Pyroptosis is now characterized as a type of regulated, lytic cell death executed by the pore-forming activity of the N-terminal domains of GSDM family proteins in the plasma membrane (14). This process is triggered by infectious stimuli [such as pathogen-associated molecular patterns (PAMPs)] or non-infectious danger signals [such as damage-associated molecular patterns (DAMPs)] (15). Hallmark features include robust proinflammatory activity and pore-dependent execution (16).

Morphologically, pyroptotic cells undergo characteristic changes. The N-terminal domain of GSDM proteins oligomerizes to form 10- to 20-nm pores in the plasma membrane (17). This process induces ionic imbalances (K⁺ efflux and Na⁺/Ca²⁺ influx) and osmotic lysis, resulting in cellular swelling, organellar edema and eventual plasma membrane rupture with the release of intracellular contents and inflammatory mediators (18,19). The nucleus exhibits chromatin condensation (often marginalized) but lacks the DNA laddering characteristic of apoptosis (20). Molecularly, pyroptosis crucially depends on activated inflammatory caspases, which cleave GSDM proteins to release the active, pore-forming N-terminal domain from autoinhibition (21,22). Caspase activation is typically driven by specific inflammasome complexes upon recognition of PAMPs or DAMPs (23). These caspases also process proinflammatory cytokine precursors [pro-interleukin (IL)-1β and pro-IL-18] into their mature, active forms (24).

Functionally, pyroptosis essentially initiates and amplifies inflammatory responses and plays pivotal roles in immune defense and disease pathogenesis. Membrane rupture leads to the massive release of cellular contents [DAMPs, such as high mobility group box 1 (HMGB1) protein, adenosine triphosphate (ATP), IL-1α and IL-33] and mature IL-1β and IL-18 (25). These alarmins potently activate and recruit neighboring immune cells, amplify the inflammatory cascade and establish an immune defense mechanism for intracellular pathogen clearance (26). However, uncontrolled pyroptosis can drive intense inflammation and serve as a central pathogenic driver in diseases such as autoinflammatory disorders and atherosclerosis (27,28).

The execution of pyroptosis across all known pathways ultimately relies on the GSDM protein family, whose pore formation is the indispensable final common step. In humans, this family comprises GSDMA, GSDMB, GSDMC, GSDMD, GSDME and deafness autosomal recessive 59 (29). GSDM family members A-E consist of a highly differentiated peptide linker domain, a conserved cytotoxic N-terminus (GSDM-N) that forms pores and a conserved C-terminus (GSDM-C) that inhibits GSDM-N (30). A self-locking regulatory mechanism that consists of GSDM-C and GSDM-N ensures that the active GSDM-N domain is released only after specific cleavage by inflammatory caspases, thereby preventing cell damage caused by spontaneous activation (31). GSDM pores use an 'electrostatic sieve' mechanism, orchestrating the selective and efficient release of IL-1β/IL-18. Concurrently, pore-driven K⁺ efflux activates the NOD-like receptor family-pyrin domain-containing 3 (NLRP3) inflammasome, initiating a positive feedback loop that amplifies inflammatory responses (32).

GSDM family members exhibit distinct expression patterns and functions. GSDMD is widely expressed and constitutively highly expressed in myeloid cells (for example, neutrophils, macrophages and dendritic cells) (33). GSDMD serves as the most specific and efficient substrate for caspase-1/4/5/11 and is considered the most versatile executor (22). GSDME is widely distributed across the placenta, intestines, thyroid gland and brain (34); it can switch apoptotic stimuli to pyroptotic death stimuli (35). GSDMB is enriched in the gastrointestinal epithelium and in colorectal cancer, and can be cleaved by granzyme A (GzmA) (36,37). GSDMA and GSDMC are predominantly restricted to epithelial surfaces and the skin (38).

The multipathway nature of pyroptosis initiation reflects the flexibility of cells in response to various threats, including PAMPs, DAMPs, cytosolic disturbances and bacterial components, such as lipopolysaccharide (LPS) (Fig. 1).

The canonical caspase-1-dependent pathway is the most extensively studied pyroptosis pathway, and is primarily activated in immune cells during host defense against pathogens. PAMPs, DAMPs and cytosolic disturbances activate pattern recognition receptors (PRRs) (39). Subsequently, PRRs trigger oligomeric sensors, the adaptor protein apoptosis-associated speck-like protein (ASC) and the effector inactive pro-caspase-1 to assemble canonical inflammasomes (40). Five sensor types, NLRP3, absent in melanoma 2 (AIM2), NOD-like receptor family-pyrin domain containing 1, pyrin and NOD-like receptor c4 (NLRC4), primarily mediate the canonical pathway. The caspase recruitment domain (CARD) domain of ASC subsequently engages the CARD domain of pro-caspase-1, triggering self-cleavage of pro-caspase-1 to generate mature caspase-1 (41). Activated caspase-1 then cleaves both GSDMD to generate the pore-forming GSDMD N-terminal (GSDMD-NT) and pro-IL-1β/pro-IL-18 into their mature forms (42). After that, pores in the cell membrane form and pyroptosis occurs.

