The aortic ECM, composed predominantly of elastin and collagen, provides elasticity, tensile strength and structural support to the aortic wall (15). Disruption of ECM homeostasis weakens the aortic wall, thereby predisposing it to aneurysm formation and dissection. In addition, pathogenic variants in genes such as fibrillin-1 (FBN1) and type III collagen α1 chain (COL3A1), disrupt collagen maturation and deposition, thereby compromising the structural integrity of the aortic wall and increasing susceptibility to aortic disease (16,17). Although ECM dysfunction is strongly supported by genetic and experimental evidence, the events that link primary matrix defects to cellular and signaling abnormalities have yet to be completely elucidated.

The TGF-β signaling pathway is a key pathway that is implicated in HTAD due to its role in vascular development, ECM regulation and cellular homeostasis. Pathogenic variants affecting TGF-β pathway components are associated with thoracic aortic disease of varying severity (18–20). In mechanistic terms, these alterations may disturb the balance between canonical Smad-dependent signaling and non-canonical pathways. In particular, the downstream activation of extracellular signal-regulated kinases ½ (ERK1/2), c-Jun N-terminal kinase (JNK) and PI3K/Akt has been implicated in maladaptive aortic remodeling (21,22).

However, the precise contributions made by TGF-β signaling events remains controversial, since both increased and decreased pathway activity has been reported in the literature, suggesting the existence of subtype-specific, stage-dependent and compensatory effects.

vSMCs are essential for maintaining aortic wall integrity, vascular tone and adaptive responses to mechanical stress. Pathogenic variants in genes such as smooth muscle cell-specific myosin heavy chain (MYH)11 and actin α2, smooth muscle (ACTA2) impair the contractile function of vSMCs and ECM synthesis, leading to a reduction in the mechanical strength of the vascular wall (23,24). As a result, these mutations hinder vSMCs from appropriately sensing and responding to hemodynamic stresses, which accelerates the degradation and dilation of the aortic wall (25–27). In particular, increased rates of vSMC apoptosis have been implicated in both the initiation and progression of HTAD, further undermining aortic wall stability (28). Nevertheless, the precise contributions made by distinct vSMC phenotypes across genetic subtypes have yet to be fully elucidated.

Hemodynamic stress contributes to aortic remodeling, particularly in BAV-associated aortopathy. Valvular abnormalities may lead to the generation of disturbed flow, aortic regurgitation and increased wall shear stress (WSS), all of which may promote aortic dilation (29–31). The BAV exhibits a significant genetic predisposition, with variants in genes including members of the NOTCH1 and GATA families being implicated in non-syndromic disease (32). These variants may be predisposed not only to valve malformation, but also to intrinsic aortic wall weakness, thereby amplifying the effects of abnormal flow. Therefore, mechanistically speaking, hemodynamic stress probably acts in concert with genetic susceptibility, rather than as an isolated driver.

Considered altogether, ECM disruption, vSMC dysfunction, dysregulated signaling and abnormal hemodynamic stress form an interconnected pathogenic network in HTAD. Matrix defects may alter mechanotransduction and growth factor bioavailability; signaling abnormalities may affect vSMC phenotype and ECM remodeling; and abnormal flow may further aggravate wall degeneration in genetically susceptible aortas. In view of these defects, defining the relative contribution and hierarchy of these mechanisms will be essential in terms of improving risk prediction and developing mechanism-based therapies.

MFS, an autosomal dominant connective tissue disorder with aortic dissection represents a major life-threatening complication. The primary causative gene, FBN1, encodes fibrillin-1, the major structural component of extracellular microfibrils distributed throughout multiple tissues and organs (33). FBN1 is composed of multiple epidermal growth factor (EGF)-like domains, the majority of which are calcium-binding (cb)EGF-like domains. These motifs bind extracellular calcium ions, thereby stabilizing FBN-1 and protecting it from proteolytic degradation (34,35).

In various MFS mouse models, different degrees of FBN1 deficiency have been observed, frequently accompanied by aortic aneurysm formation, dissection and premature death (34). In patients with Marfan syndrome, distinct histopathological alterations have been identified across different affected aortic segments, which also exhibit segment-specific patterns of ECM remodeling (36). At the molecular level, mutations in FBN1 alter the secondary structure of EGF-like domains, leading to protein misfolding and impaired microfibril assembly and stability. Furthermore, structural abnormalities in cbEGF-like motifs increase the susceptibility of FBN1 to proteolytic degradation (37). Collectively, these abnormalities in microfibril architecture and the reduction in FBN1 levels compromise aortic wall integrity and promote the development of TAAD.

