Elritercept, a modified activin receptor IIA ligand trap, increased erythropoiesis and thrombopoiesis in a phase 1 trial

Hematopoiesis is a dynamic process that requires a tightly regulated, coordinated balance of proliferation and quiescence of progenitors (precursor cells) at different stages of differentiation to replenish pools and ensure rapid responses to blood cell depletion or cytopenias. Ineffective hematopoiesis occurs when bone marrow fails to produce adequate amounts of mature functional blood cells and often results in multiple cytopenias that include anemia, thrombocytopenia, and circulation of multiple dysplastic cell lineages in a single patient. Historically, administration of exogenous erythropoietin (EPO) has served as a primary therapeutic strategy for treating anemia, and glycosylated forms of thrombopoietin have been used to treat thrombocytopenia.

The transforming growth factor β (TGF-β) superfamily, which comprises >30 ligands and multiple receptor types to transduce their signals, plays a crucial role in regulating biological processes of virtually every tissue and system in the body, including hemostasis and hematopoiesis.¹ These factors likely work in concert with one another to balance hematopoietic differentiation, maturation, and quiescence of progenitors (precursor cells), so that pools of mature cells are maintained at appropriate levels in circulation. Although specific effects of different TGF-β ligands and their interactions continue to be elucidated, signaling through type II activin receptors is known to have an inhibitory effect on erythropoiesis. Clinical studies determined that 2 distinct type II activin receptor ligand traps, sotatercept (activin type IIA receptor [ActRIIA] extracellular domain fused to the Fc region of immunoglobulin G1 [IgG1]) and luspatercept (a modified ActRIIB extracellular domain fused to the Fc region of IgG1; designed to avoid inhibition of activin A), both increased red blood cell (RBC) production in healthy participants and ameliorated anemia in patient populations with various etiologies.^2-6 Although treatment with both luspatercept and sotatercept increased erythropoiesis, no changes in thrombopoiesis were reported.

Elritercept is a recombinant fusion protein comprising a modified activin receptor type IIA extracellular domain fused to a human IgG1 Fc. Elritercept acts as a soluble ligand trap to inhibit activin A and other select TGF-β superfamily ligands including activin B, growth differentiation factor 8 (GDF-8), and GDF-11. In preclinical studies, a research form of elritercept (RKER-050) increased both RBC and platelet production in healthy and diseased rodent models in vivo^7-9 and in vitro studies demonstrated that RKER-050 promoted differentiation across both early and terminal stages of erythropoiesis and thrombopoiesis¹⁰, which supports a differentiated pharmacological profile.

This article describes a phase 1 first-in-human clinical trial of elritercept (KER-050-01). The study was designed to evaluate the safety, tolerability, and pharmacokinetic (PK) profile of elritercept. Additionally, the study evaluated pharmacodynamic (PD) markers of hematopoiesis, including RBC parameters and platelets, to explore the potential of elritercept to treat patients afflicted with disorders such as myelodysplastic syndromes (MDS) and myelofibrosis (MF) that are characterized by ineffective hematopoiesis and multiple cytopenias.

This was a phase 1, randomized, double-blind, placebo-controlled, 2-part dose-escalation study conducted at 2 centers in Australia. The primary objective of the study was to evaluate the safety and tolerability of subcutaneous (SC) administration of single and multiple doses of elritercept in healthy postmenopausal women. Secondary objectives included the assessment of PK and PD effects on markers of hematopoiesis and activin inhibition.

In part 1 (single ascending dose [SAD]), participants were randomly assigned 4:1 to receive either a single dose of elritercept (0.05, 0.5, 1.5, or 4.5 mg/kg) or placebo (saline). After a single dose in part 1, participants were evaluated for safety and PK/PD effects through day 84. In part 2 (multiple ascending doses [MAD]), participants were randomly assigned 4:1 to receive 2 SC injections every 4 weeks of either 0.75 mg/kg elritercept or placebo (saline), and participants were subsequently evaluated for safety and PK/PD effects through day 112.

Safety was evaluated by a safety review committee (SRC), which comprised the medical monitor, the principal investigator, and other members of the investigational team as deemed appropriate. In SAD, escalation to the next dose level proceeded only after the SRC reviewed cumulative, blinded safety and PK data when at least 8 participants completed day 15 in the previous dose level and assessed the protocol-defined dose-limiting toxicities (DLTs; provided in Supplemental Materials). The initiation of MAD as well as the selected dose level and frequency (0.75 mg/kg every 4 weeks) were based on the SRC recommendations after the review of safety and PK data from applicable cohorts in SAD.

