Erythropoiesis-stimulating agents in oncology: a study-level meta-analysis of survival and other safety outcomes

Abstract

Background: Cancer patients often develop the potentially debilitating condition of anaemia. Numerous controlled studies indicate that erythropoiesis-stimulating agents (ESAs) can raise haemoglobin levels and reduce transfusion requirements in anaemic cancer patients receiving chemotherapy. To evaluate recent safety concerns regarding ESAs, we carried out a meta-analysis of controlled ESA oncology trials to examine whether ESA use affects survival, disease progression and risk of venous-thromboembolic events.

Methods: This meta-analysis included studies from the 2006 Cochrane meta-analysis, studies published/updated since the 2006 Cochrane report, and unpublished trial data from Amgen and Centocor Ortho Biotech. The 60 studies analysed (15 323 patients) were conducted in the settings of chemotherapy/radiochemotherapy, radiotherapy only treatment or anaemia of cancer. Data were summarised using odds ratios (ORs) with 95% confidence intervals (CIs).

Results: Results indicated that ESA use did not significantly affect mortality (60 studies: OR=1.06; 95% CI: 0.97-1.15) or disease progression (26 studies: OR=1.01; 95% CI: 0.90-1.14), but increased the risk for venous-thromoboembolic events (44 studies: OR=1.48; 95% CI: 1.28-1.72).

Conclusion: Though this meta-analysis showed no significant effect of ESAs on survival or disease progression, prospectively designed, future randomised clinical trials will further examine the safety and efficacy of ESAs when used according to the revised labelling information.

Figure 1

Effect of erythropoiesis-stimulating agent (ESA) use on mortality. (A) Flow diagram of studies analysed. The odds ratio (OR) (95% confidence interval (CI)) for the mortality of all 60 studies and for each cancer setting is given. Pts indicates patients. (B) Meta-analysis of mortality using 60 controlled studies carried out in the settings of chemotherapy, anaemia of cancer (AoC) or radiotherapy only. This analysis included 15 323 patients, with 8343 ESA-treated patients and 6980 control patients. An asterisk denotes radiochemotherapy studies. Three chemotherapy studies are not listed, as they reported no deaths in the ESA or control arms. (C) Influence of each chemotherapy study on the mortality OR for all chemotherapy studies. This analysis calculated the mortality OR for the chemotherapy studies combined after excluding each listed chemotherapy study. Of the 47 chemotherapy studies identified, 3 were not listed in the plot as they reported no deaths in the ESA or control arms. (D) Study-level meta-analyses of mortality in 20 chemotherapy studies with long-term follow-up. This analysis included 8145 patients, with 4183 ESA-treated patients and 3962 control patients. An asterisk denotes radiochemotherapy studies.

Figure 1

Effect of erythropoiesis-stimulating agent (ESA) use on mortality. (A) Flow diagram of studies analysed. The odds ratio (OR) (95% confidence interval (CI)) for the mortality of all 60 studies and for each cancer setting is given. Pts indicates patients. (B) Meta-analysis of mortality using 60 controlled studies carried out in the settings of chemotherapy, anaemia of cancer (AoC) or radiotherapy only. This analysis included 15 323 patients, with 8343 ESA-treated patients and 6980 control patients. An asterisk denotes radiochemotherapy studies. Three chemotherapy studies are not listed, as they reported no deaths in the ESA or control arms. (C) Influence of each chemotherapy study on the mortality OR for all chemotherapy studies. This analysis calculated the mortality OR for the chemotherapy studies combined after excluding each listed chemotherapy study. Of the 47 chemotherapy studies identified, 3 were not listed in the plot as they reported no deaths in the ESA or control arms. (D) Study-level meta-analyses of mortality in 20 chemotherapy studies with long-term follow-up. This analysis included 8145 patients, with 4183 ESA-treated patients and 3962 control patients. An asterisk denotes radiochemotherapy studies.

Figure 2

Effect of erythropoiesis-stimulating agent (ESA) use on disease progression. (A) Flow diagram of studies analysed. Of the 60 controlled ESA studies that measured survival, 26 also measured a disease-progression outcome. AoC indicates anaemia of cancer. (B) Study-level meta-analysis of disease progression-related endpoints in 26 controlled studies carried out in the settings of chemotherapy, AoC or radiotherapy only. This analysis included 9646 patients, with 4905 ESA-treated patients and 4721 control patients. An asterisk denotes studies where disease progression was evaluated only as part of tumour assessment.

Figure 3

Study-level meta-analyses of venous-thromboembolic event (VTE) risk in 44 controlled studies carried out in the settings of chemotherapy, anaemia of cancer (AoC) or radiotherapy only. This analysis included 13 196 patients with 7237 ESA-treated patients and 5959 control patients.