The role of vascular pathology in SMA: implications for pathogenesis and treatment

Introduction

The concept that spinal muscular atrophy (SMA) is a multisystem disorder is now widely accepted (Yeo and Darras, 2020). However, the contribution of non-neuronal pathologies to disease progression, and their long-term clinical significance in the era of SMN-restorative therapies, remains poorly understood. As survival and life expectancy continue to improve, understanding whether these pathologies persist, contribute to ongoing morbidity or require direct therapeutic intervention has become increasingly important.

One system that has emerged as a potentially important contributor to SMA pathogenesis is the vasculature. Endothelial cells, which form the major component of the vasculature, play fundamental roles in tissue homeostasis beyond the delivery of oxygen and nutrients. Endothelial dysfunction can impair tissue perfusion, resulting in hypoxia, but may also compromise specialized vascular barriers, alter inflammatory signaling and disrupt communication between the vasculature and surrounding tissues (Ray et al., 2023; Wang and He, 2024). The consequences of these abnormalities are likely to vary between organs according to their specialized vascular architecture and metabolic demands, with the central nervous system (CNS) being particularly vulnerable because of its high energy requirements and dependence on the integrity of the blood-CNS barriers (Fang et al., 2023).

This review will summarize the current evidence for vascular pathology in SMA and discuss its potential contribution to both neurodegeneration and systemic disease. A thorough understanding of the role of the vasculature in SMA will help identify key knowledge gaps, guide future research, and inform the development of therapeutic strategies that address the multisystemic nature of SMA.

Spinal muscular atrophy

SMA is an inherited neuromuscular disorder affecting approximately 1 in 15,000 individuals (Belter et al., 2024). In more than 98% of cases, SMA is caused by mutations or deletions in the survival motor neuron 1 (SMN1) gene, resulting in insufficient production of the essential and ubiquitously expressed survival motor neuron (SMN) protein. Low levels of SMN are also generated from the paralogous survival motor neuron 2 (SMN2) gene, however, alternative splicing limits the production of full-length, functional SMN protein. These low levels of SMN2-derived protein permit survival while allowing SMA to manifest as a disease. Variation in SMN2 copy number modulates SMN protein levels and is a major determinant of disease severity (reviewed by Butchbach, 2016). In SMA patients, reduced SMN expression leads to the characteristic neuromuscular phenotype of SMA, comprising lower motor neuron degeneration, neuromuscular junction disruption, skeletal muscle denervation and atrophy, culminating in progressive weakness, respiratory insufficiency and, in its most severe forms, death during early childhood (Allardyce et al., 2024; Crawford and Pardo, 1996; Hamilton and Gillingwater, 2013).

The role of the SMN in cells and tissues

Despite its name, SMN protein is not restricted to motor neurons but is expressed in all cell types. SMN was initially characterized as an essential component of the SMN complex, which plays a central role in the biogenesis and assembly of small nuclear ribonucleoproteins (snRNPs) required for pre-mRNA splicing. However, subsequent research has established that SMN has a much broader range of cellular functions, reflecting its involvement in multiple aspects of RNA metabolism and intracellular homeostasis. These include RNA processing and splicing, mRNA transport and local translation, cytoskeletal organization, endocytosis, autophagy, ubiquitination, mitochondrial function and cellular bioenergetics (reviewed by Chaytow et al., 2018; Ottesen et al., 2023). Recent work further highlights the complexity of SMN2 regulation and the multiple molecular mechanisms that influence the production of functional SMN protein, including cis- and trans-acting regulators of exon 7 inclusion and epigenetic control of SMN2 expression (Virata et al., 2025).

The breadth of these functions suggests that SMN acts as an important regulator of cellular homeostasis rather than serving a single, specialized role in motor neurons. Importantly, the cellular response to SMN restoration is also not uniform. Patient-derived fibroblast studies have demonstrated substantial inter-individual variation in the response to SMN2 splice-modifying therapies, with baseline SMN levels, SMN2 copy number and age influencing treatment response (Signoria et al., 2024). Together, these findings highlight the complexity of SMN biology and suggest that the consequences of SMN deficiency, and its restoration, may be highly dependent on cellular and tissue context.