The non-canonical caspase-4/5-dependent pathway in humans and the non-canonical caspase-11-dependent pathway mice responds directly to cytosolicLPSfrom Gram-negative bacteria (30). These caspases are upregulated by type I interferon signaling and are activated upon LPS binding without inflammasome assembly to directly cleave GSDMD to induce pyroptosis (42). Moreover, K⁺ efflux mediated by GSDMD pores indirectly activates the NLRP3 inflammasome-caspase-1 canonical pyroptosis pathway (22). This indirect mechanism is recognized as central to non-canonical pathway-induced mature cytokine release and secondary inflammatory amplification. Caspase-4/5/11 cannot directly process pro-IL-1β/pro-IL-18 (43,44). However, recent evidence has demonstrated that LPS-activated human caspase-4 and -5 can cleave pro-IL-18 at sites identical to those targeted by caspase-1 (24). These findings challenge conventional perception, suggesting that non-canonical caspases may have the ability to directly address specific cytokine precursors under specific circumstances.

In the caspase-3-mediated pathway, caspase-3, traditionally designated as the key executioner of apoptosis, can initiate pyroptotic switching under specific stimuli. In GSDME-expressing cells, activated caspase-3 mediates specific cleavage of GSDME, generating its N-terminal pore-forming domain fragment. Subsequent plasma membrane pores and pyroptosis occur thereby executing a modality shift from apoptosis to pyroptosis (35). This process plays a particularly important role in tumor cell killing by chemotherapeutic agents.

In the caspase-8-mediated pathway, caspase-8 is a multifunctional initiator of caspase that promotes pyroptosis through multiple mechanisms; it can directly cleave GSDMD or GSDMC, or activate GSDME via caspase-3, leading to pyroptosis. This process, characterized by the absence of canonical IL-18 release coupled with compensatory IL-1α secretion, has been described as 'incomplete pyroptosis' (45). For instance, Yersinia infection triggers the canonical pyroptosis pathway through direct caspase-8 activation, where caspase-8 cleaves GSDMD (46). Under hypoxic conditions in the tumor microenvironment, tumor necrosis factor-α (TNFα) stimulation causes caspase-8 to cleave GSDMC, shifting the mode of cell death from apoptosis to pyroptosis (47). In caspase-1/11-deficient macrophages, the NLRP3 inflammasome activates caspase-3/8, which subsequently cleaves GSDME to induce pyroptosis. GSDMB can be activated by caspase-3/6/7.

The Gzm-mediated pathway represents a mechanism through which adaptive immune cells (cytotoxic T lymphocytes and natural killer cells) directly induce pyroptosis in target cells. Gzms, serine proteases delivered into target cells via perforin, can directly cleave GSDM proteins. The Gzm family comprises five known human members: GzmA, GzmB, GzmH, GzmK and GzmM (48). Among them, GzmA is the most abundant, while GzmB is distinguished by its potent proapoptotic activity (49,50). GzmB induces pyroptosis through both caspase-3-dependent and caspase-3-independent mechanisms: It rapidly activates caspase-3, which cleaves GSDME, and can also directly cleave GSDME itself (51). By contrast, GzmA directly hydrolyzes and cleaves GSDMB, triggering its pore-forming activity (37). This Gzm-mediated pyroptosis effectively amplifies inflammation in contexts such as antitumor immunity (52).

Accumulating evidence directly implicates pyroptosis activation in aneurysmal lesions, particularly in VSMCs. Both human AA tissues and mouse models consistently exhibit elevated levels of activated pyroptotic executors, including GSDMD-NT accumulation and mature IL-1β and IL-18, specifically in VSMCs within the lesion wall (53,74). Furthermore, components of the NLRP3 inflammasome are upregulated in AA tissues, with activation markers such as ASC speck formation and cleaved caspase-1 detected specifically in VSMCs (54,58). Functional studies using NLRP3- or caspase-1/11-deficient mice have revealed markedly attenuated vascular dilation, reduced VSMC death and decreased disease incidence across multiple aneurysm models (57,58,75,76). Crucially, abdominal AAs (AAAs) are associated with increased TNFα levels. Wang et al (53) observed cell swelling and pore formation in the VSMC membrane under TNFα stimulation by scanning electron microscopy, indicating the pyroptotic lytic death of VSMCs during mouse AAAs. This direct and rapid mode of cell death leads to substantial and irreversible loss of VSMCs, disrupting the cellular foundation required for maintaining aortic wall tension and structural integrity. Thus, pyroptosis represents a key mechanism driving vascular wall thinning, dilation and eventual rupture.

The profound impact of pyroptosis on aneurysmal pathology extends far beyond the direct killing of VSMCs; it triggers and perpetuates a storm of inflammation within the aortic wall.

