Hematological and clinical improvements with elritercept (KER-050, TAK-226) at the recommended Phase 2 dose (RP2D) in patients with myelofibrosis (MF) receiving ruxolitinib: Updated results from the Phase 2 restore trial

Abstract Background Anemia is a hallmark of MF, contributing to fatigue, transfusion dependence, and reduced survival. Janus kinase (JAK) inhibitors like ruxolitinib improve splenomegaly and symptom burden but are associated with anemia and thrombocytopenia, which often limit their use and lead to suboptimal dosing and reduced clinical benefit. Existing anemia treatments offer modest efficacy without survival benefit and at times, limited effect on symptoms, underscoring the need for therapies that restore or enhance hematopoiesis and improve outcomes. Elritercept (KER-050, TAK-226) is an investigational, modified activin receptor type IIA/IgG1 fusion protein designed to selectively bind and block downstream signaling of transforming growth factor-β (TGF-β) superfamily ligands, including activin A, activin B, growth differentiation factor 8 (GDF8), and GDF11. By blocking the signaling of these ligands, elritercept aims to restore functional hematopoiesis. We report updated findings from the ongoing Phase 2 RESTORE trial (NCT05037760). Methods RESTORE is a Phase 2 open-label study evaluating the safety, tolerability, pharmacokinetics, and efficacy of elritercept as monotherapy (Arm A) or in combination with ruxolitinib (Arm B) in pts with MF and anemia (transfusion-dependent [TD; ≥3 RBC units over 12 wks] or non-transfusion-dependent [NTD; <3 RBC units over 12 wks with hemoglobin [Hgb] <10g/dL]). Here we report data for pts in Arm B, receiving elritercept starting at the RP2D of 3.75 mg/kg in combination with ruxolitinib. Analyses included treatment effects on hematological parameters, spleen size, symptoms (Myeloproliferative Neoplasm Symptom Assessment Form Total Symptom Score [MPN-SAF TSS], Patient-Reported Outcomes Measurement Information System [PROMIS] Fatigue short form 7a T score, and gene mutations. Results As of April 3, 2025, 38 pts receiving ruxolitinib were enrolled in Arm B and treated with elritercept starting at the RP2D of 3.75 mg/kg. At baseline, median age was 71 yrs (range: 45–80), median duration of current ruxolitinib treatment was 44 wks (range: 18–631) and median time from MF diagnosis was 2.3 yrs (range: 0.4–18.0). Hgb, platelet and absolute neutrophil counts were 8.46 g/dL (range: 4.4–10.0), 127 x109/L (range: 45–416) and 3.71 x109/L (range: 1.17–35.7), respectively. The median RBC transfusion requirement was 4 units/12 wks (range: 0–17), with 60.5% of pts classified as TD. Dynamic international prognostic scoring system (DIPSS) risk category scores were 10.5% (4/38) intermediate-1, 68.4% (26/38) intermediate-2, and 21.1% (8/38) high. 55.3% had spleen volume ≥450 cm³, the median MPN-SAF TSS was 16.0 (range: 0–55), and the median PROMIS Fatigue T score was 53.0 (range: 39.6–66.3). The median duration of elritercept add-on treatment was 34 wks (range: 5–74). The most common adverse events (AEs) were diarrhea (18.4%, all grade 1/2; median duration 5.5 days) and thrombocytopenia (13.2%, grade 3 in 10.5%). No pts discontinued elritercept due to an AE. By wk 24, among TD pts with at least 12 wks of Hgb/RBC transfusion recorded, 30.4% (7/23) achieved RBC transfusion independence (TI), 17.4% (4/23) achieved TI with a concurrent mean Hgb increase ≥1.5 g/dL, and 56.5% (13/23) experienced ≥50% reduction in RBC transfusions from baseline over any consecutive 12-wk period. In NTD pts, 38.5% (5/13) achieved a mean Hgb increase ≥1.5 g/dL over any 12-wk period. In the combined group of TD and NTD pts with an evaluable assessment at wk 24, 14.3% (2/14) achieved spleen volume reduction ≥25%, 19% (4/21) achieved ≥50% reduction in MF-SAF TSS, and 36% (9/25) experienced ≥5-points improvement in PROMIS Fatigue T score. Additionally, 36.8% (14/38) of pts achieved mean platelet increase of ≥30 × 10⁹/L over any consecutive 12-wk period within the first 24 wks. JAK2 V617F variant allele frequency (VAF) reduction of ≥50% was observed in 3 of 18 evaluable pts at wk 24, with sustained reductions observed at wk 52. For the remainder of pts with available data, the VAF remained unchanged at wks 24 and 52 with further follow-up ongoing. Conclusions Elritercept as an add on treatment at the RP2D demonstrated clinically meaningful improvements in anemia, thrombocytopenia, spleen volume and symptoms in MF pts receiving ruxolitinib. Emerging data suggest potential effects on clonal architecture and hematopoietic restoration, even in advanced disease. These findings support elritercept as a differentiated therapy in the treatment of pts with MF.