Primary G-CSF Prophylaxis During the First Two Cycles Only or Throughout All Chemotherapy Cycles in Breast Cancer Patients at Risk of Febrile Neutropenia
1. Febrile neutropenia Febrile neutropenia is a serious side effect of cancer treatment.
Febrile neutropenia predisposes to serious and sometimes life-threatening infections. Given
the seriousness, the majority of febrile neutropenia patients are admitted to hospital for
treatment with intravenous antibiotics. Thus, chemotherapy-induced febrile neutropenia can
result in death, increased hospitalisations and intravenous antibiotic use. In addition, it
can lead to significant chemotherapy dose modifications, which may theoretically be of
concern for treatments with curative intent. Clearly, prevention of chemotherapy-induced
febrile neutropenia is a medical priority.
2 Prophylaxis of febrile neutropenia To prevent chemotherapy-related febrile neutropenia,
prophylaxis with antibiotics and granulocyte colony-stimulating factor (G-CSF) have proven
efficacious [1-3]. The use of the G-CSF was shown to shorten the duration of neutropenia,
resulting in reduction of the incidence of febrile neutropenia, hospitalisation and use of
intravenous therapeutic antibiotics by approximately 50%. .Antibiotics may, however, lead to
emergence of microbial resistance. G-CSF has only few side effects, but is expensive.
Previous guidelines recommended primary G-CSF prophylaxis to prevent febrile neutropenia,
but only for patients at substantial risk of febrile neutropenia (more than 40%). In 2006,
updated G-CSF guidelines conclude that primary G-CSF prophylaxis has clinical benefits for
and should be offered to patients at a more than 20% risk of febrile neutropenia.
Most of the recent EORTC-reviewed evidence indicate that primary and secondary G-CSF
prophylaxis had no significant impact on survival, despite the fact that G-CSF prophylaxis
enabled to maintain chemotherapy dose and dose intensity. In the GEICAM 9805 study, primary
G-CSF support did not lead to improved median dose intensity or cumulative dose when
compared to secondary G-CSF prophylaxis, although the percentage of patients who completed
six cycles of therapy increased significantly.
3. Developments in the adjuvant treatment of breast cancer After the introduction of taxanes
in the treatment of advanced breast cancer in the late nineties, these agents have now also
been introduced in the (neo) adjuvant setting. Many adjuvant trials with a positive outcome
have recently been reported.
Based on the available data with many positive and few negative trials, one can consider the
use of taxanes as standard of care in the adjuvant setting in node-positive breast cancer,
although the preferred schedule, concurrent or sequential, with or without
dose-densification, needs to be clarified from ongoing trials.
Martin et al. report on the results of a randomized adjuvant trial comparing TAC with FAC
for high-risk N0 breast cancer patients, the GEICAM 9805 trial. Comparisons are reported for
toxicity and quality of life. Only secondary G-CSF prophylaxis was allowed or even mandatory
after an episode of febrile neutropenia. After a protocol amendment, patients in the TAC arm
received primary G-CSF prophylaxis. So, in addition to the upfront planned comparison of TAC
versus FAC, the unplanned comparison of TAC-pre versus TAC-post amendment is reported. This
comparison is very interesting, as it is to our knowledge the first comparison of primary
versus secondary G-CSF prophylaxis ever reported. The use of primary G-CSF prophylaxis
significantly reduced the incidence of febrile neutropenia associated with TAC chemotherapy.
The percentage of patients with febrile neutropenia in one or more cycles was 24.6% and 6.5%
in TAC-pre and TAC-post, respectively. Of note, in the TAC-pre group 71.1% of patients had
received secondary G-CSF prophylaxis during on average 4 cycles of treatment. In the FAC
group, only 2.3% of patients had febrile neutropenia as per protocol definition.
So, the combined use of anthracyclines and taxanes come with considerable increased risk of
febrile neutropenia. Perez concluded in an editorial in the New England Journal of Medicine
that on the basis of the available data, one can consider TAC (docetaxel, adriamycin,
cyclophosphamide) to be a standard of care, as is the dose-dense regimen of doxorubicin and
cyclophosphamide followed by paclitaxel, for patients with node-positive breast cancer. With
this regimen, prophylactic growth-factor support is necessary to ameliorate myelosuppression
and febrile neutropenia.
4. Consequences for health care costs We have recently reported that, even for patients with
a considerable risk of febrile neutropenia, primary G-CSF prophylaxis comes with
considerable extra costs in the Netherlands: on average € 5,123 per patient who have a
baseline risk of febrile neutropenia of 32%. The updated guidelines and changes in daily
clinical practice will have a significant impact on our health care resources. In the
Netherlands several thousands of patients with many tumour types will require primary G-CSF
prophylaxis, leading to an cost increase of at least € 10-20 million per year.
5. Rationale for the present study It has been demonstrated that there is a higher risk of
febrile neutropenia for the first chemotherapy cycle compared to subsequent cycles in small
cell lung cancer patients. Also in advanced breast cancer the majority of first observed
episodes of febrile neutropenia occur in the initial chemotherapy cycles. With primary G-CSF
prophylaxis the absolute neutrophil count nadir is less deep and of shorter duration in
later cycles compared to the first cycle. This may suggests that there may be a priming
effect of G-CSF to subsequent cycles, emphasizing that administration of G-CSF early in the
course of treatment might be important.
So, irrespective of tumour type or chemotherapy regimen, the risk of febrile neutropenia is
highest during the first two cycles of chemotherapy. Thereafter, the risk rapidly declines,
and the benefit of G-CSF largely seems to disappear.
So, in order to improve the cost-effective administration of primary G-CSF prophylaxis, it
is justified to assess whether G-CSF prophylaxis can be limited to the first two
chemotherapy cycles as compared to the current practice of continuous G-CSF prophylaxis.
Allocation: Randomized, Endpoint Classification: Efficacy Study, Intervention Model: Parallel Assignment, Masking: Open Label, Primary Purpose: Prevention
number of febrile neutropenia episodes costs per treatment arm
18 weeks (all chemotherapy cycles)
Vivianne CG Tjan-Heijnen, MD PhD
Maastricht University Medical Center
Netherlands: The Central Committee on Research Involving Human Subjects (CCMO)