First, as aforementioned, cellular components and mature and highly bioactive IL-1β and IL-18 secreted from VSMCs undergoing pyroptosis function as potent alarm signals and chemoattractants, robustly recruiting circulating monocytes, neutrophils, T lymphocytes and other inflammatory cells into the vascular wall, particularly at the medial-adventitial junction (77,78). IL-1β and IL-18 directly activate ECs to generate adhesion molecules, facilitating leukocyte rolling, adhesion and transmigration (79). Released ATP acts as an agonist for the purinergic receptor p2x-ligand-gated ion channel 7 (P2X7) receptor on macrophages, thereby activating them, whereas HMGB1 strongly promotes proinflammatory effects via receptor for advanced glycation end product/toll-like receptor (TLR) signaling (80,81).

Second, pyroptosis establishes a positive feedback loop that aggravates vascular inflammation. Infiltrating immune cells, particularly macrophages, are strongly activated by pyroptosis-derived factors (especially IL-1β and IL-18) and DAMPs, polarizing toward a proinflammatory M1 phenotype (82). These cells subsequently produce and release large amounts of proinflammatory cytokines (such as TNFα, interleukin-6 and IL-1β), chemokines (such as monocyte chemotactic factor/c-c motif chemokine ligand 2) and reactive oxygen species (ROS) (83,84). Crucially, they may undergo pyroptosis themselves, leading to further release of DAMPs and cytokines. This creates a self-sustaining and amplifying vicious cycle of 'pyroptosis-inflammation-more pyroptosis/inflammation' (72,85). This escalation of the inflammatory milieu not only further damages and kills additional VSMCs (through pyroptosis or other death modalities) but also profoundly affects the other aforementioned key pathological processes, ultimately driving pathological vascular remodeling through the sustained release of proinflammatory cytokines and proteases (86). As shown by Zhang et al (85), exosomal lncRNA plasmacytoma variant translocation 1 (PVT1), which originates from M1 macrophages, promotes human aortic vascular smooth muscle cell (HA-VSMC) pyroptosis and inflammation via the miR-186-5p/HMGB1 pathway, upregulating NLRP3 and GSDMD expression, thereby exacerbating AAA progression (85).

Pyroptosis does not occur in isolation but is intricately interwoven with other pathological processes in AAs, collectively driving disease progression. The first of these processes is VSMC phenotypic switching. The study by Sun et al (87) demonstrated that NLRP3 inhibition prevents platelet-derived growth factor subunit B-induced phenotypic modulation in HA-VSMCs (87). Similarly, the genetic deficiency of NLRP3 attenuates angiotensin II (Ang II)-induced phenotypic switching and vascular remodeling in mouse VSMCs (88). NLRP3 activation contributes to the phenotypic transition and proliferation of rat VSMCs in hypertension (89). Inflammatory mediators and cytokines released during pyroptosis can induce surviving VSMCs to shift from a contractile to a synthetic phenotype (87). This transition impairs contractile function and promotes the secretion of proinflammatory factors and matrix metalloproteinases (MMPs), further disrupting vascular homeostasis and accelerating inflammation and matrix degradation. The second process is ECM degradation. The inflammatory microenvironment triggered by pyroptosis plays a central role in promoting ECM degradation. High levels of cytokines (such as IL-1β and TNFα) upregulate multiple MMPs (including MMP-2 and MMP-9), disrupting the balance between MMPs and their endogenous inhibitors (57). The NLRP3 inflammasome may further directly activate pro-enzymes through cleavage of inhibitory domains (53). Studies have demonstrated that NLRP3 or caspase-1 deficiency significantly reduces elastin degradation and MMP activation during early mouse AAA formation (57,76,90). Activated MMPs degrade critical structural components of the aortic wall, namely, elastic fibers and collagen, resulting in a precipitous loss of tensile strength, decreased wall compliance, and ultimately constituting the biomechanical basis for dilation and rupture. Infiltrating inflammatory cells, such as neutrophils and macrophages, are also major sources of MMPs (91). The third pathological process in AAs is EC dysfunction. Pyroptosis-induced EC death and associated inflammatory factors compromise endothelial integrity, increase vascular permeability, and facilitate plasma leakage and leukocyte extravasation, exacerbating vascular inflammation and edema. Furthermore, pyroptosis-related EC injury may cause dysregulation of the secretion of vasoactive factors (92). A fourth process is crosstalk with other programmed cell death pathways. VSMCs within the aneurysmal microenvironment often undergo multiple forms of programmed cell death simultaneously or sequentially. For instance, in Ang II-induced mouse AAA models, markers of pyroptosis (GSDMD activation), apoptosis (caspase-3 cleavage) and necroptosis (mixed lineage kinase domain-like protein phosphorylation) have been detected concurrently in VSMCs (53,93,94). Complex molecular crosstalk occurs among these pathways. The coexistence and interaction of these death mechanisms likely synergistically amplify VSMC loss and inflammatory responses, thereby accelerating aneurysmal progression. The differences and connections between them will be elaborated later in the review.

