BACKGROUND: Erythropoiesis-stimulating proteins (ESPs) are indicated for the treatment of chemotherapy-induced anemia (CIA). Evidence-based guidelines and systematic reviews of the management of CIA do not yet include all currently approved ESPs or all of the clinically relevant benefits and risks of ESPs. OBJECTIVES: The aims of this work were to provide up-to-date assessments of the clinical efficacy and effectiveness (ie, transfusions and quality-of-life [QoL] benefits) and safety (ie, risk of venous thromboembolism [VTE] and all-cause or treatment-associated death) of epoetin-alfa, epoetin-beta, and darbepoetin-alfa for the treatment of CIA in cancer patients with hemoglobin<11 g/dL. We also considered the impact of differences in study design, patients, and treatments on the results. METHODS: A systematic review of the literature was performed to identify and analyze English-language studies (controlled trials and prospective uncontrolled studies with >or=300 patients) published between 1980 and July 2005. The databases searched were MEDLINE and the Cochrane Library. Relevant abstracts from the last 2 annual meetings of the American Society of Clinical Oncology, American Society of Hematology, and European Society for Medical Oncology were also included. Studies were selected, using predefined eligibility criteria. Two reviewers had to agree on all included and excluded studies, and on all data extracted from each accepted study before they were entered into a relational database. Meta-analyses were performed to quantify benefit and risk outcomes. RESULTS: In total, 40 studies including 21,378 patients were eligible for analysis. Each ESP was found to have efficacy relative to standard care or placebo. The odds ratio (OR) for transfusions in studies of epoetin versus controls was 0.44 (95% CI, 0.35-0.55) and of darbepoetin versus controls was 0.41 (95% CI, 0.31-0.55). Patients receiving ESPs experienced a significant improvement in QoL; the mean difference in Functional Assessment of Cancer Therapy-Fatigue score for ESPs versus controls was 0.23 (95% CI, 0.10-0.36; P=0.001). The frequency of VTE and death was not significantly different between ESPs and control (VTE OR, 1.41 [95% CI, 0.81-2.47]; all-cause mortality OR, 1.00 [95% CI, 0.69-1.44]). CONCLUSIONS: This analysis of key clinical benefits and risks of epoetin and darbepoetin in the treatment of CIA found no clinically relevant differences between these drugs.