In physiological or pathological conditions, neovascularization occurs via vasculogenesis, arteriogenesis, and angiogenesis, each differing in initiation, process, and outcomes. Vasculogenesis refers to the formation of the primary vascular system from mesodermal progenitors during the early stages of embryonic development (Hutchings et al., 2021). Arteriogenesis, driven by hemodynamic forces, involves positive remodeling of pre-existing collateral arterioles—narrow, high-resistance vessels—into larger conductance arteries to compensate for occluded trunks (Helisch and Schaper, 2003; Moraes et al., 2013). Occurring in tissues with pre-existing collaterals, it generates structurally and functionally intact arteries, thereby restoring perfusion to ischemic regions (Wang et al., 2025). Angiogenesis, distinct from the above, involves the migration and proliferation of differentiated endothelial cells to form new capillary networks from existing vessels, encompassing sprouting and intussusceptive subtypes (Semenza, 2014). Sprouting—budding and branching from existing vessels—predominates in forming small vascular plexuses and plays a key role in neovascularization of atherosclerotic plaques (Mitsos et al., 2012). Mechanisms related to neovascularization are illustrated in Figure 1.

Mature capillaries consist of interconnected endothelial cells on the inner side and tightly arranged pericytes, along with an extracellular matrix (ECM), on the outer side, which collectively ensure normal blood circulation (Bergers and Song, 2005). Under physiological conditions, endothelial cells stay in a quiescent state, regulating blood flow to provide nutrients and oxygen for tissues. When triggers like hypoxia exist, the biosynthesis of pro-angiogenic signals increases, thereby activating angiogenesis. This process encompasses stages such as cell proliferation, migration, differentiation, maturation, and lumen formation, each featuring unique endothelial cell functions (Xiang et al., 2024). The basic process is as follows: A variety of angiogenic factors induce vascular dilation and enhance permeability by disrupting intercellular junctions and degrading the ECM. Then, endothelial cells (stalk cells and tip cells) migrate with the ECM serving as a scaffold, facilitating the formation of branches. Eventually, branch fusion, ECM deposition, and pericyte maturation result in the formation of functional new blood vessels. The molecular mechanism of angiogenesis is depicted in Figure 2. Angiogenesis is extensively involved in various physiological and pathological processes such as the AS, tumorigenesis, and tissue repair. Its purpose, regulatory mechanisms, and biological effects vary across different diseases. In brief, angiogenesis is a key driver of tumor progression and represents a pathological uncontrolled proliferation. It aims to provide nutrients and pathways for tumor growth and metastasis, with more complex regulatory mechanisms—tumor cells continuously secrete pro-angiogenic factors while inhibiting anti-angiogenic signals (Liu X. et al., 2025). Angiogenesis during wound healing is a physiological repair process with temporal specificity (Huang et al., 2025). Its goal is to restore blood supply to damaged tissues, support granulation tissue formation and epithelialization, and naturally terminate once wound healing is completed. In AS, angiogenesis is a pathological process primarily occurring within atherosclerotic plaques. While it intends to supply oxygen to hypoxic plaque tissues, it exacerbates plaque instability and rupture risk.

One of the characteristics of vulnerable plaques is a large necrotic core covered by a thin fibrous cap, and the apoptosis of smooth muscle cells (SMCs) and macrophages is the leading cause (Silvestre-Roig et al., 2014; Yin et al., 2024). SMCs are the primary component of the fibrous cap, promoting its thickening by synthesizing matrix components, including collagen and elastin. In the fibrous cap of vulnerable plaques, the apoptosis of SMCs is exacerbated, which reduces the synthesis of matrix proteins—contributing to the formation of a thin fibrous cap and the expansion of the necrotic core (Rotllan et al., 2015). The apoptosis of SMCs is mainly induced by the continuous infiltration of inflammatory cells and protease-mediated changes in the extracellular environment. Additionally, the accumulation of degradation products from the fibrous cap matrix can also induce SMC apoptosis (von Wnuck et al., 2006). Macrophage apoptosis occurs in all stages of AS. In the early stage, it helps reduce plaque burden; however, in the late stage, it impairs plaque stability (Deng et al., 2025). As the disease progresses, a large number of apoptotic macrophages cannot be cleared promptly and accumulate in the plaque, leading to the expansion of the necrotic core and exacerbation of local inflammatory responses (Gautier et al., 2009). In the late stage of AS, the accumulation of free cholesterol (caused by impaired esterification function) and exposure to oxidized cholesterol trigger the unfolded protein response (UPR) (Moore and Tabas, 2011). When the UPR is chronically activated, it induces macrophage apoptosis (Thorp et al., 2009). Furthermore, impairment of macrophage autophagy can also induce excessive apoptosis (Liao et al., 2012).