Studies have indicated that cell-specific deletion of GSDMD in either VSMCs or macrophages ameliorates AAA development and progression (74,86). In various AAA induction models, Gsdmd^−/− mice consistently exhibited significant and concordant protective effects. Specifically, VSMC-specific GSDMD knockout substantially attenuated aneurysm formation, aortic dilation, VSMC death, inflammatory responses and ECM degradation (74). Research by Ye et al (86) further confirmed that macrophage-specific GSDMD deletion also delayed mouse AAA progression (86). These genetic findings are strongly supported by pharmacological interventions: Compounds that specifically inhibit GSDMD pore formation or oligomerization confer therapeutic benefits in AAA mouse models (65). Together, these results underscore GSDMD as a highly attractive and precise therapeutic target for inhibiting pyroptosis in aneurysms.

GSDMD plays a uniquely destructive role in aneurysms, as its ability to mediate pyroptosis is rapid and lytic. Although VSMCs in AAs may be exposed to multiple cell death signals, GSDMD-driven pyroptosis is rapid and lytic, causing complete cellular disintegration and the rapid release of cellular contents. By contrast, apoptosis is generally slower and non-lytic (via phagocytic clearance) and tends to be anti-inflammatory or immunologically silent (12,95). Thus, pyroptosis poses a greater threat due to its acute and inflammatory nature, leading to a substantial net loss of structural cells. Furthermore, as a 'master switch' that initiates and amplifies lethal inflammatory storms, the GSDMD pore-mediated release of inflammatory mediators occurs explosively and at high local concentrations, resulting in a proinflammatory effect far exceeding that of other secretory mechanisms (96). As previously discussed, these mediators further amplify inflammatory signaling and cellular damage (97). Consequently, GSDMD pores essentially control the release of the most potent initial triggers (IL-1β/IL-18) and critical DAMPs in the pro-aneurysmal inflammatory cascade (86). Without GSDMD pore formation, even upon upstream inflammasome activation, the release efficiency of these key factors is substantially compromised, resulting in a failure to ignite or sustain the intense inflammatory storm necessary for driving aneurysm progression (98,99). This establishes GSDMD as an indispensable hub and amplifier connecting upstream danger sensing (inflammasome activation) with widespread downstream inflammatory destruction (immune cell infiltration, activation and cytokine storm).

In addition to GSDMD, other GSDM family members, notably GSDMB and GSDME, and potentially GSDMA/GSDMC, have been implicated as pyroptotic executors in aneurysmal vascular remodeling processes (64,100,101).

These paralogs share a common execution mechanism, yet their upstream activating signals are markedly different. GSDMB is cleaved by lymphocyte-derived GzmA at Lys244 and is upregulated in human AAA tissues in parallel with caspase-4 (102). A functional study has demonstrated that GSDMB knockdown attenuates macrophage pyroptosis via the mitochondrial ROS (mtROS)-dynamin-related protein 1 (Drp1)-caspase-4 axis, subsequently suppressing VSMC phenotypic switching and apoptosis, and implicating GSDMB in AA pathogenesis through immune-VSMC crosstalk (64). By contrast, GSDME activation is intimately linked to the apoptosis-pyroptosis continuum: In both human and murine aneurysmal tissues, GSDME mediates non-canonical pyroptosis and modulates inflammatory responses and senescence following VSMC phenotypic transition (100). Moreover, AMP-activated protein kinase-mediated phosphorylation of GSDME suppresses its proinflammatory activity, revealing a regulatory interface between metabolic homeostasis and cell death (103). Although direct evidence linking GSDMA or GSDMC to AAs is currently lacking, their non-canonical activation mechanisms in other systems, including streptococcal toxin-mediated GSDMA cleavage and GzmB-induced GSDMC pyroptosis in melanoma cells, offer exciting clues for potential alternative activation pathways in the vasculature (104,105). Furthermore, the non-proteolytic activation modes observed in lower eukaryotes, such as dimer dissociation-dependent activation of Tricho-GSDMs in Trichoplax adhaerens, suggest at an even broader diversity of GSDM regulatory mechanisms that may extend beyond current paradigms (106).

At the cellular level, emerging evidence has revealed a preliminary division of labor among GSDM family members within the aortic wall. GSDMD is ubiquitously expressed in macrophages, T lymphocytes, VSMCs and ECs, and functions as a multilineage pyroptosis executor (65,74,86,107). GSDMB is predominantly enriched in macrophages, indirectly shaping VSMC fate through immune modulation (64), and GSDME is preferentially upregulated in VSMCs, potentially serving as a molecular switch that governs the apoptosis-pyroptosis transition (100). This functional segregation suggests that GSDMD and GSDMB may synergistically orchestrate immune defense in AAs, whereas GSDME governs VSMC death modality decisions. Although direct evidence for functional redundancy or complementarity among GSDM members remains elusive, these observations provide a conceptual framework for deciphering the spatiotemporal coordination of multicellular and multipathway death programs during AA progression.

