The pathological process of fibrosis is initiated by the death of parenchymal cells, which is triggered by mechanical, chemical or pathogenic injury. The subsequent release of damage-associated molecular patterns (DAMPs) activates the local innate immune response, which in turn leads to the activation of NOD-like receptor family pyrin domain containing 3 inflammatory vesicles via the Toll-like receptor 4 (TLR4)/NF-κB signaling axis (25-29). This process involves dual immune recruitment. On one hand, the monocyte-macrophage system polarizes into pro-inflammatory M1 phenotypes (secreting TNF-α, IL-1β and IL-6) and pro-fibrotic M2 phenotypes (secreting TGF-β) (7,30). On the other hand, neutrophils infiltrate the injured area through the CXCL8/CXCR2 chemotaxis pathway (28). Pro-inflammatory cytokines (for example, TNF-α and IL-1β) released by these immune cells, interact with ROS generated by NADPH oxidase to create an oxidative stress microenvironment, thereby sustaining a persistent inflammatory state (31) (Fig. 1).

The hallmark of the proliferative phase is the activation and differentiation of fibroblasts into α-SMA-positive myofibroblasts, stimulated by growth factors such as TGF-β and platelet-derived growth factor (18,32-34). These cells not only exhibit abnormal proliferation and migration characteristics but also overproduce ECM components, including type I and III collagen, and fibronectin (9,35). Concurrently, the upregulation of matrix metalloproteinase inhibitors suppresses matrix metalloproteinase (MMP) activity, resulting in an imbalance in ECM metabolism (9,35,36). At this stage, the TGF-β/Smad signaling pathway fulfills a central regulatory role (37,38). In addition, mechanical stress further sustains myofibroblast activation via the integrin-YAP/TAZ axis, ultimately leading to the formation of early fibrotic lesions characterized by disorganized tissue architecture (39,40) (Fig. 1).

The remodeling phase is characterized by tissue stiffening and destruction of organ structures due to excessive ECM deposition. Disruption of the tissue structure leads to upregulation of hypoxia-inducible factor-1α (HIF-1α), which in turn promotes an increase in pro-fibrotic factors such as connective tissue growth factor and vascular endothelial growth factor. In addition, the mechanical stress generated by the stiffening of the ECM continuously activates myofibroblasts through a positive feedback loop (41-43). Meanwhile, ECM degradation products act as DAMPs that reactivate the TLR/NF-κB signaling pathway, establishing a cycle of 'inflammation-fibrosis-reinflammation' (44,45). This stage is accompanied by abnormal angiogenesis and the replacement of functional parenchymal cells, ultimately leading to irreversible organ dysfunction (46) (Fig. 1).

Under chronic injury conditions, the upregulation of transferrin receptor 1 (TFR1) facilitates the endocytosis of extracellular ferric iron, which is subsequently reduced to ferrous iron (Fe²⁺) by six-transmembrane epithelial antigen of the prostate 3 (54-56). Simultaneously, the autophagy pathway mediated by nuclear receptor coactivator 4 (NCOA4) accelerates the degradation of ferritin, thereby releasing free Fe²⁺ into the cytoplasm (57-59). Subsequently, Fe²⁺ undergoes the Fenton reaction with hydrogen peroxide to release highly reactive hydroxyl radicals. These radicals attack polyunsaturated fatty acids (PUFAs), thereby triggering lipid peroxidation (60,61) (Fig. 2A).

Acyl-CoA synthetase long chain family member 4 (ACSL4) catalyzes the enzymatic reaction that links PUFAs, such as arachidonic acid and linolenic acid, to CoA, producing acyl-CoA variants such as arachidonoyl-CoA and adrenoyl-CoA (62-65). These acyl-CoAs are subsequently esterified into phospholipids by lysophosphatidylcholine acyltransferase 3 (66,67). These phospholipids, enriched with PUFAs, are highly susceptible to peroxidation. A large number of hydroxyl radicals generated via the Fenton reaction attack PUFAs in phospholipids, generating lipid radicals. These radicals react with oxygen to form lipid peroxyl radicals, which further propagate the chain reaction by abstracting hydrogen from adjacent lipid molecules, generating LOOH (62,66,68). Accumulation of lipid peroxides leads to decreased cell membrane fluidity and increased permeability, ultimately triggering leakage of cell contents and cell death (Fig. 2B).