The requirement for SMN also varies between tissues and across the lifespan. SMN expression is highest during embryonic and early postnatal development, with a progressive decline into adulthood in both humans and mice, suggesting important developmental functions (Coovert et al., 1997; Jablonka and Sendtner, 2017; Ramos et al., 2019). While high levels are expected within the CNS, substantial expression has also been reported in peripheral organs including the heart, liver and kidneys, in a tissue- and age-dependent manner (Groen et al., 2018; Yeo and Darras, 2021). This distribution, together with the diverse cellular functions of SMN, provides a biological basis for the widespread tissue pathology observed in SMA and supports the concept that SMN deficiency may directly affect multiple cell types, including endothelial cells.

SMA as a multisystem disorder

Accumulating evidence from patients and preclinical models has demonstrated that SMA extends beyond the neuromuscular system and should now be considered a multisystem disease. In addition to motor neuron loss and skeletal muscle atrophy, pathological changes have been described in the cardiac, vascular, musculoskeletal, immune, gastrointestinal, urinary and reproductive systems (reviewed by Hamilton and Gillingwater, 2013; Yeo and Darras, 2020; Shababi et al., 2014; Yeo and Darras, 2021). Recognizing these non-neuronal manifestations is essential to understand the full spectrum of SMA pathology and to evaluate whether current SMN-restorative therapies adequately address disease beyond the motor unit.

Of these extra-neuronal pathologies, vascular pathology has emerged as an increasingly well-characterized feature of SMA. Vascular abnormalities have been reported across multiple tissue and organ systems, suggesting that impaired vascular development and maintenance may be a fundamental consequence of SMN deficiency.

Importance of the vascular system

The vascular system is composed of an extensive network of endothelial cell-lined blood vessels that ramify through every organ and tissue, delivering oxygen and nutrients while removing metabolic waste (Alberts et al., 2002). Beyond this fundamental role, endothelial cells and the vascular bed are highly specialized, adopting tissue-specific phenotypes that regulate vascular permeability, inflammatory cell trafficking and angiogenesis to meet the functional demands of individual organs (Mou et al., 2024). Each vascular bed therefore displays remarkable structural and functional heterogeneity, ranging from the highly restrictive blood-CNS barriers, through the continuous capillaries found in many peripheral tissues, to the fenestrated capillaries of the kidney and liver (Pluta et al., 2025). As vascular functions differ considerably between organs, the pathological consequences are highly tissue-specific (Ajoolabady et al., 2024). Understanding this vascular heterogeneity is therefore essential when considering how endothelial and vascular pathologies could contribute to the diverse manifestations of SMA.

Vascular abnormalities may contribute to neuromuscular dysfunction in SMA

Within the neuromuscular system, which is the primary site of pathology in SMA, vascular abnormalities have been identified in the spinal cord and skeletal muscle of both preclinical models and patients, detailed in Tables 1, 2, respectively.

Table 1. Vascular changes in SMA preclinical mouse models.

PM, postmortem; BSCB, blood spinal cord barrier; PECAM-1, platelet endothelial cell adhesion molecule-1; SMCs, smooth muscle cells; VWF, Von Willebrand factor; TSA, Trichostatin A; HDAC, histone deacetylase inhibitor; ASO, antisense oligonucleotide; scAAV9, self-complementary adeno-associated virus serotype 9.

Table 2. Clinical phenotypes of vascular deficits in SMA patients.

PM, postmortem; BSCB, blood spinal cord barrier; OCT, optical coherence tomography; CECs, circulating endothelial cells.

Spinal cord

The vascular supply of the CNS is essential for normal neuronal development, function and survival, ensuring adequate delivery of oxygen and nutrients while maintaining the highly specialized blood brain barrier (BBB) and blood spinal cord barrier (BSCB) that protect the neural microenvironment (Herculano-Houzel, 2011). As degeneration of spinal motor neurons underlies the clinical manifestation of SMA, disruption of spinal cord vascular integrity would be expected to exacerbate neuronal dysfunction and accelerate disease progression.

