Targeted filgrastim support in patients with early-stage breast carcinoma: toward the implementation of a risk model.

BACKGROUND: Severe neutropenia, a common consequence of chemotherapy, may result in infectious complications and hospitalizations. Preventive treatment with colony-stimulating factors is limited because of the inability to predict which patients will develop neutropenic complications. To the authors' knowledge, the current study is the first large prospective validation of a risk model in patients with early-stage breast carcinoma.

METHODS: Patients with Stage I-III breast carcinoma who were receiving adjuvant chemotherapy (n=624) were assigned to risk groups based on first-cycle absolute neutrophil count (ANC) nadirs of <0.5 x 10(9)/L. Filgrastim (a recombinant human granulocyte-colony-stimulating factor) was administered from Cycle 2 onward to high-risk patients. Dose intensity and rates of neutropenic complications, including febrile neutropenia and hospitalization resulting from it, were calculated for each group and compared. High-risk patients were matched by chemotherapy regimen, stage of disease, age, and baseline ANC to historic-control patients and outcomes were compared within the matched pairs.

RESULTS: Both risk groups were found to have a similar proportion of patients receiving >85% of the dose intensity (95.8% vs. 94.4%). The rate of febrile neutropenia and hospitalization in the low-risk group (n=264) was 2.6% (95% confidence interval [95% CI], 0.7-4.5%) and 0.8 (95% CI, -0.3-1.9%), respectively. The high-risk group was 2.6 times more likely to receive a full dose of chemotherapy, but no higher risk of neutropenic complications was reported compared with the matched controls.

CONCLUSIONS: The risk-related prophylactic administration of filgrastim facilitated the delivery of planned chemotherapy to the high-risk group of patients. However, further research is needed to confirm the results obtained in the current study in a randomized trial, if feasible, and in other chemotherapy and disease settings.