The study was conducted in accordance with the recommendations of the Declaration of Helsinki, Good Clinical Practice regulations, and applicable regulatory requirements. All participants provided written informed consent before study participation. The study was entered in the Australian New Zealand Clinical Trial Registry (ACTRN12619000318189). The study protocol, information, and consent form were approved by The Alfred Ethics Committee in Melbourne, Australia.

Elritercept drug substance was manufactured by Speed-to-Clinic (STC, Newton, MA). Elritercept drug product was manufactured by Berkshire Sterile Manufacturing (Lee, MA). Elritercept drug product was provided at a nominal concentration of 55 mg/mL for SC injection. The placebo used in this study was sterile normal saline (0.9% sodium chloride) for SC injection.

This study enrolled healthy postmenopausal women, aged 45 to 75 years, with a body mass index of 18.5 to 32 kg/m², who understood and signed written informed consent. Participants were eligible if they had 12 months of spontaneous amenorrhea or 6 months of spontaneous amenorrhea with serum follicle-stimulating hormone levels >40 IU/L or were at least 6 weeks after surgical bilateral oophorectomy with or without hysterectomy. Participants were excluded if they had taken any drugs that may affect bone turnover, including estrogen, androgen, parathyroid hormone (PTH), bisphosphonates, calcitonin, anabolic steroids, and selective estrogen receptor modulators within 3 months of study entry or any other investigational drugs within 1 month before dosing. Key inclusion/exclusion criteria are provided in Supplemental Materials that accompany this manuscript.

Safety parameters were assessed at baseline and regularly throughout the study. They included adverse events (AEs), physical examinations, vital signs, weight, electrocardiogram (participants in part 1 also underwent continuous cardiac telemetry on days 1-2), hematology, chemistry, urinalysis, coagulation profile, and anti-drug antibodies (ADAs). Only AEs that were newly occurring or increasing in severity or frequency during or after study drug administration were analyzed. AEs were coded using the Medical Dictionary for Regulatory Activities Version 22.0. The severity of AEs was graded by the investigator according to the National Cancer Institute Common Terminology Criteria for Adverse Events, Version 5.0. Serious AEs (SAEs) were defined in accordance with International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use Good Clinical Practice guidelines. The causality of AEs was provided by the investigator according to the protocol defined 5-tier system (not related, unlikely, possible, probable, and highly probable). Protocol-specified DLTs can be found in the Supplemental Materials that accompany this manuscript.

Serum samples for ADA analysis were collected at baseline and days 1, 15, and 84 for part 1 and days 1, 15, 42, and 112 for part 2; they were assessed using a bridging format electrochemiluminescence Meso Scale Discovery assay that was validated for use in measuring anti-elritercept antibodies in clinical serum samples. The sensitivity of the assay was 5.06 ng/mL, and the inter-batch coefficient of variation was ≤20%. All participants, clinical site staff, and study team members were masked to the participant's treatment assignment of elritercept or placebo.

Serum samples for the determination of elritercept concentrations were collected before doses and up to day 84 in SAD and up to day 112 in MAD. Serum elritercept concentrations were assessed using a validated, solid-phase enzyme-linked immunosorbent assay using a human monoclonal antibody directed against the extracellular domains of ActRIIA/B (anti-ActRIIA/B) to capture elritercept in human serum and a biotinylated human anti-ActRIIA/B monoclonal antibody to detect captured antigen. The assay range was from 1.17 to 13.33 μg/mL. The inter-batch coefficient of variation was ≤25.0%.

The following PK parameters were calculated, using standard noncompartmental or compartmental methods as implemented in Phoenix WinNonlin, Version 8.1 (Certara, Inc, Princeton, NJ): area under the concentration–time curve (AUC), AUC from time 0 to last (AUC_0-last), peak concentration (C_max), time to peak concentration (T_max), elimination half-life (t_1/2), apparent clearance (Cl/F), and volume of distribution (V_z/F). In MAD only, PK assessments also included AUC from time 0 to infinity (AUC_0-∞) and AUC from time 0 to tau of the dosing interval (AUC_0-tau).

PD assessments of hematopoiesis included reticulocytes, RBC count, hemoglobin (Hb), hematocrit, and platelets, which were collected before dose and up to day 84 in SAD and up to day 112 in MAD.

