This article is mentioned in:

Introduction

Head and neck cancer (HNC) is the seventh most common type of cancer worldwide, and head and neck squamous cell carcinoma (HNSCC) occurs mainly on the mucosal surfaces of the oral cavity, sinus cavities, pharynx and larynx, accounting for >90% of HNC cases (1,2). Smoking, drinking and viral infection are the main risk factors for HNC (3). In recent years, the incidence of HNC has shown a downward trend due to the decline in tobacco use and improved treatment options (1,4). Early detection, screening, vaccination and follow-up of HNC are essential factors affecting patient prognosis (5). Currently, HNC treatment is comprehensive and is based on surgery. However, the treatment leads to various complaints/symptoms including swallowing and speech difficulties and loss of taste, which directly affect the quality of life of patients (6). Compared with individuals with other localized diseases, patients with HNC have a poorer 5-year survival rate and >50% of patients with HNC are first diagnosed with locally advanced disease (7). In total, <20% of patients experience a long-lasting response to targeted immunotherapies for HNC, even though these treatments have increased patient survival (4). Therefore, it remains essential to explore new and more effective treatment strategies for HNC.

Ferroptosis is an iron-dependent non-apoptotic form of regulated cell death (RCD) driven by the fatal accumulation of excessive lipid peroxidation (LPO) and iron toxicity to cell membranes (8-11). Different physiological conditions and pathological stresses may trigger ferroptosis (12), and it is considered one of the most widespread forms of CD (13). There is a strong link between the occurrence and development of numerous diseases, including aging, neurodegeneration, infectious diseases, autoimmune diseases, cancer and ferroptosis (14). Ferroptosis is associated with multiple oncoproteins, tumor suppressors and oncogenic signaling pathways, and multiple cancer-related genes and signaling pathways are involved in ferroptosis (13). Different cancer types have varying sensitivities to ferroptosis (15). Inducing ferroptosis has the potential not only to inhibit cancer but also to promote its development (9). When ferroptosis-promoting activity exceeds the capacity of the ferroptosis defense system, ferroptosis occurs in cancer cells, resulting in cancer suppression. Conversely, when the mechanism that inhibits ferroptosis exceeds the cellular activity that promotes ferroptosis, cancer cells evade ferroptosis, thus promoting cancer development and even metastasis (10). Therefore, exploring specific weaknesses in cancer cells that are susceptible to ferroptosis could contribute to more personalized targeted treatments for cancer (16), and the selective induction of ferroptosis could be a new strategy for cancer treatment (17).

Cystine-glutamate reverse transporter (system Xc-), composed of the transporter protein subunit solute carrier family 7 member 11 (SLC7A11) and the regulatory subunit solute carrier family 3 member 2 (SLC3A2), is an important target for regulating ferroptosis, and its activity is mainly determined by SLC7A11 (18). SLC7A11 further regulates the synthesis of downstream glutathione (GSH), which is the reducing substrate for GSH peroxidase 4 (GPX4) activity (19,20). GPX4, a selenium protein, is one of the important signaling molecules that inhibit ferroptosis (21,22). These ferroptosis regulators are involved in the invasion and migration processes of HNSCC cells, and their expression levels can serve as prognostic indicators for HNSCC (23). For instance, ferroptosis can be inhibited by SLC7A11 and its expression level is positively correlated with the clinical stage of patients with HNSCC (24). Glutamine blockade can increase the level of LPO and the Fe2+ concentration in cells and lead to the accumulation of reactive oxygen species (ROS), thereby promoting ferroptosis in HNSCC cells (25,26). High expression of GPX4 suppresses ROS and LPO levels, ultimately inhibiting the sensitivity of HNSCC cells to ferroptosis. Conversely, GPX4-mediated inhibition of ferroptosis can improve the therapeutic efficacy for HNSCC (24,27). Increased expression of system Xc-is associated with poor prognosis in HNSCC. Knocking down system Xc-induces ferroptosis in HNSCC, while IL-6 reverses ferroptosis by upregulating system Xc-through its involvement in LPO (28). Furthermore, ferroptosis can reverse drug resistance in HNSCC cells or enhance their sensitivity to chemotherapy drugs (29). In addition to the aforementioned examples, there is a large number of reports related to ferroptosis and HNC. Therefore, further exploration of the role of ferroptosis in HNC may provide new insights for the diagnosis and treatment of HNC.

