Considerations With Luspatercept Failure in Lower-Risk MDS

For patients with lower-risk myelodysplastic syndromes (LR-MDS) and symptomatic anemia, the armamentarium has expanded significantly in recent years. The addition of luspatercept (Reblozyl) and imetelstat (Rytelo) to a landscape once defined almost entirely by erythropoiesis-stimulating agents (ESAs) has given clinicians meaningful options beyond watchful waiting and transfusion support. With these new options, clinical discourse now often focuses on when to move from one targeted agent to the next.

In a virtual Case-Based Roundtable event, Andrew Brunner, MD, hematologist-oncologist at Massachusetts General Hospital in Boston, Massachusetts, led a discussion with his fellow oncologists on strategies for treating patients with luspatercept-refractory LR-MDS.

Register today to join a Case-Based Roundtable near you.

CASE SUMMARY

• 68-year-old woman with low-risk MDS with symptomatic anemia
• No del(5q) detected; ring sideroblast-positive
• Comorbidities: mild chronic kidney disease (stage 2) and osteoarthritis
• Baseline labs (month 4): hemoglobin 8.3 g/dL, serum erythropoietin (EPO) 420 mU/mL, white blood cells 2200/µL, absolute neutrophil count 1200/µL, platelets 108,000/µL, serum ferritin 149 mcg/L, transferrin saturation 23%
• Month 4: initiated luspatercept 1 mg/kg every 3 weeks; hemoglobin 8.3 g/dL
• Month 6: luspatercept increased to 1.33 mg/kg every 3 weeks; hemoglobin 8.2 g/dL
• Month 8: luspatercept increased to 1.75 mg/kg every 3 weeks; hemoglobin 8.5 g/dL
• Month 9: red blood cell (RBC) transfusions required twice per month; increasingly symptomatic with fatigue; hemoglobin 8.3 g/dL despite maximum dose

Event Recap

The event began with a discussion of considerations for ensuring that frontline luspatercept has truly failed before committing to a second-line switch. One participant said he would first repeat a bone marrow biopsy with next-generation sequencing to rule out disease progression to a higher-risk MDS category or transformation toward AML, as either would shift the treatment algorithm away from anemia-directed therapy entirely and toward hypomethylating agents or transplant evaluation

Brunner suggested that serial serum EPO monitoring after luspatercept may help inform subsequent treatment selection. In his view, a marked rise in endogenous EPO after luspatercept could indicate a lower likelihood of benefit from adding an ESA and favor consideration of imetelstat.

A participant said that his group routinely repeats iron studies in patients who relapse on luspatercept, as clinically silent gastrointestinal blood loss can mimic disease progression. He described patients with previously unrecognized iron deficiency (for example, ferritin around 25 ng/mL) who experienced a secondary response after the bleeding source and iron deficiency were addressed. Similarly, another participant described a patient whose acquired copper deficiency, once corrected, allowed continued benefit from luspatercept for more than a year.

IMerge Data for Imetelstat

Brunner next reviewed the IMerge (NCT02598661) phase 3 trial, which enrolled 178 patients with confirmed MDS, no del(5q), low- or intermediate-risk by IPSS (International Prognostic Scoring System), prior ESA refractory or intolerant disease or EPO greater than 500 mU/mL, and transfusion dependence of at least 4 RBC units over 8 weeks.

Patients were randomized 2:1 to imetelstat 7.5 mg/kg intravenously every 4 weeks or placebo. The primary end point was RBC transfusion independence for at least 8 consecutive weeks. Imetelstat achieved that threshold in approximately 1 in 5 patients, a rate that held up over longer follow-up, with the placebo's apparent early response rate attenuating to roughly 2% by the end of the study. Among responders, the median duration of transfusion independence stretched to approximately 1 year.¹

Brunner emphasized that the median hemoglobin rise of 3.5 g/dL seen with imetelstat was clinically meaningful, but patients and clinicians alike needed to be counseled on timing: the separation between imetelstat and placebo curves did not become apparent until around week 9, requiring patience through the second and third cycles before drawing conclusions about efficacy.¹

Post hoc pooled analyses of IMerge also showed that prior luspatercept exposure did not significantly diminish imetelstat's clinical activity, though prior hypomethylating agent (HMA) use did reduce response rates. Brunner noted that imetelstat appeared effective regardless of the line of therapy in which it was used, and that response to or failure of a prior ESA did not predict imetelstat outcomes.²

Safety of Imetelstat

The safety profile of imetelstat drew considerable attention, particularly around treatment-emergent cytopenias. Brunner described the pattern as resembling that of lenalidomide, with the highest incidence of thrombocytopenia and neutropenia occurring during the first 2 cycles, followed by gradual recovery, though not necessarily back to baseline.

Importantly, post hoc analyses from the pooled IMerge studies showed that greater early declines in platelet and neutrophil counts during the first two treatment cycles were significantly associated with greater hemoglobin improvement.³ Brunner noted that these early cytopenias often identify patients who ultimately derive the greatest benefit and should be monitored and managed rather than viewed as an automatic reason to discontinue treatment.

A participant described a patient whose platelets dropped to the low 20s but who recovered with each cycle and was maintained on intermittent eltrombopag dosing without bleeding complications.

What Do Guidelines Recommend After Luspatercept Failure?

The National Comprehensive Cancer Network Guidelines for Myelodysplastic Syndromes (version 3.2026) designate imetelstat as the preferred second-line option for LR-MDS patients following luspatercept failure, with ESA plus granulocyte colony-stimulating factor and lenalidomide listed as other recommended options.⁴

Participants generally affirmed this hierarchy in practice, though the conversation made clear that no firm consensus definition of luspatercept failure exists. The boundary between continued monitoring and confirmed failure remains one that each practice navigates with some degree of individual calibration, whether defined by completing two cycles at the maximum dose of 1.75 mg/kg without response, hemoglobin trajectory over time, or transfusion frequency.

Register today to join a Case-Based Roundtable near you.

DISCLOSURES: Andrew M. Brunner has received consulting fees from Acceleron, Agios, BMS/Celgene, Gilead, Keros Therapeutics, Novartis, Takeda, and Taiho.

REFERENCES:

1. Platzbecker U, et al. Imetelstat versus placebo in patients with lower-risk myelodysplastic syndromes who have relapsed or are refractory to erythropoiesis-stimulating agents (IMerge): a double-blind, randomised, placebo-controlled, phase 3 trial. Lancet. 2024;403(10423):249-260.

2. Komrokji RS, et al. Clinical activity of imetelstat by prior therapy in low-risk myelodysplastic syndromes: analyses from pooled IMerge data. Blood Cancer J. 2025;15(1):217. https://doi.org/10.1038/s41408-025-01444-0

3. Zeidan AM, et al. Treatment-emergent cytopenias and clinical response to imetelstat. Presented at: 2025 American Society of Hematology Annual Meeting. Abstract 490.

4. NCCN. Clinical Practice Guidelines in Oncology. Myelodysplastic Syndromes, version 3.2026. Accessed February 10, 2026. https://www.nccn.org/professionals/physician_gls/pdf/mds.pdf