Efficacy and safety of luspatercept in the treatment of anemia in allo-HSCT for patients with hematological diseases

Abstract Background: Persistent anemia affects up to 30% of long-term survivors after allogeneic hematopoietic stem-cell transplantation (allo-HSCT). Current treatments have limited efficacy and substantial side effects. Luspatercept is a first-in-class erythroid maturation agent, with several encouraging results reported in small retrospective studies. We therefore conducted a relative larger, rigorously defined retrospective analysis to clarify the efficacy and safety of luspatercept in adults with anaemia persisting after allo-HSCT. Methods: This retrospective analysis included 96 consecutive patients treated at our center between November 2022 and December 2024. The eligibility criteria were post-transplant anemia (hemoglobin <65g/L despite ≥4 weeks of standard care), complete donor chimerism (≥95%), underlying disease in complete remission, and no evidence of haemolysis, bleeding or nutritional deficiency. Diagnosis included AML (47), ALL (30), aplastic anemia (15), MDS (3), and myelofibrosis (1). All patients with haematological malignancies received myeloablative conditioning regimens based on fludarabine/busulfan/melphalan or fludarabine/busulfan/TMI. The patient with aplastic anaemia received an immuno-myeloablative regimen. The primary end-point was erythroid response defined as a sustained (≥14 consecutive days) rise in haemoglobin ≥15 g/L from baseline without red-cell transfusion. Secondary endpoints included hemoglobin ≥80 g/L, time to response, and safety profiles. Results: Grafts were obtained from matched unrelated donors (n=16), matched sibling donors (n=16) and haplo-identical family donors (n=64). ABO matching was complete in 53 patients (55.2 %), major mismatched in 19 (19.8 %) and minor mismatched in 24 (25.0 %). Luspatercept was first administered a median of 21 days post-transplant (range 9–30). The median dose was 0.83 mg/kg (0.66–1.25) and most patients (n=47, 95.9%) received a single injection and only 2 patients (4.1%) were given a second dose. Erythroid response was achieved in 45 of 49 patients (91.8 %) after the first dose, with a median time to response of 11 days (2–26). The haemoglobin level rose rapidly from a median of 55 g/L (35–65) at baseline to 65 g/L (44–94) one week after the first injection (p<0.001). Neutrophil counts remained stable (6.10 vs 6.52×10⁹/L; p=0.085), whereas the platelet counts increased significantly from 44 (7-143) to 71 (7-304)×10⁹/L (p<0.001). Of note, for those patients without response after 1st dose of luspatercept, one developed PRCA and another with late response, all responded to the 2nd dose of treatment. While other 2 patients developed secondary graft failure and died eventually due to infections. By day 56 post-transplant, 44 patients (89.8%) in the luspatercept arm had achieved a haemoglobin ≥80 g/L, while 37 patients (78.7%) in the control group. Since luspatercept was started at a median of 21 days after allo-HSCT, we chose day +21 as the baseline for the control group for comparative analysis. The absolute increments in haemoglobin were consistently significant with luspatercept (Week +1: 10 vs 6 g/L, p=0.043; Week +2: 25 vs 15 g/L, p=0.001). Paired BM aspirates (n=14) showed no significant change in the proportion of basophilic or polychromatic erythroblasts before and after luspatercept treatment. In contrast, the fraction of orthochromatic erythroblasts increased significantly (p=0.008). Peripheral reticulocyte counts also rose sharply from a mean of 11.9 (4.5–102.8) × 10⁹/L at baseline to 148.4 (17.3–537.4)×10⁹/L one week after luspatercept (p<0.001), consistent with accelerated terminal maturation and the marrow shift toward late-stage erythropoiesis. During the observation period, the most common adverse events were similar between the luspatercept and control groups. All events were grade 1–2 and no patient experienced grade 3-4 toxicity. Importantly, the incidence of post-transplant viral viraemia (CMV/EBV) and acute GVHD did not differ between the two groups. Conclusion: In our study, the early administration of luspatercept within day 30 after allo-HSCT achieved a quick erythroid response and high response rate, without increasing the incidence of CMV/EBV re-activation and acute GVHD. Luspatercept is a promising treatment for post-allo-HSCT anemia, and f​urther prospective, randomized, controlled clinical trials are needed in the future.