Ferroptosis

What is ferroptosis?

Ferroptosis is defined as iron-dependent RCD caused by unrestricted LPO and plasma membrane rupture (8). It was originally considered to be a metabolically regulated form of CD activated by small molecules (11,30). When the body's defense system is unable to detoxify ferroptosis-promoting activity, the excessive accumulation of lipid peroxides on the cell membrane ultimately leads to cell membrane rupture and ferroptotic CD (9,30). Ferroptosis must fulfill the conditions of CD and LPO, and be rescued by ferroptosis inhibitors at the same time (21). Importantly, LPO without CD is different from ferroptosis, which requires careful distinction (21). The process of ferroptosis involves an increase in free radicals, fatty acids and enzymes associated with LPO (31). Fe2+ reacts with lipid peroxides to form hydroxyl radicals, which further react with polyunsaturated fatty acids (PUFAs), leading to the accumulation of lipid peroxides (32). By LPO, hydroxyl radicals target and damage lipids in cell membranes (16). LPO leads to disruption of the integrity of cell membranes and makes them unstable (16). GSH depletion, reduced cellular uptake of cystine and iron-dependent LPO are the main features of ferroptosis (32), and the mechanism of its occurrence may involve mainly free radical chain reactions or enzymatic oxygenation (21). Free intracellular iron or iron-containing enzymes react with oxygen and lipids (generally lipids containing PUFAs) to promote the formation of membrane lipid peroxides, resulting in ferroptosis (33).

Ferroptosis is dependent on intracellular iron and differs in nature from autophagy, pyroptosis and necroptosis (34,35). Ferroptosis does not involve cellular mechanisms that mediate apoptosis or necroptosis (30). Ferroptotic cells do not exhibit typical apoptotic features such as chromatin condensation or the formation of apoptotic bodies, but usually exhibit necrosis-like morphological changes (11,22). Ferroptosis does not result in cytoplasmic or organelle swelling, or membrane rupture (36). Morphologically, ferroptosis is characterized by reduced mitochondrial volume, broken mitochondrial outer membrane, reduced or absent mitochondrial ridge, normal-sized nucleus and no nuclear concentration (37). These abnormalities are caused by the loss of selective permeability of the plasma membrane due to the occurrence of membrane LPO and oxidative stress (12).

Organelles associated with ferroptosis

Different organelles regulate different types of RCD by sensing stress signals (38). Multiple subcellular organelles are involved in the onset of ferroptosis and generate signals associated with iron accumulation, lipid synthesis and LPO (38). Organelles such as lysosomes, the Golgi apparatus and mitochondria are important sources of unstable iron (39). Mitochondrial ROS feed back to the endoplasmic reticulum (ER), accelerating the unfolded protein response (40). Excessive accumulation of unfolded proteins leads to ER stress (ERS) (41). ERS activates activating transcription factor 4, thereby upregulating the downstream core effector molecule glutathione-specific γ-glutamylcyclotransferase 1 (CHAC1) and enhancing its transcription. CHAC1 further degrades GSH, amplifying oxidative damage through a positive feedback mechanism (42). Excessive ERS makes ER difficult to restore its function, which leads to increased iron accumulation in cells, thereby affecting the oxidative stress state of cells and leading to ferroptosis (41). The ER is the initiating and most important site of LPO during ferroptosis (14,43). Inactivation of GPX4 and depletion of GSH can trigger LPO of the ER membrane. Activation of LPO leads to the release of Ca2+, which in turn causes mitochondrial calcium overload, further amplifying oxidative stress, and ultimately inducing mitochondrial ROS production and subsequent ferroptosis (44). Following the commencement of ferroptosis, LPO first appears in the membrane of the ER before it spreads to the membranes of other organelles and ultimately to the entire plasma membrane (39). The ER, which has a multifaceted role in the regulation of ferroptosis, can increase or decrease ferroptosis sensitivity through the PUFA or monounsaturated fatty acid (MUFA) pathways, respectively (33).