When efferocytosis is impaired, these protective functions are weakened, thereby contributing to the development of AS (Schrijvers et al., 2005). Efferocytosis refers to the process by which phagocytes clear apoptotic cells, thereby preventing the accumulation of apoptotic cell debris and triggering the reprogramming of phagocytes toward an anti-inflammatory phenotype. In the early stages of AS, dying cells can be rapidly cleared; however, as the disease progresses, efferocytosis becomes defective (Yurdagul, 2021). Since chemotactic and phagocytic signals are usually abundant in advanced plaques, the defect in efferocytosis may result from impaired phagocytic function of macrophages within the lesion, rather than a lack of signals that attract phagocytes.

ECM remodeling constitutes an indispensable step in the formation of vulnerable atherosclerotic plaques. The ECM is a complex reticular scaffold structure composed of collagen fibers, elastin, fibronectin, laminin, and other components. These components are distributed in the basement membrane or interstitial matrix, providing structural support for tissues. In the early stage of AS, the ECM consists of laminin, polymerized type Ⅰ collagen, type Ⅲ collagen, type Ⅳ collagen, a small amount of fibronectin, and elastin fibers. In the late stage of AS, the ECM comprises monomeric type I collagen, fibronectin, fibrinogen, vitronectin, and osteopontin (Yurdagul et al., 2016; Ström et al., 2004). ECM remodeling affects plaque stability primarily by regulating the mechanical strength of the fibrous cap and vascular function. Collagen is the main component of the ECM and is particularly abundant in atherosclerotic plaques, among which type Ⅰ collagen accounts for the highest proportion (approximately 60%) (Holm et al., 2019). Studies on sirtuin 1 (SIRT1) have confirmed the importance of type Ⅰ collagen in plaque stability. Specifically, SIRT1 deacetylates regulatory factor X5 (RFX5), thereby relieving the inhibitory effect of RFX5 on the promoter activity of the collagen α2 chain (Xia et al., 2012). This process increases the expression of type Ⅰ collagen and prevents plaque rupture. Elastin enables blood vessels to adapt to changes in hemodynamics. In the intima of healthy arteries, the content of soluble elastin is high; however, its content decreases in ruptured plaques (Maedeker et al., 2016). A study has verified that inhibiting microRNA-181b can significantly increase elastin expression and stabilize plaques (Di Gregoli et al., 2017).

ECM remodeling is primarily driven by proteases, which represent a key process leading to the reduced mechanical strength of the fibrous cap (Newby, 2012). MMPs are a family of zinc-dependent endopeptidases that can degrade almost all other ECM components (Galis and Khatri, 2002; Li et al., 2014). In vulnerable plaques, in addition to endothelial cells, T cells, macrophages, and SMC can all increase the production of MMPs (Bäck et al., 2010). In AS, the expression of MMP-1 and MMP-8 is significantly upregulated, and their levels are even higher in vulnerable plaques (Molloy et al., 2004; Müller et al., 2014). As collagenases, MMP-1 and MMP-8 can degrade fibrillar collagen (Libby, 2013). In a model of AS with MMP-8 knockout, the plaque area was significantly reduced (Laxton et al., 2009). MMP-2 and MMP-9, as major gelatinases, can degrade a variety of ECM proteins, including collagen (Siefert and Sarkar, 2012). Compared with stable plaques, MMP-2 and MMP-9 in vulnerable plaques are 20 times higher (Guo ZY. et al., 2018). The aforementioned studies indicate that MMPs can regulate plaque stability by remodeling the ECM.