Recently, Allardyce et al. (2026) demonstrated important vascular and BSCB abnormalities in post-mortem tissue from patients with severe early-onset SMA. Endothelial cells, the major component of blood vessels, demonstrated significantly reduced expression of von Willebrand factor (VWF), which indicated endothelial dysfunction. Ultrastructural analyses revealed endothelial cell oedema, extensive vacuolization, and disruption of plasma membrane integrity. Importantly, pathological changes extended beyond the endothelium to other cellular components of the BSCB, with surrounding pericytes displaying vacuolization and closely apposed astrocytic end feet containing abnormal accumulation of filaments. Importantly, these structural abnormalities were accompanied by functional impairment of the BSCB, demonstrated by extravasation of fibrinogen into the spinal cord parenchyma and widespread activation of microglia, indicative of a neuroinflammatory response (Allardyce et al., 2026).

Evidence from preclinical models suggests that these vascular abnormalities arise early in disease pathogenesis. “Taiwanese” SMA mice, which model severe disease, show a decreased vascular density throughout both the gray and white matter of the spinal cord before the onset of neuromuscular symptoms (Somers et al., 2016). Vascular deficits progressively worsened with disease progression and were associated with increased neuronal hypoxia, indicating impaired tissue perfusion. In addition, a subsequent study found that NSC34 neurons with low levels of SMN protein were significantly more likely to die when subjected to a hypoxic environment (Hernandez-Gerez et al., 2020). Furthermore, altered expression of the tight junction protein claudin-5 suggested compromised BSCB integrity (Somers et al., 2016). This presence of vascular abnormalities before symptom onset suggests that vascular dysfunction is likely to contribute to, rather than result from, motor neuron degeneration.

Taken together, the evidence suggests that endothelial dysfunction and BSCB disruption may be integral features of SMA pathology within the spinal cord. By impairing tissue perfusion, promoting neuronal hypoxia and permitting entry of circulating inflammatory mediators into the CNS, these abnormalities are likely to establish a hypoxic and neuroinflammatory microenvironment that accelerates motor neuron dysfunction and degeneration.

Skeletal muscle

At the distal end of the motor unit, the skeletal muscle microvasculature plays a critical role in delivering oxygen and nutrients required to support muscle metabolism, maintenance and repair (Kusters and Barrett, 2015).

In skeletal muscle from both severe (Taiwanese) and milder (SMNΔ7) SMA mouse models, capillary density was significantly reduced and accompanied by decreased vascular branching, reduced network complexity and increased vessel caliber. These abnormalities were observed across both fast- and slow-twitch muscles, irrespective of the degree of neuromuscular junction denervation, suggesting that vascular pathology develops independently of motor neuron loss. Although absent at birth, these vascular defects emerged during early postnatal development, but preceded neuromuscular pathology (Somers et al., 2012, 2016). In addition, supporting components of the vasculature were also affected, with significant reductions in vascular pericyte coverage observed in both SMA mice and patient muscle biopsies (Zhou et al., 2015b). Importantly, these findings were validated in historical skeletal muscle biopsies from patients with Type I and Type II SMA. Here, a significant reduction in the capillary-to-muscle fiber ratio, a widely accepted measure of muscle vascularization, was observed. Furthermore, this declined approximately tenfold between birth and 3 years of age (Somers et al., 2016). This progressive loss of muscle vascularization suggests that vascular pathology is likely to proceed throughout disease progression rather than representing a developmental defect alone.

These findings suggest that impaired angiogenesis and defective vascular maturation are significant features of skeletal muscle pathology in SMA. Reduced capillary density and disruption of the supporting microvasculature are likely to impair tissue perfusion, limiting oxygen and nutrient delivery and compromising the metabolic support required for muscle maintenance and repair (Mukund and Subramaniam, 2019). These vascular abnormalities have the potential therefore, to contribute directly to muscle dysfunction, independent of motor neuron degeneration.

Vascular pathology extends beyond the neuromuscular system

Beyond the neuromuscular system, vascular abnormalities are reported in multiple systemic organs, detailed in Tables 1, 2, respectively.

Heart

Cardiovascular abnormalities represent one of the most common non-neuromuscular comorbidities, with reports of congenital cardiac defects (including atrial and ventricular septal defects) and hypoplastic left heart syndrome being consistently reported in patients with SMA (Møller et al., 1990; Rudnik-Schöneborn et al., 2008; Wijngaarde et al., 2017). This has led to medical insurance claims among SMA patients in the United States most commonly citing cardiovascular abnormalities (Lipnick et al., 2019).

