3.1 Antioxidant metabolism

The balance of antioxidant metabolism is dynamically maintained mainly by GPX4, coenzyme CQ10, and dihydroorotate dehydrogenase (DHODH), all of which contain a metabolite molecule that exists in a chemically reduced and oxidized state that can inhibit ferroptosis (Figure 2A). GPX4 is a selenoprotein that reduces hydroperoxide in the cell membrane and uses GSH as an essential cofactor [35]. Glutamine-cysteine synthase activity, cysteine concentration, and GSH feedback inhibition directly regulate GSH synthesis, thereby, affecting the enzymatic activity of GPX4 [47-49]. System XC and the trans-sulfurization pathway are the two primary sources of cysteine.

The cystine/glutamate transport complex XC, comprising of heterodimers SLC7A11 (xCT) and SLC3A2 (4F2hc), takes up cystine (cystine) in a 1:1 form and excretes intracellular glutamate [13, 50, 51]. Researchers have established that SLC7A11 is the key hub for regulating “iron overload ferroptosis” [52], and activating SLC7A11 expression in cardiomyocytes can create “dead” resurrection [53]. Many factors regulate the expression of XC in the system, including targeted reduction of SLC7A11 levels, inhibition of SLC7A11 degradation and promotion of SLC7A11 expression; indicating that the XC system can determine the tolerance of ferroptosis (Table 2) [54-101].

Glutamate is another important raw material required for the synthesis of GHS, and its uptake mainly depends on SLC38A1 and SLC1A5 [102]. GCLC(glutamate-cysteine ligase catalytic subunit) catalyzes the first step in glutathione synthesis through the linkage of cysteine and glutamate. However, when cysteine is insufficient, GCLC promotes the synthesis of γ-glutamyl peptides (γ-Glu-AAs), eliminating glutamate, and preventing ferroptosis [103]. Overall, both the biosynthesis/uptake of selenocysteine [56, 104] and the synthesis of GSH [65, 67] affect the homeostasis of GPX4, indicating that GPX4 plays a pivotal role in regulating ferroptosis.

Mitochondrial metabolism is the main source of cellular ROS, and the decomposition of glutamine, the raw material of GSH, is thought to regulate ferroptosis by providing α-ketoglutarate (α-KG) in the mitochondrial tricarboxylic acid (TCA) cycle [93, 105]. Inhibition of the mitochondrial TCA cycle, knockout of mitochondrial voltage-dependent anion channel 2/3 (VDAC2/3), or suppression of electron transfer chain (ETC) reduces lipid peroxide accumulation and ferroptosis, indicating that ferroptosis involves abnormal ROS production [11, 55, 106]. Blocking the mitochondria strongly inhibits ferroptosis; however, when GPX4 is pharmacologically inhibited, ferroptosis can be performed independently of mitochondria. When cysteine is inadequate, mitochondrial metabolism promotes rapid depletion of glutathione and subsequent lipid ROS production and ferroptosis [106], suggesting that the role of mitochondria in ferroptosis may be related to the upstream and downstream; however, the exact role of mitochondria in ferroptosis remains unclear.

Recent studies have revealed that FSP1-CoQ is an antioxidant system that is parallel to the GPX4 pathway and acts only on GPX4-depleted cells. Ferroptosis inhibitor protein 1 (FSP1), formerly known as apoptosis-inducing factor mitochondrial-related 2 (AIFM2), was identified as a ferroptosis inhibitor. FSP1 is recruited to the plasma membrane through N-terminal myristoylation as an oxidoreductase, reduces ubiquinone (CoQ10) which is another product of mevalonate metabolism to the lipophilic free radical scavenger panthenol (CoQ₁₀H₂), which limits the accumulation of lipid ROS in the membrane without GPX4 [85, 86]. Isopentenyl pyrophosphate, a metabolite of mevalonate, is essential for producing selenoproteins, GPX4 and ubiquinone. Another product, squalene synthase, is a regulator of oxidation levels and is used to accumulate ROS to induce ferroptosis [58]. Regarding the involvement of squalene synthase in the mevalonate pathway, a rare lymphoma cell lacks an enzyme involved in cholesterol synthesis. Cells without this enzyme will accumulate squalene and the accumulation of squalene can increase the antioxidant capacity of a cell, thereby, providing benefits to cancer cells by a phenomenon known as an anti-ferroptotic property [107]. The specific inducers of ferroptosis, ferroptosis inducer derived from CIL56 (FIN56) [58] and statins [108], can inhibit the CoQ10 and mevalonate pathways to mediate cell ferroptosis. In addition to this pathway, a recent study identified another GPX4-independent ferroptosis blocking pathway involving GTP cyclohydrolase 1 (GCH1), which is a rate-limiting step in the production of tetrahydrobiopterin (BH4). BH4 inhibits ferroptosis by facilitating the formation of CoQ10 and blocking the peroxidation of specific lipids [79].

