Mitochondria are the central organelles for cellular energy metabolism. They are responsible for the tricarboxylic acid (TCA) cycle, oxidative phosphorylation (OXPHOS), and the generation of ROS. Consequently, mitochondrial metabolism plays a pivotal role in the regulation of ferroptosis. Studies have shown that cysteine deprivation can lead to mitochondrial membrane hyperpolarization and accumulation of lipid peroxides, thereby inducing ferroptosis (37, 38). Inhibiting the TCA cycle or OXPHOS can alleviate this process (38). Pyruvate is a key substrate for the TCA cycle. It is converted to acetyl-CoA via pyruvate dehydrogenase (PDH), thereby driving the TCA cycle. In pancreatic ductal adenocarcinoma, pyruvate dehydrogenase kinase 4 (PDK4) enhances resistance to ferroptosis by inhibiting pyruvate oxidation.This reduces acetyl-CoA production and fatty acid synthesis (39). Furthermore, under hypoxic conditions, lactate dehydrogenase (LDH) promotes the conversion of pyruvate to lactate. This process diminishes the support of glucose aerobic oxidation for ferroptosis. Lactate can also activate the sterol regulatory element-binding transcription factor 1 (SREBP1)/stearoyl-CoA desaturase 1 (SCD1) axis. This stimulates the generation of monounsaturated fatty acids (MUFAs). MUFAs replace polyunsaturated fatty acids in the cell membrane, alleviating lipid peroxidation and thereby resisting ferroptosis (40).

In glucose metabolism, key glycolytic enzymes HK2 and PKM2 finely regulate ferroptosis sensitivity by modulating metabolic flux and redox homeostasis. HK2 not only drives glycolysis but also influences ROS production and mitochondrial function through its localization at mitochondria, thereby participating in the regulation of cell death processes (41, 42). PKM2 exhibits a dual role: its tetrameric form promotes glycolytic flux and may support antioxidant synthesis, whereas the dimeric form can translocate to the nucleus and modulate stress-responsive genes; studies have confirmed that PKM2 can directly regulate ferroptosis by affecting mitochondrial homeostasis (43, 44), highlighting its potential as a metabolic checkpoint in cancer therapy (45).

The pentose phosphate pathway (PPP) is a crucial branch of glucose metabolism, generating NADPH and pentose phosphates (Figure 2). As an electron donor, NADPH reduces oxidized glutathione (GSSG) to reduced glutathione (GSH) under the action of glutathione reductase (GR). GSH is an essential cofactor for GPX4, which inhibits ferroptosis by eliminating lipid peroxides (46). In tumor cells, even under ample oxygen supply, there is a preference for generating energy via glycolysis, a phenomenon known as the Warburg effect. The PPP not only provides precursors for nucleotide synthesis but also produces substantial NADPH, supporting rapid tumor cell proliferation (47). Additionally, NADPH is involved in GSH synthesis, aiding tumor cells in resisting oxidative damage (48). In pancreatic cancer, upstream stimulatory factor 2 (USF2) binds to the promoter of pyruvate kinase M2 (PKM2), upregulating PKM2 expression. This subsequently increases intracellular GSH levels and GPX4 activity, thereby inhibiting ferroptosis (43). Recent studies have also indicated that mucosal melanoma cells can acquire resistance to ferroptosis by taking up lactate. This enhances the PPP pathway, upregulates GPX4 and FSP1, and increases levels of NADH, NADPH, and GSH (49).

Ferroptosis, an iron-dependent form of programmed cell death, is closely associated with various amino acid metabolic processes (Figure 3). Among these, glutamine metabolism plays a central role in the regulation of ferroptosis. After entering the cell via the transporter SLC1A5, glutamine is converted to glutamate by glutaminase (GLS). Glutamate can then be further catalyzed by glutamate dehydrogenase 1 (GDH1) to generate A-ketoglutarate (A-KG), thereby driving glutaminolysis. Intracellular glutamate accumulation inhibits system Xc– activity, reducing cystine uptake, which leads to impaired glutathione (GSH) synthesis and diminished antioxidant capacity. Elevated endogenous glutamate levels cause an increase in intracellular Ca²⁺ concentration, subsequently activating the adenylate cyclase (ADCY10)-protein kinase A (PKA) signaling axis. This phosphorylates and inhibits GFPT1, thereby weakening YAP protein stability and the transcription of its downstream target gene FTH1, ultimately promoting ferritinophagy, lipid peroxidation, and enhanced ferroptosis susceptibility (50). Furthermore, A-KG can be further metabolized to 2-hydroxyglutarate, inducing mitochondrial ROS accumulation and activating the p53 pathway, collectively promoting ferroptosis (51). Notably, GLS2, an isoenzyme of GLS, exhibits dual roles in liver cancer by suppressing tumors and promoting ferroptosis (52). Therefore, targeting key nodes in glutamine metabolism, such as SLC1A5, GLS, or GDH1, has become a significant strategy for regulating ferroptosis (53, 54).