Preclinical studies have identified pronounced vascular abnormalities in the myocardium of the Taiwanese and SMNΔ7 mouse models. Here, capillary density was significantly reduced at early symptomatic stages of disease, demonstrated by reduced expression of the endothelial markers platelet endothelial cell adhesion molecule-1 (PECAM-1) and VWF, and associated oxidative stress (Maxwell et al., 2018). In a separate study, coronary arteries show a loss of smooth muscle cells in the tunica media associated with a pre-symptomatic decrease in arterial wall thickness and remodeling, however endothelial cell integrity appeared unaffected as demonstrated by levels of endothelial nitric oxide synthase (e-NOS) (Shababi et al., 2012). In this study, the systemic restoration of SMN using self-complementary AAV9 gene therapy only partially rescued coronary artery morphology, improving arterial wall thickness, however, myocardial capillary density remained significantly reduced (Shababi et al., 2012). This incomplete rescue of cardiac microvasculature following SMN-restorative therapy raises the possibility that vascular abnormalities may persist despite correction of SMN deficiency by gene therapy. Reduced myocardial capillary density would be expected to impair perfusion and compromise cardiac function (Giacobbe et al., 2026). Consistent with this, echocardiographic assessment of the milder Smn2B/– SMA mouse model demonstrated increased systolic function and global longitudinal strain, indicative of elevated cardiac workload and physiological stress (Nair et al., 2025).

Kidneys

The renal vasculature is characterized by specialized fenestrated capillaries that form the selective filtration barrier required to maintain systemic homeostasis (Mou et al., 2024). In severe Taiwanese SMA mouse models, the decreased expression of PECAM-1 indicated a marked reduction in renal vascular density (Allardyce et al., 2020). This reduction was particularly pronounced within the renal cortex, the primary site of glomerular filtration, and was accompanied by reduced capillary network complexity within nephrons. This would be predicted to impair glomerular filtration and overall renal function. Consistent with these findings, alterations in renal biomarkers have been reported in SMA patients, including changes in serum creatinine, cystatin C, sodium, glucose and calcium concentrations, supporting the presence of clinically relevant renal involvement in SMA (Nery et al., 2019). These findings suggest that vascular dysfunction has the potential to impair renal function through disruption of the specialized microvascular network required for filtration.

Gastrointestinal tract

The gastrointestinal tract is a highly vascularized organ, receiving approximately 20% of resting cardiac output to support digestion and nutrient absorption (Granger et al., 2015). In severe SMA mouse models, capillary density was significantly reduced in the duodenum and ileum and was accompanied by increased enteric neuronal density and macrophage infiltration (Sintusek et al., 2016). Gastrointestinal dysfunction is common in SMA patients and is often attributed to impaired smooth and skeletal muscle function (Corsello et al., 2021). However, impaired intestinal perfusion has the potential to contribute to these symptoms. Once again, systemic restoration of SMN using antisense oligonucleotide (ASO) therapy only partially restored intestinal vascular density in preclinical models (Sintusek et al., 2016). Furthermore, gastrointestinal complications continue to be reported in patients receiving SMN-targeted therapies (Yu and Liu, 2024; Zhang et al., 2024; Zhong et al., 2023).

Spleen

The spleen is a highly vascularized lymphoid organ that plays essential roles in immune surveillance, blood filtration and erythrocyte turnover (Lewis et al., 2019). Unlike other peripheral organs, histological analyses revealed no significant differences in vascular architecture or vessel density of the spleen (Thomson et al., 2017). However, functional vascular deficits have been reported, with SMA mouse models demonstrating reduced splenic perfusion, diffuse hypoxia and decreased peak systolic flow velocity in the splenic artery (Deguise et al., 2020; Hernandez-Gerez et al., 2020). As splenic growth and development are dependent on an adequate blood supply, perfusion deficits could contribute to the reduced spleen size observed in SMA models and the gross pathologies observed in patient spleens (Thomson et al., 2017).