FSP1 inhibits ferroptosis by producing ubiquinol but its activity is limited to the cell membrane. Thus, it is still unclear whether the mitochondria use a similar mechanism to produce ubiquinol to repair the oxidative damage caused by mitochondrial membrane lipids. The latest research proposes a third protection system: DHODH, an enzyme located on the outer surface of the inner mitochondrial membrane and is responsible for oxidizing dihydroorotic acid (DHO) in the inner mitochondrial membrane to orotic acid (OA). DHODH reduces CoQ to CoQH2 [80, 81]. Therefore, DHODH is a protective mitochondrial lipid system that protects against oxidative damage. The following three systems have unique subcellular locations: GPX4 in the cytoplasm and mitochondria, FSP1 on the plasma membrane, and DHODH on the mitochondria. Damage to one system will force cells to rely more on the other, while simultaneous loss of all three protective systems will trigger ferroptosis induced by lipid peroxidation. The above results indicate the existence of three parallel antioxidant systems in cells, namely GPX4, FSP1-CoQ, and DHODH, which precisely regulate the metabolism of cellular antioxidant substances to prevent ferroptosis.

3.2 Iron metabolism

As one of the “bioelements,” iron in an oxygen-rich environment can generate ROS through the Fenton reaction with different types of phospholipid hydroperoxides and lipid (fatty acid) hydroperoxides [28]. Fe³⁺ circulating in the blood is transported to the cells by binding to transferrin (TF) and through endocytosis mediated by the transferrin receptor (TFRC). After Fe³⁺ is taken into cells, it is reduced to Fe²⁺ by the action of metal reductase Six-transmembrane epithelial antigen of the prostate 3(STEAP3) on the ferrosomes. The divalent metal transporter (DMT1) mediates the release of Fe²⁺ from ferrosomes to the labile iron pool (LIP) medium (Figure 2B) [53, 67, 109, 110]. Recent studies have revealed that transferrin in the liver inhibits liver damage, fibrosis, and cirrhosis by regulating ferroptosis [111], and the iron homeostasis pathway driven by sterol regulatory element binding protein-2 (SREBP2) contributes to cancer progression, drug resistance and metastasis [112]; indicating that iron homeostasis is essential for organ survival.

Most iron can be stored in ferritin or heme. Ferritin includes two subunits: FTL (light chain) and FTH1 (heavy chain) [113]. The nuclear receptor co-activator 4 (NCOA4) directly binds to FTH1 to form the ferritin complex targeting lysosomes for “ferritinophagy,” promoting the degradation of ferritin and increasing intracellular free iron during iron deficiency [89, 90]. As another common protein containing iron in cells, Fe²⁺ released by heme under the catalysis of heme oxygenase-1 (HO-1) accumulates in cardiomyocytes and induces ferroptosis [114] which may be activated by NRF2 and BAY [57, 91]. Ferroportin (FPN1) is the only cellular efflux channel for iron which is degraded by binding to hepcidin secreted by the liver, resulting in a decrease in cellular iron output (Figure 2B) [92]. However, most cells do not have an effective mechanism to export iron, resulting in an increase in LIP levels when the amount of iron exceeds the storage capacity [46]. Regardless of the pathway, the increase in intracellular free iron increases the induction of ferroptosis by LIP.