Mitochondria are metabolic hubs that regulate ferroptosis, and contain factors that both drive and defend against ferroptosis (45,46). Mitochondrial quantity and quality modulate the activity of ferroptosis inducers, and mitochondria regulate ferroptosis by influencing mitochondrial ROS, mitochondrial iron, mitochondrial DNA and the tricarboxylic acid cycle (38). Ferroptosis is induced by excessive ROS and Fe2+, and is inhibited by intact and normal mitochondrial function (47). Musashi-2 deficiency can lead to intracellular redox imbalance, mitochondrial damage and Fe2+ accumulation, thereby inducing ferroptosis (48). When ROS causes the activation of heat shock protein 90α (Hsp90α), Hsp90α competitively binds to Kelch-like ECH-associated protein 1 (Keap1), allowing nuclear factor erythroid 2-related factor 2 (NRF2) to escape the binding of Keap1 and be released. NRF2 is then transported to the cell nucleus and activates the GPX4 pathway, thereby inhibiting ferroptosis (49,50). ROS interferes with cellular homeostasis and when iron homeostasis is disrupted, the expression of iron-related proteins is altered (51,52). Mitochondria are also the site of synthesis of metabolites of coenzyme Q (CoQ or CoQ10), which are resistant to ferroptosis (33). In addition, mitochondria regulate ferroptosis by influencing processes including amino acid metabolism, lipid metabolism and energy metabolism (53).

Mitochondria differ from other organelles in that they possess mitochondrial DNA (54). Mitochondria maintain cellular and organismal homeostasis by participating in ATP and ROS production, metabolic precursor synthesis, calcium regulation, immune signaling and apoptosis (54). However, mitochondria are highly susceptible to dysfunction, and oxidative stress is promoted by mitochondrial damage and dysfunction, which in turn contributes to ferroptosis (55). Ferroptosis occurs when there is an imbalance between lipid hydroperoxide detoxification and accumulation, and environmental stresses may skew this imbalance in favor of lipid ROS accumulation (56). The accumulation of excess free iron within the mitochondrial membrane allows LPO of the mitochondrial membrane, which also contributes to the occurrence of ferroptosis (Figs. 1 and 2) (57).

Figure 1 Effects of different organelles on ferroptosis. TCA cycle, tricarboxylic acid cycle; mtROS, mitochondrial reactive oxygen species; mtDNA, mitochondrial DNA; PUFA, polyunsaturated fatty acid; CoQ, coenzyme Q; ER, endoplasmic reticulum; MUFA, monounsaturated fatty acid; ARF1, ADP ribosylation factor 1; ROS, reactive oxygen species; NO, nitric oxide.

Figure 2 Activation pathways for ferroptosis. Synthesis and peroxidation of PUFA-PLs, iron metabolism and mitochondrial metabolism are major prerequisites for ferroptosis. TF, transferrin; TFR1, transferrin receptor 1; STEAP3, six transmembrane epithelial antigens of prostate 3; TRPML1, transient receptor potential mucolipin 1; DMT1, divalent metal transporter 1; LIP, labile iron pool; ROS, reactive oxygen species; PUFA, polyunsaturated fatty acid; ACSL4, acyl coenzyme A synthetase long chain family member 4; CoA, coenzyme A; LPCAT3, lysophosphatidylcholine acyltransferase 3; ALOX, arachidonate lipoxygenase; PUFA-PLs, PUFA-containing phospholipids; PUFA-PL-OOH, phospholipid hydroperoxides.

Mitochondria have complex functions. In addition to being important generators of ROS, mitochondria prevent ferroptosis by preserving oxidative defenses and the cellular redox equilibrium (58). Mitochondrial self-regulation (including mitochondrial fission, mitochondrial fusion and mitochondrial autophagy) affects ferroptosis, and metabolic imbalances in self-regulation promote or inhibit ferroptosis (55). The mitochondrial pathway influences ferroptosis in multiple ways and is linked to the treatment of ferroptosis-related diseases. MitoTEMPO, a mitochondrial-targeted antioxidant, can prevent mitochondrial LPO as well as maintain the integrity of mitochondria and epithelium (59). Mitoquinone acts as a ROS scavenger, maintaining mitochondrial integrity and function, and the cell survival rate (60). Targeting the mitochondrial dynamic protein dynamin-related protein 1 in combination with ferroptosis inducers may be beneficial for the treatment of oral squamous cell carcinoma (61). Therefore, more in-depth studies on the role of mitochondria in ferroptosis are needed. It is esteemed that the mitochondrial pathway can be used to develop potential targets for innovative therapeutic interventions related to ferroptosis.

