Background: Recent clinical trial data have generated safety concerns regarding ESA use, particularly as to whether the results observed in certain patients with breast cancer, lung cancer, and hematologic malignancies are generalizable to all patients with those tumor types. It is critical that the results of these studies are appropriately integrated into the total body of evidence in order to enable balanced conclusions to be drawn. We conducted a systematic literature review to assess the safety and efficacy of ESA therapy in patients with breast cancer, lung cancer, and hematologic malignancies.
Methods: The search was aligned with that conducted by the Cochrane Collaborative group (Bohlius 2006) and was therefore limited to randomized, controlled clinical trials in cancer patients that compared the use of ESAs with either observation (RBC transfusion as required) or placebo and that collected survival data (either long-term or during the study period). Results from trials were meta-analyzed using Comprehensive Meta-Analysis v 2.2 software. Heterogeneity between studies was evaluated using a chi-square test (Q). Random and fixed-effects models were used to estimate an overall odds ratio and 95% confidence interval; the inconsistency statistic, I2, is also provided.
Results: For breast cancer, seven studies were identified. Meta-analysis of data from these studies indicated an overall neutral risk of death for patients receiving ESAs, despite the large contribution (weighted at about 40% of the overall result) of the BEST study results (Leyland-Jones, 2005) (Table). Five ESA-controlled lung cancer studies were identified in CIA, one in radiotherapy, and one in anemia of cancer (AOC). In each of the studies in CIA, the ESA groups had neutral survival risks relative to the control group. Meta-analysis of the CIA, radiotherapy, and AOC lung cancer data combined indicated an overall neutral risk of death for patients receiving ESAs (Table). For hematologic malignancies, eight studies were identified (one of which had no deaths). Meta-analysis indicated an overall neutral risk of death for patients receiving ESAs (Table).
Conclusions: The results of some studies have suggested a detrimental effect on survival associated with ESA use in certain patients with breast, lung, and hematological malignancies. However, results do not appear to be generalizable to similarly designed, randomized controlled trials conducted in the same tumor types. Additionally, when these studies are integrated into the total body of evidence, ESAs have a neutral effect on survival in these patients.
