The economic value of primary prophylaxis using pegfilgrastim compared with filgrastim in patients with breast cancer in the UK

Abstract

Febrile neutropenia (FN) is a serious adverse event associated with myelotoxic chemotherapy that predisposes patients to life-threatening bacterial infections. Prophylaxis with granulocyte colony-stimulating factors (G-CSFs) from the first cycle of chemotherapy is recommended by the 2006 American Society of Clinical Oncology, 2008 National Comprehensive Cancer Network and 2006 European Organisation for Research and Treatment of Cancer guidelines when the overall risk of FN is approximately 20% or higher. Once-per-cycle pegfilgrastim and daily filgrastim are two commonly used G-CSFs with different dosing schedules and associated costs. To evaluate the cost effectiveness of pegfilgrastim versus filgrastim primary prophylaxis in women with early-stage breast cancer receiving chemotherapy in the UK. A decision-analytic model was constructed from the UK NHS perspective with a lifetime study horizon. The model simulated three clinical scenarios: scenario 1 assumed that pegfilgrastim and filgrastim had differential impact on the risk of FN; scenario 2 assumed additional differential impact on FN-related mortality; and scenario 3 assumed additional differential impact on chemotherapy relative dose intensity (RDI) with long-term survival effects. The base-case population included 45-year-old women with stage II breast cancer receiving four chemotherapy cycles, with an FN risk of approximately 20% or higher. Model inputs, including FN risk, FN case-fatality, RDI, impact of RDI on survival and utility scores, were based on a review of the literature and expert panel validation. Using data from the literature, it was estimated that the absolute risk of FN associated with pegfilgrastim was 5.5% lower than with 11-day filgrastim (7% vs 12.5%), and 10.5% lower than with 6-day filgrastim (7% vs 17.5%). Costs were taken from official price lists or the literature and included drugs, drug administration, FN-related hospitalizations and subsequent medical costs. Breast cancer mortality and all-cause mortality were obtained from official statistics. The main outcome measures were the costs ( pound, year 2006 values) per percentage decrease in (absolute) FN risk, per FN event avoided, per life-year gained (LYG), and per QALY gained. Model robustness was tested using deterministic and probabilistic sensitivity analyses. Pegfilgrastim was cost saving compared with 11-day filgrastim ( pound 3196 vs pound 4315). Compared with 6-day filgrastim, pegfilgrastim was associated with a cost of pound 4200 per FN event avoided, or pound 42 per 1% decrease in absolute risk of FN, in scenario 1. In scenario 2, pegfilgrastim provided 0.055 more LYGs or 0.052 more QALYs at a minimal cost increase of pound 441 ( pound 3196 vs pound 2754) per person, yielding an incremental cost-effectiveness ratio (ICER) of pound 8075/LYG or pound 8526/QALY. In scenario 3, when all potential benefits of G-CSF were considered, the ICER became pound 3955/LYG or pound 4161/QALY. Results were most sensitive to the relative risk of FN for 6-day filgrastim versus pegfilgrastim. In this UK analysis, pegfilgrastim appears to dominate 11-day use of filgrastim. The value of pegfilgrastim versus 6-day filgrastim at pound 4161-8526/QALY was very favourable compared with the commonly used threshold in the UK. In this setting, primary prophylaxis with pegfilgrastim may be cost effective compared with filgrastim.