The development of intrapulmonary shunting caused by luspatercept in low grade myelodysplastic syndrome: a case report

Introduction

Myelodysplastic syndromes (MDS) are a diverse group of bone marrow disorders characterized by clonal disruption of hematopoietic stem cells which leads to ineffective hematopoiesis and myeloid dysplasia. Patients manifest with variable cytopenias and, rarely, transform into acute leukemia. Symptomatic anemia and progression to transfusion dependency is common and can increase mortality in elderly patients with co-morbidities such as kidney and heart disease (1). When erythropoiesis is severely dysfunctional, the need for repeat transfusions increases a patient’s risk for iron overload which can lead to numerous organ toxicities affecting the brain, heart, or liver. Most patients presenting with MDS are low risk as defined by the Revised International Prognostic Scoring System (IPPS-R) and anemia is the most common manifestation at this stage (2). Low risk patients with erythropoietin (EPO) levels <500 U/L may be treated with erythropoietin stimulating agents (ESA) as first-line therapy (3).

Not all patients qualify for ESA therapy and many patients do not respond well to ESA therapy alone. The rate of relapse into transfusion dependency in studies are cited as high as 40% after two years of treatment (4). New research has led to the approval of drugs that target the transforming growth factor-β (TGF-β) pathway and can be used to treat anemia in low-risk MDS patients. Luspatercept, first approved in 2020, is a recombinant fusion protein that binds certain members of the TGF-β family of ligands, which in turn reduces down-stream SMAD signaling. This reduction in SMAD signaling blocks the inhibitory signals which prevent red cells from fully maturing—thus leading to an overall increase in successful erythropoiesis (5). Common side effects of luspatercept include fatigue, headache, arthralgias and upset stomach as well as hypertension and an increased incidence of venous thromboembolic events. Up until now, there have been no reports of a patient developing pulmonary arteriovenous malformations (PAVM) or an intra-pulmonary shunt on luspatercept. We present the first documented occurrence of a de novo intrapulmonary shunt secondary to the development of a PAVM in a low-risk MDS patient being treated with luspatercept. In depth cellular biology of the TGF-β and SMAD pathways are discussed connecting the likely pathophysiology behind this case with a literature review focusing on this class of medication that target TGF-β. We present this case in accordance with the CARE reporting checklist (available at https://acr.amegroups.com/article/view/10.21037/acr-2025-135/rc).

Case presentation

Investigations

A 77-year-old female with a history of MDS being treated with luspatercept was referred to the advanced lung disease clinic at Inova Fairfax Hospital for evaluation of new onset hypoxemia. No personal or family history of chronic obstructive pulmonary disease (COPD), heart failure or genetic diseases. The patient was a lifelong non-smoker, had no occupational exposures, and no known pulmonary risk factors. She was diagnosed approximately 15 years ago with a low-grade MDS with refractory anemia and >15% ringed sideroblasts (MDS-RARS) and SF3B1 mutation. Her only cytopenia was anemia and her hematology team discovered early on in her disease course that she suffered from secondary hemochromatosis from chronic iron supplementation in the setting of her mild anemia. She was initially started on darbepoetin alfa and lenalidomide was added that same year which increased her baseline hemoglobin (hgb) to around 10 g/dL from 8 g/dL. She continued this therapy until her hgb started to decrease slowly. At that time, the decision was made to switch her to luspatercept (1 mg/kg subcutaneously every three weeks) after declining response. Within three months of starting luspatercept (cycle 4), she developed exertional dyspnea with resting oxygen saturation of 90–92% on room air, declining further to 80–85% despite increasing home oxygen. She reported no chest pain, cough, hemoptysis, or systemic symptoms. hgb stabilized around 10 g/dL while on luspatercept.

She was then referred to the advanced lung disease clinic for shortness of breath and hypoxemia after she had undergone over three months of luspatercept injections. She was noted by her hematology team that her resting oxygen saturation levels were in the low 90 s on numerous occasions while she was at her scheduled injection appointments. On presentation she was visibly weak and out of breath with a normal cardiovascular exam. Lung exam revealed no wheezing, or crackles. A bubble echocardiogram was done showing a normal right ventricle (RV) and left ventricle (LV) size and function, but agitated saline was seen emanating from the pulmonary veins suggestive of an intrapulmonary shunt (Figure 1).

