Background Spaepen et al. (The Oncologist 2008;13:596–607) reported lower costs to achieve similar outcomes with darbepoetin alfa (DARB) compared with epoetin alfa (EPO-A) and epoetin β (EPO-B) for similar patient profiles in the treatment of chemotherapy-induced anaemia. Data were derived from the IMS Hospital Disease Database, a longitudinal database in secondary care unique to Belgium.

Purpose The objectives of this study were to assess the applicability of the Spaepen et al. analysis in the German setting, and to evaluate differences in cost between ESAs in Germany.

Materials and methods To adapt the Belgian data to Germany, differences in epidemiology and treatment patterns were examined. Costs per patient in the Belgian dataset were replaced with German-specific unit costs (Euro 2011) for drugs, outpatient visits, hospitalisations and blood transfusions. Adjustments were made for tumour-specific incidence, chemotherapy use, setting of care (hospital vs retail) and frequency of DARB administration. Costs were analysed using a mixed-effects model stratifying for propensity score quintiles as in Spaepen 2008. Data sources included Eurostat, national cancer registries, IMS sales data, and reimbursement and treatment guidelines.

Results The German and Belgian populations were comparable in terms of age, gender, ESA use and blood transfusions. After adjusting for treatment-related factors, total (mean±SE) DARB costs (€7,237±516) were 17% lower compared with EPO-A (€8,720±408; p=0.0004) and 14% lower compared with EPO-B (€8,392±544; p=0.0385). Anaemia-related costs per patient were not statistically different between DARB (€2,893±140), EPO-A (€2,940±78; p=0.7628) or EPO-B (€2,529±153; p=0.0736). Mean duration of treatment was significantly shorter for DARB (43.22±2.37 days) compared with EPO-A (54.15±1.23 days; p<0.0001) and EPO-B (54.62±2.51 days; p=0.0010).

Conclusions Total costs were significantly lower in patients receiving DARB compared to EPO-A or EPO-B whereas anaemia-related costs were not significantly different. By using published epidemiologic and treatment pattern data, it was possible to adapt the Belgian Hospital database to the German setting.