Lysosomes promote ferroptosis mainly through the activation of autophagy, release of lysosomal cathepsin, and accumulation of lysosomal iron or nitric oxide (38). Fe2+ promotes LPO through the Fenton reaction, and the release of lysosomal iron may cause LPO to spread to other membrane systems (62). When cathepsin B is translocated from lysosomes to the nucleus, it can cleave DNA or histone H3, thereby potentially altering the expression of ferroptosis-related genes (38). The Golgi apparatus influences the sorting and transport of cellular cargo during ferroptosis (38). Lipid peroxide generation can be decreased by the Golgi-associated small GTPase ADP ribosylation factor 1 (Fig. 1) (63). The sensitivity to ferroptosis varies depending on the fatty acids contained in the lipid droplets (LDs) (33). The degradation and storage of LDs influence susceptibility to ferroptosis, with the number of LDs increasing in the early stages of ferroptosis but decreasing in the final stages (38). Peroxidase regulates ferroptosis by affecting lipid synthesis, redox homeostasis, as well as steroid and peptide hormone biosynthesis and signaling (38). In addition, other organelles affect ferroptosis.

The development or function of different organelles has different effects on the occurrence and development of ferroptosis. Therefore, it is necessary to consider the effects of different organelles on the occurrence of physiological or pathological ferroptosis or the development of therapies related to ferroptosis.

Activation of ferroptosis

PUFA-containing phospholipid (PUFA-PL) synthesis pathway

Cellular membranes are composed of lipid bilayers, of which phospholipids are the fundamental building blocks (64). PUFAs integrate into the cell membrane and affect its fluidity and flexibility (16). The lipids that are most vulnerable to peroxidation during ferroptosis are PUFAs (65). The accumulation of PUFA oxides is a marker of ferroptosis (15). However, free PUFAs do not directly drive ferroptosis; they must be esterified into membrane phosphatidylethanolamine or cholesteryl esters to become lethal upon peroxidation (16). The quantity and location of PUFAs in the phospholipid bilayer influence the cellular sensitivity to ferroptosis (36). PUFAs are activated by acyl coenzyme A synthetase long chain family member 4 (ACSL4) (64). Among the ACSL family members, ACSL4 and ACSL3 have the strongest correlation with ferroptosis (21). ACSL4 is an important proferroptotic gene and the upregulation of ACSL4 is a marker of ferroptosis (15,66). ACSL4 is required for the execution of ferroptosis, and its expression determines its susceptibility to ferroptosis by shaping the composition of cellular lipids (21,67). ACSL4 links free PUFAs with coenzyme A (CoA) to produce PUFA-CoA (68). Lysophosphatidylcholine acyltransferase 3 (LPCAT3) re-esterifies PUFA-CoA to phospholipids, which then generate PUFA-PLs (69). PUFA-PLs form phospholipid hydroperoxides (PUFA-PL-OOH) under the action of unstable iron and catalytic enzymes, ultimately leading to the accumulation of lipid peroxides and ferroptosis (Fig. 2) (32,64,70).

The PUFA-PL synthesis pathway is also related to disease treatment. For example, ACSL4 promotes the sensitivity of human pancreatic cancer cells to chemotherapy; silencing ASCL4 helps to weaken the invasive ability of breast cancer cells (71). In the Balb/c male mouse model of reproductive damage, di-(2-ethylhexyl) phthalate can alter the expression of ACSL4 in the testicular tissue of mice, further affecting the synthesis of PUFA-PL and MUFA-PL, potentially reshaping lipid metabolism (72). In the tamoxifen-inducible skeletal muscle-specific LPCAT3 knockout mouse model, LPCAT3 deficiency prevented the accumulation of lipid peroxides and 4-hydroxynonenal (4HNE), while limiting the uptake of PUFAs into phospholipids limits the generation of lipid peroxides (LOOH) (73). Therefore, establishing targeted treatment strategies based on the different target sites of the PUFA-PL synthesis pathway may provide a new treatment option for related diseases. However, PUFA treatment alone is usually not sufficient to induce effective ferroptosis (68). This needs to be explored more thoroughly in future studies.

