Granulocyte-Colony Stimulating Factor Use in Young Fit CLL Patients Receiving Fludarabine-Cyclophosphamide-Rituximab Frontline: Impact on Outcomes, Relative Dose Intensity and Toxicities

fludarabine-cyclophosphamide-rituximab (FCR) is the standard upfront immunochemotherapy for young fit CLL patients. The clinical benefit of growth factors support is yet unclear, despite recently published possible impact on outcomes in a series of 32 patients [Grüber M, 2011]. The use of G-CSF is not recommended in the CLL8 trial, outside febrile neutropenia, but often applied outside clinical trials to prevent toxicities and achieve good relative dose-intensity (RDI). We retrospectively assessed, in an Hematology healthcare network, the impact of G-CSF on survivals (PFS, TTNT, OS), outcomes (RDI, minimal residual disease (MRD), overall response rates (CR/CRi), and toxicities (grade 3–4 neutropenia, fever, hospitalizations).

PATIENTS AND METHODS: among 101 patients treated with FCR frontline, three groups of patients are considered: group 1 (no G-CSF), group 2 (primary prophylaxis with pegfilgrastim after each course of FCR), and group 3 (patients of group 1 initially, but who were treated with G-CSF due to at least one episode of grade 4 neutropenia (at the discretion of the physician)). Respectively, toxicities have been assessed in 24/28/13 patients, and outcomes in 45/23/23 patients. Pretreatment characteristics were well balanced between G-CSF-naïve and -treated patients (IgVH, Binet stage, del11q). No del17p patient was included in this series. Planned RDI for F and C were calculated before 1^st cycle of FCR according to age (-20% if >65y), and creatinine clearance (-25% if <60ml/mn). Average RDI (ARDI) actually prescribed to patients were assessed at the last cycle, in mg/m²/week (6xFCR=24 wks) to include dose delays in the calculation of RDI.

median age in the cohort of 101 FCR treated pts was 60y (21–83y), 68% were males. 20% had >65y, 13% had creatinine clearance <60ml/mn, CIRS-G comorbidity scores were: 0 (25.5%), 1 (26.5%), 2 (19%), 3 (10%), 4 (8%), 5 (5%), 6 (2%), ≥7 (4%). Planned RDI was ≤75% standard FCR doses (due to age, CrCL, or physician's choice) in 12% of cases, and ARDI was further decreased ≥20% of initial planned RDI in 25% of patients. Peripheral blood 4-color flow MRD wad undetectable in 49% of patients.

results are summarized in Table 1. The use of G-CSF on curative intent for grade 4 neutropenia induced an increase in rates of CRi and prolonged neutropenia at the end of therapy. When used at the time of neutropenia (d15-d21 after FCR cycle), stimulation with G-CSF may be deleterious due to the prescription of the next FCR at d28. Median PFS, TTNT, OS were not significantly improved by the use of G-CSF (prophylactic or curative). G-CSF did not impact on MRD levels neither. MRD eradication was the strongest parameter linked to PFS/TTNT.

results are summarized in Table 2. The use of prophylactic G-CSF (group 2) significantly reduced the rate of neutropenia grade 3–4, and tended to decrease the need for antibiotics given for fever. A dose modification (>10%) was observed in 26% vs 33.3% in patients receiving prophylactic G-CSF or not, respectively. A planned RDI <75% standard FCR doses was linked to reduced PFS and TTNT, but not OS. Prophylactic G-CSF did not prevent a decrease >20% of planned RDI at the end of therapy.

Our data suggest that prophylactic G-CSF use after FCR decreases toxicities but does not impact on outcomes. We plan to study G-CSF impact on FCR results in an older, less fit population (FORTIS phase III trial).

Ysebaert:Roche: Membership on an entity's Board of Directors or advisory committees, Research Funding.