As aforementioned, AA progression involves the concurrent or competing activation of multiple regulated cell death pathways, among which pyroptosis, apoptosis, necroptosis and ferroptosis exhibit distinct triggers, execution mechanisms, inflammatory consequences and cell-type preferences, collectively constituting a complex regulatory network that governs vascular cell fate.

Pyroptosis, the central focus of this review, has previously been characterized in detail in AAs. By contrast, apoptosis plays a dual role in AA pathobiology that is highly context dependent. As a non-lytic death modality, apoptosis is initiated by intrinsic mitochondrial pathways (Bcl-2 associated X protein/Bcl-2 antagonist/killer 1-caspase-9) or extrinsic death receptor pathways [fas cell surface death receptor (Fas)-fas ligand-caspase-8], culminating in caspase-3/7-mediated cellular dismantling and phagocytic clearance without eliciting inflammation (108-110). During early AA formation, physiological apoptosis contributes to vascular remodeling; however, in progressive stages, excessive VSMC apoptosis leads to medial thinning and elastic fiber depletion, whereas uncleared macrophage apoptosis may undergo secondary necrosis, exacerbating local inflammation (111). Notably, apoptosis-pyroptosis transformation has recently garnered attention. For instance, caspase-8 can initiate apoptosis and under certain conditions, cleave GSDMD or activate NLRP3. Apoptotic bodies generated during apoptosis may be phagocytosed by macrophages, triggering further inflammation and pyroptosis. In addition, cytokines released during pyroptosis can induce apoptosis in neighboring cells (112,113). This mechanism was preliminarily validated in aneurysmal VSMCs (64).

Necroptosis, a prototypical type of necrosis pathway, is triggered by death receptors (tumor necrosis factor receptor 1 and Fas) or PRRs (TLR3/4) under conditions of caspase-8 inhibition, which are executed via the receptor interacting serine/threonine kinase 1 (RIPK1)-RIPK3-mixed lineage kinase domain-like cascade to form plasma membrane pores and release cellular contents, with the intensity of inflammation intermediate between that of apoptosis and pyroptosis (114,115). In AAs, necroptosis has been documented in both macrophages and VSMCs and is associated with MMP activation and vascular wall remodeling (116). However, its shared upstream signals with pyroptosis (for example, RIPK1 involvement in inflammasome regulation) and similar lytic morphology suggest potential functional overlap or compensatory mechanisms.

Ferroptosis is characterized by iron-dependent lipid peroxidation, resulting from inactivation of the system X_c⁻/glutathione/glutathione peroxidase 4 antioxidant axis or disruption of the ferroptosis suppressor protein 1/CoEnzyme Q10 pathway (117-120). Morphologically, it features mitochondrial shrinkage with initially preserved plasma membrane integrity, culminating in membrane rupture and the release of DAMPs (121). In AAs, iron deposition and lipid peroxidation products are markedly elevated; VSMC ferroptosis has been shown to be involved in medial degeneration and potentially cross talks with pyroptosis. Ferroptosis-derived lipid peroxides activate the NLRP3 inflammasome, while pyroptosis-released heme exacerbates iron overload (122).

In summary, in the aneurysmal context, pyroptosis, apoptosis, necroptosis and ferroptosis do not occur in isolation but instead form an integrated programmed cell death network through shared triggers, competition for caspase resources and reciprocal modulation of inflammatory outputs. Pyroptosis is a network hub with a potent proinflammatory capacity; apoptosis promotes basal cellular turnover but can be switched to pyroptosis to amplify pathology, while necroptosis serves as an alternative lytic pathway when caspases are compromised, and ferroptosis is involved in positive feedback with pyroptosis through lipid peroxidation and oxidative stress. The cell-type preferences, temporal dynamics and functional consequences of these four death modalities in AAs are summarized comparatively in Table II (38,86,122-132).

