Real-world hemoglobin (Hb) outcomes of patients with lower-risk myelodysplastic syndromes (LR-MDS) receiving first-line (1L) luspatercept or 1L erythropoiesis-stimulating agents (ESAs) in the US Free

Background:

More than 80% of patients with LR-MDS have anemia, which is associated with a higher risk of cardiac events, death, and lower quality of life (QoL). The treatment goal for anemic patients with LR-MDS is to maintain Hb between 10 and 12 g/dL. While ESAs have been the mainstay anemia treatment in LR-MDS, 68% of patients receiving ESAs experience treatment failure (Fonseca et al. Clin Lymphoma Myeloma Leuk. 2024). Such patients often receive red blood cell (RBC) transfusions for anemia management, but prolonged RBC transfusion dependence (RBC-TD) can worsen disease prognosis. Luspatercept was approved as 1L anemia treatment for LR-MDS by the US FDA in August 2023 based on data from the COMMANDS trial. Patients in the trial achieving Hb ≥10 g/dL experienced durable RBC transfusion independence, and improved QoL regardless of therapy (Santini et al. Blood. 2024; Oliva et al. HemaSphere. 2024). This real-world study aims to assess Hb outcomes of patients with LR-MDS receiving 1L luspatercept or ESA after 1L luspatercept approval in the US.

Methods:

Data were collected between November 2024 and March 2025 from a retrospective medical records' review of patients with LR-MDS receiving 1L luspatercept or ESAs. Eligible adult patients had an IPSS/IPSS-R-defined diagnosis of LR-MDS and initiated 1L luspatercept or ESA between August 28, 2023 and July 31, 2024 (treatment initiation date = index date). Patient characteristics and outcomes were descriptively analyzed. Among patients with baseline (BL) Hb <10 g/dL before 1L treatment initiation and known Hb during the first 6 months of treatment, time to achieving Hb ≥10 g/dL was estimated using Kaplan-Meier (KM) analysis and multivariable Cox regression.

Results:

Data were collected for 418 patients (luspatercept cohort, n=213; ESA cohort, n=205).Median age at index for patients in the luspatercept and ESA cohorts was 68.4 and 66.6 years, respectively; most patients were male (55.9%; 61.0%) and White (68.5%; 67.8%), with a median follow-up of 10.8 and 11.2 months, respectively. In the luspatercept and ESA cohorts, IPSS/IPSS-R status was intermediate-1/intermediate for 23.5% and 12.7% of patients, respectively; 46.9% and 12.7% had an SF3B1 mutation and 88.3% and 81.5% had ECOG 0/1. Of patients with known ring sideroblast and SF3B1 mutation status, 52.0% (89/171) of patients in the luspatercept cohort and 81.5% (119/146) of patients in the ESA cohort were RS negative. BL serum erythropoietin was <200 IU/L for 37.1% (63/170) of patients in the luspatercept cohort and 74.0% (111/150) of patients in the ESA cohort.

KM and multivariable analyses were performed in 394 patients with known follow-up Hb and BL Hb <10 g/dL (luspatercept cohort, n=194; ESA cohort, n=200). Among these, 45.4% of patients in the luspatercept cohort and 42.5% of patients in the ESA cohort had BL Hb <8 g/dL, and 22.7% and 17.0%, respectively, were RBC-TD at BL (defined as ≥2 RBC units/transfusions per 8 weeks). More patients in the luspatercept cohort vs the ESA cohort achieved Hb ≥10 g/dL during the first 6 months of treatment (69.6% vs 58.0%) with a quicker response time (median [95% CI], 2.3 months [2.1-3.0] vs 5.3 months [3.2-not estimable]; P=0.003). The adjusted hazard ratio (aHR) for achieving Hb ≥10 g/dL with luspatercept vs ESA was 1.44 (95% CI, 1.1-2.0; P=0.023), indicating that patients receiving 1L luspatercept had a 44% higher chance of achieving Hb ≥10 g/dL at any given time vs patients receiving 1L ESA. In the same model, patients with BL Hb ≥8 g/dL had a higher chance of achieving Hb ≥10 g/dL (aHR, 2.0 [95% CI, 1.5-2.7]; P<0.001), and BL RBC-TD patients had a lower chance (aHR, 0.5 [95% CI, 0.3-0.8]; P=0.001).

Conclusions:

During the first 6 months of treatment,patients treated with 1L luspatercept vs 1L ESA showed a greater and faster response rate of achieving Hb ≥10 g/dL, demonstrating the significant clinical benefit of 1L luspatercept for Hb improvement in patients with LR-MDS. Controlling for 1L treatment and patient characteristics, a statistically significant likelihood of Hb improvement was seen for patients with BL Hb ≥8 g/dL. However, patients with BL RBC-TD had a significantly lower probability of achieving Hb ≥10 g/dL. These findings underscore the importance of early anemia treatment initiation in LR-MDS before patients' Hb drops below 8 g/dL and/or they become RBC-TD which may improve Hb outcomes and optimize response to therapy.