Physycians' refusal to continue the study

Granulocyte colony stimulating factor (G-CSF) is frequently used among patients with cancer including those with haematological malignancies.

Filgrastim is a recombinant human CSF whose biological activity is similar to that of endogenous G-CSF.

In the treatment of chemotherapy-induced neutropenia in patients with various types of cancer CSFs significantly reduced the time to neutrophil recovery and length of hospitalization.

Granulocyte colony stimulating factors (G-CSFs) stimulate the proliferation and differentiation of myeloid progenitor cells, improve cell survival and affect some end-cell functions, through binding to G-CSF receptors present on all cells of the neutrophilic granulocyte lineage. Filgrastim (recombinant G-CSF) is frequently used among patients with cancer including those with haematological malignancies. In-vivo studies and studies in healthy people show that SC administration of CSF results in lower peak but more prolonged and stable levels of G-CSF as compared with Intravenous (IV) administration, with similar or higher neutrophil counts. It is safe to assume that IV administration of G-CSFs would be more comfortable to patients when hospitalized, especially during or after chemotherapy when most patients have a central catheter and are thrombocytopenic. However, it is necessary to ensure that the same effects are obtained with both methods of administration.

Objectives: To compare the time to neutropenia resolution with Intravenous (IV) versus Subcutaneous (SC) filgrastim administration among patients with acute leukemia, lymphoma or multiple myeloma in hospital. Secondarily, the investigators aim to assess comparative rates of infection, adverse effects and patients' satisfaction.

Methods: The investigators plan a randomized controlled trial comparing the effects of IV versus SC filgrastim (Neupogen®) given as per clinical indication on neutrophil counts in hospitalized patients. The investigators will include patients hospitalized in haemato-oncology ward starting filgrastim for the treatment of chemotherapy-induced neutropenia. The investigators will compare SC vs. IV filgrastim, both given as a single daily dose of 5 mcg/kg (rounded to 300 mcg or 480 mcg). No blinding will be used. Patients will be approached to obtain informed consent and randomized to the mode of filgrastim administration after the decision to administer the drug has been made. Patients will be crossed over to the alternative study arm on the subsequent chemotherapy course, if filgrastim is clinically indicated.

Outcomes:

Primary efficacy: Time to stable neutrophil recovery, defined as the number of days from start of filgrastim (day 1) until the neutrophil count has reached >500/mcL for 3 consecutive days.

Primary safety: 30-day mortality or documented infection (CDI, MDI, bacteremia or probable/ proven IFI, see definitions below) within the chemotherapy course (before or after neutrophil recovery).

Secondary outcomes will include rates of infection, fever days, hospital stay, patient's satisfaction, other clinical endpoints and adverse events.

The investigators will assess the distribution pattern of the time to neutrophil recovery and compare groups using Student's t-test or the Mann-Whitney U test, as appropriate. The investigators will construct Kaplan-Meier curves for time to neutrophil recovery and compare treatment arms using a two-tailed log rank test. Dichotomous outcomes will be compared using a chi-square test. A sample of 96 patients with AML (48 in each group) was calculated to demonstrate equivalence allowing a 2-day difference between treatment arms (power of 90%, alpha 0.05).

Interim analysis and stopping rules: We will conduct interim analyses for safety assessment after every 50 patients recruited. Stopping rules will be based on the primary safety outcome (p<0.1 for stopping) and deaths alone (p<0.2 for stopping).

Inclusion Criteria:

• Patients hospitalized in haemato-oncology ward starting filgrastim for the treatment of chemotherapy-induced neutropenia.
• Will include patients with acute myelogenous leukemia (AML), acute lymphoblastic leukemia (ALL), aggressive lymphoma or multiple myeloma.
• Will include both patients with or without a documented infection at the time of CSF initiation. Initiation of filgrastim treatment will follow the 2006 ASCO guidelines (departmental routines).

Exclusion Criteria:

• The investigators will exclude patients receiving CSFs for their primary disease (e.g. aplastic anemia, myelodysplastic syndromes) and pregnant women.