Erythropoiesis-Stimulating Agents and the Risk of Thrombosis in Cancer-Associated Anemia

CHICAGO–Patients with cancer are at risk for developing anemia. The treatment of cancer-associated anemia remains a challenge, given that the most effective therapy for it (erythropoiesis-stimulating agents [ESAs]) can also increase the risk for thromboembolic events. Results from a large meta-analysis presented at the 2016 ASCO Annual Meeting provide new evidence about the association between ESA treatment, thrombosis, and survival – and how risks for these events differ by tumor type.

Vineeth Sukrithan, MD, of the Jacobi Medical Center/Albert Einstein College of Medicine in New York, and authors expanded on data from a meta-analysis of 91 randomized clinical trials, encompassing 20,102 patients, published in 2012.2 The earlier meta-analysis found that ESAs reduced the need for red blood cell transfusions, but increased the risk of thromboembolic events in patients with cancer-associated anemia.

In  the latest report, Dr. Sukrithan and researchers analyzed data from several of the trials included in the Tonia et al. analysis, plus three more randomized, controlled trials comparing ESAs with placebo or control in specific solid tumor types, to determine if thromboembolic or mortality risks differed by tumor type. Trials with inadequate survival data were excluded.

A total of 34 randomized, controlled trials were included in the analysis, with a total of 9,808 patients with cancer and anemia.

The authors calculated combined thrombosis rates and odds ratios (ORs) for survival and also performed sensitivity analyses by tumor type (head/neck, breast, lung, and pelvic).

As with the earlier meta-analysis, Dr. Sukrithan and co-authors found that treatment with ESAs nearly doubled the thrombosis rate, independent of tumor type (OR=1.93; p<0.01). Patients with breast, lung, and pelvic malignancies had significantly higher risks for thrombosis (TABLE).

Treatment with ESAs also slightly increased mortality risk, but this finding was not statistically significant (OR=1.09; p=0.12). The odds of survival, however, were markedly different according to tumor type. Survival was notably lower in breast and head/neck cancers, and there was no clear survival effect in lung cancer and in pelvic malignancies (TABLE).

“These results indicate that survival detriments are not explained by thrombosis, in that this risk is equal in all cancers analyzed,” Dr. Sukrithan and authors concluded. “These findings suggest that, in certain cancers, ESAs may play a role in tumor promotion while, in others, ESAs may not have an activating role in tumor growth.”


    1. Sukrithan V, Gralla RJ. Do risks with the use of erythropoietin stimulating agents (ESAs) differ by tumor type? Implications concerning survival and thrombosis based on meta-analysis (MA). Abstract #10023. Presented at the 2016 ASCO Annual Meeting, June 6, 2016; Chicago, IL.
    2. Tonia T, Mettler A, Robert N, et al. Erythropoietin or darbepoetin for patients with cancer. Cochrane Database Syst Rev. 2012;12:CD003407.

TABLE. Thrombosis and Survival Rates Among Patients With Cancer-Associated Anemia Treated With Erythropoiesis-Stimulating Agents
Randomized Controlled Trials (Patients) Odds Ratio (95% CI) p Value
Overall effect 27 (8,259) 1.93 (1.6-2.34) <0.01
Head/neck 3 (794) 2.17 (0.89-5.3) 0.09
Breast 8 (4,518) 1.94 (1.46-2.57) <0.01
Lung 7 (1,747) 1.82 (1.35-2.48) <0.01
Pelvic 9 (1,200) 2.24 (1.19-4.21) 0.01
Overall effect 32 (9,692) 1.09 (0.98-1.2) 0.12
Head/neck 5 (1,397) 1.23 (0.98-1.53) 0.07
Breast 9 (4,713) 1.20 (1.03-1.39) 0.02
Lung 10 (2,417) 0.94 (0.73-1.2) 0.6
Pelvic 8 (1,165) 1.06 (0.85-1.33) 0.6