Figure 1 Echocardiogram with bubble study from 6/2023. (A) The study before the bubbles were added. (B) Bubbles in the right atrium. (C) Bubbles opacifying the entire left atrium and left ventricle. LA, left atrium; LV, left ventricle; RA, right atrium; RV, right ventricle.

Diagnosis

At the initial encounter, the patient was worked up for connective tissue disease with suspect interstitial lung disease as an alternate diagnosis outside of this intrapulmonary shunt. Pulmonary embolism was also excluded given the risk associated with luspatercept use. Notably, she had an echocardiogram without a bubble study done in 2020 well before her Luspatercept was started that showed preserved cardiac function and no notable structural defects. Spirometry and lung volume measurements were unremarkable, however, her diffusion capacity of the lungs for carbon dioxide (DLCO) was low at 50%. A computed tomography (CT) angiogram of the chest excluded pulmonary embolism, and a subsequent pulmonary angiogram did not reveal any macroscopic PAVM. Echocardiography demonstrated normal left and right ventricular function. A contrast (bubble) study revealed early opacification of the left atrium consistent with a large intrapulmonary shunt (grade 3). Initial autoimmune disease labs were negative. The patient was on significant amounts of home oxygen with frequent desaturations between 82–88% on 6–8 L. Physical exam revealed tachypnea without wheezes, crackles, or murmurs. The only new variable that was introduced at the time of her hypoxemia starting was the luspatercept. Given the known association of this drug class with vascular malformations and the involvement of TGF-β/SMAD pathway in the pathophysiology of PAVM formation, the decision was made to stop her luspatercept in December of 2023.

Treatment and outcomes

The only medical intervention for her hypoxemia that was performed was the cessation of the luspatercept injections. No steroids or antibiotics were given. Within a month, her oxygen requirements disappeared. The hematology team has since switched her to Azacitidine and the FDA was contacted to report the adverse event. All procedures performed in this study were in accordance with the ethical standards of the institutional and/or national research committee(s) and with the Declaration of Helsinki and its subsequent amendments. Written informed consent for publication of this case report and accompanying images was not obtained from the patient or the relatives after all possible attempts were made.

Cellular mechanism of action

Luspatercept exerts its effects on the TGF-β and SMAD pathway. The TGF-β signaling pathway plays many roles in cellular growth and differentiation and has a large influence on immune function and overall homeostasis. This key pathway is responsible for physiological functions in embryonic development and cellular maturation with the immune and vascular systems (6). The TGF-β superfamily of ligands was named for the numerous ligands that can influence this pathway by binding to a TGF-β receptor. Ligand activation of the transmembrane TGF-β receptors triggers a phosphorylation cascade leading to downstream activation of the SMAD pathway (7). These phosphorylated SMAD proteins then translocate inside the nucleus to act as a transcription factor which regulates the genes leading to the numerous cellular and physiological functions.

In MDS, these pathways cause dysregulation of hematopoiesis. The TGF-β ligands activin, bone morphogenetic proteins (BMP), and growth differentiation factor 11 (GDF11) activate SMAD 2/3 and mediate inhibitory mechanisms of erythroid maturation. This SMAD2/3 pathway leads to impaired erythroid differentiation and is constantly active and overexpressed in MDS CD34⁺ cells (8). Under normal cellular conditions, SMAD7 acts as an important negative feedback inhibitor of the SMAD2/3 pathway and thus tightly regulating red cell maturation. However, patients with MDS have a decrease in SMAD7 expression that leads to this unregulated activation of the TGF-β ligand and SMAD2/3 inhibitory effect on erythropoiesis. Furthermore, the TGF-β ligand GDF11, a negative regulator of late-stage erythropoiesis, is also increased in MDS which further contributes to the decrease in mature red cell production (9). Figure 2 depicts the effects MDS alone and MDS + luspatercept have on the TGF-β-SMAD pathway. Structurally, luspatercept is a fusion protein consisting of human immunoglobulin G1 and modified human activin IIB receptor. It binds to the TGF-β superfamily of ligands and prevents the activation of the inhibitory TGF-β-SMAD2/3 pathway thus promoting late-stage erythroid maturation and development (10). The expression of ligands GD11 and activin are also reduced with treatment.

Figure 2 A visual representation of the effects MDS and MDS + luspatercept have on the TBG-β-SMAD pathway as it related to erythroid maturation. Created in: https://BioRender.com. MDS, myelodysplastic syndrome; TGF-β, transforming growth factor-β.