Inflammation plays a crucial role in plaque rupture, and vulnerable plaques exhibit obvious signs of inflammation (Aukrust et al., 2011). During the progression of AS, macrophages, dendritic cells (DCs), and mast cells can affect plaque stability through phenotypic changes and the secretion of cytokines. An extensive cohort study involving 200 human carotid plaques reported an association between CD163⁺ macrophages and the vulnerable plaque phenotype (Bengtsson et al., 2020). In vulnerable plaques, the number of mature dendritic cells is significantly increased. Due to the interaction between dendritic cells and regulatory T cells (Tregs), the migration of DCs and their adhesion to endothelial cells are directly inhibited, thereby impairing plaque stability (Dietel et al., 2013). The frequency of CD117⁺ mast cells is higher in unstable plaques than in stable ones (Joo et al., 2020). Mast cells are activated by binding to IgE attached to the Fc epsilon receptor (FcεR), which leads to the release of cytoplasmic granules containing proinflammatory factors, histamine, and proteases (Kritikou et al., 2019). As a crucial driver of AS, the understanding of inflammatory responses has been progressively deepened. Some researchers have proposed an inflammation-based 3-category classification of atherosclerotic plaques, namely active plaques, dormant plaques, and inactive plaques, analogous to the activity states of volcanoes (Miceli et al., 2025). Even clinically “stable” plaques may exhibit silent yet persistent immunometabolic and thromboinflammatory activities, thereby elevating residual cardiovascular risk. This evolving paradigm endorses immunomodulation as the cornerstone of precision cardiovascular medicine.

Oxidative stress serves as a vital initiating and promoting factor in the development of vulnerable plaques. Oxidative stress refers to a state where the body, upon exposure to various stimuli, generates excessive amounts of highly reactive molecules. When the degree of oxidation exceeds the scavenging capacity of antioxidants, an imbalance occurs between the oxidative and antioxidant systems, leading to tissue damage. Reactive oxygen species (ROS) are a group of oxygen-containing molecules or oxygen atoms with unpaired electrons, including superoxide anions, hydroxyl radicals, and hydrogen peroxide. Among these, superoxide anions are produced in the largest quantity. Studies have shown that the size of AS lesions is closely correlated with the rate of ROS production; inhibiting oxidative stress and ROS generation can suppress the formation of vulnerable plaques (Zhu et al., 2024). In vitro studies have shown that inhibiting ROS production can effectively prevent macrophage apoptosis, thereby reducing the risk of atherosclerotic plaque rupture (Fang et al., 2021). In vivo studies have demonstrated that targeted scavenging of ROS in AS model mice using nanoparticle therapy can significantly inhibit AS progression, providing a novel approach for the diagnosis and treatment of vulnerable plaques (Dai et al., 2022). Key catalytic enzymes involved in oxidative stress include nicotinamide adenine dinucleotide phosphate (NADPH) oxidase, xanthine oxidase, and endothelial nitric oxide synthase (NOS). In human atherosclerotic plaques, superoxide anions and their leading producers—NADPH oxidase and xanthine oxidase—are present in large quantities (Förstermann et al., 2017). Inhibiting nitric oxide (NO) derived from NOS is an effective strategy for reducing foam cell formation and limiting the progression of atherosclerotic plaques (Roy et al., 2021).

