Improvement of Underlying Disease Pathophysiology of Ineffective Erythropoiesis in Non-Transfusion-Dependent (NTD) Patients with β-Thalassemia Receiving Luspatercept: Biomarker Analysis from the BEYOND Trial

Background: Ineffective erythropoiesis and chronic hemolytic anemia are hallmarks of β-thalassemia which ultimately lead to extramedullary hematopoiesis (EMH), iron overload and other complications. Patients with hemoglobin levels chronically < 10 g/dL are at higher risk for clinical complications and mortality. A significant proportion of patients treated with luspatercept on the BEYOND study (NCT03342404) achieved ≥ 1.0 g/dL hemoglobin increases from baseline, with a manageable safety profile. This increase of hemoglobin is associated with improvement in health-related quality of life. (Taher et al, Lancet Haematol. 2022; 9(10):e733-e744).

Aim: To report biomarker-correlated data demonstrating the benefit of luspatercept on underlying ineffective erythropoiesis, iron homeostasis, and hemolysis.

Methods: Serum samples and MRI data were collected from 145 patients (luspatercept [n = 96] and placebo [n = 49]) enrolled in the multi-center, randomized, double-blind, placebo-controlled phase 2 BEYOND trial for this study. Samples were analyzed for levels of hepcidin, erythroferrone (ERFE), erythropoietin (EPO), soluble transferrin receptor (sTFR), fetal hemoglobin (HbF), serum ferritin, growth differentiation factor (GDF) 11, GDF15 and LDH by ELISA at Intertek (San Diego, CA). Liver iron content (LIC) was assessed by R2 MRI. MRI/ultrasound was used to determine spleen volumes.

Results: Luspatercept responders, (patients who achieved hemoglobin increase ≥ 1.0 g/dL from baseline) had significantly lower baseline levels of ERFE (P = 0.0065), EPO (P = 0.00019), sTFR1 (P = 0.00043), GDF15 (P = 4.4e^-5), and HbF (P = 0.00016), and higher hepcidin levels (P = 0.022), compared with non-responders. Baseline bilirubin, serum ferritin, LIC, and spleen volume were not significantly different between responders and non-responders. Levels of GDF11, a member of the TGFβ superfamily of ligands targeted by luspatercept, were significantly reduced at week 48 in the treatment group compared to baseline (P = 0.001), though not different between responders and non-responders; no such reduction was observed in the placebo group. In the luspatercept arm, bilirubin and LDH did not significantly change from baseline to week 48 in either responders (P = 0.289) or non-responders (P = 1.000). However, reticulocytes increased significantly in responders (P = 0.028) without any increase in HbF (P = 0.82), suggesting increased effective erythropoiesis and stabilization of hemolysis. Luspatercept responders had significant reductions in spleen volume at week 48 compared with baseline (P < 0.001), suggesting reduced EMH. Spleen volumes at week 48 of NR in both arms were not significantly different. Placebo patients had a significant increase in serum ferritin levels at week 48 (P = 0.030), likely from ongoing ineffective erythropoiesis and dysregulated iron homeostasis. At week 48, serum ferritin levels were significantly decreased in luspatercept responders (P = 0.003), but not in non-responders (P = 0.480). No significant changes in LIC were observed in either the placebo or luspatercept groups at week 48 compared to baseline.

Conclusions: Analysis of biomarkers suggests that patients with NTD β-thalassemia receiving luspatercept had an improvement in ineffective erythropoiesis, the primary underlying pathophysiologic mechanism. In responders to luspatercept, increase in reticulocytes, without an increase in HbF, suggests more efficient maturation and differentiation in the bone marrow. Furthermore, the reduction in spleen volume suggests that this efficient improvement in marrow erythropoiesis leads to reduction in EMH. At the time of analysis, LIC remained unchanged, which in combination with reduction in serum ferritin, suggests that the improvement in ineffective erythropoiesis may also have improved dysregulated iron homeostasis, with reduction in iron absorption and no further iron loading. Therefore, this analysis shows that luspatercept treatment results in improved ineffective erythropoiesis, reduced EMH and improved iron homeostasis.