Discussion

Luspatercept first became widely recognized after an incidental increase in hgb was noticed in post-menopausal women taking sotatercept, another TGF-β ligand trap of the same drug family with the same mechanism, was approved for treatment of pulmonary arterial hypertension (11). In 2017, PACE-MDS was the first phase II multicenter study testing luspatercept dosing in lower risk MDS patients with or without transfusion dependency (12). The COMMANDS trial, now published, demonstrated luspatercept’s efficacy as first-line therapy in transfusion-dependent lower-risk MDS, irrespective of ring sideroblast status, cementing its role in treatment algorithms. However, vascular side effects are underrecognized (13). Sotatercept, a drug in the same TGF-β/SMAD ligand family as luspatercept, has been shown to promote the formation of telangiectasias but locations were not specified and there have been no reports yet of a direct connection between sotatercept and PAVMs. The TGF-β-SMAD pathway has wide ranging influence over many cellular and physiological processes and the most prominent one when discussing this case is the growth of blood vessels. TGF-β signaling has been well connected to vascular disorders and diseases. It was proven in laboratory studies on mice that inhibition or loss of TGF-β signaling can cause dysfunction capillary formation and creation of poorly structured telangiectasias and arteriovenous malformations (14). For example, in a genetically engineered Smad4-endothelial knockout model, postnatal loss of SMAD4 caused neonatal retinal AVMs, increased endothelial cell proliferation and size, abnormal mural cell coverage, and disrupted artery-vein marker expression (15). Other Animal models with Acvrl1 (ALK1) or Eng (Endoglin) mutations—both key components in TGF-β signaling—develop vascular fragility, hemorrhage, and age-dependent AVMs, characterized by capillary telangiectasias and impaired pericyte coverage (16). Clinicians should avoid conflating telangiectasias with AVMs, as the entities have distinct presentations and risk implications, however, the mechanism involved in both their formation can be linked back to the TGF-β/SMAD pathway causing vascular growth dysfunction. Telangiectasias are born from localized endothelial hyperproliferation and fragile capillary networks, often in post-capillary venules, that can form due to reduced ALK1/SMAD1/5/8 signaling or excessive ALK5/SMAD2/3 signaling (17). This can cause destabilization of small vessels, making them prone to dilation and bleeding without forming high-flow shunts. On the other hand, ALK1/ACVRL1 or endoglin (ENG) loss leading to impaired SMAD1/5/8 signaling can lead to uncontrolled angiogenesis and vessel enlargement, promoting high-flow shunts such as PAVMs (18). Both of these vascular disorders arise from TGF-β/SMAD pathway dysregulation, but the severity and flow characteristics of the vascular lesions differ and only telangiectasias, so far, have been directly linked to this family of drugs. However, a recent 2025 AJRCCM case series reported new intrapulmonary shunts in PAH patients treated with sotatercept, raising early concerns that this class of medication’s vascular remodeling effects may rarely manifest as PAVM-like lesions like we describe here in this case (19). In the phase 3 STELLAR trial testing the efficacy and safety of sotatercept and standard therapy for patients with symptomatic pulmonary arterial hypertension, over 10% (17/138) of patients developed telangiectasias vs. 3% (5/140) in the placebo group (20). Telangiectasias were also noted in over 22% (93/409) of patients according to preliminary data from the long term, open label SOTERIA study which had a median exposure of 462 days (21). Because of the risk of developing telangiectasias, this drug is contraindicated in patients with a previous history of telangiectasis. This risk profile and side effect was not found in patients during the luspatercept clinical trials.

Mutations in the TGF-β and SMAD network have been identified as the culprit in many diseases related to vascular dysfunction including hereditary hemorrhagic telangiectasia (HHT), primary pulmonary hypertension, and Marfan’s syndrome. While the focus of this signaling pathway in luspatercept and MDS is on hematopoiesis and erythroid maturation, the widespread influence the TGF-β-SMAD pathway has on vascular formation cannot be ignored when discussing this case. Patients with HHT have mutations along the TGF-β-SMAD pathway that can lead to dysfunctional blood vessel creation and roughly half of HHT patients can develop PAVM (22). HHT patients with PAVM are at risk for spontaneous hemorrhage due to the fragility of the vessels involved and strokes from an embolus forming in the vasculature. Figure 3 depicts the pathophysiology of the creation of PAVM from disruption of the TGF-β-SMAD pathway. Echocardiography is the method of choice for confirming PAVM and shunt physiology. A delay of three or more cardiac cycles and the appearance of bubbles in the pulmonary vein signal a high probability of extra-cardia shunting: identical to what our patient’s echocardiography showed.