Iron metabolism pathway

Iron is an essential chemical element involved in a variety of physiological and pathological processes in the body, and abnormalities in iron metabolism are involved in a wide range of diseases (74). Imbalances in iron metabolism (including absorption, utilization, storage and export) can lead to ferroptosis (11). Fe3+ binds to transferrin (TF), and subsequently binds to TF receptor protein 1 and forms a complex on the cell membrane, which then forms Fe2+ with the help of six transmembrane epithelial antigens of prostate 3 (69,75). Fe2+ is then delivered into the cytosolic labile iron pool (LIP) by transient receptor potential mucolipin 1 or divalent metal transporter 1 (DMT1) (69). DMT1 is responsible for Fe2+ absorption in the intestine (74). The LIP is responsible for regulating cellular iron homeostasis (74). When the level of free iron in cells is high, LIP induces ferroptosis more frequently (65).

Ferric ion overload is the main cause of lipid ROS accumulation (76,77). ROS are generated mainly by excess Fe2+ through the Fenton reaction (78). Excess ROS produce oxidative stress when redox metabolism is disturbed (79). The redox state of the cell is altered by changes in the level of intracellular iron, which further affects LPO (74,80). LPO can be divided into two pathways: i) Nonenzymatic and ii) enzymatic LPO (78). Iron is usually involved in ferroptosis by initiating the nonenzymatic Fenton reaction and acting as an important cofactor for arachidonate lipoxygenases (ALOX) and P450 oxidoreductase (POR) (Fig. 2) (32). POR, a flavin protein, promotes ROS production and transfers electrons to cytochrome P450 enzymes during LPO. POR also produces superoxide radicals that induce LPO and ferroptosis (70,78). Iron accumulation, and cytoplasmic and lipid ROS induction, are characteristics of ferroptosis (81). Notably, not all iron accumulation leads to ferroptosis and ferric ions are not the only conditions that induce ferroptosis (36,82). In addition to iron, metal ions such as zinc and copper can also induce ferroptosis under certain conditions (70).

Iron metabolism is an important indicator for determining the occurrence of ferroptosis. Related inhibitors and inducers that regulate iron absorption and transport are not only common methods for studying ferroptosis, but also therapeutic strategies for certain ferroptosis-related diseases (83). Ferroptosis inhibitors can regulate free iron levels, thereby protecting cells from the effects of ferroptosis (84). For instance, in the female Wistar rats spinal cord injury model, deferoxamine can chelate iron ions and inhibit the Fenton reaction, thereby inhibiting ferroptosis, which helps promote spinal cord injury repair (85). The synthesis of polyamines is related to iron overload in ferroptosis. Polyamine supplements enhance the sensitivity of lung cancer cells to radiotherapy and chemotherapy by inducing ferroptosis, which indicates that altering the response of cancer cells to ferroptosis may offer a new approach for cancer treatment (86). It is possible to overcome the resistance of sorafenib in the treatment of hepatocellular carcinoma by altering the iron ions that affect the occurrence of ferroptosis in the in situ mouse model of liver cancer (87). The aforementioned examples fully demonstrate the feasibility of iron metabolism pathways in ferroptosis induction therapy strategies.

Inhibition of ferroptosis

Eliminating lipid peroxides and maintaining cellular redox balance can antagonize ferroptosis (88). The inhibition of ferroptosis can be divided into two pathways: i) Inhibition of the activation system of ferroptosis and ii) enhancement of the inhibition system of ferroptosis. It is possible to influence or weaken the occurrence of ferroptosis by regulating or inhibiting PUFA-PL synthesis, iron metabolism and mitochondrial function. For instance, iron chelating agents such as deferoxamine or deferiprone can prevent the transfer of electrons from intracellular iron to oxides, thereby inhibiting the activation of ferroptosis (89). Reducing the production of ROS and LPO by promoting the activity of the ferroptosis inhibition system may also inhibit the occurrence of ferroptosis. Common ferroptosis inhibition systems include the GPX4-GSH, ferroptosis suppressor protein 1 (FSP1)-ubiquinone (CoQH2), dihydroorotate dehydrogenase (DHODH)-CoQH2, GTP cyclohydrolase 1 (GCH1)-tetrahydrobiopterin (BH4) and membrane-bound O-acyltransferase (MBOAT)-1/2-MUFA systems. The selection criteria for these two approaches of inhibiting ferroptosis differ in terms of their focus on basic research and clinical application. In basic research, inhibitors are typically validated through gene knockout or overexpression experiments to confirm their inhibitory phenotypes, and their target sites are identified in experiments (90,91). For example, lipocalin 2 knockout reduces ferroptosis in lung cancer cachexia (92); high mobility group AT-hook 2 overexpression inhibits ferroptosis in pancreatic cancer cells (93). In clinical applications, factors such as bioavailability, half-life, side effects and ability to cross the blood-brain barrier must be considered. These factors can ensure patient safety and reduce risks while providing treatment (94-96). The screening of ferroptosis inhibitors for basic research and clinical application is detailed in Table I (97-105).

