Oncology Research Fellow
Michael D Green
Department of Medical Oncology
Royal Melbourne Hospital
Breast cancer is the leading malignancy in women in western countries.(1) The majority of breast cancers are diagnosed early (ie, without distant metastases), and standard treatment involves resection of the primary cancer with either mastectomy or breast-conserving surgery followed by adjuvant therapy. In patients at high risk of developing subsequent recurrence, adjuvant treatment with systemic chemotherapy results in significant prolongation of both disease-free and overall survival.(2) For women with metastatic breast cancer, chemotherapy also has an established role both in palliation of symptoms and improvement of survival.
Neutropenia is a common sequel of chemotherapy and can be graded on a scale of 0–4 with the National Cancer Institute’s Common Terminology Criteria for Adverse Events.(3) Immunosuppression as a result of neutropenia may result in significant clinical consequences, such as febrile neutropenia (FN) or overwhelming sepsis. FN is defined as a fever of >-38.3°C or a temperature of 38.0°C for >-1h in the setting of a neutrophil count of <1.0 x 10(9)/l. This review will discuss the management of FN in the treatment of breast cancer, with particular emphasis on prevention and treatment.
Prevention of neutropenia
Neutropenia can be minimised by a reduction in the chemotherapy dose or by increasing the time interval between each cycle. While this is a common practice in the treatment of metastatic breast cancer where the ultimate goal is palliation, it may come at a cost of reduced efficacy when employed in the adjuvant setting.
Bonadonna and Valagussa first demonstrated a dose–response effect in the adjuvant treatment of breast cancer in a retrospective review, and showed that delivery of <85% of standard dose of chemotherapy results in compromised survival.(4) Prospective studies have also confirmed dose intensity and dose density (shortening the time interval between each cycle) to be important in optimising survival outcome in the adjuvant treatment of breast cancer.(5,6) Unfortunately, dose intensity and density are also associated with increased myelosuppression, and the resultant neutropenia is a major dose-limiting toxicity. The last decade has also seen the introduction of taxanes into clinical practice. While they have proven to be a highly active therapy in breast cancer treatment, studies have also shown them to be significantly myelotoxic agents, particularly when used in combination with other chemotherapeutic agents.
The emergence of the recombinant granulocyte colony-stimulating factor (GCSF) filgrastim more than a decade ago revolutionised the management of chemotherapy-induced neutropenia. In a pivotal clinical trial in small-cell lung cancer, it was demonstrated to reduce the incidence, duration and severity of grade IV neutropenia as well as lowering the incidence of hospitalisation and FN.(7) The use of filgrastim in adjuvant chemotherapy of breast cancer is perhaps more appropriate in settings where neutropenia is expected to occur at a significant rate. This may include high-dose chemotherapy with stem cell support, combination taxane-based chemotherapy or a dose-dense regimen.
High-dose chemotherapy with stem cell support has largely fallen out of favour in the management of breast cancer following the lack of significant benefit demonstrated in several trials.(8)
Recent data suggest that taxane-based combination chemotherapy results in improved disease-free and overall survival in the adjuvant treatment of women with lymph node-positive breast cancer over standard therapy.(9–11) However, the incorporation of taxanes has also resulted in a significant increase in the incidence of myelosuppression. The Breast Cancer International Research Study Group (BCIRG) recently undertook a randomised study comparing 5-FU–doxorubicin–cyclophosphamide (FAC) standard chemotherapy with docetaxel–doxorubicin– cyclophosphamide (TAC).(12) In the TAC arm, 24.7% of patients developed FN, compared with only 2.5% of patients in the FAC arm. The study protocol allowed the use of filgrastim in subsequent cycles in all cases of FN. Retrospective subgroup analysis showed that the rate of FN per cycle without filgrastim was 6% (TAC) and 0.5% (FAC). In subgroups that were treated with filgrastim, this decreased to 3.1% (TAC) and 0.3% (FAC). These findings are mirrored by an interim analysis of a similar study conducted by the Spanish Breast Cancer Research Group (GEICAM) looking at the adjuvant treatment of high-risk lymph node-negative breast cancer with TAC versus FAC.(13) The incidence of FN was 1.3% in the FAC arm, compared with 23.8% in those treated with TAC without filgrastim. When filgrastim was added to TAC, the rate of FN was reduced to 3.5%.
