Clinical activity of luspatercept in patients with poor erythroid engraftment after allogeneic hematopoietic stem cell transplantation

After allogeneic hematopoietic stem cell transplantation (allo-HSCT), some patients continue to have severe anemia despite neutrophil and platelet engraftment, and persistent poor erythroid recovery often leaves them dependent on red blood cell transfusion. Management in this setting remains difficult. To examine the clinical activity of luspatercept in patients with anemia related to poor erythroid engraftment after allo-HSCT, focusing on hemoglobin (Hb) recovery, transfusion outcomes, and changes in bone marrow erythropoiesis. We retrospectively reviewed 12 patients treated at our center between January 2023 and December 2025 for anemia related to poor erythroid engraftment after allo-HSCT. All had achieved myeloid and megakaryocytic engraftment and complete donor chimerism confirmed by short tandem repeat analysis, but their peripheral blood Hb remained <80 g/L up to 2 months after transplantation. Luspatercept was given in addition to conventional therapy. We collected baseline characteristics, Hb levels before and after treatment, time to response, overall response, transfusion outcomes, changes in bone marrow erythroid status, and adverse events. Continuous variables were summarized as median (range), and categorical variables as counts and percentages. Exploratory analyses used the Wilcoxon test and Spearman correlation analysis. For Hb and marrow assessment, the post-treatment value generally referred to the measurement recorded at the first documented erythroid response. Because Hb was generally measured weekly, time to response was defined as the interval from the first luspatercept administration to the first observation of an Hb increase exceeding 15 g/L. Response was confirmed by reviewing the subsequent 14-day period, during which no red blood cell transfusion was administered. Twelve patients were included. The median age was 35.5 years (range, 21-61 years); 5 were male and 7 were female. The underlying diseases were acute myeloid leukemia (AML, n = 5), acute lymphoblastic leukemia (ALL, n = 3), aplastic anemia (AA, n = 3), and myelodysplastic syndrome (MDS, n = 1). Eleven patients (91.7%) achieved an erythroid response, and 1 did not. Among responders, the median time to response was 18 days (range, 7-75 days). Median Hb increased from 44 g/L (range, 31-70 g/L) before treatment to 113.5 g/L (range, 43-142 g/L) after treatment, corresponding to a median increase of 65 g/L (range, 7-98 g/L). None of the 11 responders required further red blood cell transfusions after treatment. Before treatment, bone marrow erythroid proliferation was active in 5 patients (41.7%) and decreased in 7 (58.3%); after treatment, the corresponding numbers were 11 (91.7%) and 1 (8.3%). Responders with active baseline erythroid proliferation had a shorter time to response than those with decreased erythroid proliferation ( P = 0.034). Baseline erythroid proportion was negatively correlated with time to response (rho = -0.65, P = 0.032), and showed a positive trend with Hb increase (rho = 0.55, P = 0.066). Adverse events observed during luspatercept treatment included abdominal distension, diarrhea, rash, and edema; all were grade 1 except edema, which was grade 1-2. In this small retrospective cohort, luspatercept demonstrated clinical activity and acceptable tolerability in patients with anemia related to poor erythroid engraftment after allo-HSCT. Eleven of 12 patients achieved an erythroid response and remained free from red blood cell transfusion during follow-up. Across the entire cohort, the median follow-up after the last luspatercept administration was 604 days (range, 18–983 days). Among responders, the median duration of follow-up after the last luspatercept administration was 706 days (range, 207–983 days). Baseline marrow erythroid findings may be associated with response kinetics, but this exploratory observation requires validation in larger studies.