Phase I/II trial of human recombinant granulocyte-colony-stimulating factor (filgrastim) and escalating doses of cyclophosphamide, mitoxantrone, and 5-FU in the treatment of advanced breast cancer

Purpose: We performed a phase I/II dose-escalation trial of cyclophosphamid e, mitoxantrone, and 5-fluorouracil (CNF) in combination with human recombi nant granulocyte-colony-stimulating factor (G-CSF, filgrastim) in patients with advanced breast cancer. The objectives of this trial were (1) to gain experience with filgrastim given to patients with advanced breast cancer an d receiving standard-dose CNF, and (2) to determine the maximum tolerated d ose of CNF that could be given with filgrastim support by incremental dose escalation of two components of the CNF regimen, cyclophosphamide and mitox antrone. Methods: Four patients who had received prior therapy for advanced disease received standard-dose CNF with filgrastim support. Sequentially e nrolled patients who had received no prior chemotherapy for advanced diseas e were treated with standard-dose CNF without filgrastim (5 patients), stan dard-dose CNF with filgrastim (15 patients), or were entered into sequentia l cohorts of 3-6 patients to be treated with increasing doses of CNF with f ilgrastim support (29 patients). Results: The maximum tolerated doses that could be given with filgrastim support were 1500 mg/m(2) cyclophosphamide, 30 mg/m2 mitoxantrone, and 500 mg/m(2) 5-FU. Overall, 7 complete (14%) and 13 partial responses (26%) were observed. Despite the use of filgrastim, re peated cycles of CNF at doses of 2000 mg/m(2) cyclophosphamide. 25 mg/m(2) mitoxantrone, and 500 mg/m(2) 5-FU could not be given because of neutroprni a and thrombopenia, Among 18 patients with bidimensionally measurable disea se there were 3 complete (17%) and 5 partial (28%) responses. The median pr ogression-free survival of all patients was 236 days (34 weeks). Conclusion : The use of filgrastim allows CNF to be given at approximately twice the d ose intensity of "standard"-dose CNF, Because nonhematopoietic toxicity was not dose-limiting, further dose escalation of this regimen might be possib le with mon effective hematopoietic support. The response rate and survival of patients treated in this study were within the range expected with stan dard-dose chemotherapy.