The initiation of pyroptosis in AAs is governed by a multilayered network of upstream signaling cascades (Fig. 3). This architecture can be deconstructed into four hierarchical tiers, including the triggering level, transduction level, effector level and regulatory level. At the triggering level, a repertoire of membrane receptors, including α7 nicotinic acetylcholine receptor (α7nAChR), cannabinoid receptor type 1, P2X7, endothelin receptor type A and Ang II's type 1 receptor, senses danger signals within the aneurysmal microenvironment (55,57,62,67,72). At the transduction level, signals from receptors are transduced to the intracellular level via the modulation of key second messengers, such as cyclic adenosine monophosphate (cAMP)/protein kinase A, nuclear factor κ-light-chain-enhancer of activated B cells (NF-κB), p38 mitogen-activated protein kinase (p38)/extracellularly regulated protein kinases (ERK), phosphatidylinositol 3-kinase (PI3K)/protein kinase B (AKT), nuclear factor erythroid 2-related factor 2 (Nrf2)/heme oxygenase 1 and Ca²⁺ influx. Next, these intracellular signals converge on the effector level, namely the NLRP3 inflammasome assembly (53-56,67,69,73). Parallel kinase pathways, such as p38/ERK and PI3K/AKT, are engaged by effectors such as caspase recruitment domain family member 10 deficiency, further amplifying inflammasome assembly and crosstalk with the endoplasmic reticulum and mitochondrial stress (71). Non-canonical pyroptosis in AAs involves caspase-4/11-mediated GSDMB/GSDMD cleavage, which is driven by the mtROS-Drp1 or PI3Kγ-cAMP axis (64,69). Superimposed upon this core cascade is the regulatory level mediated by epigenetic mechanisms, where lncRNAs (such as PVT1) and miRNAs (such as miR-186-5p, miR-146a-5p, miR-17-5p and miR-26a) fine-tune signaling intensity by targeting key nodal molecules, including tumor necrosis factor receptor-associated factor 6 (TRAF6), thioredoxin interacting protein, HMGB1 and krüppel-like factor 4 (63,70,73,85). Furthermore, the cluster of differentiation 95 ligand-cluster of differentiation 95 axis induces caspase-8-mediated pyroptosis, whereas the kelch-like ECH-associated protein 1-Nrf2 pathway provides antioxidative counter-regulation (53,66).

The aforementioned levels are illustrated by specific examples in AA; for example, at the triggering and signal transduction level, TRAF6 promotes macrophage pyroptosis and accelerates mouse AAAs through NLRP3 inflammasome engagement (61). The endothelin-1/Ca²⁺ axis similarly drives VSMC pyroptosis and mouse AAA formation via NLRP3 and GSDMD activation (55). Conversely, α7nAChR activation suppresses NLRP3 and GSDMD activity, limiting pyroptosis and attenuating mouse AAAs (62). At the regulatory level, exosomal miR-17-5p from adipose-derived mesenchymal stem cells alleviates mouse AAAs by inhibiting thioredoxin-interacting protein-NLRP3 signaling and GSDMD-NT expression (63).

Collectively, these intertwined upstream signals precisely orchestrate canonical and non-canonical pyroptosis pathways in vascular cells, fueling the sustained inflammation and matrix degradation that characterize aneurysm progression. Related studies are shown in Table I (53-73).

Inflammasomes, particularly NLRP3, are attractive upstream targets.

MCC950 is a potent and selective small-molecule inhibitor of NLRP3; it binds to the NACHT domain of NLRP3, inhibiting its ATPase activity and thereby preventing inflammasome oligomerization and activation (152). Notably, MCC950 suppresses NLRP3 activation through both canonical and non-canonical pathways (153). In apolipoprotein E-deficient (ApoE^−/−) mice fed a Western diet, MCC950 delayed atherosclerosis progression and reduced IL-1β and IL-18 secretion (133). Although direct evidence in AA models needs accumulation, its potent anti-inflammatory effects suggest considerable therapeutic potential in AAs, including the alleviation of vascular inflammation and VSMC dysfunction. MCC950 has entered clinical trials for several autoinflammatory diseases; however, its development was halted in one trial due to hepatotoxicity, and structural optimization efforts are ongoing to improve its safety profile (154-156). Similar to MCC950, CY-09 and 3,4-meth ylenedioxy-β-nitrostyrene (MNS) also bind directly to the NACHT domain of NLRP3, competitively inhibiting ATP binding and blocking NLRP3 activation (134,157). CY-09 has been shown to alleviate aortic valve stenosis and calcification by suppressing proinflammatory cytokine expression through NLRP3 inhibition, whereas MNS improved outcomes in murine colitis (134,157). The natural compound curcumin modulates NLRP3 inflammasome activation by regulating NF-κB signaling, attenuating K⁺ efflux, blocking Ca²⁺ influx and preventing ASC oligomerization with NLRP3, and has been demonstrated to reduce endothelial damage and atherosclerosis in mice (136,158).

Beyond NLRP3, specific inhibitors for AIM2 or NLRC4 are less developed but represent a future direction. For instance, pristimerin has been shown to inhibit AIM2 and ameliorate tendonopathy in mice (137).

Direct caspase inhibition provides a strategy to block pyroptosis execution.

