2026 ASH®
Annual Meeting & Exposition December 12-15, 2026 ▪ New Orleans, Louisiana

Paper Details

Abstracts, authors, affiliations, and disclosures.

Monday, December 14, 2026
06:00 p.m. - 08:00 p.m.
MCCNO - Hall F
Technology and Procedures, Molecular Testing, Omics Technologies, Gene Editing, Treatment Considerations, Combination Therapy, Drug Development, Supportive Care, Therapy Sequence, Biological Therapies, Transplantation, Immunotherapy, Monoclonal Antibody Therapy, Adverse Events, Artificial Intelligence (AI), Deep Learning, Machine Learning, Biological Processes, Pathogenesis, Genomics, Hematopoiesis, Molecular Biology, Clinical Practice (Health Services and Quality), Clinical Guidelines, Patient Safety, Quality Improvement, Diseases, Bone Marrow Failure Syndromes, Aplastic Anemia, Immunology, Neutropenia, Myeloid Malignancies, Acute Myeloid Malignancies, Chronic Myeloid Malignancies, Measurable Residual Disease, Research, AML, Translational Research, MDS, MPN, CMML, Clinical Research, Clinical Trials, Epidemiology, Health Outcomes Research, Patient-Reported Outcomes, Real-World Evidence, Survivorship, Study Population, Human, Adult, Elderly
PAPER

5616 Erythroid response and post-failure survival with luspatercept by line of therapy in lower-risk myelodysplastic syndromes

Ian Bouligny, M.D., Emmanuel Almanza, MD2, Alex Bataller, MD, PhD3, Alexandre Bazinet, MD, MSc4, Kelly Sharon Chien, MD5, Danielle Hammond, MD6, Mahesh Swaminathan, MD7, Guillermo Montalban-Bravo, MD8, Guillermo Garcia-Manero, MD9
1 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
2 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
3 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
4 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
5 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
6 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
7 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
8 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
9 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States

Background

Luspatercept is approved for anemia in lower-risk MDS (LR-MDS), but registration trials enrolled predominantly first-line or second-line populations after erythropoiesis-stimulating agent (ESA) failure. Whether erythroid response is preserved when luspatercept is given later in the treatment sequence, and what happens after luspatercept failure, are unresolved.

 

Methods

This was a retrospective study of LR-MDS patients treated at MD Anderson Cancer Center (2014–2025) who received luspatercept. Primary endpoint was overall survival (OS) from the date of luspatercept failure to the date of death or last follow-up across treatment lines. Failure was classified from predefined EFS events: recurrent cytopenias, return of transfusion dependence, blast elevation, or lack of response. Secondary endpoints included the composite IWG 2018 hematologic improvement–erythroid (HI-E) response rates across treatment lines and baseline predictors of IWG 2018 response. Firth penalized logistic regression and Cox regression were used for these analyses.

 

Results

Among 146 LR-MDS patients, the median age at luspatercept initiation was 75.3 y (IQR 69.0–79.1) and 61% were male. Baseline ring sideroblasts were ≥15% in 71%, 61% had prior ESA, 73% were transfusion-dependent, 14% had therapy-related MDS, and 3% had marrow blasts ≥5%. SF3B1 was mutated in 64% of first-line and 67% of later-line episodes. The median IPSS-R was 3.0 (IQR, 2.0–3.5) and the median IPSS-M was -0.5 (IQR, -0.9–0.3). Eighty-eight were treated first-line, 38 were treated second line, and 24 were treated in the third or later line; 4 were retreated with luspatercept after luspatercept failure.

 

IWG 2018 HI-E achievement fell with increasing line of therapy: 68% (95% CI, 56–79) at line 1, 42% (95% CI, 23–63) at line 2, and 22% (95% CI, 6–48) at line ≥3 (116 evaluable; trend p=0.00015; line 1 vs later OR 4.06, 1.73–9.90, p=0.0005). Median cycles delivered decreased across lines, though not significantly different (7, 6, 5, respectively; p=0.47). Among 72 evaluable first-line patients, 49 (68%) of whom achieved HI-E, the strongest predictors of response were the SF3B1 mutation (OR 3.88, 1.21–13.7; p=0.022), baseline hemoglobin (OR 1.60, 1.07–2.53, p=0.021), and serum EPO ≤200 IU/L (OR 12.3, 1.74–149; p=0.011; 82% vs 20% response, n=27). Transfusion dependence (OR 0.17, 0.04–0.56; p=0.003) and increasing IPSS-M score (OR 0.28, 0.10–0.64; p=0.002) predicted lack of response. In a two-covariate model (n=59), SF3B1 (OR 3.47, 1.01–13.1; p=0.048) and transfusion dependence (OR 0.18, 0.03–0.70; p=0.012) remained independent predictors. IWG 2018 HI-E response tracked the SF3B1 lesion rather than ring-sideroblast morphology: SF3B1-mutant/RS+ 88%, SF3B1-mutant/RS− 100% (n=3), SF3B1-wild-type/RS+ 50%, wild-type/RS− 61%. By SF3B1 codon, HI-E was 83% for K700E, 92% for non-K700E, and 60% for wild-type.

 

At a median follow-up time of 31.2 months, median OS from first-line initiation was 42.2 months (86% at 12 mo, 72% at 24 mo). Across all lines, 74 patients met failure criteria (40 refractory, 34 relapsed/progressed). Of 74 failures, 44 received subsequent therapy: hypomethylating agents in 23 (52%), imetelstat in 6 (14%), lenalidomide in 4 (9%), and trial agents in 11 (25%). Median post-luspatercept failure OS was 21.6 months overall; 23.6 months after first-line failure, 23.9 months after second-line failure, and 9.2 months at the third or later line (log-rank p=0.002). Line ≥3 luspatercept versus lines 1–2 was associated with worse HR 2.99 (1.58–5.67; p=0.0008), persisting after separate adjustment for age (HR 3.17; 95% CI 1.64–6.10; p=0.0006), transfusion dependence (HR 2.41; 95% CI, 1.20–4.81; p=0.0129), and IPSS-M (2.44, 95% CI, 1.04–5.72; p=0.041). Independent predictors of failure were line of therapy (OR 2.33, 1.12–6.14; p=0.020) and IPSS-M (OR 2.73, 1.25–6.96; p=0.0097).

 

Conclusions

Post-failure survival diminishes beyond second line and erythroid response to luspatercept in LR-MDS declines with line of therapy. SF3B1 mutation, transfusion independence, and low serum EPO identify patients likely to respond by IWG 2018 criteria, with response tracking SF3B1 rather than sideroblast morphology. These data argue for prospective evaluation of sequencing strategies after failure.

Ian Bouligny, M.D.: No relevant conflicts of interest to declare.