Real-world thrombotic and survival outcomes with erythropoiesis-stimulating agents in cancer-related anemia: A propensity-matched cohort Study

Abstract

Abstract Background: Erythropoiesis stimulating agents (ESAs) are frequently used to manage cancer related anemia by increasing hemoglobin levels and reducing transfusion requirements. However, their use has been consistently linked to an increased risk of thrombotic complications. Multiple randomized trials and meta-analyses have reported higher rates of venous and arterial thromboembolism including deep vein thrombosis (DVT), pulmonary embolism (PE), and arterial events among ESA treated cancer patients compared to controls. Despite these findings, real world data evaluating thrombotic outcomes in diverse cancer populations across different ESA formulations remain limited. This study aims to assess the incidence and patterns of thrombotic events associated with ESA use in a contemporary oncology cohort. Methods We conducted a retrospective cohort study using the TriNetX Research Network to evaluate the impact of ESAs on thrombotic outcomes in patients with cancer-related anemia from January 1st, 2015, to July 31st, 2024. Adult patients (aged ≥18 years) who received ESA therapy were propensity score matched 1:1 with patients who did not, balancing for age, sex, cancer type, and comorbidities. Those with a history of thrombotic events or anticoagulant use prior to the index treatment were excluded. Outcomes assessed included the incidence of DVT, PE, new anticoagulant use, and survival at 1-year. Risk differences (RD), relative risks (RR), odds ratios (OR), and Kaplan-Meier survival analyses with hazard ratios (HR) 95% confidence intervals (CIs) were used to compare event rates between groups. Results Initially 1,613 patients were identified in the ESA treatment group and 141,439 in the no-ESA treatment group. Prior to matching, ESA users were older on average (78.2 vs 74.3 years) with a higher prevalence of CKD (57% vs 19%), diabetes (33% vs 24%), and significantly lower hematocrit, hemoglobin, and erythrocyte counts. After matching, covariates were well balanced between groups, with standardized differences mostly < 0.05. A total of 1,606 patients were included in each matched cohort, with the majority having prostate, breast, bladder, colon, and kidney cancer. There was no statistically significant difference in the incidence of DVT (1.5% vs 2.1%; p=0.1851) or PE (22.0% vs 25.0%; p=0.0562) between the ESA and no ESA groups. However, new anticoagulant usage was significantly lower in patients who received ESA therapy (6.3% vs 10.2%; p<0.0001), with a RR of 0.62 (95% CI 0.49–0.78). Survival at 1-year was also significantly higher in the ESA group (73% vs 69%, p = 0.004, HR = 0.82, 95 CI: 0.72-0.94). Conclusion In this large real-world oncology cohort, ESA use was not associated with a statistically significant increase in the risk of DVT or PE. Contrary to prior concerns, overall anticoagulant use was higher among non-ESA patients. Additionally, 1-year survival was higher with ESA usage. These findings suggest that, with appropriate patient selection, ESA therapy may be safely utilized for cancer-related anemia without substantially increasing thrombotic risk. Further studies are warranted to explore the underlying mechanisms and to validate these observations across specific cancer subtypes and ESA formulations.