VX-765 is an orally bioavailable, reversible and selective caspase-1 inhibitor; it is a prodrug that is hydrolyzed in vivo to its active form, VRT-043198 (159). By inhibiting caspase-1, VX-765 reduces the production of IL-1β and IL-18, thereby mitigating inflammatory tissue damage. Li et al (138) demonstrated that VX-765 attenuated plaque formation and inflammation in an atherosclerosis model. In a vascular calcification model, it alleviated VSMC calcification via suppression of STAT3 signaling (139). Moreover, VX-765 protects against acute myocardial infarction by mitigating pyroptosis, hypoxia-induced cardiomyocyte injury, oxidative stress, apoptosis and inflammation (140). These findings suggest its potential benefit in AA-related vascular remodeling. Notably, VX-765 has been evaluated in clinical trials for psoriasis (NCT00205465) and epilepsy (NCT01048255; EudraCT 2011-004156-19), although its long-term use may raise safety concerns.

Other caspase-1 inhibitors also represent promising therapeutic candidates. For instance, Ac-YVAD-CHO is a peptide-based irreversible caspase-1 inhibitor widely used in basic research; it reduces the plasma levels of the caspase-1-dependent cytokines IL-1β and IL-18 in endotoxemic rats. Inhalation of Ac-YVAD-CHO concurrently downregulates inducible nitric oxide synthase gene expression in alveolar macrophages and cyclooxygenase-2 expression in lung tissues, supporting the potential of inhaled caspase inhibitors for treating inflammatory diseases (141). However, owing to its peptide nature, Ac-YVAD-CHO has poor pharmacokinetic properties and is unsuitable for systemic administration, although it serves as a valuable lead compound for inhibitor design.

Targeting caspases involved in the non-canonical pyroptosis pathway is an emerging area of research. Several molecules have been reported to directly bind and inhibit caspase-4/11. For instance, oxidized 1-palmitoyl-2-arachidonyl-phosphatidylcholine binds directly to caspase-4/11, competing with LPS to suppress pyroptosis, IL-1β release and septic shock (142). However, its potential as a drug candidate remains limited. Serine protease inhibitor B1 (SERPINB1) functions as a critical endogenous inhibitor of pyroptosis; its suppression promotes spontaneous activation of caspase-1/4/5/11, while its presence helps minimize inflammatory damage (160). Specifically, SERPINB1 restricts caspase activity by inhibiting CARD oligomerization and enzymatic activation. Choi et al (160) demonstrated that its C-terminal CARD-binding motif suppresses the activation of pro-caspase-1/4/5/10. Furthermore, a study has indicated that SERPINB1 attenuates pathological cardiac hypertrophy and remodeling through its antipyroptotic effects (143). These findings offer a rationale for the development of novel inhibitors that mimic this mechanism.

Additional inhibitors targeting caspases implicated in pyroptosis include Ac-DEVD-CHO, caspase-6b and Z-IETD-FMK.

Ac-DEVD-CHO is a specific caspase-3 inhibitor that suppresses endothelial apoptosis and promotes cerebrovascular spasms. Zhou et al (161) demonstrated that selective caspase-3 inhibition blocks downstream signaling of caspase-8, even when caspase-8 is activated within the caspase cascade. Caspase-6b, an isoform generated by alternative splicing of the melanin-concentrating hormone receptor 2 gene, acts as a natural caspase-6-specific inhibitor; it selectively prevents caspase-6 autoactivation without affecting its preactivated form (162).

Z-IETD-FMK is a broad-spectrum, irreversible caspase inhibitor commonly used in scientific research; it mitigates the interplay between NLRP3 inflammasome activation, pyroptosis and apoptosis, thereby conferring protection against oxidative stress-induced lung injury (163). In a CaCl₂-induced mouse AAA model, intraperitoneal administration of Z-IETD-FMK reduced the incidence of AAAs. This protective effect is attributed to the role of the CD95/CD95L system in recruiting inflammatory cells to injury sites and promoting the phenotypic switching of aortic smooth muscle cells via caspase-8 signaling, processes that are involved in AAA pathogenesis (66). However, owing to its lack of selectivity and associated toxicity, Z-IETD-FMK is unsuitable for clinical use.

Directly inhibiting GSDMD pore formation is considered a highly specific strategy that potentially blocks pathological inflammation while sparing upstream immune surveillance.

These inhibitors function by covalently modifying specific cysteine residues on GSDMD, thereby blocking its oligomerization and pore formation.

Necrosulfonamide (NSA) covalently binds to GSDMD (Cys191/Cys192 in humans/mice) to inhibit oligomerization and pore formation, thereby inhibiting inflammasome-dependent pyroptosis (164). A distinctive feature of NSA is its ability to directly target GSDMD while preserving other cell death pathways, such as those involved in TLR signaling-mediated cell death (165). However, studies indicate that it also impedes inflammasome priming and caspase-1 activation (166-168). Owing to its extreme hydrophobicity and immunocyte toxicity, NSA exhibits unfavorable pharmacokinetics in vivo. To address this limitation, Boersma et al (169) developed a porous nanoparticle-based delivery system for targeted NSA delivery to phagocytes, improving its targeting and safety. Recent evidence has demonstrated that NSA ameliorates Ang II-induced AAA progression in ApoE^−/− mice (65).

