Use of darbepoetin alfa and epoetin alfa in clinical practice in patients with cancer-related anemia

Abstract

Background: Patients with cancer may receive erythropoiesis-stimulating agents (ESAs), including darbepoetin alfa (DA) or epoetin alfa (EA), to treat cancer-related anemia (CRA). DA and EA differ, however, with respect to their assumed duration of effect and thus their approved frequency of dosing, complicating direct comparison of their doses and costs. Objective: The objective of this study was to examine, from the perspective of a third-party payer, patterns of use and costs of DA and EA in patients with CRA, using episode-based methodology to account for differences in assumed duration of effect and frequency of dosing with these products. Methods: Using a large US health insurance claims database, we identified all patients with cancer who received ESAs between January 1, 2005, and June 30, 2005 (study period). For each such patient, we identified all unique episodes of care (EOCs) with DA or EA, and then compared mean weekly dose and cost of ESA therapy within these EOCs, which were calculated using the ratio of total dose received and total cost of ESA therapy, respectively, to total EOC duration; only the first EOC for each patient was considered. EOCs were assumed to begin on the date of first ESA admin-istration within the study period, and end on the date of final ESA administration (within the episode) plus an assumed duration of effect based on the ESA received and corresponding dose. We also estimated the ratio of mean weekly dose of EA (in units) to mean weekly dose of DA (in micrograms) (EA/DA weekly dose ratio). Multivariate regression was used to control for differences in baseline characteristics of EA and DA patients. Results: We identified a total of 1226 patients with complete EOCs with ESAs (EA, 381; DA, 845). DA patients were more likely to have had evidence of receipt of chemotherapy (54% vs 47% for EA; P = 0.02); they also had more comorbidities (mean Charlson comorbidity scores, 4.3 and 3.9, respectively; P < 0.01). Estimated mean (95% CI) weekly dose within EOCs was 97 μg (94-99) for DA, and 43,184 U (40,181-46,589) for EA; EA/DA weekly dose ratio was 445:1. Adjustment for differences in patient characteristics yielded a slightly lower ratio (403:1). Results were sen-sitive to the exclusion of EOCs consisting of a single administration of ESA therapy and/or the addition of an assumed duration of effect to the final ESA dose administered. Conclusions: Cost comparisons of DA and EA are sensitive to the assumed duration of effect added to the final dose of ESA therapy, especially for EOCs with relatively few administrations.