Real-world comparison of thrombotic events in patients with Myelodysplastic Syndromes treated with luspatercept versus erythropoietin-stimulating agents using trinetx database

Abstract Background: Luspatercept, an erythroid maturation agent, has shown efficacy in improving anemia in both β-thalassemia and lower-risk myelodysplastic syndromes (MDS). In the BELIEVE trial, which led to its approval in transfusion-dependent β-thalassemia, thromboembolic events were notably higher in patients with risk factors such as splenectomy. More recently, the COMMANDS trial demonstrated superior transfusion independence with luspatercept over erythropoiesis-stimulating agents (ESAs) in ESA-naive, lower-risk MDS patients. However, limited data exist regarding comparative thrombotic risk in real-world MDS populations. We conducted a multicenter, real-world study to evaluate thrombotic event rates associated with luspatercept versus ESAs in MDS patients without prior venous thromboembolism (VTE). Methods: We queried the TriNetX Research Network, a federated real-time electronic health record (EHR) database covering over 100 million patients across major healthcare organizations. Adults (≥18 years) with a diagnosis of MDS (ICD-10-CM code D46) and no prior history of deep vein thrombosis (DVT) or pulmonary embolism (PE) were included. Patients were stratified based on treatment with luspatercept or ESAs (epoetin alfa or darbepoetin alfa). Propensity score matching (PSM) using a 1:1 nearest-neighbor algorithm without replacement was performed to balance demographics and clinical covariates. Primary outcomes included the incidence of DVT, PE, and composite VTE events within 12 months of therapy initiation. Results: A total of 6,448 patients with MDS and no prior VTE were identified: 5,698 received ESAs and 750 received luspatercept. The overall mean age was 72.5 ± 13.5 years, with 48.1% male. After PSM, 741 patients were included in each treatment arm. The matched cohort had a mean age of 76.7 ± 10.7 years, and 41.3% were female. In the ESA group, 58/741 (7.8%) developed DVT compared to 27/741 (3.6%) in the luspatercept group (Hazard Ratio [HR] 1.4; 95% CI, 0.9–2.3). PE occurred in 27 ESA-treated patients (3.6%) versus 13 luspatercept-treated patients (1.8%) (HR 1.3; 95% CI, 0.6–2.7; p=0.941). The overall composite VTE rate was 10.0% (74/741) in the ESA group and 4.9% (36/741) in the luspatercept group (HR 1.3; 95% CI, 0.9–2.0; p=0.732). Conclusion: In this large, multicenter real-world analysis, we observed a numerically lower incidence of thrombotic events with luspatercept compared to erythropoietin-based agents in patients with MDS, though differences did not reach statistical significance. These findings support the thrombotic safety profile of luspatercept observed in clinical trials and warrant further prospective validation.