropenia and if chemotherapy did not differ substantially between the CSF and control arms; 148 studies (16,839 patients or treatment cycles) met inclusion criteria. The primary endpoint was the median short-term (3–6 weeks) all-cause mortality rate; secondary endpoints were infection and febrile neutropenia (FN) incidences.

Median short-term mortality rates were similar for the CSF and control groups, although infection-related deaths were slightly less common with CSF (see Table 1). Duration of FN and time that absolute neutrophil counts remained lower than 500 cells/mm³ were shorter in the CSF group than in the control group by 1 and 4 days, respectively. Interestingly, CSF showed no benefit in patients with high risk for treatment-associated FN; in fact, CSF benefit was greater in patients for whom FN risk was 20%.

Comment: This meta-analysis represents an important effort to identify all published randomized clinical trials in which the benefit of CSFs in support of chemotherapy and stem-cell transplantation was evaluated. Although no effect on short-term or infection-related mortality rates was observed, fewer documented infections and substantially fewer episodes of FN were noted with CSFs. Surprisingly, the benefit of CSF use was greater for patients with low risk for FN (<20%), as opposed to those with higher risk. This result contrasts with National Comprehensive Cancer Network and American Society for Clinical Oncology guidelines, which recommend use of CSF support when the chemotherapy regimen used or patient risk factors predict risk for FN will be >20%. The NCCN guidelines also recommend that clinicians "consider CSF" for patients with intermediate risk (10%–20%) for FN. Of note, the ASCO guidelines characterize fewer episodes of FN as an ". . . important clinical outcome that justifies the use of CSFs, regardless of impact on other factors, when the risk of FN is approximately 20% and no other equally effective regimen that does not require CSFs is available."

Formal analyses of quality of life and effect of CSFs in less-intensive regimens, such as those identified in this meta-analysis, should help refine the current guidelines. Assessing the optimal CSF dosing schedule and its effect on chemotherapy dose intensity or dose density and on treatment response and cure rates is important. A negative mortality effect was not apparent for CSF use in this analysis; whether certain patients with specific tumor types benefit more or less from CSFs remains to be addressed.

— Michael E. Williams, MD

Published in Journal Watch Oncology and Hematology October 23, 2007

Citation(s):

Sung L et al. Meta-analysis: Effect of prophylactic hematopoietic colony-stimulating factors on mortality and outcomes of infection. Ann Intern Med 2007 Sep 18; 147:400.

• Original article (Subscription may be required)
• Medline abstract (Free)