The sample size of this phase 1 dose-escalation study was sufficient to evaluate safety, tolerability, and PK based on clinical considerations. Baseline was defined as the last assessment before the first administration of either elritercept or placebo. Demographics and safety evaluations were analyzed by descriptive statistics using data from all participants. Placebo participants from all cohorts in SAD were combined, whereas placebo participants from the single cohort for MAD were presented separately. There was no pooling of SAD and MAD participants.

PK and PD parameters were evaluated by descriptive statistics using data from all participants who received at least 1 dose of the study drug and had at least 1 post-dose measurement. Dose proportionality was assessed using the power model for serum PK parameters C_max and AUC. Accumulation of multiple doses was assessed by the ratios of C_max (day 29) to C_max (day 1) and AUC_tau (day 29) to AUC_tau (day 1). PD exploratory biomarkers were assessed using descriptive statistics, as well as mean percent change from baseline evaluations.

Demographic and baseline characteristics of all participants enrolled in this study are presented in Supplemental Table 1 and Table 1, respectively. All participants were healthy postmenopausal women, meeting the 12 months of spontaneous amenorrhea criteria. In SAD, the mean age was 60.4 years (range, 51-71). Most participants were White (90.0% across SAD and MAD studies). The mean and median body mass index across treatment groups were 24.31 to 25.65 kg/m² and 23.35 to 26.60 kg/m² in SAD and MAD cohorts, respectively. There were no notable differences among groups, except the placebo (SAD and MAD) and 0.05 mg/kg cohorts were younger than the higher dose cohorts. Enrollment in the 4.5 mg/kg cohort was stopped after 6 participants based on a review of the ongoing data by the SRC, which revealed that 3 participants administered the 4.5 mg/kg dose experienced a treatment-emergent AE (TEAE) of hypertension associated with increases in Hb.

Single and multiple doses of elritercept were generally well tolerated at all dose levels tested, ranging from 0.05 to 4.5 mg/kg as a single dose and multiple doses of 0.75 mg/kg (Table 1). There were no SAEs or DLTs, as well as no deaths, in elritercept-treated participants.

In SAD, the most common drug-related TEAEs (≥10% participants) by the Medical Dictionary for Regulatory Activities preferred terms after elritercept administration included increased Hb, hypertension, nausea, and upper respiratory tract infection, compared with placebo. All TEAEs reported after the administration of elritercept were mild or moderate in severity, and there were no treatment-emergent SAEs. After multiple elritercept administrations in MAD, headache and increased Hb had the highest incidence of all TEAEs. No adverse trends were observed with respect to temperature, respiratory, or heart rates. There were no clinically meaningful changes in total white blood cell counts at any dose. Two participants who received a single dose of elritercept had a neutrophil count of <1.5 × 10⁹ cells per liter; the decrease resolved in 1 participant and returned to near normal levels (1.4 × 10⁹ cells per liter) in the other participant by the end of the study. Other than in these 2 participants, no changes in neutrophils occurred, and no changes in lymphocytes were observed. There were also no clinically meaningful post-dose abnormalities observed via clinical chemistry or urinalysis after either single or multiple doses of elritercept.

In the 4.5 mg/kg cohort of SAD, 4 of 6 participants (66.7%) reported a TEAE of hypertension after a single dose (in all cases, hypertension was grade 1, transient, and considered related to the study drug); the event resolved in all 4 participants. Three of these 4 participants had a TEAE of increased Hb, with an increase of at least 3 g/dL from baseline, and the event of hypertension was resolved with the decrease in Hb.

In SAD, grade 1 TEAEs occurred in 1 participant (12.5%) who received placebo and 12 participants (40.0%) who received a single dose of elritercept. Grade 2 TEAEs occurred in 1 participant (12.5%) who received placebo and 7 participants (23.3%) who received elritercept. Grade 2 TEAEs that occurred only in participants receiving a single dose of elritercept included increased Hb (n = 4 [66.7%] at 4.5 mg/kg), nausea (n = 1 [12.5%] at 0.5 mg/kg), and fatigue (n = 1 [12.5%] at 0.5 mg/kg). In MAD, grade 1 TEAEs occurred in 2 participants (100%) who received placebo and 6 participants (75.0%) who received elritercept. Grade 2 TEAEs that occurred only in participants receiving multiple doses of elritercept (n = 2 [25.0%] at 0.75 mg/kg) included headache (n = 2 [25.0%]) and muscle spasms (n = 1 [12.5%]). No participants who received either single or multiple doses of elritercept experienced a grade 3, 4, or 5 (fatal) TEAE.