Furthermore, studies have shown that ferroptosis can also occur within the mitochondria (69-71). Mitochondria serve as a critical hub for intracellular iron metabolism, participating in the synthesis of heme, iron-sulfur clusters (Fe-S), and iron storage. Mitochondrial membranes are rich in PUFAs, making them highly susceptible to attack by hydroxyl radicals, which in turn triggers mitochondrial rupture and the release of ROS (72). Ruptured mitochondria release a substantial amount of iron ions, which subsequently catalyze the generation of hydroxyl radicals via the Fenton reaction (73). These newly generated hydroxyl radicals, together with ROS, further assault the mitochondrial membrane, thereby establishing a self-amplifying vicious cycle (65,74,75) (Fig. 2C).

Cells contain multiple antioxidant systems capable of eliminating ROS and lipid peroxides, thereby inhibiting the onset of ferroptosis. These systems are primarily divided into four pathways, with the GPX4-glutathione (GSH) system playing a leading role in eliminating cytotoxic LOOH. Furthermore, the ferroptosis suppressor protein 1 (FSP1)-ubiquinol (CoQH2) system, the dihydroorotate dehydrogenase (DHODH)-CoQH2 system and the GTP cyclase hydrolase 1 (GCH1)-biosynthesis of tetrahydrobiopterin (BH4) system have been shown to inhibit ferroptosis by eliminating ROS within the cell (14,15,54,76).

The GPX4-GSH system serves as the central antioxidant system, dynamically regulating lipid peroxidation via GPX4 and its associated metabolic network (77,78). GPX4 utilizes reduced GSH as an electron donor to reduce cytotoxic LOOH into non-toxic lipid alcohols, thereby halting lipid peroxidation chain reactions and preserving membrane integrity (79,80). The biosynthesis of GSH begins with the uptake of cystine, which is mediated by the solute carrier family 7 member 11 (SLC7A11)-dependent cystine/glutamate antiporter. Intracellular cystine is reduced to cysteine, which is then combined with glutamate by glutamate-cysteine ligase to form γ-glutamylcysteine, and finally converted to GSH via glutathione synthetase (81-83). This metabolic pathway provides essential reducing equivalents for GPX4 activity. In terms of transcription factors, nuclear factor erythroid 2-related factor 2 (Nrf2) enhances antioxidant capacity by binding to antioxidant response elements to upregulate GPX4 and SLC7A11 expression (84-86). Conversely, tumor protein p53 (p53) promotes ferroptosis by suppressing SLC7A11 transcription and depleting GSH (87-89) (Fig. 2D).

The FSP1-CoQH2 pathway, as a GPX4-independent pathway, also plays a crucial role in the antioxidant process. FSP1, an NADH-dependent oxidoreductase, reduces coenzyme Q10 (CoQ10) to CoQH2, which directly neutralizes lipid peroxidation radicals and terminates chain reactions (90-92). FSP1 expression is positively regulated by Nrf2 and negatively regulated by p53 (93-95) (Fig. 2C).

The DHODH-CoQH2 system safeguards mitochondrial redox homeostasis by coupling pyrimidine biosynthesis with electron transport. DHODH is an enzyme localized to the inner mitochondrial membrane; it catalyzes the oxidation of dihydro-lactate to lactate, transferring electrons to CoQ10 in the process and generating CoQH2 (96,97). Under conditions of elevated ROS, this system enhances CoQH2 production to eliminate the burst of ROS and preserve mitochondrial function (98,99) (Fig. 2C).

The GCH1-BH4 axis is another intracellular ferroptosis defense pathway that operates independently of GPX4. GCH1 is the rate-limiting enzyme in the biosynthesis of BH4; it catalyzes the conversion of GTP to dihydropterin triphosphate, thereby initiating the de novo synthesis of BH4 (100,101). BH4 blocks lipid peroxidation chain reactions by directly scavenging lipid radicals, while also promoting the reduction of CoQ10 to CoQH2, thereby enhancing membrane antioxidant capacity and synergistically inhibiting the onset of ferroptosis (102-104) (Fig. 2C).

Since ROS burst is a central driver of ferroptosis, the following sections will systematically discuss strategies for regulating intracellular ferroptosis at three levels: Modulating iron metabolism pathways, modulating lipid peroxidation pathways and enhancing antioxidant systems.