Beyond these structural abnormalities, studies conducted on MFS models and patient tissues have revealed the critical yet complex role of TGF-β in the progression of TAAD. Accumulating evidence indicates that enhanced TGF-β signaling, particularly through the non-canonical (Smad-independent) pathway, is a major driver of aortic disease. Inhibition of TGF-β signaling in myeloid cells attenuates aortic aneurysm formation in MFS mouse models, whereas TGFB1 expression has been reported to be elevated in patients with MFS (38,39). Studies suggest that FBN1 plays an essential role in sequestering active TGF-β via latency-associated peptide through its interaction with potential TGF-β binding proteins, thereby limiting TGF-β activation (Fig. 2A) (40,41). The loss of FBN1 integrity may therefore increase TGF-β activity, consequently resulting in excessive TGF-β activation and overactivation of multiple downstream non-canonical signaling cascades, including the ERK1/2 and JNK1 pathways (Fig. 2B). Losartan, an angiotensin II (AngII) type 1 receptor blocker, has been shown to exert protective effects against aneurysm formation by suppressing non-canonical ERK1/2 signaling (42). Collectively, dysregulation of the TGF-β signaling pathway and aberrant ERK1/2 activation are considered key contributors to aortic aneurysm development in MFS.

LDS is an autosomal dominant aortic aneurysm syndrome characterized by arterial aneurysms and tortuosity, craniofacial abnormalities, skeletal features and cutaneous manifestations. Although its clinical manifestations are similar to those of MFS, typical distinguishing features of LDS include a bifid uvula, arterial tortuosity and wide-set eyes. The current gene-based classification acknowledges seven LDS subtypes (43). The two most common forms, LDS1 and LDS2, are caused by pathogenic variants in TGF-β receptor 1 (TGFBR1) and TGFBR2, respectively, whereas variants in the genes SMAD3, TGFB2, TGFB3, SMAD2 and importin 8 have also been identified in affected individuals (44–46). Collectively, these genes encode ligands, receptors or intracellular mediators of the TGF-β signaling pathway, thereby rendering TGF-β dysregulation the central pathogenic feature of LDS.

Clinical studies have consistently shown enhanced TGF-β signaling in LDS aortic tissue across subtypes, including increased levels of TGF-β expression, SMAD2 phosphorylation and TGF-β target gene expression (19,46). However, these findings were paradoxical, as the majority of the reported TGFBR1 and TGFBR2 mutations have been shown to reduce receptor serine/threonine kinase activity (18). Notably, increased levels of phosphorylated SMAD2 have been observed both in LDS patient aortic tissue with loss-of-function TGFBR1 variants and in LDS mouse models (47,48).

Taken together, these observations suggest that cell-autonomous defects in canonical TGF-β signaling may trigger the compensatory upregulation of TGF-β ligand production, which, in turn, both promotes the paracrine activation of neighboring cells and enhances non-canonical signaling, thereby amplifying maladaptive aortic wall remodeling (49). Therefore, LDS-associated mutations may attenuate canonical TGF-β signaling while driving the compensatory activation of non-canonical pathways, and the uncovering of this mechanism helps to explain the paradoxical increases in TGF-β expression that are observed in LDS, underscoring the complex role of TGF-β pathway dysregulation in the pathogenesis of HTAD (Fig. 2C).

vEDS is one of the most severe forms of EDS and is caused by heterozygous pathogenic variants in the gene COL3A1. Arterial dissection or rupture, including aortic rupture, is the leading cause of mortality in affected individuals (50). COL3A1 encodes the α1 chain of type III collagen. Of note, three of these chains assemble into homotrimeric type III collagen molecules, which are characterized by a repeating Gly-X-Y amino acid sequence. Type III collagen is abundant in blood vessels and hollow organs, where it provides tensile strength and tissue flexibility (51).

The majority of pathogenic COL3A1 variants in vEDS involve glycine substitutions within the collagen triple-helical domain, which results in the disruption of trimer formation and the destabilization of type III collagen. In addition to this structural mechanism, however, recent studies have suggested that altered stress-activated signaling may modify the severity of the disease (52–54). In vEDS modeled mice, genetic ablation of the signaling protein mitogen-activated protein kinase kinase 6 (Map2k6), a p38-activating kinase, was found to increase the risk of aortic rupture, and this was associated with reduced activation of p38 and enhanced protein kinase C/ERK phosphorylation (53). Taken together, these findings suggested that maladaptive crosstalk between ERK and MAP2K6/p38 signaling may exacerbate matrix-driven vascular fragility.

In conclusion, vEDS is primarily driven by COL3A1-mediated type III collagen instability, which results in profound arterial and visceral fragility. Emerging evidence has demonstrated that signaling modifiers may further influence rupture susceptibility, thereby linking ECM defects to broader stress-response pathways in syndromic HTAD.

