Prospective, randomized trial of sequential interleukin-3 and granulocyte- or granulocyte-macrophage colony-stimulating factor after standard-dose chemotherapy in cancer patients

Abstract

BACKGROUND AND OBJECTIVE: Several in vitro and animal studies have shown that IL-3 primes hematopoietic stem cells to become more sensitive to later acting growth factors. We wanted to compare the toxicity and the synergistic stimulatory effect of interleukin-3 (IL-3) followed by granulocyte colony-stimulating factor (G-CFS) or granulocyte-macrophage colony-stimulating factor (GM-CSF) on white blood cell (WBC) and platelet counts, after standard-dose chemotherapy (CT) in patients with solid tumors. DESIGN AND METHODS: Fifty consecutive cancer patients with thrombocytopenia and/or leukopenia registered during a previous course of CT were randomized to receive, after the following course, IL-3 (10 microg/kg/day, s.c., day 1-5) followed by G- or GM-CSF (5 microg/kg/day, day 6-8). RESULTS: The nadir of WBC in the cycles supported with the combination of IL-3 and G-CSF was significantly higher than that observed in the CT cycles not supported by growth factors (p < 0. 005). Furthermore, severe leukopenia was abrogated in all the cycles supported with IL-3+G-CSF, while in the cycles without cytokines, this event was registered in 62.5% of the cases (p < 0.0005). Finally, the recovery of WBC was achieved a mean of 4 days earlier in the cycles supported with IL-3+G-CSF. As for thrombocytoprotection, no significant differences were evidenced, but severe thrombocytopenia was abrogated in all the cycles supported by IL-3+G-CSF (p < 0.05). Furthermore, platelet recovery after CT was achieved on average 3.5 days earlier in the IL-3+G-CSF group than in the previous cycles. The nadir of WBC count in the cycles supported by the combination of IL-3 and GM-CSF was significantly higher than that observed in the CT cycles not supported by growth factors (p < 0.005). Furthermore, severe leukopenia was abrogated in 40% of the cycles supported by IL-3+GM-CSF, while in the cycles without cytokines, this event was registered in 80% of the cases (p < 0.005). Finally, the recovery of WBC was achieved a mean of 3.5 days earlier in the cycles supported by IL-3+GM-CSF. As far as thrombocytoprotection is concerned, there were no significant differences in the nadir between the cycles supported by the association IL-3+GM-CSF and the cycles not supported by cytokines. However, severe thrombocytopenia was registered in 20% of the cycles not supported by growth factors but in only 10% of the cycles supported by IL-3+GM-CSF (p < 0.05). Furthermore, platelet recovery after CT was achieved on average 3 days earlier in the IL-3+GM-CSF group. The combination of IL-3 and G-CSF would appear to be more effective than the combination of IL-3 and GM-CSF in the control of both severe thrombocytopenia and leukopenia. Indeed, severe leukopenia was abrogated in all the cycles in arm A, but only in 40% of the cycles in arm B (p < 0.0005). Furthermore, considering a platelet count below 49