3.3 Lipid metabolism

Lipid production is the basis of survival and lipid peroxidation is a biomarker of ferroptosis [115]. Lipid peroxidation in the process of ferroptosis requires acyl-CoA synthetase long-chain family member 4 (ACSL4), lysophosphatidylcholine acyltransferase 3 (LPCAT3), and arachidonic acid 15-lipoxygenase (ALOX15) (Figure 2C). Quantitative lipidomics has found that arachidonic acid (AA) and epinephrine (AdA), which are phosphatidylethanolamines (PEs), are necessary for ferroptosis [116]. In addition, it has recently been found that enhancing TPD52-dependent lipid storage or blocking ATG5, and RAB7A-dependent lipid degradation both prevent RSL3-induced lipid peroxidation and subsequent ferroptosis, indicating that the balance between lipid synthesis, storage, and degradation can be manipulated by ferroptosis [117]. ACSL4 expression is regulated by the Hippo signaling pathways [100], radiotherapy [118], and the lactic microenvironment [88]. Located in the endoplasmic reticulum, LPCAT3 is mainly expressed in the metabolic tissues such as liver, fat, and pancreas [119], by specifically inserting acyl groups in phosphatidylcholine (PC) and PE-CoA produces PE-AA [120]. Lack of LPCAT3 can lead to a significant decrease in cell membrane AA levels and accumulation of cytoplasmic lipid droplets, which leads to lethality and intestinal cell damage in mice [120], indicating that LPCAT3 is essential for cell survival. Another interesting study recently defined the saturation of ether lipids as a determinant of ferroptosis. It was also found that in the differentiation of cardiomyocytes and nerve cells, increased levels of polyunsaturated ether phospholipids (PUFA-ePLs) are accompanied by increased sensitivity to ferroptosis, and selective downregulation of highly unsaturated PUFA-ePLs levels may be an important approach for cells to escape ferroptosis [101]. The above findings are of great importance to the rational design and use of ferroptosis-targeted drugs in disease treatment.

Lipoxygenases (LOXs) can oxidize PUFAs into their corresponding hydroperoxides. In mammalian cells, linoleic acid and AA are the most abundant polyunsaturated fatty acid substrates in LOXs [42, 119]. PUFAs are the lipids that are most prone to peroxidation during ferroptosis. Therefore, pretreatment of cells with PUFAs containing heavy hydrogen isotope deuterium at the peroxidation site (D-PUFA) can prevent PUFA oxidation and block ferroptosis [121]. Acetyl-coenzyme A carboxylase (ACC) is an enzyme involved in the fatty acid synthesis and directly synthesizes PUFAs from acyl-CoA. PE-AA is a polyunsaturated fatty acid PUFA, which can be used in LOX15 and Fe²⁺ catalytic production of oxidized phospholipid ox-PE, contributes to the development of ferroptosis [59, 98]. The mutation of the key protein iPLA2β for phospholipid remodeling leads to a decrease in the activity of its hydrolyzed oxidized phospholipid ox-PE, leading to the accumulation of oxidized phospholipids in dopaminergic neurons [99]. Evidence has shown that antioxidant metabolism, iron levels and lipid metabolism play an extremely important role in hypertrophy, and steady-state changes on either side induce ferroptosis (Table 2). However, when and how lipid peroxidation regulates ferroptosis and non-ferroptotic cell death in different ways remain to be discovered.

4.1 Energy metabolism pathways

Recent studies have indicated that energy metabolism pathways are involved in the regulation of ferroptosis. Because of the Warburg effect [122], cancer cells are extremely dependent on glucose and have low productivity. Insufficient energy metabolism in the cell leads to a decrease in ATP content and activates AMPK, an important hub that senses and regulates the balance of cell energy metabolism. ACC1 and ACC2 are two enzymes that catalyze the synthesis of malonyl-CoA from acetyl-CoA, promoting fatty acid synthesis and inhibiting fatty acid oxidation. AMPK inhibits ACC to reduce the synthesis of PUFAs, ultimately inhibiting ferroptosis [123]. The above results indicate that the AMPK-ACC-PUFA pathway activated by energy metabolism reduces the synthesis of polyunsaturated fatty acid PUFAs, thereby inhibiting ferroptosis (Figure 3A). AMPK phosphorylation also mediates the phosphorylation of BECN1, the main component of the phosphatidylinositol 3-kinase (PI3K) complex, which promotes BECN1-SLC7A11 complex formation, system XC inhibition, and subsequent ferroptosis [72]. These results indicate that tumor cells avoid the occurrence of ferroptosis to a certain extent through metabolic reprogramming. Inhibition of AMPK-ACC-PUFA signaling may produce antitumor effects by targeting this pathway in different cell types of the tumor and its niche, by enhancing the efficacy of immunotherapies and potentiating chemotherapy treatments. However, therapeutic interference with AMPK-ACC-PUFA signaling may also result in unexpected adverse effects; therefore, these inhibitors should be used with care.