Table I Screening for ferroptosis inhibitors.

Table I

Screening for ferroptosis inhibitors.

System	Basic research tool	Clinical candidate	Key considerations	(Refs.)
GPX4-GSH	RSL3	Altretamine	Monitor selenium levels	(97-99)
FSP1-CoQH2	iFSP1	FSP1-siRNA	Immunotherapy	(100)
DHODH-CoQH2	Brequinar	Leflunomide, triflunomide	Targeted mitochondria	(101,102)
GCH1-BH4	GCH1 expression vector	Sapropterin	Monitoring of BH4	(103,104)
MBOAT1/2-MUFA	MBOAT1/2 CRISPRa	N/A	Sex hormone regulation	(105)

[i] GPX4-GSH, glutathione peroxidase 4-glutathione; RSL3, RAS-selective lethal 3; FSP1-CoQH2, ferroptosis suppressor protein 1-ubiquinone; iFSP1, FSP1 inhibitor; DHODH-CoQH2, dihydroorotate dehydrogenase-ubiquinone; GCH1-BH4, GTP cyclohydrolase 1-tetrahydrobiopterin; MBOAT1/2-MUFA, membrane-bound O-acyltransferase1/2-monounsaturated fatty acid; CRISPRa, clustered regularly interspaced short palindromic repeats activation.

GPX4-GSH system

System Xc-plays a role in transporting small molecular nutrients during ferroptosis (35,106). Since the light chain encoded by SLC7A11 is a specific subunit of system Xc-, there is usually a positive correlation between its expression level and the activity of system Xc- (107). System Xc-facilitates the exchange of cystine and glutamate across the plasma membrane, and when cystine enters the cell, it is reduced to cysteine (106). Ferroptosis is usually associated with the depletion of cysteine (108). Cysteine is a key precursor of GSH synthesis and has an impact on GSH synthesis (70). GSH synthesis is slower during episodes of ferroptosis, and maintenance of GSH levels during this period is dependent on system Xc- (70).

GSH, a tripeptide composed of glycine, cysteine and glutamate, is the most abundant nonprotein thiol in mammalian cells (21). As a major endogenous antioxidant, GSH acts as an electron donor in the reduced state, and two molecules of GSH are oxidized and condensed to oxidize glutathione (21,22). GSH is a cofactor and preferred substrate for GPX4, an enzyme that converts toxic LOOH to benign lipid alcohols (21,33,109). GPX4 converts and reduces reactive PUFA-PL-OOH to nonreactive and nonlethal PUFA phosphatidyl alcohols (32), and GPX4 also converts phospholipid hydroperoxides (PLOOH) to the corresponding phospholipid alcohols, making it a key negative regulator of ferroptosis (Fig. 3) (16). GPX4, at the center of ferroptosis control, is the only enzyme that is ubiquitous and capable of reducing LOOH (21). GPX4 functions as a PLOOH, and its expression or activity is influenced by GSH and selenium (16,22). GPX4 is universally expressed in vivo and its expression is directly related to selenium metabolism (21). Selenium, a key component of GPX4, is the most important trace element in ferroptosis inhibition (21). Selenocysteine is the active site of GPX4 and supplementation with the trace element selenium may help prevent tissue damage and disease associated with ferroptosis (35). Several ALOX genes (particularly ALOX15) regulate GPX4 inhibition-induced ferroptosis (78).