Besides glutamine, arginine metabolism also plays a complex role in ferroptosis regulation. On one hand, arginine can be metabolized via the urea cycle to generate fumarate. As an electrophile, fumarate covalently binds to glutathione, leading to intracellular GSH depletion and thereby promoting erastin-induced ferroptosis (55). On the other hand, the metabolic pathway mediated by argininosuccinate synthase 1 (ASS1) can antagonize ferroptosis. ASS1 promotes the reductive carboxylation of glutamine, reducing its entry into the TCA cycle for oxidative metabolism, consequently lowering the accumulation of mitochondria-derived lipid ROS. Simultaneously, ASS1 participates in the arginine-succinate metabolic axis, activating the mTORC1–SREBP1–SCD5 signaling pathway, which promotes monounsaturated fatty acid synthesis, enhances the anti-peroxidation capacity of the cell membrane, and ultimately confers resistance to ferroptosis in tumor cells (56). Thus, the role of amino acid metabolism in regulating ferroptosis is dependent on the metabolic context and molecular mechanisms. Different metabolic enzymes and microenvironments can lead to divergent cell fates, providing a theoretical basis and potential targets for intervening in ferroptosis via specific amino acid pathways.

Glutamine is imported via SLC1A5 and metabolized to glutamate and A-ketoglutarate through GLS/GDH1. This process promotes ferroptosis through multiple mechanisms: inhibiting system Xc⁻, depleting GSH, activating p53, and suppressing the GFPT1-YAP-FTH1 signaling axis via the ADCY10/PKA pathway. In contrast, the metabolic pathway mediated by ASS1 confers resistance to ferroptosis by promoting the reductive carboxylation of glutamine and activating the mTORC1-SREBP1-SCD5 axis to drive monounsaturated fatty acid synthesis. However, arginine can also be metabolized via the urea cycle to generate fumarate, which acts as an electrophile to deplete glutathione, thereby promoting ferroptosis.

Despite distinct execution mechanisms, ferroptosis and apoptosis interact at multiple levels. They share upstream stress signals; for instance, the p53 tumor suppressor can transcriptionally repress SLC7A11 to promote ferroptosis while also initiating classical apoptosis (57). Mitochondria serve as a key hub: outer membrane permeabilization and collapse of membrane potential are hallmark apoptotic events that also cause ROS bursts, exacerbating ferroptosis (58). In myocardial ischemia/reperfusion injury, Tanshinone IIA concurrently inhibits both caspase-3-mediated apoptosis and GPX4 degradation-triggered ferroptosis by targeting VDAC1, highlighting the mitochondrial channel’s coordinating role (59).

Their relationship is dynamic and can be mutually restrictive. Activated caspases can suppress ferroptosis by cleaving glutaminase, thereby limiting glutamate availability essential for System Xc⁻ function (60). Therapeutically, co-induction can overcome resistance. In breast cancer, DNAJC12 concurrently suppresses doxorubicin-induced apoptosis and ferroptosis via the HSP70-AKT axis, while combining inhibitors for both pathways fully reverses chemoresistance (61). Similarly, in oral squamous cell carcinoma, melatonin synergizes with erastin to induce both apoptosis and ferroptosis via ROS promotion (62). Several studies indicate that the efficacy of traditional chemotherapeutics partly stems from their ability to activate both pathways (63).

Ferroptosis and necroptosis often act synergistically in acute tissue injury, amplifying inflammation and damage. Metabolic crosstalk is central: activation of the key necroptotic kinases RIPK1/RIPK3 can cause energy crisis and ROS bursts, creating conditions favorable for ferroptosis (64). Conversely, severe oxidative stress from ferroptosis can upstream activate the RIPK3-MLKL pathway (65).

In acute kidney injury, a “wave-like death” model proposes that initial ferroptosis acts as an “initiator,” releasing DAMPs that subsequently activate necroptosis in neighboring cells as an “amplifier” to propagate death signals, a process termed “necroinflammation” (66). In acute respiratory distress syndrome (67), intestinal ischemia/reperfusion injury (68), and calcium oxalate crystal-induced kidney injury (69), ferroptosis, necroptosis, and pyroptosis frequently co-occur, forming a highly interconnected, functionally redundant network. Inhibiting one pathway often leads to “compensatory” death via others, presenting a therapeutic challenge and pointing to the need for multi-pathway targeting.