Angiogenesis is an inevitable consequence of the progression of plaque burden. Meanwhile, the immature neovessels can induce intraplaque hemorrhage, expand the necrotic core, and exacerbate plaque burden. As mentioned earlier, the hypoxic microenvironment within the large necrotic core of vulnerable plaques can activate angiogenic pathways. While the neovessels formed via angiogenic pathways provide oxygen and nutrients to the plaque, they also facilitate the influx of lipids, leukocytes, and red blood cells. This influx leads to the expansion of the necrotic core and intraplaque hemorrhage, which increases the plaque burden and impairs plaque stability (Moreno et al., 2012; Kwon et al., 2015). Intraplaque hemorrhage is a critical determinant of plaque instability and is closely associated with the occurrence of future ischemic events (Faghani et al., 2025). It accelerates plaque progression by increasing the size of the necrotic core and the overall plaque volume, resulting from the accumulation of lipid-rich red blood cell membranes within the plaque (Zhao et al., 2022). Such hemorrhage originates from the rupture and leakage of structurally immature neovessels. Studies have found that 80% of intraplaque neovessels exhibit poor integrity, with loose endothelial junctions, insufficient pericyte coverage, and a lack of SMC support, making them prone to leakage (Sluimer et al., 2009). In human carotid plaques, macrophages are typically found surrounding hemorrhagic areas. These macrophages participate in the phagocytosis of red blood cells and iron, and can also secrete VEGF, further enhancing the permeability of neovessels (Guo L. et al., 2018). During the leakage process, blood cells infiltrate into the plaque core, and their cell membranes release free cholesterol. This free cholesterol promotes the formation of cholesterol crystals, which can erode the fibrous cap and protrude into the arterial lumen, potentially triggering embolism or thrombosis (Abela et al., 2009). Undoubtedly, the aforementioned blood cells, cholesterol, and other substances extravasated through immature neovessels further increase the burden on phagocytes within the plaque, exacerbate macrophage apoptosis, and augment plaque burden.

Both angiogenesis and plaque tissue matrix remodeling rely on the regulation of MMPs, making MMPs the mediator that bridges these processes. As previously mentioned, a variety of MMPs are involved in the formation of vulnerable plaques; they also act as driving factors for angiogenesis, while neovessels can, in turn, accelerate vascular remodeling by affecting MMPs. A range of angiogenic factors can induce the activation of MMPs. Stimulated by VEGF and bFGF, endothelial cells secrete vesicles containing MMP-2 and MMP-9, thereby regulating tip cell migration and lumen formation (Taraboletti et al., 2002). After inhibiting the expression of MMP-2, the proliferation and migration capabilities of endothelial cells decrease, and the growth of the new capillary network is suppressed (Cheng et al., 2007). MMP-1 can promote the expression of VEGFR2 and the proliferation of endothelial cells; it exerts this effect by stimulating serine/threonine protein kinases and activating the transcription factor nuclear factor-κB (Mazor et al., 2013). MMP-7 can regulate the VEGF pathway and indirectly promote angiogenesis by degrading soluble VEGFR1 (Libby et al., 2011; Ito et al., 2009). The aforementioned MMPs involved in angiogenesis can accelerate the degradation of the fibrous cap, thereby exacerbating plaque instability or rupture (Leclercq et al., 2007). At the same time, cells that extravasate through immature neovessels can release TNF-α, IL-8, and other cytokines, which further stimulate endothelial cells to produce MMP-2, MMP-8, and MMP-9, thereby regulating the processes of angiogenesis and vascular remodeling (Li et al., 2005; Akbari et al., 2023).