In SAD, 3 participants tested positive for ADA on a single occasion, with 2 participants having a positive test on day 15 (0.5 and 4.5 mg/kg). The third participant tested positive at the end of the study on day 84 (1.5 mg/kg) and tested negative on follow-up assessment after the end of study. The titers were low, and subsequent time points failed to show a positive test in the ADA assay. Positive treatment-emergent ADA results were not related to hypersensitivity, immunologic TEAEs, or other safety concerns. PK exposures were unlikely to be affected by ADA positivity. No participants in MAD tested positive in the ADA assay.

The PK profiles of elritercept by dose are displayed in Figure 1. Serum concentrations for elritercept increased in a dose-related manner and eliminated monoexponentially at essentially the same rate for the 0.5 to 4.5 mg/kg doses after SC administration (Figure 1 inset). After SC administration, elritercept was absorbed to the central compartment with a mean T_max of 142.4, 108.5, and 143.8 hours for 0.5, 1.5, and 4.5 mg/kg, respectively. Mean C_max ranged from 4.23 to 27.02 μg/mL, and mean AUC_0-∞ ranged from 2172.1 to 17 184.8 hour × μg/mL across the dose ranges of 0.5 to 4.5 mg/kg; and the t_1/2 was ∼12 days.

The C_max and AUC of elritercept were linear after SC administration of single doses ranging from 0.5 to 4.5 mg/kg (Figure 2A-B, respectively). Dose-proportionality analysis indicated a proportional increase in systemic exposure, as determined by the regression of log-transformed parameters with log-transformed dose.

After 2 doses of elritercept at 0.75 mg/kg in MAD, serum PK of elritercept was generally comparable with that observed in SAD. Accumulation was not evident after the second dose based on comparisons of day 29 exposure with day 1 exposure, with a mean day 29/day 1 C_max ratio of 1.080 and a mean day 29 AUC_tau/day 1 AUC_0-last ratio of 1.138.

Administration of single elritercept doses in SAD resulted in generally dose-dependent increases from baseline in reticulocytes, RBCs, Hb, and platelets as early as day 2. Similar results were observed in MAD, with 2 administrations of 0.75 mg/kg elritercept.

The mean changes from baseline in reticulocytes, RBC, Hb, and platelets from SAD over time is shown in Figure 3, respectively. Increases in reticulocytes were observed starting as early as day 2 and peaked on day 15 in all elritercept dose groups. Dose-dependent increases in RBCs and Hb were observed starting as early as day 2 (4.5 mg/kg dose) through day 4. Overall RBCs and Hb peaked between day 15 (0.5 and 1.5 mg/kg doses) and day 29 (4.5 mg/kg dose) in elritercept dose groups. In the highest dose group (4.5 mg/kg), rapid and generally sustained increases in reticulocytes were observed by days 15 to 21, likely providing a continuous supply of immature RBCs for further maturation into RBCs. This resulted in RBC and Hb increases observed through day 29, which was sustained for the duration of the study (day 84). These early increases in erythroid markers are supportive of maturation of late-stage erythrocyte precursors and their release into circulation. After single doses of elritercept, a change in Hb ≥1.5 g/dL at any time point during the study was observed in 1 of 8 participants (12.5%) within the 0.05 mg/kg dose group, 1 of 8 (12.5%) in 0.5 mg/kg, 4 of 8 (50%) in 1.5 mg/kg, and 4 of 6 (66.7%) in 4.5 mg/kg, compared with 0 of 8 within the placebo control group. Consistent with observations in SAD, administration of multiple SC doses (0.75 mg/kg/dose) of elritercept in MAD resulted in increases in reticulocytes, RBCs, and Hb. Further supporting these observations in SAD and MAD groups was 1 participant who demonstrated a rapid rise in reticulocytes, as well as a biphasic kinetic response in RBC (characterized by an initial increase in the first 7 days followed by a secondary increase starting at day 21), after a single SC administration of 4.5 mg/kg (Figure 4). These responses suggest that the administration of elritercept increased pools of early erythroid precursors that could be mobilized for continued differentiation to mature RBCs.

