A multicenter study comparing granulocyte-colony stimulating factors to antibiotics for primary prophylaxis of taxotere/cyclophosphamide-induced febrile neutropenia in patients with early-stage breast cancer.

7001 Background: Docetaxel-cyclophosphamide (TC) adjuvant chemotherapy is commonly used in patients with early stage breast cancer (EBC). Due to the risk of febrile neutropenia (FN) with TC, primary prophylaxis with either ciprofloxacin (cipro) or granulocyte-colony stimulating factors (G-CSF) is recommended. Despite significant differences in costs (7-120 $US/course [cipro] vs. 2100-7000 $US/dose [G-CSF]) and toxicity profiles, optimal primary FN prophylaxis is unknown. We performed a pragmatic randomised trial comparing the superiority of G-CSF to cipro. Methods: EBC patients receiving TC chemo were randomized to receive cipro or G-CSF as primary FN prophylaxis. The primary outcome is a composite of either treatment-related hospitalisations or FN. Secondary outcomes included: chemo dose reductions, delays, discontinuations and incidence of C. difficile infections. Primary analysis was performed with the intention to treat (ITT) population. Results: 455 eligible patients were randomized to cipro (227) or G-CSF (228). 37/227 (16.3%) patients on cipro had a hospitalization, compared with 25/228 (11.0%) on G-CSF (Fisher’s exact test p-value=0.10). Relative risk (RR) of hospitalization for patients on G-CSF:0.68, 95%CI=0.42 to 1.09. Patients on cipro were statistically significantly more likely to be hospitalized for FN (30/227, 13.2%) vs 9/228 (4.0%) patients on G-CSF(p<0.001). RR of developing FN and being hospitalized for patients on G-CSF: 0.44, 95%CI=0.26 to 0.76. There was no significant difference between groups for chemo dose reductions, delays, and C. difficile rates. Twenty patients on cipro (8.8%) and 9 on G-CSF (3.9%) discontinued chemo early (p=0.036). RR of discontinuing chemo: 0.43, 95%CI=0.19 to 0.96. Conclusion: G-CSF was superior to cipro at reducing FN. While a trend towards reduced hospitalizations was also observed with G-CSF, it did not attain statistical significance. However, as 18 patients would need to be treated with G-CSF to prevent one hospitalization compared to cipro, this would suggest a cost of over $100000 $US to prevent a hospitalization. A formal cost-effectiveness analysis will be performed. Clinical trial information: NCT02173262, NCT02816112 .