Ferroptosis and pyroptosis are closely linked through inflammation. Lipid peroxides accumulated during ferroptosis act as DAMPs, activating the NLRP3 inflammasome (70). Activated caspases-1/4/5/11 then cleave gasdermin D (GSDMD); its N-terminal fragment forms pores in the plasma membrane, executing pyroptosis and releasing pro-inflammatory cytokines like IL-1β, thereby coupling oxidative damage to intense inflammation (71).

In epilepsy models, ferroptosis and pyroptosis jointly contribute to disease progression via oxidative stress and neuroinflammation, with intersecting pathways (72). This connection is crucial in cancer immunotherapy. CD8⁺ T cells can not only induce apoptosis via the granzyme-perforin pathway but also trigger tumor cell ferroptosis by releasing IFN-γ, which downregulates SLC7A11 (73). Furthermore, inflammatory signals from ferroptotic cells can remodel the tumor microenvironment, converting “cold” tumors into “hot” ones, thereby enhancing the efficacy of immune checkpoint inhibitors (74, 75). In glioblastoma, ferroptosis, pyroptosis, and autophagy form a complex regulatory network influencing tumor progression and treatment response (76).

As two novel metal-dependent regulated cell death pathways, ferroptosis (iron-driven lipid peroxidation) and cuproptosis (excess copper leading to aggregation of lipoylated TCA cycle proteins and proteotoxic stress) have distinct core mechanisms (77, 78). Despite this, significant metabolic interactions exist.

In liver cancer therapy, ferroptosis inducers were found to enhance cuproptosis induced by copper ionophores. The mechanism involves inhibition of mitochondrial protease-mediated degradation of FDX1 (a key cuproptosis protein) and reduced synthesis of glutathione (a major copper chelator) via cystine uptake blockade, collectively promoting copper-dependent lipoylated protein aggregation (79). In cardiovascular diseases, copper can directly promote ferroptosis by inhibiting GPX4. Both ferroptosis and cuproptosis can contribute to PANoptosis—a coordinated cell death involving pyroptosis, apoptosis, and necroptosis—through ROS generation, exacerbating chemotherapy-related cardiotoxicity (80). Epigenetic regulation, such as DNA methylation and non-coding RNAs, has also been identified to modulate key proteins in both ferroptosis and cuproptosis, offering new perspectives for co-targeting (81, 82). Notably, copper metabolism also influences immune cell function, such as T cell activation and macrophage polarization, and may synergize with ferroptosis to enhance immunogenic cell death and reshape the tumor immune microenvironment, thereby providing a rationale for combined immunomodulatory strategies.

Nuclear factor erythroid 2-related factor 2 (NRF2) is a central regulator of the antioxidant response and plays a core role in inhibiting ferroptosis (91). Under oxidative stress, NRF2 escapes KEAP1-mediated degradation, translocates to the nucleus, and coordinates cellular redox homeostasis, iron metabolism, and lipid peroxidation defense. It does this by activating genes containing antioxidant response elements (AREs), thereby constructing a multi-layered ferroptosis resistance network.

Mechanistically, NRF2 orchestrates key facets of ferroptosis through coordinated regulation of multiple pathways. In glutathione metabolism, it directly upregulates SLC7A11 to enhance cystine uptake for GSH synthesis (92) and activates genes for GSH de novo synthesis, supporting GPX4 function (91). Regarding iron metabolism, NRF2 induces expression of FTH1 and heme oxygenase-1, promoting iron storage and detoxification to limit the labile iron pool and reduce Fenton reactions (3, 93). Additionally, NRF2 modulates lipid metabolic enzymes like ACSL4 and SCD1, altering membrane phospholipid composition and reducing susceptibility to lipid peroxidatio (88, 94–96).

Pathophysiologically, aberrant NRF2 pathway activation is closely linked to ferroptosis resistance in cancer cells. In models of ovarian and liver cancer, NRF2 signaling enhances resistance to inducers like erastin and sorafenib (97, 98), whereas its loss or inhibition increases sensitivity. Notably, oncogenic mutations such as KRASG12D and BRAFV619E can directly induce NRF2 expression, building a robust antioxidant defense system, which offers new insights into tumor drug resistance.

NRF2 activity itself is subject to precise, multi-layered control. p62 accumulation competitively binds KEAP1, promoting NRF2 stabilization and nuclear translocation. DPP9 cooperates with p62 to enhance NRF2 stability, mediating sorafenib resistance. mTORC1 and disulfiram/copper (DSF/Cu) activate p62 phosphorylation, strengthening KEAP1 inhibition and leading to NRF2 accumulation and ferroptosis resistance (99, 100). Conversely, the transcription factor BACH1 antagonizes NRF2 by suppressing genes like SLC7A11 and FTH1, thereby promoting ferroptosis (87, 101) (Figure 4).