Figure 3 Disruption of the TGF-β-SMAD pathway leading to formation of telangiectasias in the pulmonary vasculature creating PAVM. Created in: https://BioRender.com. PAVM, pulmonary arteriovenous malformations; TGF-β, transforming growth factor-β.

Sporadic cases of PAVM without underlying HHT are extremely rare and over 80% of patients diagnosed with PAVM end up also being diagnosed with HHT (23). Luckily, our patient never suffered from a paradoxical embolism or hemorrhage while she was hypoxemic. Cirrhosis, trauma, and rare infections have been deemed risk factors or causes of PAVM in case reports but never has there been a report focusing on a luspatercept induced PAVM.

To date, this is the only case of a patient with MDS treated with luspatercept developing hypoxemia from a PAVM arising as an off-treatment effect. An increased risk of thromboembolic events has been associated with luspatercept but uncommon. In the phase 3 trial of luspatercept in patients with transfusion dependent β-thalassemia, eight patients (3.6%) had a confirmed thromboembolic event vs. one patient in the placebo group (24). There have been no prior documented cases of patients who take luspatercept that subsequently develop hypoxemia rapidly after treatment in the setting of a new shunt. Based on observed increases in the rate of telangiectasias among patients on sotatercept, a related compound, and the temporal relationship, it is probable that luspatercept was the cause of this patient’s PAVM with resulting shunt and hypoxia. Given the immediate resolution of her symptoms following discontinuation of luspatercept, the decision was made to notify the FDA to inform them of this potential adverse reaction.

Conclusions

This case rests on the pathophysiology of TGF-β-SMAD signaling disruption and the timeline of events with a confirmed intrapulmonary right to left shunt. For the year while this patient was taking luspatercept, we hypothesize that a PAVM was created which was driving this patient’s hypoxemia. The swift resolution of her symptoms and her shunt after discontinuing the medication strongly support this. With the medication no longer disrupting the TGF-β/SMAD signaling pathway, endothelial disruption and angiogenesis halted causing regression of PAVM and, therefore, her symptoms. Given the connection this family of medications has with the TGF-β/SMAD pathway and abnormal vascular formation, it is reasonable to discuss hypoxemia and PAVM formation as rare, but possible side effects of luspatercept.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the CARE reporting checklist. Available at https://acr.amegroups.com/article/view/10.21037/acr-2025-135/rc