Approximately 20% of cases of non-syndromic TAAD have exhibited familial aggregation, and this condition, which is typically inherited in an autosomal dominant manner, is termed FTAAD (55). Compared with sporadic disease, FTAAD generally presents at an earlier stage and is associated with more rapid aneurysm growth. The most significant FTAAD-associated genes include ACTA2, MYH11, MYLK, protein kinase cGMP-dependent 1 (PRKG1) and lysyl oxidase (LOX) (3,56–59). With the exception of LOX, these genes mainly affect the vSMC contractile apparatus, leading to impaired contraction, defective mechanosensing and phenotypic remodeling, whereas LOX variants are primarily responsible for compromising ECM crosslinking and biomechanical strength.

The contractile apparatus of vSMCs consists of two myosin heavy chains, two essential light chains and two regulatory light chains (RLC) (25,60). Concerning the genes involved in these processes (and as mentioned above), the SMC-specific filament, α-actin, is encoded by ACTA2, whereas the myosin heavy chain is encoded by MYH11. During vSMC contraction, intracellular Ca²+ levels increase, and Ca²+ binding to calmodulin activates MYLK, which subsequently phosphorylates RLC, resulting in the contractile force of vSMCs (60). By contrast, during relaxation, × cGMP-dependent protein kinase I (PKG-I), encoded by PRKG1, activates myosin light chain phosphatase, leading to RLC dephosphorylation and vSMC relaxation. Notably, MYLK and PRKG1 encode key kinases involved in the regulation of smooth muscle contraction and relaxation, respectively.

Among these genes, ACTA2 is the most common mutation associated with FTAAD. Mutations that disrupt arginine 179 (p.R179H) and arginine 258 (p.R258C) have been shown to correlate with aortic events (61). In the ACTA2 knockout mouse model, aortic α-SMA expression was reduced, accompanied by more severe aortic dilatation, which was associated with increased levels of reactive oxygen species and enhanced NF-κB signaling in vSMCs, as well as increased sensitivity to exogenous AngII (62,63). Concurrently, downregulation of α-SMA reduces the expression of integrins involved in cell-matrix adhesion and impairs the interaction between vSMCs and the ECM, thereby weakening aortic wall contractility (64,65).

Mutations in MYH11 are less common, but likewise impair vSMC function. vSMC phenotypic switching from a contractile to a synthetic state has been observed in the aortas of patients with TAAD carrying mutations in both ACTA2 and MYH11 (24,26,66). This phenotypic transition is considered an important mechanism contributing to TAAD pathogenesis. Furthermore, histological staining of aortic sections from patients carrying MYH11 mutations revealed enhanced TGF-β signaling, which may be associated with the pathogenesis of their TAAD. By contrast, the upregulation of TGF-β in ACTA2 mutations exhibited a less pronounced effect on TAAD (67).

Pathogenic variants in MYLK and PRKG1 are relatively rare; however, both have been associated with severe TAAD by disrupting vSMC contractile function (68). Insufficient MYLK activity reduces phosphorylation of RLC proteins, thereby impairing vSMCs' contractile function. A recent study has further shown that MYLK overexpression can reverse the transition of vSMC from a contractile phenotype to a secretory phenotype, and suppress the TGF-β signaling, ultimately attenuating TAAD progression (69). In addition, the p.Arg177Gln mutation in PRKG1 alters the structure of its protein PKG-I, leading to its overactivation, which results in reduced phosphorylation of RLC in vSMC. This ultimately leads to a reduction in vSMC contractility, thereby predisposing to aneurysm and dissection (70).

LOX encodes a copper- dependent LOX, which catalyzes the oxidative deamination of lysine and hydroxylysine residues. In the ECM, LOX facilitates the formation of covalent crosslinking between collagen and elastin, as well as the precipitation of an insoluble matrix, making it an indispensable component of tissue development and pathological repair (71–73). Both missense and loss-of-function LOX variants have been associated with extensive aortic and arterial aneurysmal disease, which is accompanied by connective tissue manifestations (74,75). Reduced LOX activity, along with a reduction in elastin within the ECM, is hypothesized to impair the elasticity and tensile strength of the aorta, thereby inducing the occurrence of TAAD.

Other ECM-associated genes affecting the ECM have also been implicated in FTAAD, including COL3A1 and microfibril-associated glycoprotein 5 (MFAP5), encoding a protein involved in the interaction within FBN1 in the ECM. Loss-of-function variants in MFAP5 may likewise represent a potential cause for the development of FTAAD (76).