Therapeutically, targeting the NRF2 pathway has become a key strategy for reversing tumor ferroptosis resistance. Inhibitors like ML385 significantly enhance the anti-tumor efficacy of ferroptosis inducers combined with chemotherapy, radiotherapy, or immunotherapy (102, 103). However, a major challenge remains achieving tumor-specific targeting due to NRF2’s vital protective role in normal tissues (Figure 4).

In summary, NRF2 integrates diverse metabolic and stress signals to establish a sophisticated transcriptional and metabolic defense network. It serves as a critical regulatory hub for ferroptosis susceptibility. A deeper understanding of its regulatory mechanisms and tumor-specific activation will provide a crucial foundation for developing novel ferroptosis-targeted therapies.

As a paramount tumor suppressor, p53 exhibits a complex and sophisticated “double-edged sword” role in regulating ferroptosis, with its function being highly dependent on cell type, stress signal intensity, and microenvironmental context (85, 104).

The most established pro-ferroptotic function of p53 involves the transcriptional repression of cystine uptake. P53 directly binds to the SLC7A11 promoter, repressing its transcription and consequently limiting extracellular cystine import (4). This action impedes intracellular glutathione (GSH) synthesis, thereby weakening GPX4 activity and sensitizing cells to lipid peroxidation accumulation and ferroptosis. Notably, transcription-independent functions of p53 also contribute to ferroptosis regulation. p53 can facilitate the nuclear translocation of the deubiquitinase USP7, leading to reduced levels of histone H2B monoubiquitination at lysine 120 (H2Bub1), an epigenetic mark crucial for SLC7A11 transcriptional activation. This decrease in H2Bub1 further reinforces SLC7A11 suppression, deepening the commitment to ferroptosis (105). Beyond SLC7A11, p53 can indirectly promote lipid peroxidation by activating other downstream targets. For instance, p53 transcriptionally upregulates spermidine/spermine N1-acetyltransferase 1 (SAT1), driving polyamine metabolism, a process associated with increased arachidonate 15-lipoxygenase (ALOX15) activity, collectively fostering lipid hydroperoxide generation (106). Simultaneously, p53-induced expression of glutaminase 2 (GLS2) enhances glutaminolysis, potentially influencing ferroptosis sensitivity by altering the intracellular redox state or providing precursors for lipid synthesis (85, 107). These parallel pro-death pathways collectively enable p53 to effectively induce ferroptosis, fulfilling its tumor-suppressive function.

Conversely, under specific cellular contexts or stress levels, p53 demonstrates clear anti-ferroptotic effects. This protective role is partly mediated via transcription-independent mechanisms. Nuclear p53 can interact with and sequester dipeptidyl peptidase 4 (DPP4), preventing its translocation to the plasma membrane. As membrane-localized DPP4 is required for NADPH oxidase-dependent lipid peroxidation, p53-mediated nuclear sequestration effectively inhibits ferroptosis initiation (19). Alternatively, classical transcription-dependent functions also contribute to the protective mechanism. P53 can induce the expression of cyclin-dependent kinase inhibitor 1A (CDKN1A/p21), leading to cell cycle arrest. This proliferative halt may indirectly reduce basal levels of cellular anabolism and reactive oxygen species (ROS) generation, thereby enhancing resistance to ferroptosis (108). Furthermore, p53 has been reported to suppress mitophagy. In certain contexts, p53 binds to and sequesters Parkin in the cytoplasm, hindering its translocation to damaged mitochondria and thus inhibiting their clearance. Since damaged mitochondria are a significant source of ROS, preventing their autophagic removal may help maintain mitochondrial fitness and reduce pro-ferroptotic oxidative stress signals (109) (Figure 5).

As core effectors of the Hippo signaling pathway, YAP and TAZ play a critical yet complex role in regulating ferroptosis. Notably, their function exhibits a distinct context-dependent duality, influenced by factors such as cell type, tumor microenvironment, and oncogenic background.

In most scenarios, YAP/TAZ activation promotes ferroptosis through multiple mechanisms. Specifically, YAP can increase intracellular iron concentration by transcriptionally upregulating transferrin receptor expression (110). Concurrently, YAP activation promotes lipid peroxidation by modulating arachidonate lipoxygenase 3 (111). More significantly, YAP/TAZ enhance cellular sensitivity to ferroptosis by upregulating ACSL4 expression (112). Consistently, studies in renal fibrosis also link YAP activation to elevated ACSL4 levels and increased ferroptosis susceptibility (113). TAZ, on the other hand, indirectly regulates the expression of ROS-producing NADPH oxidases via the angiopoietin-like 4 (ANGPTL4)-NOX2 axis and the epithelial membrane protein 1 (EMP1)-NOX4 axis, further promoting lipid peroxidation and ferroptosis (114). Additionally, the YAP/p53 axis has been demonstrated as essential for cytoglobin-induced lipid peroxidation and ferroptosis (115) (Figure 6).

