Assessing risks and benefits of erythropoiesis stimulating agents (ESAs) by tumor type in advanced solid tumor patients (ST): A meta-analysis of 16 randomized controlled trials (RCTs) including 4,646 patients

Raftopoulos, H.; Gralla, R. J.; Bria, E.

NlmCategory="UNASSIGNED">9541 Background: ESAs as supportive care agents remain controversial. Caution involving efficacy and safety resulted in 2 ODAC meetings and 2 FDA label revisions, including a black box warning; 2 Cochrane Group meta-analyses (MA) found ESA benefits to be only a 1-1.5g/dl Hb rise and a 1 unit transfusion reduction (Bohlius JNCI 2005, 2006). Quality of life and symptom benefit claims are no longer permitted, with only a 3-4% advantage (Osterborg JCO 2002) questioning clinical relevance. Such symptom improvement is as likely due to disease parameters or response to therapy as it is to modest Hb effects (Wisloff Eur J Hem 2005). To assess further potential risks of ESAs, we initiated a data review addressing objections to the Cochrane MA. We conducted an abstracted data MA using these criteria: phase III RCTs comparing ESAs to placebo or control; trials with a single tumor type or closely related cancer, to reduce heterogeneity (Het); trials in ST. Adjuvant or neoadjuvant studies were excluded due to inadequate survival data. CMA software v2 (Biostat, NJ) calculated combined thrombosis rates and survival HRs, using fixed and random effects models including Het testing. Sensitivity analyses by tumor type were performed. See table Conclusions: Benefits of ESAs in patients with ST are modest and risks substantial. This analysis exploring risks by disease type shows that significant thrombotic risk extends across all tumor types. Detrimental survival outcomes appear to be related to disease type, suggesting the possibility of different mechanisms contributing to this risk. While overall there is a significant negative impact on survival, breast and head and neck cancers account for this. Continued ESA use requires consideration of these outcomes and available alternatives (observation or transfusion). Patients should receive detailed information concerning potential risks and small benefits of ESAs. [Table: see text] [Table: see text].