Real-world treatment patterns and outcomes of first-line luspatercept in patients with myelodysplastic syndromes in the United States

Abstract

Abstract Background: Patients with lower-risk myelodysplastic syndromes (LR-MDS) who experience anemia require strategies to reduce transfusion dependency (TD) in order to enhance their quality of life and survival. Based on the phase 3 COMMANDS trial, luspatercept received approval in August 2023 in the United States for the first-line (1L) treatment of anemia in adults with LR-MDS who may require red blood cell (RBC) transfusions and were naive to erythropoiesis-stimulating agents (ESAs). Since then, limited real-world data have been reported on 1L luspatercept use. Additionally, no data currently exist on the real-world effectiveness of second-line (2L) ESAs following 1L luspatercept discontinuation. Herein, we describe patient characteristics, clinical outcomes, and treatment patterns of patients with MDS receiving 1L luspatercept and receiving 2L ESA following 1L luspatercept discontinuation in a real-world setting. Methods: This retrospective, observational US cohort study used the Flatiron Health electronic record database. Eligible patients received a MDS diagnosis after January 1, 2020, were ≥18 yrs old at diagnosis, with known ring sideroblast (RS) status, and were included in the Flatiron database with ≥2 visits. TD was defined as patients requiring ≥1 RBC units during 16 wks prior to treatment initiation, while non-TD (NTD) was defined as patients who did not receive any transfusions during those 16 wks. Outcomes included achievement of transfusion independence (TI; no RBC transfusions for ≥12 wks in Wks 1-24 of treatment) and modified erythroid hematologic improvement (mHI-E; per IWG 2018 definition) for ≥16 wks in Wks 1-24. Treatment patterns included dose escalations and treatment discontinuations. Clinical outcomes were also reported for patients who received 2L ESA after 1L luspatercept discontinuation. Results: In total, 320 patients received 1L luspatercept; the median age was 78 yrs (95% CI, 53-85). Of the patients with available data for each characteristic, 31.4% (96/306) were TD; 27.2% (87/320) had RS- status; and 66.5% (115/173) had serum erythropoietin (sEPO) <200 U/L. Median baseline hemoglobin (Hb) level was 9 g/dL (95% CI, 7-11). Median time from diagnosis to luspatercept initiation was 1.6 mo (range, 0-131.2). At a median follow-up of 8.8 mo, TI was achieved in 61.5% (59/96) of TD patients. Data were consistent across subgroups, with 53.3% (16/30) of RS- patients and 69.2% (18/26) of patients with sEPO <200 U/L achieving TI. mHI-E response was assessed for 293 patients with available data, with 50.9% (n=149) achieving a response. mHI-E response per subgroup was 42.9% (33/77) in RS- patients and 59.5% (66/111) in patients with sEPO <200 U/L. Luspatercept was initiated at 1 mg/kg dose in98.9% (n=263) out of 266 patients with a documented initial dose. Overall, 27.5% (88/320) of patients required a dose escalation. Of those patients, 86.4% (76/88) escalated from 1 to 1.33 mg/kg and 52.3% (46/88) escalated from 1.33 to 1.75 mg/kg. Luspatercept was discontinued in 28.1% (90/320) of patients. Of those patients, 36.7% (33/90) discontinued before receiving the maximum luspatercept dose (1.75 mg/kg), primarily due to lack of anemia response (33.3%; 11/33). Following 1L luspatercept discontinuation, 25.6% (23/90) of patients initiated 2L ESA therapy, 13 of whom were TD. At a median follow-up of 5.7 mo, TI ≥8 wks in Wks 1-24 of treatment was achieved in 38.5% (5/13) of TD patients. Of 19 patients with available data for mHI-E response assessment, 26.3% (n=5) achieved a response ≥8 wks. Conclusions:In real-world practice, patients with MDS treated with 1L luspatercept achieved TI and mHI-E, with response rates comparable to those observed in the registrational clinical trial. Therapeutic benefits with 1L luspatercept were observed across all subgroups, including RS- and sEPO <200 U/L populations. Overall, 27.5% of patients required dose escalation and 36.7% discontinued treatment before receiving the maximum luspatercept dose recommended by its prescribing information, primarily due to lack of response. Additionally, 2L ESA remained an effective treatment following 1L luspatercept discontinuation. These findings highlight the clinical benefit of luspatercept, demonstrating its effectiveness as a treatment option for patients with MDS in real-world clinical practice.