Abstract P1-15-01: Patterns of granulocyte colony stimulating factor (G-CSF) use in elderly breast cancer (BC) patients receiving myelosuppressive chemotherapy

Abstract

Abstract Background: Febrile neutropenia (FN) is a common and potentially serious complication of myelosuppressive chemotherapy treatment in cancer patients. Oncology guidelines recommend primary G-CSF prophylaxis (PPG) in patients with a high risk of developing FN, which is risk >20% based on myelotoxicity of the regimen itself or from a combination of the therapy, older age, comorbidities and disease characteristics (Lyman Cancer 2011). Current patterns of G-CSF use and FN occurrence among elderly patients receiving myelosuppressive chemotherapy for BC have not been previously reported. To determine this, we performed a retrospective analysis using a subset of the Medicare 5% database. Methods: The Medicare 5% claims data set (includes a representative 5% systematic sample of Medicare beneficiaries) was used to identify BC patients age 65+ initiating chemotherapy between 7/1/2003 and 6/30/2009. Chemotherapy courses were identified for each patient; only the first course was used for this analysis. Using the National Comprehensive Cancer Network guidelines on Myeloid Growth Factors (NCCN V1.2012), chemotherapy course regimens were classified as high risk (HR) or intermediate risk (IR) for FN. Duration of first cycle was from date of first chemotherapy claim to the chemotherapy claim at day 21 or later, which defined the first day of the second cycle, etc., to a maximum of 9 cycles. First administration of G-CSF [filgrastim (NEUPOGEN®) or pegfilgrastim (Neulasta®)] was classified as either PPG (within first 5 days of first cycle), secondary prophylaxis (within first 5 days of second or subsequent cycles), or reactive (day 6 or later of first or subsequent cycles). FN assessed during the chemotherapy course was defined as hospitalization with a code for neutropenia in any position. Results: 885 courses with high FN risk and 1046 with IR FN risk were identified. The HR cohort was younger (71.4 vs 74.5 yrs) and had fewer comorbidities than the IR cohort. Selected aspects of G-CSF use patterns are summarized in the table. Among HR courses, 11.8% had ≥1 FN hospitalization and 2.1% had 2+; among IR courses 5.6% had ≥1 and 0.4% had 2+. Conclusion: NCCN recommends PPG be used with HR regimens and older age (notably >65 yr), an important risk factor for developing severe neutropenic complications. Despite this, PPG was used for elderly breast cancer patients in only 52% of chemotherapy courses with high risk of FN and in 10% of IR courses. More than 10% of patients with a HR regimen had an FN hospitalization. Careful attention to FN risk factors, including regimen and patient age, is needed when planning treatment strategy. HR regimens: TAC (389); dose dense AC→T (345); docetaxel+trastuzumab (61); doxorubicin+docetaxel (50); doxorubicin+paciltaxel (21); docetaxel q14(19). IR regimens: CMF classic (481); paclitaxel q21 (337); docetaxel q21 (94); paclitaxel+ trastuzumab (87); FEC (47). Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P1-15-01.