4.2 HCAR1/MCT1-SREBP1-SCD1 pathway

In addition to forming a microenvironment that is prone to tumor metastasis, the lactate produced by tumor metabolism may also inhibit the ferroptosis of tumor cells through the HCAR1/MCT1-SREBP1-SCD1 pathway, providing potential for its metastasis and development (Figure 3B). Lactate produced by tumors enhances lipid synthesis by inhibiting glycolysis Moreover, lactate induces the expression of HCAR1 (lactate receptor) and MCT1 (lactate transporter). MCT1-mediated lactate uptake promotes the production of ATP in hepatocellular carcinoma (HCC) cells, increasing the AMP/ATP ratio and phosphorylating AMPK, leading to the upregulation of sterol regulatory element-binding protein 1 (SREBP1) [88, 124]. SREBP1 is a transcription factor that regulates the expression of multiple lipid synthesis-related genes, including stearoyl-CoA desaturase 1 (SCD1), which converts saturated fatty acids into monounsaturated fatty acids (MUFAs), thereby, inhibiting ferroptosis [87]. Lipidomic testing found that the knockdown of MCT1 upregulates the expression of ACSL4, reduces the production of phospholipids containing MUFA, and downregulates the expression of another ferroptosis inhibitor, CoQ10, downstream of SCD1 [88]. The above results indicate that lactate induces the formation of monounsaturated fatty acids through the HCAR1/MCT1-SREBP1-SCD1 pathway which helps liver cancer cells to resist lipid peroxidation induced by oxidative stress.

4.3 PI3K-AKT-mTOR pathway

Regardless of the energy stress and lactate signaling pathway activated by tumor metabolism, the master growth regulator of the cell, mTORC1, and another different mTOR kinase complex, mTORC2, have also been shown to be involved in the regulation of ferroptosis (Figure 3C). Many cancers require mTORC1 to maintain their ability to regulate protein synthesis and cell growth, coordinating nutrition and energy supply, to ensure that they undergo unlimited cell division, whereas mTORC2 mainly promotes cell proliferation and survival through phosphorylation of AKT [125]. Both mTORC1 and mTORC2 are involved in ferroptosis. mTORC2 interacts with SLC7A11 to directly phosphorylate serine at position 26 of SLC7A11, inhibiting its transporter activity [126]. High cell density activates LATS1/2 kinase in the Hippo pathway, subsequently phosphorylating a component of mTORC1, leading to inactivation of mTORC1 and preventing mTORC1 from degrading SLC7A11 in the lysosome [75]. Activated PI3K-AKT-mTOR is one of the most mutated signaling pathways in human cancers, and activation of the PI3K-AKT-mTORC1 signaling pathway ensures that cancer cells resist ferroptosis through SREBP1-mediated MUFA production [87].

Excessive metabolism and proliferation burden lead to higher oxidative stress in cancer cells. NFE2-related factor 2 (NRF2) is a transcription factor that integrates cell stress signals and directs various transcription elements [127]. However, during periods of increased oxidative stress or mutations in KEAP1, CUL3 or NRF2, NRF2 is no longer ubiquitinated and degraded, enabling translated NRF2 to be transported to the nucleus to activate the transcription of genes containing antioxidant response elements (ARE) [128, 129]. In addition, mTORC1 promotes the binding of p62 and Keap1 by phosphorylating p62, leading to the degradation of Keap1 and accumulation of NRF2 [87]. Many antioxidant defense proteins involved in iron and ROS metabolism, including SOD1, GPX4, glutathione synthesis enzymes (GCL and GSS), HO-1 and FTH1, are all induced by NRF2 transcription [128-130]. Therefore, the PI3K-AKT-mTORC1 signaling pathway plays a critical role in maintaining cellular redox homeostasis through NRF2-mediated signal transduction and SREBP1/SCD1-mediated MUFA synthesis (Figure 3C).