Peripheral vasculature

In the peripheral circulation, distal digital necrosis has been reported in severe patient cases, both in conjunction with, and independently of congenital heart malformations (Araujo et al., 2009; Carrasco et al., 2019; Maeda et al., 2019; Matesanz et al., 2020; Rudnik-Schöneborn et al., 2010; Salmin et al., 2019; Weissman et al., 2021). At the ultrastructural level, skin from these digits showed extensive necrosis of the dermis and epidermis, with thrombotic occlusion of small blood vessels, vascular wall thickening, inflammatory infiltration of the dermal papillae and features of vasculitis, which together are consistent with impaired tissue perfusion and vascular remodeling (Matesanz et al., 2020; Rudnik-Schöneborn et al., 2010; Weissman et al., 2021). In parallel, distal necrosis affecting the ears, tail and hindlimbs were among the first vascular phenotypes described in SMA mouse models. In the Taiwanese and Burgheron mouse models, distal necrosis was observed predominantly in milder phenotypes with prolonged survival, suggesting that peripheral vascular pathology develops as disease progresses (Bogdanik et al., 2015; Hsieh-Li et al., 2000).

Together, these findings demonstrate that vascular pathology in SMA extends well beyond the neuromuscular system and affects multiple peripheral organs and tissues (Figure 1). Although the manifestations differ between tissues, a consistent pattern emerges of impaired vascular development, aberrant microvascular density, and disruption of specialized vascular functions. Such abnormalities have the ability to compromise tissue homeostasis and contribute to organ-specific pathology, supporting the idea that vascular pathology is most likely to be a direct consequence of SMN deficiency rather than a secondary effect of motor neuron degeneration.

SMN depletion directly impairs endothelial cell function

Experimental studies using isolated endothelial cells have demonstrated that SMN depletion directly impairs endothelial function, supporting the hypothesis that vascular pathology arises, at least in part, through cell-autonomous mechanisms. Using both SMN-depleted human umbilical vein endothelial cells (HUVECs) and aortic endothelial cells isolated from severe Taiwanese SMA mice, Zhou et al. (2022) demonstrated impaired endothelial cell migration and network formation, characterized by reduced tube formation and vessel branching. As endothelial migration and tube formation are fundamental processes in angiogenesis (Lamalice et al., 2007), these findings provide a plausible potential mechanistic explanation for the impaired vascular development in SMA. Although the molecular pathways linking SMN deficiency to endothelial dysfunction remain incompletely understood, these studies provide compelling evidence that endothelial dysfunction is a direct consequence of SMN depletion. Further investigation into the mechanisms underlying endothelial dysfunction will be essential to understand how vascular pathology develops and to identify therapeutic strategies to preserve vascular health.

Biomarkers of vascular damage suggest ongoing pathology in SMA patients

While evidence from experimental models and post-mortem studies have established our current understanding of vascular pathology in SMA, it does not allow us to assess the extent or progression of these abnormalities in patients living with SMA. The identification of non-invasive biomarkers capable of detecting ongoing vascular injury is essential to establish the clinical relevance of vascular pathology and to monitor vascular health following therapeutic intervention.

Optical coherence tomography

Optical coherence tomography (OCT) is a non-invasive imaging method that enables visualization of the retinal vasculature (Zeppieri et al., 2023). As an extension of the CNS, the retina provides a valuable surrogate for assessing CNS vascular health (Kashani et al., 2021). In patients with Type II and III SMA, OCT revealed significant reductions in retinal vascular density and network complexity compared with age-matched controls (Zhou et al., 2022). As alterations in the retinal vasculature are widely recognized as biomarkers of cerebrovascular pathology in neurological disease (Mainster, 1990), these findings provide in vivo evidence that vascular abnormalities are present in patients living with SMA.

Studies in the Taiwanese SMA mouse model provide mechanistic insight into these clinical observations. The primary retinal vascular network forms postnatally in mice, providing a unique opportunity to study developmental angiogenesis (Stahl et al., 2010). Zhou et al. (2022) demonstrated delayed retinal vascular development, characterized by reduced radial outgrowth of the vascular network from the optic nerve, decreased vessel complexity and reduced smooth muscle cell coverage, indicative of impaired angiogenesis and defective vessel maturation. Once again, these abnormalities preceded the onset of neuromuscular pathology and suggest that vascular dysfunction is an independent and early pathological event.