Peer Review File: Available at https://acr.amegroups.com/article/view/10.21037/acr-2025-135/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://acr.amegroups.com/article/view/10.21037/acr-2025-135/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All procedures performed in this study were in accordance with the ethical standards of the institutional and/or national research committee(s) and with the Declaration of Helsinki and its subsequent amendments. Written informed consent for publication of this case report and accompanying images was not obtained from the patient or the relatives after all possible attempts were made.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Stauder R, Valent P, Theurl I. Anemia at older age: etiologies, clinical implications, and management. Blood 2018;131:505-14. [Crossref] [PubMed]
2. Sangerman MA, Pomares H, Alonso E, et al. Impact of red blood cell transfusion burden status in patients with lower-risk myelodysplastic syndromes. Blood 2019;134:3013.
3. Fenaux P, Santini V, Spiriti MAA, et al. A phase 3 randomized, placebo-controlled study assessing the efficacy and safety of epoetin-α in anemic patients with low-risk MDS. Leukemia 2018;32:2648-58. [Crossref] [PubMed]
4. Park S, Grabar S, Kelaidi C, et al. Predictive factors of response and survival in myelodysplastic syndrome treated with erythropoietin and G-CSF: the GFM experience. Blood 2008;111:574-82. [Crossref] [PubMed]
5. Vinchi F, Platzbecker U. Luspatercept: A peaceful revolution in the standard of care for myelodysplastic neoplasms. Hemasphere 2024;8:e41. [Crossref] [PubMed]
6. Deng Z, Fan T, Xiao C, et al. TGF-β signaling in health, disease, and therapeutics. Signal Transduct Target Ther 2024;9:61. [Crossref] [PubMed]
7. Miyazawa K, Miyazono K. Regulation of TGF-β Family Signaling by Inhibitory Smads. Cold Spring Harb Perspect Biol 2017;9:a022095. [Crossref] [PubMed]
8. Suragani RN, Cadena SM, Cawley SM, et al. Transforming growth factor-β superfamily ligand trap ACE-536 corrects anemia by promoting late-stage erythropoiesis. Nat Med 2014;20:408-14. [Crossref] [PubMed]
9. Verma A, Suragani RN, Aluri S, et al. Biological basis for efficacy of activin receptor ligand traps in myelodysplastic syndromes. J Clin Invest 2020;130:582-9. [Crossref] [PubMed]
10. Fenaux P, Kiladjian JJ, Platzbecker U. Luspatercept for the treatment of anemia in myelodysplastic syndromes and primary myelofibrosis. Blood 2019;133:790-4. [Crossref] [PubMed]
11. Sherman ML, Borgstein NG, Mook L, et al. Multiple-dose, safety, pharmacokinetic, and pharmacodynamic study of sotatercept (ActRIIA-IgG1), a novel erythropoietic agent, in healthy postmenopausal women. J Clin Pharmacol 2013;53:1121-30. [Crossref] [PubMed]
12. Platzbecker U, Germing U, Götze KS, et al. Luspatercept for the treatment of anaemia in patients with lower-risk myelodysplastic syndromes (PACE-MDS): a multicentre, open-label phase 2 dose-finding study with long-term extension study. Lancet Oncol 2017;18:1338-47. [Crossref] [PubMed]
13. Della Porta MG, Garcia-Manero G, Santini V, et al. Luspatercept versus epoetin alfa in erythropoiesis-stimulating agent-naive, transfusion-dependent, lower-risk myelodysplastic syndromes (COMMANDS): primary analysis of a phase 3, open-label, randomised, controlled trial. Lancet Haematol 2024;11:e646-58. [Crossref] [PubMed]
14. Goumans MJ, Liu Z, ten Dijke P. TGF-beta signaling in vascular biology and dysfunction. Cell Res 2009;19:116-27. [Crossref] [PubMed]
15. Crist AM, Lee AR, Patel NR, et al. Vascular deficiency of Smad4 causes arteriovenous malformations: a mouse model of Hereditary Hemorrhagic Telangiectasia. Angiogenesis 2018;21:363-80. [Crossref] [PubMed]
16. Atri D, Larrivée B, Eichmann A, et al. Endothelial signaling and the molecular basis of arteriovenous malformation. Cell Mol Life Sci 2013;71:867-83. [Crossref] [PubMed]
17. Shovlin CL. Pulmonary arteriovenous malformations. Am J Respir Crit Care Med 2014;190:1217-28. [Crossref] [PubMed]
18. Mendez PL, Obendorf L, Jatzlau J, et al. Atheroprone fluid shear stress-regulated ALK1-Endoglin-SMAD signaling originates from early endosomes. BMC Biol 2022;20:210. [Crossref] [PubMed]
19. Mitchell S, Gomez-Rojas O, Mathavan A, et al. Development of New Intrapulmonary Shunts in Pulmonary Arterial Hypertension Patients Treated with Sotatercept. Am J Respir Crit Care Med 2025;211:1300-2. [Crossref] [PubMed]
20. Hoeper MM, Badesch DB, Ghofrani HA, et al. Phase 3 Trial of Sotatercept for Treatment of Pulmonary Arterial Hypertension. N Engl J Med 2023;388:1478-90. [Crossref] [PubMed]
21. Preston IR, Badesch D, Ghofrani HA, et al. Late Breaking Abstract-A long-term follow-up (LTFU) study of sotatercept for pulmonary arterial hypertension (PAH). European Respiratory Society 2023;62:OA739.
22. Meier NM, Foster ML, Battaile JT. Hereditary hemorrhagic telangiectasia and pulmonary arteriovenous malformations: clinical aspects. Cardiovasc Diagn Ther 2018;8:316-24. [Crossref] [PubMed]
23. Kaufman CS, McDonald J, Balch H, et al. Pulmonary Arteriovenous Malformations: What the Interventional Radiologist Should Know. Semin Intervent Radiol 2022;39:261-70. [Crossref] [PubMed]
24. Cappellini MD, Viprakasit V, Taher AT, et al. A Phase 3 Trial of Luspatercept in Patients with Transfusion-Dependent β-Thalassemia. N Engl J Med 2020;382:1219-31. [Crossref] [PubMed]