The time course of changes in platelets after single doses of elritercept is shown in Figure 5, with dose-dependent increases generally peaking between days 7 and 15. The majority of participants (≥ 75%) in the 2 highest dose groups (1.5 mg/kg [6/8 (75%)] and 4.5 mg/kg [6/6 (100%)]) demonstrated a maximum overall change of ≥30 × 10⁹ cells per liter, compared with those who were treated with placebo (4/8 [50%]). Similar changes of ≥30 × 10⁹ cells per liter were observed after 2 doses of elritercept in the MAD population (0.75 mg/kg/dose [5/8 (62.5%)]), compared with placebo control (1/2 [50%]).

Elritercept is a recombinant modified ActRIIA fusion protein that inhibits signaling by select TGF-β superfamily ligands, including activin A, activin B, GDF-8, and GDF-11. Findings from this phase 1 study in healthy postmenopausal women demonstrate that elritercept had a favorable safety and tolerability profile at all single and multiple doses tested, a PK profile conducive to monthly dosing, and elicited rapid, sustained, and dose-dependent increases in RBCs, Hb, and platelets.

AEs in elritercept-treated participants were mild or moderate in severity, and no discontinuations or DLTs occurred. Grade 1 hypertension was reported in 4 of 6 participants (66.7%) who received the highest dose tested of 4.5 mg/kg, of whom 3 had an increase in Hb >3.0 g/dL over baseline. Hypertensive events were mild in severity, reversible, and resolved as Hb levels returned to baseline. The observed hypertension was likely secondary to increases in red cell mass from a starting point of normal Hb, which would not be expected in patients who are afflicted with disorders such as MDS and MF that are characterized by lower red cell mass concomitant with anemia. ADA-positive serum samples were generally infrequent, low in titer, transient in nature, and no discernable relationship to PK or PD responses was apparent.

Administration of elritercept resulted in dose-proportional increases in C_max and AUC. The t_1/2 of 12 days and the lack of accumulation seen after the second every 4 weeks dose in MAD support monthly dosing. Administration of elritercept also led to increases in RBCs, Hb, and platelets that were rapid, dose dependent, and sustained from a single dose (1.5 and 4.5 mg/kg doses). A dose-proportional increase in the proportion of participants who had increases in Hb greater than 1.5 g/dL was observed.

The kinetics of elritercept on PD markers of hematopoiesis were consistent with broad regulation of both erythropoiesis and thrombopoiesis. Erythropoiesis is a multiple-stage process that occurs over ∼3 weeks, with proliferation from the earliest erythroid committed precursor (BFU-E) to differentiation into mature RBCs. Administration of exogenous EPO as a therapeutic agent acts to stimulate early erythroid precursor cells. Consistent with its mechanism, EPO agents result in delayed increases in reticulocytes, RBCs, Hb, and hematocrit in the bloodstream. In contrast, elritercept acted rapidly, with increased reticulocytes, RBCs, and Hb being observed at all doses administered (Figure 3). The time course of reticulocytes observed in this study was characterized by rapid increases (day 2) that remained elevated 29 days after the dose. RBCs and Hb were also increased as early as day 2, continued to rise till day 29, and were sustained through day 84 at the highest doses evaluated (1.5 and 4.5 mg/kg). Interestingly, one 61-year-old postmenopausal female in this study provided further insight into elritercept’s putative mechanism of action (Figure 4). This participant demonstrated a rapid rise in reticulocytes that was consistent with the elritercept-treated SAD and MAD cohorts, as well as a biphasic kinetic response in RBCs, after a single SC administration of 4.5 mg/kg. The time course of reticulocytes and the biphasic response in RBCs in this participant suggest that the administration of elritercept led to increased pools of early erythroid precursors that could be mobilized for continued differentiation to mature RBCs. Taken together with preclinical studies using a research version (RKER-050),^7-11 these results in healthy postmenopausal women collectively support a differentiated mechanism of elritercept to stimulate the terminal maturation of late-stage erythroid precursors as well as the maturation of early-stage erythroid progenitor populations that can be mobilized for a sustained upregulation of erythropoiesis.