Circulating endothelial cells

Under physiological conditions, endothelial cells remain tightly attached to the vessel wall, with few detaching to become circulating endothelial cells (CEC). However, endothelial injury promotes their detachment, with increased CEC numbers representing an established biomarker of vascular damage (Erdbruegger et al., 2006). Patients with SMA (Types I–III) have significantly increased CEC numbers, which are correlated with clinical disease severity. Furthermore, SMA patients also demonstrate a reduced capacity for endothelial repair, indicated by decreased numbers of functional endothelial progenitor cell, assessed by their ability to generate colony-forming units (Zhou et al., 2022). Together, this indicates that vascular homeostasis is tipping toward injury (Deanfield et al., 2007).

These findings demonstrate that vascular abnormalities are not confined to experimental models or end-stage disease but can be detected in patients living with SMA. The continued development and validation of vascular biomarkers will be essential to define and monitor vascular involvement in the post-therapeutic landscape.

Implications: Could microvascular dysfunction exacerbate neurodegeneration in SMA?

Vascular dysfunction has the potential to exacerbate neurodegeneration in SMA through at least two complementary mechanisms: impaired tissue oxygenation and disruption of the BSCB.

Impaired tissue oxygenation

Given the exceptionally high metabolic demands of the CNS, even modest reductions in tissue perfusion have the potential to promote neuronal dysfunction and degeneration. Indeed, the CNS consumes approximately 20% of the body's resting oxygen despite accounting for only a small proportion of total body mass, making it particularly susceptible to hypoxic injury (Eltzschig and Carmeliet, 2011; Hwang and Lee, 2011; Silver and Erecitiska, 1998).

The abnormal vascular architecture, loss of key endothelial proteins including PECAM-1 and VWF, and ultrastructural endothelial damage described in the spinal cord of SMA patients and preclinical models, are ideally positioned to compromise tissue perfusion and oxygen delivery. Consistent with this, an increased number of hypoxic neurons are present in the brain and spinal cord of early symptomatic SMA mouse models, alongside evidence of hypoxia in peripheral tissues including skeletal muscle, kidney, spleen and liver (Hernandez-Gerez et al., 2020; Somers et al., 2016). Evidence consistent with hypoxia have also been reported in patient autopsy tissue, where markers of oxidative stress were identified in the brain and spinal motor neurons of patients with Type I SMA (Hayashi et al., 2002). Although motor neurons are inherently sensitive to oxygen deprivation, SMN-deficient neurons display an even greater susceptibility to hypoxic stress, resulting in metabolic dysfunction, cell-cycle arrest and ultimately cell death (Hernandez-Gerez et al., 2020; Watzlawik et al., 2015). Hypoxia, through vascular dysfunction, may therefore represent an additional pathological insult capable of amplifying neuronal vulnerability in SMA and accelerating neurodegeneration driven directly by SMN depletion in motor neurons.

Disruption of the BSCB

In addition to impaired oxygen delivery, vascular pathology may exacerbate neurodegeneration through disruption of the BSCB. The BSCB is essential for maintaining a specialized neural microenvironment by regulating the entry of circulating molecules, inflammatory mediators and immune cells into the spinal cord (Chopra et al., 2021). Loss of barrier integrity therefore has the potential to expose vulnerable motor neurons to neurotoxic blood-derived components and promote chronic neuroinflammation (Jin et al., 2021; Winkler et al., 2011).

Post-mortem analysis of the spinal cord from patients with severe SMA demonstrated ultrastructural damage to all cellular components (endothelial cells, basement membrane, pericytes and astrocytes) of the BSCB, which was associated with functional impairment of the BSCB demonstrated by extravasation of fibrinogen into the spinal cord parenchyma, accompanied by widespread microglial activation (Allardyce et al., 2026). Clinical pharmacokinetic studies demonstrated rapid appearance of intrathecally administered nusinersen in the systemic circulation following cerebrospinal fluid injection, suggesting increased leakage across the BSCB (Chiriboga et al., 2016). Similarly, elevated cerebrospinal fluid total protein and albumin quotient (QAlb), which are established clinical indicators of barrier dysfunction, have been reported in both adult and pediatric SMA patients receiving nusinersen (Müschen et al., 2021; Wurster et al., 2019).

Collectively, these findings suggest that disruption of the BSCB may represent an additional mechanism through which vascular pathology could further accelerate neurodegeneration in SMA by exposing the spinal cord to circulating blood-derived molecules and promoting neuroinflammation.

