Long-Term Response Analysis of Transfusion Independence in Erythropoiesis Stimulating Agent-Naive Patients with Very Low-, Low-, or Intermediate-Risk Myelodysplastic Syndromes Treated with Luspatercept Vs Epoetin Alfa in the COMMANDS Trial

Introduction Luspatercept has demonstrated durable clinical efficacy in patients (pts) with lower-risk myelodysplastic syndromes (MDS) who are red blood cell (RBC) transfusion dependent (TD). In the phase 3 COMMANDS trial (NCT03682536), in erythroid stimulating agent (ESA)-naive pts with lower-risk TD MDS, pts in the luspatercept arm had a significantly longer duration of response; with a median (95% CI) cumulative duration (sum of all ≥ 12-week RBC-transfusion independent [TI] response episodes [week 1-end of treatment (EOT)]) of 154.7 weeks (118.4-not estimable [NE]) compared with 91.1 weeks (73.1-123.9) in the epoetin alfa arm (P = 0.0016, Santini V, et al, EHA Library 2024. P785). Overall duration of response was also longer in the luspatercept arm for prespecified subgroups including pts who were ring sideroblast (RS)+ and RS- (Della Porta MG, et al. Lancet Haem 2024. In press). The aim of this analysis was to assess the long-term clinical value of luspatercept in COMMANDS, including extended RBC-TI and cumulative response. Methods Eligible pts were aged ≥ 18 years, had Revised International Prognostic Scoring System very low-, low-, or intermediate-risk MDS, were ESA-naive, and required RBC transfusions. Pts were randomized 1:1 to receive luspatercept 1.0 mg/kg (titration up to 1.75 mg/kg) subcutaneously once every 3 weeks or epoetin alfa 450 IU/kg (titration up to 1050 IU/kg; maximum total dose 80,000 IU) subcutaneously once weekly. MDS disease status was assessed after 24 weeks; pts with clinical benefit (transfusion reduction ≥ 2 RBC units/8 weeks from baseline [BL] for any 8-week period in the 12 weeks prior to assessment and absence of disease progression) could continue treatment until unacceptable toxicity, withdrawal of consent, or other discontinuation criteria. Cumulative duration of RBC-TI was defined as the sum of all durations of RBC-TI for ≥ 12 or ≥ 24 weeks from week 1 to EOT. Results BL characteristics were generally similar between luspatercept and epoetin alfa arms (Platzbecker U, et al. Lancet 2023). At data cutoff (Sept 22, 2023), 81/182 (44.5%) pts in the luspatercept arm and 50/181 (27.6%) pts in the epoetin alfa arm achieved a longest single RBC-TI period lasting ≥ 1 year, from week 1-EOT (common risk difference 16.7, 95% CI 7.2-26.1, P = 0.0003; odds ratio [OR] 2.2, 95% CI 1.4-3.5). Among pts with RBC-TI ≥ 12 weeks (luspatercept, 139/182 [76.4%]; epoetin alfa, 101/181 [55.8%]), cumulative RBC-TI ≥ 1 year was achieved by 90/139 (64.7%) and 55/101 (54.5%) pts in the luspatercept and epoetin alfa arms, respectively. In all, 55 (30.2%) pts in the luspatercept arm and 25 (13.8%) pts in the epoetin alfa arm achieved a longest single RBC-TI period lasting ≥ 1.5 years (week 1-EOT; common risk difference 16.3, 95% CI 8.0-24.6, P < 0.0001; OR 2.8, 95% CI 1.6-4.8). The ≥ 1.5 years RBC-TI benefit with luspatercept vs epoetin alfa was observed across prespecified subgroups, including those who were RS- (24.5% vs 16.0%; OR 1.7, 95% CI 0.6-4.6), had BL serum erythropoietin ≤ 200 U/L (35.9% vs 15.3%; OR 3.1, 95% CI 1.8-5.5), and had non-mutated SF3B1 (21.5% vs 13.9%; OR 1.7, 95% CI 0.7-4.2). Progression to high-risk MDS occurred in 4 (2.2%) pts in the luspatercept arm and 6 (3.4%) pts in the epoetin alfa arm. Exposure-adjusted incidence rate (95% CI) for progression to high-risk MDS per 100 patient-years was 98.5 (37.0-262.5) in the luspatercept arm and 117.5 (52.8-261.6) in the epoetin alfa arm. In total, 7 (3.8%) pts in the luspatercept arm and 8 (4.4%) pts in the epoetin alfa arm progressed to acute myeloid leukemia (AML) after randomization. Median time to AML progression was NE in either arm (P = 0.9621, HR 1.0, 95% CI 0.4-2.9). Conclusions In this analysis of long-term responses in the COMMANDS trial, luspatercept continued to demonstrate sustained and durable clinical benefit vs epoetin alfa, maintaining responses of ≥ 1.5 years across multiple clinically relevant subgroups including RS− status, low BL serum erythropoietin, and non-mutated SF3B1. Progression rates to high-risk MDS and AML were low in both treatment arms. These data support luspatercept as the treatment of choice in TD ESA-naive pts with lower-risk MDS.