4.4 Cadherin-NE2-Hippo-YAP pathway

Apart from the mTORC1 pathway which controls tumor cell growth, Hippo controls organ size by regulating cell proliferation and apoptosis, and E-cadherin, maintains cell polarity and is also involved in ferroptosis. The Hippo pathway is a kinase chain composed of a series of highly conserved protein kinases and transcription factors [131]. E-cadherin (ECAD) is an important regulator of inter-cell interconnection in epithelial cells and is positively correlated with cell density [132]. Knockout of the Hippo element or overexpression of YAP inhibits the reduction of cell proliferation caused by the binding of E-cadherin to the cell surface. The E-cadherin/catenin complex acts as an upstream regulator of the Hippo signaling pathway in mammalian cells and inhibits the activity of YAP in the nucleus, directly regulating the Hippo pathway [133].

Recent studies have also revealed that neighboring cells can also activate the NF2 and Hippo signaling pathways through ECAD-mediated interactions, thereby, inhibiting the ferroptosis of tumor cells (Figure 3D). TEA domain transcription factor 4 (TEAD4), which interacts with Yes-associated protein (YAP) and activates its function, binds to the promoter regions of the ferroptosis markers TFRC and ACSL4. This effect was inhibited by the phosphorylation of YAP. Therefore, an increase in cell density leads to a decrease in the expression of TFRC and ACSL4. This phenomenon can be reversed by knocking out ECAD, NF2 or inhibiting the phosphorylation of YAP, thereby promoting ferroptosis [100]. In addition, LATS1/2 is a key component of Hippo and inhibits mTORC1 by phosphorylating Ser606 of Raptor, an important component of mTORC1, realizing the interaction between the Hippo signaling pathway and mTORC1 [75]. The above results indicate that the intercellular mechanism may be a vital defense function of cells which resists oxidative stress and ferroptosis.

Unlimited growth is a major feature of tumors, accompanied by massive consumption of energy, “self-toxic” density, and violent biochemical metabolism. Thus, ferroptosis was significantly induced in tumor cells. However, in the natural state, few tumor cells lose the ability to develop infiltration due to ferroptosis, and there must be a special mechanism to antagonize ferroptosis. Some studies have indicated that the classic energy metabolism pathways, density effects and biochemical reaction metabolites in tumor cells interact closely with ferroptosis and significantly inhibit cell ferroptosis. Therefore, inhibition of ferroptosis may be the result of tumor cells achieving a large amount of growth. In summary, a better understanding of the crosstalk between ferroptosis and classic cancer signaling pathways could help developing new compounds to target specific ferroptosis activation mechanisms or particular effectors, increasing selectivity and decreasing the clinical drawbacks of current inhibitors.

6.1 Tumor immune checkpoint inhibitor therapy

Traditional NK(natural killer cell) cells and innate lymphoid cells interact with cancer cells through perforin, TNF family death receptors (such as TRAIL), and immune checkpoint receptors (such as programmed cell death protein 1 [PD-1]) [180]. PD-1 is lacking on resting T cells, while high levels of PD-1 expression are detected in tumor-infiltrating T cells, indicating that immune-induced tumor PD-L1 expression may be an adaptive drug resistance mechanism for tumor cells to respond to immune challenges [181-183]. Studies have revealed that interferon has the dual function of activating and suppressing the immune system. Interferon signals in cancer cells and immune cells oppose each other to establish a regulatory relationship that limits adaptability and natural immune killing. Therefore, blocking PD-1 inhibitors in breast, melanoma, and CRC cells can block the interferon signal (IFNγ) and enhance the abundance of CD8⁺ T cells, and ultimately activate the immune system to attack cancer cells [184]. However, due to the complexity of the tumor microenvironment and insufficient activation of the host immune system, the systemic antitumor response rate of immune checkpoint blocking therapy for many cancers is still limited (Figure 5) [185, 186]. Ferroptosis-related lipid peroxides can be used as an identifying signal to promote the recognition, phagocytosis and processing of tumor antigens by dendritic cells, and present tumor-associated antigens to CD8⁺ T lymphocytes, activating cytotoxic T lymphocytes to enhance tumor immunotherapy. Selective enrichment of RSL-3-containing nanoparticles in tumor tissues can induce the immunogenic death of tumor cells, initiate antitumor immune responses and induce cytotoxic T lymphocytes to secrete IFN-γ. More importantly, IFN-γ and RSL-3 can block the lipid peroxide repair pathway and increase the accumulation of lipid peroxide in tumor cells. After further treatment with immune checkpoints, the infiltration of cytotoxic T lymphocytes in tumor tissues increased by four times, effectively inhibiting tumor growth and metastasis [175]. This shows that tumor immune checkpoint inhibitor therapy based on tumor cell ferroptosis is expected to provide a new strategy for improving ferroptosis-mediated tumor immunotherapy.