Implications: Could vascular pathology contribute to long-term systemic complications in SMA?

Over 10 years ago, a review posed the question: “Does ameliorating disease progression by targeting motor neurons lead to the ‘uncovering' of less severe multisystem phenotypes?” (Hamilton and Gillingwater, 2013). The widespread vascular abnormalities described here provide further support for this concept. Historically, severe SMA was associated with a limited life expectancy, meaning that patients often did not survive long enough for chronic, multisystem pathologies to become clinically apparent. However, the availability of SMN-restorative therapies has dramatically improved survival (Yu et al., 2026), bringing greater attention to the long-term consequences of multisystem pathology.

Evidence from preclinical therapeutic studies suggests that prolonged survival may itself reveal previously subclinical vascular pathologies. Prior to the availability of current SMN-restorative therapies, several experimental approaches that modestly extended lifespan, resulted in distal necrosis that affected the ears, tail and hindlimbs (Foust et al., 2010; Hua et al., 2011; Narver et al., 2008; Passini et al., 2011; Riessland et al., 2010). In contrast, systemic restoration of SMN using ASOs or self-complementary AAV9 gene therapy delayed or prevented the development of digital necrosis, particularly when administered pre-symptomatically (Bogdanik et al., 2015; Hua et al., 2010, 2011; Zhou et al., 2013). These findings suggest that vascular pathology may emerge as survival is prolonged, but also that it could be targeted by systemic upregulation of SMN protein.

The heart, kidneys, liver and gastrointestinal tract are all highly vascularized organs that depend on specialized microvascular networks to maintain tissue homeostasis (Mou et al., 2024). Experimental studies have demonstrated vascular abnormalities affecting each of these organs, including reduced capillary density, impaired vascular remodeling and altered tissue perfusion. Although the long-term clinical consequences of these abnormalities remain unknown, persistent vascular dysfunction could compromise organ function over time, particularly as patients age.

Importantly, several experimental studies reviewed here suggest that aspects of vascular pathology are only partially rescued following SMN-restorative therapy, and that this may be dependent upon the route and mechanisms of delivery. Whether residual vascular abnormalities contribute to chronic cardiovascular, renal or gastrointestinal dysfunction in treated individuals remains to be determined. Longitudinal studies incorporating vascular biomarkers and functional assessment of peripheral organs will therefore be essential to establish the long-term clinical significance of vascular pathology in SMA.

Do current SMN-restorative therapies adequately target vascular pathology?

The introduction of SMN-restorative therapies has transformed the clinical management of SMA, dramatically improving survival, motor function and quality of life (Yu et al., 2026). However, all four currently approved therapies were developed to restore SMN levels in motor neurons, thereby halting neuronal loss. Whether these therapies also restore SMN expression in extra neuronal tissues, including the vasculature, and reverse vascular pathology remains largely unknown.

The first approved therapy, nusinersen (Spinraza; Biogen, Cambridge, MA), is an ASO administered intrathecally to increase production of full-length SMN protein from the SMN2 gene (Neil and Bisaccia, 2019). Due to its route of administration, nusinersen is largely confined to the CNS and should have limited access to peripheral tissues. While there is evidence for leakage of nusinersen through the BSCB into the systemic circulation, it is rapidly metabolized, excreted and therefore not biologically active (Chiriboga et al., 2016). Although this strategy effectively targets spinal motor neurons, it is unlikely to directly address endothelial and vascular pathology in systemic organs. Supporting this concept, preclinical studies using analogous ASOs demonstrated greater improvements in survival when treatment was delivered both systemically and to the CNS than when administered to the CNS alone, suggesting that correction of peripheral pathology contributes to therapeutic benefit and additional systemic delivery may be beneficial (Hua et al., 2011, 2015; Zhou et al., 2015a).

In contrast, the gene replacement therapy onasemnogene abeparvovec (Zolgensma; AveXis/Novartis, Chicago, IL) which delivers a functional SMN1 gene systemically, allowing widespread distribution throughout the body while also crossing the BBB and BSCB via the AAV9 vector (Bitetti et al., 2022). In principle, this broader biodistribution offers the potential to target both central and peripheral pathology (Reilly et al., 2022). However, despite its ability to access the CNS, AAV9 is a relatively poor transducer of endothelial cells, limiting transgene delivery to the vasculature and possibly therapeutic effect (Merkel et al., 2017). Consistent with this, experimental studies demonstrated only partial rescue of vascular pathology following systemic SMN restoration with AAV9 (Shababi et al., 2012).