Conversely, in certain specific cancer types, YAP/TAZ exhibit an opposing, ferroptosis-suppressive role. For instance, in sorafenib-resistant hepatocellular carcinoma, YAP/TAZ induce SLC7A11 expression in a TEAD-dependent manner, helping cells resist ferroptosis (86) (Figure 6).

For cancers where YAP/TAZ inhibit ferroptosis, various targeted inhibitors have shown promising therapeutic potential. Mevalonate pathway inhibitors, such as zoledronic acid, suppress YAP/TAZ nuclear translocation by maintaining them in a phosphorylated state (116). Verteporfin inhibits the transcriptional activity of YAP/TAZ by disrupting their interaction with TEAD. The BET protein inhibitor JQ-1 suppresses YAP-mediated transcription by targeting BRD4, a key component of the YAP/TAZ-TEAD complex (116). In esophageal squamous cell carcinoma, arsenic-iron oxide conjugate nanocomposites effectively degrade YAP, significantly sensitizing tumor cells to radiotherapy and chemotherapy (117). These inhibitors disrupt the YAP/TAZ signaling pathway through distinct mechanisms, successfully inducing ferroptosis in specific cancer types and offering novel strategies to overcome therapy resistance (Figure 6).

YAP signaling can suppress ferroptosis by targeting the ferritin light chain and SLC7A11. Inhibition of YAP/TAZ activity can be achieved through several approaches: mevalonate pathway inhibitors (e.g., zoledronic acid) maintain YAP/TAZ in a phosphorylated state; verteporfin inhibits the interaction between YAP/TAZ and TEAD, thereby preventing DNA binding and transcription; and JQ1 inhibits BRD4, a component of the YAP/TAZ-TEAD complex. Conversely, YAP activation promotes ferroptosis by upregulating the expression of ACSL4 and TFRC. Additional regulators of YAP-mediated ferroptosis include NOX2, NOX4, ALOXE3, and cytoglobin.

The role of CD8⁺ T cells in tumor immunotherapy extends beyond direct tumor cell killing to include the regulation of a novel cell death modality—ferroptosis. Research shows that CD8⁺ T cells activated by immunotherapy can downregulate the expression of the key cystine transporter SLC7A11 (a component of system Xc⁻) in tumor cells by releasing interferon-gamma (IFN-γ). This limits cystine uptake, leading to lipid peroxide accumulation and inducing ferroptosis in tumor cells (136). Radiotherapy and immunotherapy can act synergistically to enhance tumor lipid oxidation and ferroptosis by suppressing SLC7A11, improving tumor control outcomes (137). Furthermore, IFN-γ from CD8⁺ T cells cooperates with fatty acids like arachidonic acid in the TME, driving tumor cell ferroptosis by upregulating the key lipid metabolism enzyme ACSL4 (138).

Conversely, the TME can actively induce ferroptosis in CD8⁺ T cells, impairing their anti-tumor function. For instance, cholesterol in tumors can cause CD8⁺ T cell exhaustion by triggering endoplasmic reticulum stress and activating XBP1 (139). More directly, the fatty acid transporter CD36 on the surface of CD8⁺ T cells mediates excessive uptake of fatty acids from the TME, directly triggering intracellular lipid peroxidation and ferroptosis, compromising their cytokine production capability and anti-tumor function (140). Similarly, prostaglandin E2 (PGE2) can promote ferroptosis in tumor-infiltrating CD8⁺ T cells by interfering with the IL-2 signaling pathway (141). Within T cells, loss of the DEPDC5 gene leads to overactivation of mTORC1 signaling, upregulating the expression of ATF4 and xanthine oxidase, thereby spontaneously inducing ferroptosis in CD8⁺ T cells (142) (Figure 7).