Ferroptosis, a distinct form of cell death, is present throughout the entire pathological process of organ fibrosis and primarily affects parenchymal cells, macrophages and effector cells (105,106). In the early stages of tissue injury, an imbalance in redox homeostasis and disrupted iron metabolism within parenchymal cells are the primary triggers of ferroptosis. As lipid peroxidation occurs, the cell membranes of damaged cells rupture and release DAMPs, which in turn activate the immune response and amplify the inflammatory cascade (106,107). Concurrently, ferroptosis in the inflammatory microenvironment drives polarization of macrophages (108). M1 macrophages secrete pro-inflammatory factors such as TNF-α and IL-1β, further exacerbating the inflammatory response (108). M2 macrophages, on the other hand, secrete TGF-β, which induces fibroblast proliferation, activation and myofibroblast differentiation, ultimately promoting fibrosis (109,110). The following sections will provide a detailed explanation of the specific mechanisms underlying cellular ferroptosis during the fibrotic process.

The fibrotic microenvironment can also exert positive feedback regulation on ferroptosis, a process that occurs primarily during the remodeling phase of fibrosis. During the remodeling phase, excessive deposition of the ECM can cause tissue stiffening and structural damage to organs, leading to tissue hypoxia and subsequent upregulation of HIF-1α (41). Under hypoxic conditions, lactate accumulation upregulates GPX4 expression, endowing activated fibroblasts with ferroptosis resistance and thereby maintaining their profibrotic phenotype. Conversely, reducing lactate levels restores the sensitivity of the cells to ferroptosis and alleviates fibrosis (111). Previous studies have shown that during hypoxia caused by acute ischemia or injury, HIF-1α plays a tissue-protective role by reducing tissue damage and mediating a moderate inflammatory response to prevent abnormal repair (112-114). However, in the context of long-term chronic hypoxia caused by fibrosis, HIF-1α acts as a promoter of fibrosis (106,115). For example, under conditions of chronic hypoxia in the liver and kidneys, HIF-1α promotes iron release by upregulating heme oxygenase-1 (HO-1). This is accompanied by downregulation of GPX4 and SLC7A11, which exacerbates ferroptosis in parenchymal cells, thereby accelerating organ fibrosis (116-118).

Liver fibrosis is a condition in which the liver undergoes abnormal repair and progresses toward sclerosis due to persistent stimulation from various forms of chronic liver injury. The core pathological mechanism involves hepatocyte death and the activation and transdifferentiation of hepatic stellate cells (HSCs). Recent research on ferroptosis has primarily focused on the two types of cells (119,120). Under pathological conditions, iron overload, accumulation of lipid peroxides and depletion of GSH occur in the liver, ultimately triggering ferroptosis in hepatocytes (121,122). For example, in alcohol-related liver disease, TFR1 expression in hepatocytes is markedly upregulated, leading to intracellular iron overload. Furthermore, microRNA (miRNA/miR)-214 can exacerbate the accumulation of lipid oxidation products by upregulating the expression of ACSL4 (123). In metabolic fatty liver disease, elevated levels of TFR1 and reduced levels of ferroportin (FPN) jointly lead to a significant expansion of the labile iron pool (124). At the same time, impaired function of the GPX4-GSH system further exacerbates ferroptosis (125).

Subsequently, liver cells undergoing ferroptosis release DAMPs, which activate Kupffer cells and induce their polarization toward the M1 and M2 phenotypes, leading to the secretion of cytokines such as TGF-β1 and IL-6 (108). These factors further drive the activation of quiescent HSCs into myofibroblasts, which secrete large amounts of ECM, ultimately leading to the development of liver fibrosis (105). In viral hepatitis, miR-222 generated by hepatocytes infected with HBV can inhibit ferroptosis in HSCs by suppressing the expression of the TFR1 in HSCs, thereby promoting their activation (126). In the context of metabolic fatty liver disease, the accumulation of cholesterol within HSCs not only activates the HSCs but also confers resistance to ferroptosis by maintaining high levels of GPX4 expression (127).

Targeting ferroptosis to mitigate fibrosis has become a prominent research focus, with intervention strategies primarily focused on three key areas. Firstly, during the inflammatory stage, the death of parenchymal cells releases DAMPs, which trigger an inflammatory response and drive the fibrotic process (25,26). Inhibiting ferroptosis in these cells by modulating the oxidative or antioxidant systems can halt inflammation at its source, thereby impeding fibrosis progression (106,155).

