Experience with luspatercept therapy in patients with transfusion-dependent low-risk myelodysplastic syndromes in real-world clinical practice: exploring the positive effect of combination with erythropoietin alfa

Abstract

Background: Luspatercept, an inhibitor of the transforming growth factor beta (TGF-β) pathway, is a novel treatment for anemic patients with lower-risk myelodysplastic syndromes (MDS) with transfusion dependence (TD) who do not respond to erythropoiesis-stimulating agents (ESA) therapy or are not suitable candidates for this treatment. We present real-world experience with luspatercept therapy from two hematology centers in the Czech Republic.

Methods: By January 2024, 54 MDS patients (33 men, 21 women) with a median age of 74 years (range, 55-95) were treated with luspatercept ± ESA at two Charles University hematology centers in Prague and Hradec Králové. According to the WHO 2016 classification, the cohort included 32 MDS-RS-MLD, seven MDS-MLD, two patients with 5q- + ring sideroblasts (RS), 12 RARS-T, and 1 patient with CMML-0 + RS. SF3B1 mutation data were available for 45 patients. All patients were in the IPSS-R and IPSS-M lower-risk groups (except four IPSS-M high). The median follow-up was 17 months (range, 1-54). All patients were transfusion-dependent. Thirty-five (64.8%) patients had a high transfusion burden (HTB) with ≥ 4 transfusion units (TU)/8 weeks, and 19 (35.2%) had a low transfusion burden (LTB) (< 4 TU/8 weeks). The median time between diagnosis and initiation of luspatercept was 27 months (range, 4-156). ESA were used prior to luspatercept in 45 patients, and luspatercept was used as first-line treatment in nine patients. Thirty-one (61%) patients were treated simultaneously with ESA.

Results: Only patients who received luspatercept for ≥ 8 weeks (51 patients) were assessed. We evaluated the achievement of transfusion independence (TI) lasting 8, 12, 16, and 24 weeks. Thirty-two (62.7%) patients achieved TI for ≥ 8 weeks, 31 (60.7%) for ≥ 12 weeks, 29 (56.8%) for ≥ 16 weeks, and 25 (49%) for ≥ 24 weeks. Hematologic improvement (HI) without TI was achieved in six patients (11.7%). Overall, HI + TI was achieved in 38 patients (74.5%). Epoetin alfa was used simultaneously in 31 patients (60.7%). In 21 (55.2%) of all responding patients, concomitant therapy with epoetin alfa led to an improved response, with 16 reaching TI. Thirteen (25.5%) patients were nonresponders. Eight (21%) patients experienced therapy failure and became transfusion-dependent again. Optimal response required a gradual increase in the luspatercept dose to 1.75 mg/kg in up to 35 patients, with 23 responders (TI + HI). Response rates varied by transfusion burden: 79% in LTB and 50% in HTB reached TI. Of RS+ patients, 70% reached TI, while only one out of five RS- patients achieved TI. Among 39 SF3B1-positive patients, 61.6% achieved TI. In the low and very low IPSS-M groups, 86% of patients responded (TI + HI), compared to 62% in the moderate-low group. Luspatercept was well-tolerated, with no adverse events higher than grade II toxicity.

Conclusion: We have demonstrated in real-world clinical practice that luspatercept is a very effective agent, even in an unselected, pretreated, significantly TD MDS population. The effect was particularly high in the IPSS-M low and very low groups. We believe that the relatively high response rate in our patients was influenced by the frequent use of a higher dose (1.75 mg/kg) and especially by adding ESA to luspatercept in poorly responding patients.

Keywords: anemia; epoetin alfa; luspatercept; myelodysplastic syndromes; treatment.

Figure 1

Response rate related to achieving transfusion independence (TI) lasting ≥ 8, 12, 16, and 24 weeks.

Figure 2

The relationship of responses to therapy in relation to the severity of transfusion dependence prior to therapy. Shown is the percentage of achieving transfusion independence (TI), hematological improvement (HI), or percentage of nonresponders (NR) in both groups of patients with high transfusion dependence/burden ≥ 4 transfusion unit (TU)/8 weeks (HTB, 32 patients) or low transfusion dependence/burden (LTB, 19 patients) with < 4 TU/8 weeks. Assessment refers to the achievement of individual responses lasting ≥ 8 weeks.

Figure 3

The percentages of responses (assessed by achieving transfusion independence (TI), hematological improvement (HI), or no response (NR) depending on the status of the presence of ring sideroblasts (RS) in the bone marrow (RS+, RS−) and SF3B1 mutation positivity/negativity (45 patients were evaluated for the presence of the mutation). Assessment refers to the achievement of individual responses lasting ≥ 8 weeks. In RS-negative (RS−) and SF3B1-negative groups, there were no patients with HI.

Figure 4

The response rate as represented in the summary of hematological improvement (HI) plus transfusion independence (TI) in each WHO 2016 group. There was one patient with isolated 5q-, ring sideroblasts (RS), and SF3B1 mutation, and one CMML patient with ring sideroblasts (RS) and SF3B1 mutation. Assessment refers to the achievement of individual responses lasting ≥ 8 weeks.

Figure 5

Response rate as represented in the summary of hematological improvement (HI) plus transfusion independence (TI) in each IPSS-M (IPSS molecular) risk group. Assessment refers to the achievement of individual responses lasting ≥ 8 weeks.