Based on these mechanisms, enhancing the persistence of CD8⁺ T cells and protecting them from ferroptosis has emerged as a strategy to improve immunotherapy efficacy. Studies have found that the Tc9 subset (IL-9-secreting CD8⁺ T cells) reduces its own lipid peroxidation and ferroptosis by enhancing fatty acid oxidation via the IL-9/STAT3 signaling axis, allowing it to survive longer in the TME and exhibit stronger anti-tumor capabilities (143). Therapeutically, targeting HnRNP L can downregulate PD-L1 expression on tumor cells while promoting CD8⁺ T cell-mediated tumor ferroptosis, synergizing with anti-PD-1 therapy (144). In liver cancer, restoring MAT1A expression has also been shown to enhance CD8⁺ T cell activity and induce tumor cell ferroptosis (145). Collectively, these findings highlight the central role of ferroptosis in T cell immunity, providing a theoretical foundation for combining immune checkpoint blockade with ferroptosis-inducing strategies.

The occurrence of ferroptosis itself constitutes a potent cellular stress signal, actively reprogramming the functional phenotype of macrophages towards a pro-inflammatory, anti-tumor M1 direction. The molecular basis of this process lies in the core biochemical reaction of ferroptosis—lipid peroxidation. Studies confirm that in nasopharyngeal carcinoma models, the enzyme ACSL4 promotes ferroptosis by driving lipid peroxidation, which directly leads to the repolarization of tumor-associated macrophages (TAM) from a pro-tumor M2 phenotype to an anti-tumor M1 phenotype (146). Similarly, in the acidic microenvironment of breast cancer, ferroptosis activated by the ZFAND5/SLC3A2 signaling axis not only directly kills tumor cells but also acts as a polarization signal, inducing macrophage transformation into the M1 phenotype, forming a “dual-strike” mechanism against the tumor (147). More interestingly, the susceptibility of macrophages themselves to ferroptosis determines their functional state. For instance, classically activated M1 macrophages generate large amounts of nitric oxide (NO•) through high expression of inducible nitric oxide synthase(iNOS). NO• can competitively inhibit the activity of the 15-lipoxygenase/PEBP1 complex, thereby blocking the generation of the key pro-ferroptotic phospholipid HpETE-PE. This intrinsic anti-ferroptosis mechanism allows M1 macrophages to survive in the pro-inflammatory environment and continuously perform immune surveillance functions, constituting their unique cellular identity (148, 149). Therefore, the ferroptosis pathway is both a result and a significant driver of macrophage polarization.

In the tumor microenvironment, through direct and indirect intercellular communication, macrophages become key arbiters determining whether tumor cells undergo ferroptosis, with a distinctly bidirectional regulatory role. On one hand, macrophages can act as “promoters” of ferroptosis. In liver and lung cancer models, the absence of the cystine transporter xCT in macrophages triggers their own ferroptosis and inhibits M2 polarization. This change remodels the immune microenvironment, recruiting and activating more CD8⁺ T cells. The activated T cells, by secreting IFN-γ, subsequently downregulate System Xc⁻ activity on tumor cells, making them more sensitive to ferroptosis, thereby amplifying the cascade of anti-tumor immunity (150, 151). On the other hand, macrophages may also transform into “resistors” of ferroptosis, leading to therapeutic failure. Recent studies have found that when ferroptosis inducers are used, macrophages themselves undergo phospholipid peroxidation. This lipid peroxidation impairs their phagocytic function mediated by TLR2, preventing them from effectively clearing tumor cells that have undergone ferroptosis, ultimately resulting in tumor resistance to ferroptosis therapy (152). Furthermore, in some cases, low-dose ferroptosis inducers (such as erastin) fail to kill tumor cells. Instead, by activating the STAT3/IL-8 signaling axis in macrophages, they polarize macrophages into an M2 phenotype, thereby enhancing tumor invasion and metastasis, completely contrary to the therapeutic intent (153). This demonstrates that the functional state of macrophages is a core variable determining the success or failure of ferroptosis therapy.

Myeloid-derived suppressor cells (MDSCs) maintain their immunosuppressive function in the tumor microenvironment through various molecular mechanisms and exhibit significant resistance to ferroptosis. Mechanistically, the high expression of N-acylsphingosine amidohydrolase 2 (ASAH2) in MDSCs effectively reduces ROS generation by inhibiting the p53-Hmox1 signaling axis, thereby protecting cells from ferroptosis (154). Concurrently, MDSCs undergo unique lipid metabolic reprogramming. The uptake of arachidonic acid mediated by fatty acid transport protein 2 (FATP2) and the synthesis of prostaglandin E2 significantly reduce lipid peroxide accumulation, further enhancing resistance to ferroptosis (155). Additionally, MDSCs utilize their highly active Xc⁻ system to take up large amounts of extracellular cystine. While maintaining their own redox homeostasis, this leads to cystine/cysteine depletion in the tumor microenvironment, consequently impairing T cell activation and function (156). These findings reveal that MDSCs construct a comprehensive ferroptosis defense system through multiple pathways, including ASAH2, lipid metabolic reprogramming, and cystine deprivation, thereby maintaining an immunosuppressive state in the tumor microenvironment. Therefore, targeting these key pathways to induce ferroptosis in MDSCs may become an effective strategy for reversing immunosuppression and enhancing anti-tumor immune responses.

