Population-based patterns of granulocyte colony stimulating factor (GCSF) use in breast cancer (BrCa) patients receiving myelosuppressive chemotherapy.

Abstract

9625 Background: Prophylaxis with GCSF can reduce hospitalization due to neutropenic fever, but early studies show that use of GCSF is frequently suboptimal. Our aims were to 1) characterize patterns of GCSF use in a population-based cohort of BrCa patients, 2) determine the rate of neutropenia and neutropenic fever in those who received and did not receive GCSF, and 3) identify patient and physician factors associated with appropriate GCSF prophylaxis. Methods: Patients diagnosed with BrCa from January to December 2008, seen at any 1 of 5 regional cancer centers in British Columbia, Canada and treated with chemotherapy protocols that posed >20% risk of neutropenic fever were reviewed. Using regression models that adjusted for confounders, the relationship between GCSF use and 1) various patient and physician characteristics and 2) treatment outcomes, such as neutropenic fever, were analyzed. Results: A total of 525 women were included: median age was 51 years (IQR 45 to 59 years), 38% reported smoking, 50% used alcohol regularly, 62% were ECOG 0, and 26% had private health insurance. In the entire cohort, 203 (38%) patients were given GCSF. Among those treated with GCSF, 80 (39%) and 123 (61%) individuals received GCSF as primary and secondary prophylaxis, respectively. Overall, neutropenia was noted in 292 (56%) cases while neutropenic fever was experienced by 117 (22%) patients. When compared to those who did not use GCSF, patients who used GCSF experienced a lower rate of neutropenia (15 vs 49%, p<0.01) and a decreased incidence of neutropenic fever (7 vs 13%, p<0.01). In regression models, patients lacking extended medical coverage (35 vs 49%, p=0.02), poor performance status (31 vs 53%, p=0.03), and those who were evaluated at non-teaching institutions (24 vs 68% p<0.01) were less likely to receive GCSF. Patients seen at non-teaching institutions were also given primary GCSF prophylaxis less frequently (15 vs 57%, p<0.01) than those at teaching centers. Conclusions: GCSF prophylaxis was associated improved neutropenia-related outcomes. However, use of GCSF was low in this population-based cohort of BrCa patients, especially in specific marginalized subgroups.