Secondly, targeting macrophages is also a key focus of current research. Upon receiving inflammatory signals from upstream, macrophages can polarize into two phenotypes: Pro-inflammatory (M1) and pro-fibrotic (M2). These two polarization pathways are key drivers of the fibrotic process (156). Therefore, macrophages can be targeted in two ways to effectively inhibit the fibrotic process. On the one hand, inhibiting ferroptosis in macrophages can prevent their phenotypic transition. On the other hand, inducing ferroptosis can eliminate already polarized macrophages.

Additionally, inducing ferroptosis in activated effector cells, such as HSCs and fibroblasts, is a promising strategy (157). However, this strategy is currently primarily applied to liver fibrosis, with relatively limited research on fibrosis in other organs. The present review further elaborates on anti-fibrotic interventions centered on these three core strategies (Fig. 3).

Studies have revealed that macrophage ferroptosis plays a vital role in the progression of fibrosis. For instance, the expression of ferroptosis-related genes in macrophages is markedly upregulated in both pulmonary fibrosis and systemic sclerosis models (214,215). Moreover, Fei et al (216) performed single-cell RNA sequencing analysis on hypertrophic ligamentum flavum tissues and identified a population of enriched secreted phosphoprotein 1-positive macrophages. These cells are not only involved in the fibrotic process but also exhibit highly activated metabolic pathways associated with ferroptosis.

Ferroptosis drives macrophage polarization into pro-inflammatory (M1) and pro-fibrotic (M2) phenotypes by modulating classical signaling pathways (156,217). Specifically, the formation of the pro-inflammatory M1 phenotype depends on the activation of the cGAS-STING pathway, which exacerbates chronic inflammatory microenvironments through the sustained secretion of inflammatory factors, such as IL-6 and TNF-α (218,219). By contrast, the pro-fibrotic M2 phenotype is synergistically driven by the coordinated activation of the Wnt/β-catenin and JAK-STAT pathways, which subsequently activate fibroblasts via the TGF-β/Smad signaling axis to induce pathological ECM deposition (220,221).

Interventions targeting macrophage ferroptosis can be implemented through two approaches: Suppressing phenotypic polarization and eliminating polarized cells. In terms of inhibiting polarization, targeted blockade of the ATF3-CD36 pathway can activate the Nrf2/GPX4 signaling axis, thereby suppressing macrophage ferroptosis (222,223). Additionally, inhibition of calpain downregulates ACSL4 expression, thereby reducing macrophage susceptibility to ferroptosis (179,215). In polarized cell elimination, current research focuses on pro-inflammatory M1 macrophages. For example, upregulating ACSL4 or inhibiting SLC7A11 expression can specifically increase intracellular lipid peroxidation levels, thereby inducing ferroptosis in M1 macrophages (179,224). Drugs represented by glyceraldehyde 3-O-mono-β-D-glucuronide can selectively eliminate inflammatory macrophages by activating the interferon regulatory factor 1/SLC7A11 signaling pathway, which provides new insights into reversing the progression of fibrosis (224).

During the progression of fibrosis, the inflammatory microenvironment activates tissue-resident effector cells, such as fibroblasts and HSCs, leading to their abnormal proliferation and differentiation into myofibroblasts. Targeting the elimination of activated effector cells has emerged as a pivotal strategy in anti-fibrotic therapy. Currently, this strategy is mainly applied to hepatic fibrosis; it induces ferroptosis of activated HSCs via multiple pathways to alleviate fibrosis (157,225-227).

In addition to enhancing oxidative pathways, inhibiting the antioxidant system is another potential therapeutic strategy (227,250). Studies have shown that various plant extracts can induce ferroptosis in activated HSCs by regulating the GPX4-SLC7A11 system. Ginkgolic acid and fig extract can downregulate GPX4 expression, thereby inducing ferroptosis in HSCs and ultimately improving liver fibrosis (251,252). Ginsenoside Rb1 and ginsenoside Rh2 inhibit SLC7A11, disrupting the cysteine-glutamate reverse transport system and thereby inducing ferroptosis in HSCs (253,254). Furthermore, simvastatin selectively downregulates GPX4 expression in HSCs, achieving anti-fibrotic effects while avoiding hepatotoxicity, demonstrating significant translational medical value (255).