Although the concept that shortening the interval between chemotherapy cycles may attain better tumour cell apoptosis is not new,(14) the consequence of significant myelosuppression meant that it was not able to be put into clinical practice until the introduction of filgrastim in the 1990s. In the pivotal Cancer and Leukemia Group B (CALGB) 9741 study, dose-dense, two-weekly chemotherapy consisting of doxorubicin, cyclophosphamide and paclitaxel was demonstrated to be superior to conventional three-weekly treatment. The study protocol specified the use of filgrastim in the two-weekly arm, but not in the conventional three-weekly regimen. The rate of FN was low at 3% overall, with the incidence of grade IV neutropenia higher at 33% in the three-weekly arm, compared with 6% in the dose-dense group.(5)
The last update of the American Society of Clinical Oncology (ASCO) guidelines in 2000 did not recommend the routine use of colony-stimulating factors for primary prophylaxis;(15) instead, they should be considered as a secondary prophylaxis. Dose modification was considered a medically acceptable alternative. Cost analyses showed that colony-stimulating factors are cost-effective only when the FN rate is greater than 40%. No routine regimens at the time of the guideline publication had rates exceeding 15%, but the advent of taxane-based and dose-dense therapies has since demonstrated a significantly elevated FN risk over standard treatments.
While the use of filgrastim reduces the duration of neutropenia and the incidence of FN, definitive prospective data confirming that it may improve survival are still lacking.(16) It does, however, minimise the dose-limiting toxicity of neutropenia, hence facilitating the use of standard adjuvant breast cancer chemotherapy regimen(17) as well as more intensive treatment such as taxane combination or dose-dense regimens, which ultimately result in improved survival.
Pegfilgrastim (pegylated filgrastim) is produced by the binding of a polyethylene glycol moiety to the N-terminus of filgrastim. In contrast to standard filgrastim, the pegylated molecule is too large to undergo glomerular filtration and, hence, renal excretion. Its resulting prolonged half-life enables the administration of pegfilgrastim as a once-per- cycle injection. This obviously offers an attractive alternative to standard filgrastim, which requires daily dosing until the neutrophil count has reached >1.0 x 10(9)/l following the neutrophil nadir.
In several randomised studies, pegfilgrastim has been demonstrated to be at least equivalent to filgrastim in adjuvant breast cancer treatment.(18,19) In 2002, the US Food and Drug Administration approved pegfilgrastim to decrease the incidence of FN in patients with nonmyeloid malignancies receiving myelosuppressive drugs with a clinically significant incidence of FN. The role of pegfilgrastim in dose-dense protocols is currently being examined in clinical studies.
Treatment of febrile neutropenia
FN has traditionally been viewed as a medical emergency, considering the risk of significant sepsis if left untreated. Standard practice consists of immediate administration of broad-spectrum dual intravenous (IV) antibiotics.
More recently, randomised trials have demonstrated the equivalence of single-agent broad-spectrum antibiotic to dual therapy, as well as the safety of outpatient oral treatment in a carefully selected group of patients at low risk of complications. The Infectious Disease Society of America (IDSA) recently updated the guidelines for the use of antimicrobial agents in neutropenic patients with cancer.(20) For patients with FN who are at low risk of complications and morbidity (scored on an index based on features such as dehydration, extent of symptoms or age), suitable options include oral ciprofloxacin and amoxicillin– clavulanate or a single-agent IV antibiotic, such as cefepime, ceftazidime or carbapenem. If dual IV therapy is considered for patients at high risk of complications, acceptable options are an aminoglycoside with an antipseudomonal penicillin, cefepime, ceftazidime or carbapenem. Because of the increasing emergence of vancomycin-resistant organisms, the routine administration of vancomycin is generally not recommended as part of an initial empiric therapy, unless indicated by specific culture results.
- American Cancer Society. Cancer facts and figures. Atlanta: American Cancer Society; 2004.
- Lancet 1998;352:930-42.
- National Cancer Institute. Common Terminology Criteria for Adverse Events v3.0 (CTCAE) 2003.
- N Engl J Med 1981;304:10-5.
- J Natl Cancer Inst 1998;90:1205-11.
- J Clin Oncol 2003;21:1431-9.
- N Engl J Med 1991;325:164-70.
- Farquhar C, et al. (Cochrane Review). In: The Cochrane Library, Issue 2, 2004. Chichester: John Wiley.
- Martin M, et al. Breast Cancer Symposium, San Antonia (TX, USA); 2003 December 3-6: Abs 43 [poster].
- J Clin Oncol 2003;21:976–83.
- Proc Am Soc Clin Oncol 2003:Abs 12.
- Proc Am Soc Clin Oncol 2004;Abs 677.
- Proc Am Soc Clin Oncol 2004:Abs 620.
- Cancer Treat Res 1986;70:163-9.
- J Clin Oncol 2000;18:3558-85.
- Clark O, et al. (Cochrane Review). In: The Cochrane Library, Issue 2, 2004. Chichester: John Wiley.
- Oncology 1996;53:289-94.
- Ann Oncol 2003;14:29-35.
- J Clin Oncol 2002;20:727-31.
- Infect Dis 2002;34:730-51.