Pegfilgrastim Use Associated with Lower Risk of Hospitalization Than Filgrastim Use: A Retrospective US Claims Analysis.

Abstract

AbstractAbstract 3801 Objective:Patients receiving myelosuppressive chemotherapy are at risk for developing febrile neutropenia, a major dose-limiting toxicity associated with hospitalization, morbidity, and mortality. Prophylactic use of recombinant human granulocyte colony-stimulating factors (G-CSF), such as daily filgrastim and once-per-cycle pegfilgrastim, can decrease the incidence of febrile neutropenia. This study examined real-world effects of G-CSF on hospitalization risk. Methods:This retrospective U.S. claims analysis utilized data from 1/1/2004 to 2/28/2009 to examine hospitalization rates for filgrastim- and pegfilgrastim-treated patients receiving chemotherapy for non-Hodgkin's lymphoma (NHL), breast cancer, lung cancer, ovarian cancer, and colorectal cancer. Claims with the ICD-9 code for neutropenia (288) were categorized as neutropenia-related. Cycles were included if they were 20–60 days, as defined by chemotherapy claims. G-CSF use was designated ‘prophylactic’ if initiated in the first 5 days of a chemotherapy cycle, or ‘delayed’, if after day 5. Odds ratios (OR) and 95% confidence intervals (CI) were estimated by generalized estimating equations models. ORs were adjusted for potential confounders such as patient, tumor, and treatment characteristics. Healthcare utilization and costs were calculated during each cycle, as well as for emergency room, inpatient, and ambulatory visits. Cycles were considered highly myelosuppressive if patients received one or more chemotherapy agents deemed highly myelosuppressive per NCCN guidelines. Results:We identified 3,958 patients, representing 13,070 chemotherapy cycles during which G-CSF was administered (12,218 pegfilgrastim, 852 filgrastim). Most patients were female, with a mean age of 55. The most frequent cancers were breast cancer (57%), lung cancer (18%), and NHL (17%). Pegfilgrastim was used prophylactically (96% of cycles) more frequently than filgrastim (44% of cycles). Compared to chemotherapy cycles with filgrastim, those with pegfilgrastim had a decreased risk of neutropenia-related hospitalization (OR=0.33, 95% CI 0.19–0.58) and all-cause hospitalization (OR=0.56, 95% CI 0.43–0.72). Chemotherapy cycles with prophylactic initiation of either G-CSF had decreased risk of neutropenia-related hospitalization (OR=0.30, 95% CI 0.18–0.50) and all-cause hospitalization (OR=0.55, 95% CI 0.43–0.69) compared with delayed initiation of G-CSF. In subgroups of cycles with or without highly myelosuppressive chemotherapy, similar reductions in all-cause and neutropenia-related hospitalization risk were observed both with pegfilgrastim vs. filgrastim and prophylactic vs. delayed G-CSF. The two types of cycles were generally similar in patient characteristics, with the exception of proportion female, 84% vs. 67%, and baseline Deyo-Charlson comorbidity score, 4.2 vs. 5.3, for cycles with or without highly myelosuppressive chemotherapy, respectively. For all-cause utilization by cycle, the mean numbers of ambulatory visits (8.6 vs. 5.5, P<0.001) and inpatient stays (0.13 vs. 0.06, P<0.001) were greater with filgrastim as compared with pegfilgrastim, while the numbers of emergency room visits were the same (0.11 for both). For neutropenia-related utilization by cycle, there were also more ambulatory visits (1.5 vs. 0.36, P<0.001) and inpatient stays (0.02 vs. 0.01, P<0.01) with filgrastim as compared with pegfilgrastim, while mean emergency room visits were 0 for both groups. Mean total per-cycle costs (for all claims and pharmacy costs) due to all causes were similar for filgrastim and pegfilgrastim ($9,581 vs. $9,881) while mean total per-cycle costs due to neutropenia-related causes were numerically greater with filgrastim than pegfilgrastim ($1,615 vs. $1,190, P=0.054). Inpatient stays were more costly with filgrastim than pegfilgrastim for both all causes ($2,002 vs. $862, P<0.005) and neutropenia-related causes ($468 vs. $89, P=0.062). Conclusions:In this comparative effectiveness study, use of pegfilgrastim resulted in a lower risk of neutropenia-related and all-cause hospitalization compared to use of filgrastim. Inpatient stays were more frequent and more costly during cycles in which patients received filgrastim. Prophylactic use of either G-CSF was associated with a consistent reduction in hospitalization risk as compared with delayed use. [Display omitted] Disclosures:Naeim:Amgen Inc.: Consultancy. Henk:i3 Innovus: Employment; Amgen Inc.: Research Funding. Becker:i3 Innovus: Employment; Amgen Inc.: Research Funding. Chia:Amgen Inc.: Employment, Equity Ownership. Badre:Amgen Inc.: Employment, Equity Ownership. Deeter:Amgen Inc.: Employment, Equity Ownership, Research Funding.