Disulfiram (DSF), a Food and Drug Administration-approved drug for alcohol dependence, is metabolized in vivo to S-methyl-N, N-diethylthiocarbamate sulfoxide, which covalently modifies GSDMD at Cys191/Cys192, effectively inhibiting pore formation (170). DSF significantly reduces IL-1β secretion from activated macrophages and attenuates systemic inflammatory responses (171). Liao et al (54) reported that oral DSF administration in Ang II-infused mice reduces AAA incidence while ameliorating collagen deposition and elastin degradation within the aortic wall. The repurposing potential of DSF supports its rapid clinical translation. However, its multitarget nature, including inhibition of aldehyde dehydrogenase, may lead to off-target effects, necessitating careful dose and indication exploration.

Dimethyl fumarate (DMF), a drug used for multiple sclerosis, inhibits pyroptosis by succinylating GSDMD at Cys192, thereby blocking its processing and oligomerization (172). Wang et al (144) reported that DMF restored v-src sarcoma/focal adhesion kinase signaling and collagen transport in aortic VSMCs, and mitigated aortic injury in mouse aneurysm and dissection models. As an approved drug, DMF has a well-established safety profile, facilitating its translational research in AAs.

Itaconate and its derivatives function as potent cysteine modifiers (173). GSDMD Cys77 is the modification target of itaconate. Bambouskova et al (174) demonstrated that endogenous itaconate post-translationally modifies GSDMD to prevent its processing and inhibit pyroptosis. Moreover, itaconate also protects against AAA formation by inducing and activating Nrf2 expression (145).

Additional non-covalent inhibitors and natural compounds suppress pyroptosis by blocking GSDMD pore formation. C202-2729, a small molecule identified through high-throughput screening, non-covalently binds to the GSDMD N-terminal domain, preventing its membrane association and pore formation. C202-2729 has demonstrated efficacy in a sepsis model (146). Caffeic acid, a natural phenolic compound found in coffee and various fruits and vegetables, inhibits lipoxygenase activity and has antioxidant and anti-inflammatory effects (175); it has been shown to bind GSDMD and inhibit its cleavage by caspase-1, thereby reducing pyroptosis (147).

The two strategies targeting GSDMD possess distinct pharmacological profiles and clinical applications (176). Covalent inhibitors irreversibly modify active site cysteines to block pore formation, offering potent and sustained inhibition suitable for acute inflammatory conditions (165,177). However, their broad-spectrum electrophilic nature confers off-target effects, and irreversible modification may increase their immunogenicity (176,178). Non-covalent inhibitors reversibly bind the GSDMD N-terminal domain through conformational complementarity, theoretically providing higher selectivity and lower off-target toxicity, making them preferable for chronic disease intervention (179,180). However, their efficacy is highly dependent on target conformational stability; activation-induced conformational rearrangements may diminish binding affinity, posing challenges for in vivo persistence (181).

From a translational perspective, strategy selection should align with therapeutic windows. Acute progressive stages in AAs warrant covalent inhibitors with targeted delivery to mitigate off-target effects. Chronic settings favor non-covalent inhibitors with optimized pharmacokinetics through medicinal chemistry. Future directions include integrating both approaches, namely, the development of reversible covalent modifiers or non-covalent inhibitors targeting intermediate GSDMD conformations.

Given that the detrimental effects of pyroptosis largely result from the release of IL-1β and IL-18, directly neutralizing these cytokines represents another effective therapeutic strategy.

Anakinra, a recombinant IL-1 receptor antagonist, blocks the signaling of both IL-1α and IL-1β; its efficacy has been demonstrated in macrophage activation syndrome with markedly elevated IL-18 levels, and it is already used to treat autoinflammatory diseases (148). Canakinumab is a humanized monoclonal antibody against IL-1β. In a randomized trial involving postmyocardial infarction patients conducted by Ridker et al (149), canakinumab markedly reduced cardiovascular events, providing strong clinical evidence for the role of IL-1β in human vascular disease. Additionally, the IL-1 inhibitor goflikicept notably reduced recurrence rates in idiopathic recurrent pericarditis and improved patient quality of life (150). These findings indirectly support the therapeutic potential of IL-1-targeting agents in AAs. IL-18 binding protein, a natural inhibitor of IL-18, is also under development as a recombinant therapeutic.

Each of the aforementioned strategies presents distinct advantages and limitations. Upstream inhibitors (for example, MCC950) may broadly suppress multiple downstream effects but risk interfering with essential host defense mechanisms. Downstream cytokine inhibitors (for example, canakinumab) have well-defined targets but are often costly and require parenteral administration. Direct targeting of GSDMD is regarded as a strategy to precisely 'defuse the bomb', blocking pathological lytic cell death and inflammatory release while potentially preserving upstream immune sensing functions. This approach has thus become a major focus in contemporary medicinal chemistry.