6.2 Activating tumor immune cells and reversing radiotherapy resistance

Immunotherapy is a method of treating cancer by generating or enhancing the immune response to cancer [187]. CD8⁺ T cells activated by immunotherapy are traditionally thought to be through (i) perforating granzyme and (ii) Fas-FasL induces tumor cell death [188]. Studies have also established that Janus kinase (JAK) and signal transducer activator of transcription 1 (STAT1) mediate IFNγ signal activation and regulate immune responses [189]. Further research found that CD8⁺ T cell-derived IFN-γ increased the binding of STAT1 to the transcription start site of SLC7A11 and inhibited its transcription. Therefore, the lack of STAT1 in tumor cells eliminates IFN-γ-mediated downregulation of SLC7A11 and simultaneously reverses RSL3-induced lipid peroxidation and ferroptosis [190]. Erastin and RSL3 promote ferroptosis in cancer cells, except in human and mouse CD8⁺ T cells [191], suggesting that tumor cells and T cells may have different susceptibilities to ferroptosis inducers. When mice are treated with immune checkpoint inhibitors and ferroptosis inducers, they can produce a strong immune response and promote ferroptosis against tumors [190].

Radiotherapy was once considered to be one of the most vital tumor treatment methods [191, 192]. However, radiation resistance is still the main factor leading to the failure of radiation therapy, accompanied by tumor recurrence and metastasis; therefore, it is used in combination with specific chemical drugs to sensitize patients to radiotherapy. Studies have indicated that ferroptosis may crosstalk with radiotherapy. For example, radiotherapy can cause mutations in the KEAP1 gene, which upregulates the expression of SLC7A11, subsequently triggering cell resistance to ferroptosis and radiotherapy [118]. In addition, the combination of radiotherapy and immunotherapy increases cell sensitivity to ferroptosis, and cytoplasmic radiation instead of nuclear radiation has a synergistic effect with ferroptosis inducers. These treatments promote cell death by increasing sugar absorption and inducing peroxidation, rather than the typical DNA-damaging effect [193].

Recent studies have indicated that effector T cells and radiotherapy interact through ferroptosis to promote tumor clearance. CD8⁺ T cells activated by immunotherapy increase lipid peroxidation and ferroptosis through IFNγ, the latter directly sensitizing tumor cells to radiation therapy [194]. Radiotherapy activates the expression of telangiectasia mutant gene ATM to inhibit the expression of SLC7A11, limiting the uptake of tumor cystine, reducing glutathione, increasing the level of lipid peroxidation, and mediating tumor cell ferroptosis [194]. In addition, radiotherapy produces a large amount of ROS and reduces the level of ferroptosis marker 4-HNE; combining FINs to inhibit SLC7A11 or GPX4 significantly enhances the efficacy of radiotherapy and reverses radiation resistance [118], indicating that radiotherapy combined with ferroptosis-inducer FINs and immune checkpoint inhibitors may be a new strategy to eliminate radiotherapy resistance (Figure 5). It is noteworthy that ferroptosis is related to the adverse events of radiation, such as pulmonary fibrosis and granulocyte-macrophage hematopoietic progenitor cell death [195, 196].

6.3 Targeting tumor metabolism addiction

Maintaining the Warburg effect of tumor metabolism requires higher glucose uptake and metabolic activity, making tumor cells heavily dependent on antioxidant mechanisms; therefore, they are more susceptible to oxidative stress [122]. Moreover, tumor cells consume glucose and glutamine at a higher rate than normal cells, which is called glucose and glutamine addiction. New treatment methods may selectively target tumor cells by blocking their metabolism [197]. For example, the metabolic pressure of mitochondria increases the ROS level of T cells under hypoxia, causing severe T cell dysfunction and exhaustion, thereby, reducing the internal ROS of T cells and alleviating the hypoxic environment of tumor cells effectively blocks T cell immune failure and achieves a synergistic anticancer effect with tumor immunotherapy [198].