Based on the above mechanisms, the synergy between ferroptosis and immune checkpoint inhibitors constitutes a powerful positive feedback loop. Ferroptotic tumor cells act as an “in situ vaccine,” efficiently priming dendritic cell-mediated T cell responses by releasing tumor antigens and damage-associated molecular patterns (DAMPs) (157, 158). Meanwhile, by eliminating MDSCs and promoting M1 macrophage polarization, ferroptosis inducers can effectively reverse “cold tumors” into “hot tumors” infiltrated by immune cells, fundamentally improving the microenvironment for immunotherapy (159). Therefore, targeting ferroptosis defense nodes highly expressed in tumors, such as SLC7A11, has become an effective strategy to overcome primary and secondary resistance to immune checkpoint inhibitors (160). Emerging nanotherapeutic strategies (e.g., metal-organic framework nanoparticles loaded with sorafenib) not only induce tumor ferroptosis but also actively promote CD8⁺ T cell activation and tumor infiltration, achieving synergistic enhancement between ferroptosis induction and immune activation (161). Collectively, ferroptosis serves as a potent modulator of the tumor immune landscape. By inducing immunogenic cell death, reprogramming macrophage polarization, and depleting immunosuppressive cells, it can effectively convert immunologically “cold” tumors into “hot,” T-cell-inflamed microenvironments. This remodeling not only enhances intrinsic anti-tumor immunity but also establishes a strong rationale for combining ferroptosis inducers with immune checkpoint inhibitors, offering a promising strategy to overcome resistance and improve therapeutic outcomes.

Ferroptosis functions as a multifunctional immune modulator within the TME, with its net impact on antitumor immunity determined by a balance of cell type-specific effects. This duality is exemplified by CD8⁺ T cells, which can induce tumor ferroptosis (e.g., via IFN-γ) yet are themselves vulnerable to ferroptosis, compromising their effector function. Macrophages primarily act as responders; ferroptotic signals can promote immunostimulatory M1 polarization, but their functional state also dictates therapeutic outcome by influencing immunogenic clearance or pro-tumor M2 skewing. In contrast, myeloid-derived suppressor cells (MDSCs) resist ferroptosis through robust antioxidant defenses, thereby sustaining an immunosuppressive microenvironment.

This analysis highlights that the pro-immunogenic effects of ferroptosis (e.g., DAMP release, M1 polarization) largely stem from targeting tumor cells and certain innate immune cells. Conversely, its immunosuppressive effects often arise from collateral damage to CD8⁺ T cells or the fortification of immunosuppressive networks (MDSCs, M2 macrophages). Therefore, promising therapeutic strategies should aim to selectively induce ferroptosis in tumor and immunosuppressive cells while protecting or reinvigorating CD8⁺ T cells. Future combination therapies must be designed with this cellular selectivity in mind to harness the immunostimulatory potential of ferroptosis while mitigating its detrimental effects on adaptive immunity.

Small molecule inducers targeting the core ferroptosis machinery represent the most direct and extensive strategy for activating this cell death pathway. Based on their targets, they are primarily categorized into System Xc⁻ inhibitors, GPX4 inhibitors, and compounds targeting iron metabolism (2).

System Xc⁻ inhibitors work by blocking cystine uptake, depleting intracellular glutathione, and consequently removing support for GPX4 activity. First-generation inhibitors like Erastin and its derivatives, as well as the clinically used drug sulfasalazine, function by inhibiting SLC7A11 activity (1, 133). Recent studies have identified various other compounds and natural products that can indirectly induce ferroptosis by targeting the System Xc⁻ axis.

GPX4 inhibitors directly act on the key enzyme GPX4, inactivating it. RSL3 is the prototype of this class, which irreversibly inhibits GPX4 enzyme activity by covalently binding to its active site selenocysteine (23). Beyond direct inhibitors, regulating GPX4 stability is also an effective approach (162, 163).

Compounds targeting iron metabolism modulate intracellular iron homeostasis, increasing the labile iron pool to catalyze the Fenton reaction and accelerate lipid peroxidation. This includes strategies that utilize iron chelators—which in specific contexts can inhibit ferroptosis by sequestering excess iron—but more commonly involves using iron ionophores or inducing ferritinophagy to increase intracellular free iron.