Inflammation and angiogenesis in vulnerable plaques are interconnected. On the one hand, various inflammatory cells infiltrate the plaque and secrete inflammatory and growth factors, which promote the formation of new blood vessels. On the other hand, neovessels can facilitate the infiltration of blood components, such as leukocytes, into the plaque, thereby exacerbating the inflammatory response. This positive feedback loop impairs plaque stability, intensifies inflammation, and leads to the formation of more defective neovessels. Inflammation increases the local metabolic rate, characterized by elevated glucose uptake and oxygen consumption, which further aggravates local hypoxia. Hypoxia serves as a crucial stimulus for angiogenesis. A hallmark of plaques is the infiltration of inflammatory cells, such as macrophages. These cells can secrete VEGF and bFGF, actively participating in angiogenesis. Additionally, inflammatory cells can secrete classical inflammatory cytokines with pro-angiogenic functions, including IL-6, TNF-α, MCP-1, and IL-8. Beyond the aforementioned classical pro-angiogenic factors, inflammatory cells can secrete specific factors that promote angiogenesis. For instance, the macrophage-derived peptide PR39 can inhibit the ubiquitin-proteasome-dependent degradation of the HIF-1α, thereby accelerating the formation of vascular structures (Li et al., 2000). Numerous pro-angiogenic factors, in turn, can exacerbate the inflammatory response. VEGF can stimulate the migration of monocytes into the plaque (Chen et al., 2025). In hypercholesterolemic mice, infusion of low-dose VEGF increases the recruitment of monocytes from the bone marrow to the bloodstream, ultimately exacerbating the plaque burden (Celletti et al., 2001). bFGF can synergize with inflammatory factors, including TNF-α, to enhance the expression of adhesion molecules on activated endothelial cells, thereby promoting leukocyte recruitment (Zittermann and Issekutz, 2006). Intraplaque hemorrhage secondary to neovessel formation can activate platelets and promote the retention of leukocytes on the vascular endothelium, thereby further exacerbating inflammation (May et al., 2008).

Recent studies have highlighted the role of neutrophil extracellular traps (NETs) in neovascularization. In mouse models of choroidal neovascularization, the levels of NETs are elevated, and inhibition of NETs can significantly reduce the formation of new blood vessels. In vitro, NETs are capable of stimulating the proliferation, migration and tube formation of endothelial cells. The underlying mechanism involves the activation of endothelial cells via the TLR4/HIF-1α pathway, which in turn leads to the upregulation of MMP-9 and IL-1β—both of which are key mediators of angiogenesis (Zeng et al., 2023). The involvement of NETs in angiogenesis has been verified in various diseases such as tumors and pulmonary arterial hypertension (Yang et al., 2023; Aldabbous et al., 2016).

Oxidative stress and angiogenesis together accelerate the formation of vulnerable plaques. The interaction between oxidative stress and angiogenesis centers on the VEGF signaling pathway. ROS generated during arterial endothelial injury upregulate VEGF expression; for example, hydrogen peroxide can induce VEGF expression in both SMCs and endothelial cells, thereby promoting angiogenesis (Bretón-Romero and Lamas, 2014). Additionally, after endothelial cells are stimulated by ox-LDL, they produce ROS that activate angiogenesis-related signaling pathways, including those involving p38 mitogen-activated protein kinase (p38MAPK) and extracellular regulated protein kinases 1/2 (ERK1/2). Antioxidants can block the activation of these aforementioned signaling pathways, thereby inhibiting angiogenesis (Dandapat et al., 2007; Camaré et al., 2015; Kim and Byzova, 2014). ROS also affect the activation, aggregation, and phosphorylation of VEGFR2 (Colavitti et al., 2002). Furthermore, oxidized products, such as oxidized phospholipids have been shown to interact with VEGFR2 and activate angiogenesis (Birukova et al., 2012). In summary, oxidative stress promotes angiogenesis by acting on upstream and downstream components of the VEGF/VEGFR2 signaling pathway. Conversely, intraplaque angiogenesis can further activate oxidative stress. Microhemorrhage caused by intraplaque neovessels is directly associated with oxidative stress (Yunoki et al., 2009; Nagy et al., 2010). Hemorrhage from immature neovessels releases hemoglobin and iron; these substances can promote the production of free radicals and ROS (e.g., via the Fenton reaction), which in turn inactivate nitric oxide and drive lipid peroxidation (Camaré et al., 2017). Additionally, VEGF can activate NADPH oxidase in endothelial cells, thereby promoting ROS generation (Huang and Nan, 2019).