In addition to stimulating early and late stages of erythropoiesis, elritercept administration increased the production of platelets (Figure 5). The kinetics of thrombopoiesis upon elritercept administration followed the same pattern as that was observed in erythropoiesis, with increases starting at day 2 and peaking between days 7 and 15, which is supportive of engagement at multiple stages of the pathway. Overall, the lifecycle of a platelet is shorter than that of an RBC, with development from the BFU-MK stage (the earliest committed megakaryocyte precursor) to that of a circulating platelet occurring over ∼4 to 7 days and newly released platelets being removed from circulation after 5 to 7 days.¹² Therefore, the initial rise in platelets observed after elritercept administration as early as day 2, followed by the sustained increase through day 15, is consistent with elritercept affecting both early- and late-stage platelet precursor cell types. The observed increases in platelet counts of ≥30 × 10⁹ cells per liter was consistent with a threshold previously reported to provide clinically meaningful benefits to patients with thrombocytopenia.¹³ Furthermore, elritercept-mediated platelet changes did not exceed a clinically normal range (thrombocythemia was not observed), supporting the potential use as a single therapeutic agent in patients with anemia and/or thrombocytopenia.

TGF-β ligand traps with varying ligand and pharmacologic profiles have been developed and are currently approved to treat disorders with varying etiologies. Luspatercept, which was engineered to avoid binding activin A⁴ while maintaining the potent binding to activin B, GDF-8, and GDF-11, is described as a late-stage erythroid maturation agent.^14,15 Similar to luspatercept, sotatercept and elritercept also bind activin B, GDF-8, and GDF-11; however, they are also potent inhibitors of activin A. Although the ligand-binding properties of luspatercept and sotatercept differ in regard to activin A binding, the resulting pharmacology is similar in that they both act to promote erythropoiesis without increasing platelets or affecting any other hematopoietic lineage.^16-19 In fact, sotatercept was found to reduce platelet counts in patients with pulmonary arterial hypertension.^18,20 Notably, the ligand traps with similar target ligand-binding profiles (sotatercept and elritercept) have overlapping yet distinct pharmacology in that they both promote erythropoiesis, but only elritercept also increases platelet numbers. It is tempting to speculate that sotatercept only acts on the erythropoietic lineage, whereas elritercept acts on both erythropoiesis and thrombopoiesis and may act as early as the common megakaryocyte erythroid progenitor to stimulate the production of RBCs and platelets. Additional studies are needed to elucidate the connection between ligand binding and pharmacology with these ligand traps.

Multilineage dysplasias and multiple cytopenias, including severe anemia and thrombocytopenia, are common in patients with MDS and MF, in which mutations may interrupt differentiation throughout hematopoiesis. The unique mechanism of action of elritercept to promote early- and late-stage differentiation and maturation of progenitors in erythropoiesis and thrombopoiesis supports its potential to treat anemia and thrombocytopenia in patients with MDS and MF without eliciting detrimental changes in white blood cells such as neutrophils and lymphocytes.

In conclusion, elritercept had a favorable safety and tolerability profile in this phase 1 first-in-human study conducted in healthy postmenopausal female participants. The data reported here support a unique differentiated pharmacologic profile, as evidenced by robust and sustained increases in erythropoiesis and thrombopoiesis that are consistent with the stimulation of both early- and late-stage pathways. Therefore, we conclude that elritercept has the potential to ameliorate cytopenias in patients with disorders characterized by ineffective hematopoiesis and multiple cytopenias, such as anemia and/or thrombocytopenia. Consequently, elritercept is being evaluated in 2 ongoing phase 2 trials in MDS (NCT04419649) and MF (NCT05037760). Both trials use 2-part designs in which part 1 serves as a dose escalation from 0.75 mg/kg to 4.5 mg/kg (identical to the dose range described here) to confirm a dose that will be administered chronically in part 2.

The authors acknowledge those who participated in this research study, Jana Baskar (IQVIA, Australia and New Zealand), site personnel, and Christine Graham, Jen Salstrom, and Bill Aschenbach for their critical review of this manuscript.

Contribution: J.S. and J.L. contributed to the conception and design of the study; B.S. served as the principal investigator in the study; H.N. and S.B. analyzed the data; J.L. and J.T. contributed to manuscript writing; and all authors had access to study data, contributed to data interpretation, manuscript writing, editing, and critical review of the manuscript, and approved the final version of the manuscript.

Conflict-of-interest disclosure: J.L., C.R., S.B., J.T., H.N., and J.S. are current employees of or consultants for the study sponsor, Keros Therapeutics. B.S. declares no competing financial interests.

Correspondence: Jasbir Seehra, Keros Therapeutics, Inc, 1050 Waltham St, Suite 302, Lexington, MA 02421; email: jasbir@kerostx.com.