Abstract P5-10-01: Overall Survival in Breast Cancer Patients in Randomized Controlled Trials Comparing Primary G-CSF Chemotherapy Support to Control
Lyman, GH; Kuderer, NM; Crawford, J; Culakova, E; Poniewierski, MS; Huang, M; Dale, DC
Published in: Cancer Research
Background Randomized controlled trials (RCTs) have demonstrated that granulocyte colony-stimulating factor (G-CSF) is capable of reducing the risk of neutropenic complications including febrile neutropenia as wellas infection-and treatment-related mortality (Kuderer et al. JCO 2007). To address the potential impact of G-CSF support of chemotherapy on overall survival in patients with breast cancer, a systematic review of RCTs of chemotherapy with or without G-CSF support and at least 2 years of follow-up was conducted.Methods A systematic review of the medical literature identified 20 RCTs of breast cancer patients receiving chemotherapy with or without primary G-CSF support and reporting overall mortality by study arm. Data extraction from eligible studies was undertaken by two independent reviewers. Heterogeneity was assessed by Cochran's Q-statistic and the Inconsistency Index (I2). Weighted summary estimates of relative risk (RR) with 95% CIs were based on the method of Mantel and Haenszel using a fixed effects model. Planned chemotherapy relative dose intensity (RDI) with G-CSF support compared to control was meta-regressed on the natural logarithm of the RR for mortality. Publication bias was assessed on the basis of Egger's regression intercept.Results Patients in 20 eligible RCTs received chemotherapy with (n=5,938) or without (n=6,627) primary G-CSF support. Deaths were reported in 1,468 patients randomized to receive primary G-CSF and in 1,607 controls. No significant heterogeneity was observed (Q=23.53, P=.215; I2=19%). Overall RR for mortality was 0.94 [0.89-0.99; P=.029]. RR for mortality was 0.92 [0.85-0.99; P=.036] in the 11 trials of patients with locoregional disease compared to 0.98 [0.92-1.04; P=.503] in the 9 RCTs of patients with advanced disease. A nonsignificant trend was observed between median duration of follow up and RR reduction for mortality [P=.08]. A significant inverse trend was observed between chemotherapy RDI with G-CSF support and RR reduction for mortality [P=.05]. The greatest reduction in mortality was observed in the four RCTs of dose dense chemotherapy. A trend toward publication bias was observed among studies of patients with advanced disease but not locoregional disease.Conclusions Overall mortality in patients with breast cancer is reduced in those receiving chemotherapy with primary G-CSF support compared to controls. Greater impact on mortality was observed in clinical trials of early-stage disease and in studies with greater chemotherapy RDI including dose dense schedules.Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P5-10-01.