Despite the significant potential of small molecule inducers, their clinical application is often limited by poor pharmacokinetic properties, off-target toxicity due to non-specific distribution, and the complexity of the tumor microenvironment (164). Nanotechnology provides revolutionary tools to address these challenges. Engineered nanodelivery systems can significantly enhance the targeting of ferroptosis inducers, reduce side effects, and enable synergy with multiple treatment modalities (165).

Nanocarriers are widely used to efficiently encapsulate and deliver classic small molecule ferroptosis inducers, improving their water solubility, stability, and tumor accumulation. For instance, zwitterionic polymer-coated magnetic nanoparticles loaded with simvastatin were developed for treating triple-negative breast cancer (166).

Inorganic nanomaterials themselves, particularly iron-based nanomaterials, can serve dual roles as both iron sources and Fenton reaction catalysts, inherently inducing ferroptosis. They can dissociate within tumor cells to release Fe²⁺/Fe³⁺, catalyzing H₂O₂ to produce highly toxic ·OH and trigger lipid peroxidation (165, 167).

The most advanced strategies involve designing multifunctional nano-platforms that combine ferroptosis induction with other therapeutic modalities for a powerful synergistic antitumor effect. For example, Li et al. designed a core-shell nanoparticle for co-delivering copper and Erastin, successfully synergizing cuproptosis and ferroptosis to enhance cancer immunotherapy (168). Although nanocarriers hold great promise for improving targeting and reducing toxicity, their in vivo stability, potential immunogenicity, and large-scale standardized manufacturing remain critical bottlenecks for clinical translation.

Radiotherapy and chemotherapy are mainstays of current cancer treatment, but their efficacy is often limited by acquired resistance in tumor cells. Recent studies indicate that ferroptosis plays a key role in cell death induced by these therapies, particularly in drug-resistant tumor cells with mesenchymal features or enhanced antioxidant capacity.

Radiotherapy generates reactive oxygen species via ionizing radiation, initiating lipid peroxidation and subsequently inducing ferroptosis. For instance, in nasopharyngeal carcinoma, radiotherapy can upregulate GSTM3 expression, which stabilizes USP14 and inhibits GPX4, thereby enhancing radiotherapy-induced ferroptosis (169). Chemotherapeutic agents like cisplatin and oxaliplatin can also indirectly induce ferroptosis by depleting glutathione or inhibiting System Xc⁻ function.

Targeted therapies function by specifically inhibiting oncogenic signaling pathways in tumor cells, but single-agent application often fails due to pathway redundancy or feedback activation. Combining ferroptosis inducers with targeted drugs can enhance antitumor effects by synergistically regulating metabolic and oxidative stress states.

For example, in melanoma, combining an AXL inhibitor with a BRAF inhibitor significantly induces ferroptosis, suppresses protective autophagy, enhances apoptosis, and overcomes BRAF inhibitor resistance (170). In hepatocellular carcinoma, while Sorafenib can inhibit System Xc⁻ to induce ferroptosis, it simultaneously activates the Nrf2 pathway leading to resistance; Nrf2 inhibitors can block this feedback mechanism, restoring Sorafenib’s ability to induce ferroptosis (171).

Combining ferroptosis induction with immunotherapy is one of the most promising directions in current cancer therapy. Ferroptosis, as a form of immunogenic cell death, can release tumor-associated antigens and damage-associated molecular patterns, promoting dendritic cell maturation and antigen presentation, and activating tumor-specific immune responses by cytotoxic T lymphocytes (158, 172).

This immune activation provides a novel approach to overcome primary or acquired resistance to ICIs. Research shows that interferon-gamma is a key molecular bridge connecting ICIs to ferroptosis; activated CD8⁺ T cells secrete IFN-γ, which downregulates SLC7A11 expression in tumor cells, thereby inhibiting cystine uptake and enhancing lipid peroxidation and ferroptosis sensitivity (138, 173).

The combination of ferroptosis with physical therapies, particularly photodynamic therapy(PDT) and photothermal therapy (PTT), demonstrates powerful synergistic antitumor potential. PDT uses photosensitizers activated by specific light wavelengths to generate ROS, directly killing tumor cells and inducing immunogenic cell death; the produced ROS can also serve as substrates for the Fenton reaction, promoting lipid peroxidation and thereby sensitizing and enhancing ferroptosis (174, 175).

The integration of nanotechnology has significantly advanced the translational application of this combination strategy. Multifunctional nano-platforms can co-deliver ferroptosis inducers and photosensitizers/photothermal agents, enabling precise, tumor microenvironment-responsive release (167, 176). For instance, nanoparticles loaded with Fe³⁺ and the photothermal agent polydopamine, upon laser irradiation, not only generate thermal effects but also produce ·OH via the Fenton reaction, achieving simultaneous PTT and ferroptosis therapy (177, 178).