At the transcriptional level, HIF-1α can upregulate the expression of SLC7A11. Sorafenib alleviates liver fibrosis by inhibiting the HIF-1α/SLC7A11 pathway, thereby triggering ferrocytosis in HSCs (225). Conversely, the tumor suppressor protein p53 promotes ferroptosis by inhibiting SLC7A11 transcription, thereby reducing the biosynthesis of GSH (256). For example, vogonolide and artemisinin can activate p53, disrupt the antioxidant system and consequently induce ferroptosis in HSCs (257,258) (Fig. 5B).

Exosome therapy, a novel therapeutic strategy based on extracellular vesicles (exosomes), regulates intercellular communication by delivering bioactive molecules (for example, proteins, miRNAs and mRNAs) and demonstrates potential in treating diverse diseases (259,260). As critical mediators of cell-cell communication, exosomes play a role in fibrotic processes by modulating ferroptosis signaling pathways. In pulmonary fibrosis, ficolin B secreted by alveolar macrophages can promote ferroptosis in pulmonary epithelial cells via the cGAS-STING signaling pathway, thereby exacerbating the progression of pulmonary fibrosis (134). Following HBV infection, miR-222 in exosomes secreted by hepatocytes can suppress ferroptosis in HSCs by inhibiting TFR1 expression (126). Additionally, exosomes derived from renal tubular epithelial cells can activate ATF3, which both inhibits the Nrf2/GPX4 signaling pathway and promotes M2 macrophage polarization, thereby synergistically driving renal fibrosis (261).

Exosome-engineered therapies have opened up a new dimension in fibrosis treatment through a two-way precision regulation strategy of promoting effector cell death and protecting parenchymal cells. In targeting effector cell ferroptosis, mesenchymal stem cell-derived exosomes (MSC-Exos) exhibit specific therapeutic advantages. For instance, human umbilical cord MSC-Exos (hucMSC-Exos) selectively induce HSC ferroptosis by delivering the Beclin 1 (BECN1) protein to suppress GPX4 expression (262). The miR-499a-5p carried by hucMSC-Exos interacts with the transcription factor ETS proto-oncogene 1 to downregulate SLC7A11 expression, thereby disrupting GSH synthesis and facilitating the precise elimination of HSCs (263,264). Similarly, bone marrow MSC-Exos utilize miR-144-3p to silence SLC7A11, thereby effectively inhibiting HSC activation (265).

In parenchymal cytoprotective strategies, engineered exosomes exhibit multifunctional regulatory properties. Spike receptor-binding domain (S-RBD)-modified MSC-Exos inhibit ferroptosis of alveolar epithelial cells by delivering miR-486-5p, while inhibiting SMAD2 signaling and activating the Akt pathway to achieve dual 'antifibrotic and anti-inflammatory' effects (266). Menstrual blood-derived SC-Exos deliver miR-let-7 to suppress Sp3 transcription factor expression and reduce recruitment to histone deacetylase 2, thereby alleviating the inhibition of Nrf2 and ultimately suppressing alveolar epithelial cell ferroptosis (267).

Nanodelivery platforms are drug delivery systems engineered using nanotechnology to enhance drug targeting, stability and therapeutic efficacy (268,269). Nanoparticles, owing to their unique targeting capabilities and controllable drug delivery properties, have emerged as cutting-edge strategies for modulating ferroptosis to alleviate fibrosis.

As innovative carriers for regulating ferroptosis, nanoparticles can protect parenchymal cells or eliminate effector cells through targeted drug delivery, thereby intervening precisely in fibrosis. In parenchymal cell protection, silica nanoparticles loaded with allicin are able to inhibit ferroptosis in cardiac microvascular endothelial cells, significantly improving myocardial fibrosis (270). Selenium-doped silica nanoparticles mitigate myocardial ischemia-reperfusion injury by synergistically alleviating ferroptosis and mitochondrial dysfunction (271).

In the elimination of effector cells, naringenin-loaded nanoparticles selectively remove HSCs by activating autophagy-dependent ferroptosis pathways (272). Carbon nitride-based hybrid nanoparticles induce HSC ferroptosis via downregulation of the HIF-1α/SLC7A11 signaling axis (273). Beyond that, taurine-conjugated lipid nanoparticles and fluorinated peptide-lipid hybrid nanoparticles enhance HSC ferroptosis sensitivity through multi-dimensional regulation (274,275).

While nanoparticles show promise in ferroptosis-targeted therapy, their design requires optimization to improve targeting specificity and reduce off-target toxicity. Future development of multifunctional platforms (for example, stimuli-responsive release and multi-target modulation) may offer more efficient solutions for fibrosis treatment.