Cancer cells with high SLC7A11 expression are more sensitive to glutamine starvation and exhibit glucose-PPP dependence. Using glucose transporter inhibitors to limit the glucose in SLC7A11 highly expressing cancer cells leads to NADPH depletion and redox system breakdown, causing intracellular cystine accumulation and ferroptosis [199]. Aspartic acid undergoes transamination with GOT1(aspartate aminotransaminase) to produce oxaloacetate, which undergoes downstream biochemical reactions to produce NADPH. NADPH is used to support redox balance in cancer cell proliferation. Knockdown of GOT1 damages the oxidative phosphorylation of the mitochondria, prevents mitochondrial metabolism, increases the unstable iron pool, and increases cell sensitivity to ferroptosis, thereby triggering ferroptosis [96]. Furthermore, the combination of tryptophan depletion and immune checkpoint blockade synergistically enhanced T cell-mediated antitumor immunity and induced tumor cell ferroptosis [190]. In conclusion, the treatment methods for the characteristics of cancer cell metabolism may include: (1) directly blocking the activity of the SLC7A11 cystine transporter to inhibit cystine absorption, ultimately inducing lipid peroxidation and ferroptosis; (2) inhibiting glucose uptake to target the glucose dependence of SLC7A11 highly expressing cancer cells, leading to rapid cell death caused by nutrient deficiency; (3) in glutamine -dependent cancer cells, the growth of cancer cells is inhibited by using glutaminase inhibitors.

6.4 Innate transformation of anticancer macrophages

Recent studies have established that ferroptosis is directly immunogenic or triggers an inflammatory response to other forms of regulatory necrosis in the tumor microenvironment [49]. Tumor-associated macrophages (TAMs) can exhibit different activation states based on the local microenvironment stimuli. The M1 macrophage phenotype leads to tumor regression and inhibits tumor growth and occurrence [200]. M2 macrophages secrete pro-angiogenic factors and immunosuppressive cytokines to effectively inhibit the effector functions of T cells and M1 macrophages [201].

The above results indicate that TAMs play an important role in promoting tumor growth and metastasis, and inhibiting the antitumor immune response. Therefore, it is conceivable that the development of anticancer macrophage innate transformation therapy, directly targeting intracellular signaling pathways to increase the ratio of M1/M2, may be another direction for tumor therapy. For example, the protein, KRAS^G12D, released from pancreatic cancer cells undergoing ferroptosis, is further taken up by macrophages through a specific receptor for hyper-glycosylation end products, ultimately inducing STAT3-dependent fatty acid oxidation and driving the formation of M2 macrophages [66]. In addition to sulfadiazine-loaded magnetic nanoparticles (Fe₃O₄-SAS@PLT) [186], combined therapy with microparticles released from tumor cells and radiotherapy with PD-1 antibody and TGF-β inhibitor [202], repolarizes M2-TAMS to the antitumor M1-TAMS, regulating the antitumor interaction between tumor cells and TAMS [203]. Additionally, the combined treatment of oxygen-proliferative photosensitizer (PS-PDT) [204] and strong ferroptosis-inducer leads to an increase in lipid peroxidation, IFN-γ secretion, and immunity reshaping of the TME [172]. In addition, M1 macrophages are easier to activate and produce IFN-γ than M2 macrophages [205] whereas PD-1 antibody and TGF-β inhibitors synergistically enhance the immune response in the TME, leading to M1 polarization, and the increase in H₂O₂ levels in macrophages triggers a Fenton reaction with ions released from the core of the nanoparticle, which subsequently contributes to tumor cell ferroptosis [206]. In summary, anticancer macrophage innate transformation combined with ferroptosis therapy may have broad application prospects.