Among currently available therapies, risdiplam (Evrysdi; Roche, Basel, Switzerland) may offer the greatest potential to address vascular pathology. As an orally administered small-molecule SMN2 splicing modifier, risdiplam readily distributes throughout both the CNS and peripheral tissues and enters cells by passive diffusion (Fowler et al., 2022; Poirier et al., 2018). Clinical studies have demonstrated sustained increases in SMN protein levels in peripheral blood, suggesting broad tissue exposure (Messina and Sframeli, 2020). Nevertheless, whether this widespread biodistribution translates into restoration of endothelial function or reversal of vascular pathology remains unknown, and long-term vascular outcomes have yet to be investigated (Singh et al., 2020).

Overall, SMN-restorative therapies have revolutionized the treatment of SMA, yet their ability to correct vascular pathology remains uncertain. Given the evidence that endothelial dysfunction is widespread, detectable in living patients and may contribute to both neurodegeneration and systemic disease, evaluation of vascular outcomes should become an important consideration in future preclinical and clinical studies. Incorporating vascular biomarkers alongside conventional neurological outcome measures will be essential to determine whether restoration of SMN is sufficient to normalize vascular health or whether the vasculature represents an additional therapeutic target in SMA.

Future perspectives

Although considerable progress has been made in defining the vascular phenotype in SMA, important questions remain regarding clinical significance, underlying mechanisms, and therapeutic implications.

A priority for future research will be translating these findings into clinical practice. Further development and validation of non-invasive vascular biomarkers could provide valuable tools for monitoring vascular health throughout disease progression and assessing vascular responses to SMN-restorative therapies. Longitudinal studies in treated patients will be particularly important to determine whether vascular abnormalities persist, progress or are reversed following each of the currently approved therapies. Together, these approaches could establish the clinical significance of vascular pathology and determine whether it is adequately addressed by current therapies.

In addition, it remains important to define the molecular pathways through which SMN deficiency disrupts endothelial cell function and to establish how these changes affect vascular development, maintenance and barrier integrity across different organ systems. Several complementary approaches could address these questions: transcriptomic analyses and single-cell RNA sequencing to identify disease-associated changes in endothelial gene expression and cellular pathways; endothelial-specific SMN depletion models to establish a direct link between SMN depletion and vascular defects; advanced human in vitro models, (organoids, assembloids and multicellular co-culture systems) to provide more physiologically relevant platforms for study. These approaches will guide us toward a more complete understanding of the role of vascular pathology in SMA.

Finally, if vascular abnormalities persist despite SMN-restorative therapy, a careful study of the route and mechanisms by which SMN levels are restored may lead to a better understanding of how to target the vascular system. Equally, combinatorial approaches may provide an opportunity to address residual pathology and further improve long-term outcomes for individuals living with SMA.

Conclusion

The evidence reviewed here demonstrates that vascular pathology is a widespread feature of SMA, affecting both the CNS and multiple peripheral organs. Rather than representing a secondary consequence of motor neuron degeneration, accumulating evidence suggests that endothelial dysfunction and resulting vascular pathology is an intrinsic consequence of SMN deficiency, resulting in impaired angiogenesis, altered vascular architecture and compromised barrier integrity. Importantly, these abnormalities are detectable in patients living with SMA and have the potential to contribute to both neurodegeneration and long-term systemic pathology.

The remarkable success of SMN-restorative therapies has fundamentally changed the outlook for individuals living with SMA. However, whether these treatments fully restore vascular health remains uncertain. As survival continues to improve, understanding the long-term consequences of vascular dysfunction and determining whether it requires direct therapeutic intervention will become increasingly important. The vasculature deserves to be considered alongside motor neurons as a potentially important contributor to SMA pathogenesis and a target for future therapeutic development.

Generative AI statement

The author(s) declared that Generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Ravindra N Singh, Iowa State University, United States

Reviewed by: Michael Christian Virata, San Beda University, Philippines

Elisa Pagliari, University of Milan, Italy