A BAV is the most common congenital valvular heart defect, with aortic stenosis and regurgitation as its most common complications. The incidence of Stanford type A proximal aortic dilatation TAAD is significantly increased in patients with a BAV, indicating a strong association between valve morphology and aortic disease (77). BAV-related TAAD can be attributed to various factors, including genetic predispositions, alterations in the ECM of the vascular wall and hemodynamic changes.

The primary aortopathy hypothesis proposes that BAV-associated aortic dilation arises, at least in part, as an intrinsic developmental or genetic abnormality of the aortic wall. NOTCH1 variants have been shown to hinder the endothelial-mesenchymal transition of blood vessels during embryonic development, impairing the aorta's ability to respond to pulse pressure (29,78). GATA binding protein 4 (GATA4), GATA5 and GATA6, members of transcription factor family, are expressed in the mesoderm during early heart development. Variants in these genes can disrupt transcription factor activity, leading to BAV, although their direct contribution to TAAD remains less well defined (32). Additionally, variants in other genes, including LOX, roundabout guidance receptor 4 and ACTA2, may also contribute to aortic dilation through affecting ECM integrity, endothelial stability or vSMC contractile function.

The hemodynamic hypothesis holds that the BAV significantly alters the direction of blood flow and shear stress in the ascending aorta, resulting in structural changes in the aortic wall. Four-dimensional flow cardiovascular magnetic resonance imaging has revealed altered blood flow patterns and increased WSS in the aorta of patients diagnosed with a BAV (79–81). Computational models likewise have shown such features as a reduced valve opening area, eccentric flow and elevated WSS compared with tricuspid aortic valves (30,82). Collectively, these findings support a mechanistic role for abnormal hemodynamic stress in BAV-associated aortopathy; furthermore, a greater extent of elevated WSS has been associated with faster aortic dilation, suggesting that WSS may serve as a marker of disease progression (83).

However, TAAD progression may persist in certain patients with BAV even after aortic valve replacement, suggesting that both the primary aortopathy and hemodynamics hypotheses likely play a role (84). Increasing evidence suggests that their relative contribution may vary according to the aortic segment, with aortic root dilation being more strongly influenced by intrinsic genetic factors, and ascending aortic dilation being more susceptible to hemodynamic stress (85). However, definitive separation of these effects remains challenging, since most of the studies in the literature, to date, have lacked integrated genetic and flow imaging data. Future studies combining genomic profiling with advanced hemodynamic phenotyping will be essential, both to clarify their respective roles and to improve risk stratification in BAV-associated TAAD.

In MFS, β-blockers, including atenolol and propranolol, remain the standard treatment, although losartan may also be administered to slow aortic dilation in certain patients, with the efficacy of the treatment being influenced by the FBN1 genotype (84,91).

In LDS, pharmacologic management mainly relies on angiotensin receptor blockers and β-blockers to reduce blood pressure and aortic wall stress (92–94). Comparative data have suggested that losartan may be more effective at reducing pulse wave velocity and arterial stiffness, whereas atenolol may be more effective at lowering cardiac output and in the treatment of stroke, supporting individualized drug selection (95). No LDS-specific targeted therapy has yet been approved.

Because vEDS has a distinct pathogenic basis, therapies effective in other forms of HTAD may not be equally beneficial. The selective β1-blocker celiprolol appears to improve vascular integrity in vEDS animal models and is currently the preferred preventive medication, whereas losartan has been shown to have limited benefits in experimental models (96,97).

For FTAAD and BAV-associated TAAD, no etiology-specific drug therapies are available at present; their management therefore depends on antihypertensive treatment, imaging surveillance and prophylactic surgery (98,99).

Several mechanism-based approaches have shown promise in preclinical studies. Given the central role of TGF-β signaling in MFS and LDS, TGF-β type I receptor (TβRI/ALK5) kinase inhibitors have been shown to be beneficial in animal models and may represent a future targeted strategy (100,101). RNA-based therapies, including allele-specific silencing or correction of pathogenic variants such as MYH11, are also currently being investigated, although efficient vascular delivery, specificity and long-term safety remain major challenges.

Other experimental targets include inducible nitric oxide synthase 2, the inhibition of which has also been shown to reverse aortic dilation and medial degeneration in MFS animal models (102). In BAV-associated TAAD, epigenetic regulation and microRNA-based mechanisms are currently being explored as possible therapeutic targets (103–105).

Advances that are being made in disease modeling and biomarker development may further accelerate therapeutic discovery. For example, patient-derived smooth muscle cell organ-on-a-chip models can recapitulate cyclic biomechanical strain in the aortic wall, thereby providing a platform for mechanistic studies and drug screening (106). In parallel, standardized biomarker validation frameworks may facilitate the clinical application of circulating markers associated with disease progression or treatment response (107–109).