6.5 Clever use of damage-related molecular patterns (DAMPs)

Necrotic cancer cells release DAMPs in vitro which bind to receptors in various immune cells to induce the maturation of dendritic cells and the cross-induction of CD8⁺ T cells, accompanied by the production of IFN-γ, subsequently activating the adaptive immune system in the tumor microenvironment and mediating antitumor immunity [207-209]. In the process of tumorigenesis, ferroptosis has the dual effect of promoting and inhibiting tumors, which mainly depends on the release of DAMPs in the tumor microenvironment. Cell death caused by ferroptosis is characterized by the release of restricted intracellular components, including DAMPs. Once released outside the cell, these components acquire immunostimulatory properties and act as adjuvants, thereby, activating the innate and adaptive immune system by binding to pattern recognition receptors [210]. For example, in a model of cardiac injury, ferroptotic cells attract neutrophils in a TLR4-TRIF-dependent manner, which indicates that DAMPs are released when ferroptotic cell death triggers TLR4 signals [211]. These results indicate that cells undergoing ferroptosis secrete factors that strongly activate the innate immune system, as shown in several pathological and genetic models of kidney and brain ferroptosis. Phagocytosis of bone marrow-derived dendritic cells is activated by three key DAMPs (CRT, HMGB1, and ATP) in cells undergoing ferroptosis [204]. Neutrophils are also recruited through the TLR4/TRIF/1 IFN or Wnt signaling pathways induced by cancer cells undergoing ferroptosis, which are important regulators and potential therapeutic targets against cancer progression [210]. In addition, DAMP, high-mobility group protein B1 (HMGB1), plays a crucial role in DNA regulation, such as DNA repair, transcription, and replication [212]. Both type I and type II ferroptosis activators, including erastin, sorafenib, RSL3, and FIN56, can induce the release of HMGB1 [213], which in turn interacts with the immune system and regulates cancer treatment, and plays a vital role in the success of the therapy [214]. Therefore, it is speculated that the accumulation of DAMPs can trigger tissue inflammation and regulate ferroptosis in a spontaneous cascade [215], and the delicate use of DAMPs can activate positive immune feedback to inhibit tumor growth.

6.6 Targeting cancer cell differentiation plasticity

There is different drug resistance in cancer cell lines in different growth and differentiation states. For instance, melanoma can be divided into four subtypes, each of which has a specific sensitivity to ferroptosis induction which targets differentiation plasticity to improve the target and provides a treatment method for immunotherapy efficacy [216]. Tumor necrosis factor α (TNFα) and interferon γ cause the dedifferentiation of melanoma cells. Compared with the vector or cytokine control, cell death increased after Erastin or RSL3 combination treatment, which verified that the failure differentiation subtypes were the most sensitive and melanocyte subtypes had the strongest drug resistance [216].

The hypermesenchymal cell status observed in human cancer cell lines is related to the resistance of different cancer lineages to multiple treatment modalities. This treatment-resistant hyper-mesenchymal cell status relies on a drug-targetable lipid peroxidase pathway, characterized by increased lipid peroxidase activity and promotion of polyunsaturated lipid synthesis [108]. Two different types of compounds are closely related to the selective induction of cell death in epithelial cancer cell lines in a high-mesenchymal state. The first type includes RSL3, ML210 and ML162, which are known to induce ferroptosis. The second category is statins that inhibit 3-hydroxy-3-methyl-glutaryl-CoA reductase (HMGCR). The inhibitory effect of statins on HMG-CoA reductase also consumes CoQ10 and mevalonate pathways, mediating its selectivity for high mesenchymal cancer cells [108]. Therefore, targeting the differentiation plasticity of cancer cell lines and adjusting their different cell states to improve targeting and immunotherapy may be a potential cancer treatment method.

Ferroptosis is a new therapy that depends on iron-mediated lipids. Several studies have shown that ferroptosis is closely related to chemotherapy and radiotherapy [60, 217]. However, the connection between ferroptosis and cancer immunotherapy is speculative at best because the available evidence is based on very few well-conducted studies, and only few ferroptosis-related genes have been subjected to prognostic analyses [218, 219]; moreover, these analyses are unreliable because these genes possess many physiological functions beyond ferroptosis.

Nevertheless, when one treatment strategy is used alone, its systemic antitumor response rate for multiple cancers is limited. Thus, considering the characteristics of cancer cell metabolism and differentiation, we propose that a combination of the abovementioned six ferroptosis-related methods will be of great significance in cancer treatment.