Figure 3 Inhibitory systems of ferroptosis. GPX4 reduces ROS accumulation. GPX4 is expressed mainly in the cytoplasm and mitochondria, FSP1 is expressed mainly in the plasma membrane, while DHODH is expressed mainly in mitochondria. GSH, glutathione; GSSG, oxidized glutathione; GPX4, glutathione peroxidase 4; LOOH, lipid hydroperoxide; LOH, lipid alcohols; PLOOH, phospholipid hydroperoxides; PLOH, phospholipid alcohols; ROS, reactive oxygen species; CoQ10, coenzyme Q10; DHODH, dihydroorotate dehydrogenase; CoQH2, ubiquinone; FSP1, ferroptosis suppressor protein 1; GCH1, GTP cyclohydrolase 1; BH4, tetrahydrobiopterin; SLC25A22, solute carrier family 25 member 22; SFA, saturated fatty acid; SCD1, stearoyl-CoA desaturase 1; MUFA, monounsaturated fatty acid; ACSL3, acyl coenzyme A synthetase long chain family member 3; MBOAT, membrane-bound O-acyltransferase domain-containing.

System Xc- and GPX4 are two targets of ferroptosis, and their inhibition effectively induces ferroptosis (35,110). Small molecule-mediated inhibition of cystine input by system Xc-leads to iron-dependent LPO (107). Both components of the Xc-inhibitory system can limit GSH synthesis, thereby inducing ferroptosis, which ultimately leads to reduced GSH synthesis and inactivation of GPX4 (34,35,68,106). Direct inactivation of GPX4 drives ferroptosis independent of intracellular cysteine and GSH levels; thus, targeting GPX4 could be used to treat diseases associated with ferroptosis (35).

FSP1-CoQH2 system

The apoptosis-inducing factor mitochondria-associated 2, also known as FSP1, acts as an antioxidant regulator during ferroptosis (22). It is a proapoptotic gene but also has antiferroptotic effects (35). The FSP1-CoQH2 system, which is independent of the GPX4-GSH system, is considered the second ferroptosis inhibitory system that effectively prevents LPO (111). As an oxidoreductase localized in the plasma membrane, FSP1 converts CoQ10 to its reduced form, CoQH2, which prevents LPO and inhibits ferroptosis in the cell membrane (Fig. 3) (32,101). CoQH2 detoxifies lipid peroxyl radicals by acting as a lipophilic radical-trapping antioxidant (101). CoQ is reduced by the oxidoreductase FSP1, which is dependent on both FAD and NAD(P)H (33,112). Lower levels of CoQ impede the re-expression of FSP1, which may attenuate the inhibitory effect of FSP1 on ferroptosis (113). Vitamin K is a cofactor of FSP1 that prevents ferroptosis and works in the same way as CoQ does (21). CoQ and vitamin K have similar structures and functions, but CoQ is more abundant in mammals than vitamin K is, and CoQ is a major cofactor in FSP1-mediated ferroptosis prevention (21). The inhibition of CoQ10 can lead to mitochondrial respiratory dysfunction and oxidative damage. Conversely, supplementation with CoQ10 effectively inhibits ferroptosis (96).

The inhibition of FSP1 alone is generally not sufficient to induce ferroptosis due to the presence of GPX4 (21). However, the overexpression of FSP1 alone prevents ferroptosis (114). The antiferroptotic effect of FSP1 is associated mainly with cancer cells (21). FSP1 is expressed in various cancer types and prevents ferroptosis even in the absence of GPX4 (115). Inhibition of the FSP1-CoQ10-NAD(P)H pathway may be a rational strategy for the treatment of tumors associated with ferroptosis (35). Previous research has shown that the expression level of FSP1 is correlated with the prognosis and invasion of HNSCC (23).

DHODH-CoQH2 system

DHODH is an enzyme involved in the de novo synthesis of pyrimidines located on the outer surface of the inner mitochondrial membrane (101,116-118). Its activity is dependent on mitochondrial complex III function (117). As an important regulator of the antioxidant response, DHODH clears lipid peroxides from cells (119). DHODH inhibits ferroptosis by reducing ubiquinone to ubiquinol in the inner mitochondrial membrane (Fig. 3) (118). Mishima et al (114) reported that DHODH inhibitors increased sensitivity to ferroptosis via FSP1 inhibition. Mao et al (101) reported that the inhibition of DHODH could significantly increase the CoQ/CoQH2 ratio, and that DHODH inhibited ferroptosis by reducing CoQ to CoQH2 in mitochondria, and that the mode of action of DHODH on the inhibition of LPO and ferroptosis may parallel that of mitochondrial GPX4 rather than cytoplasmic GPX4 or FSP1. The authors reported that DHODH inhibitor treatment did not affect the expression of GPX4, ACSL4, SLC7A11 or GSH, and that DHODH and FSP1 may be two relatively independent systems involved in the inhibition of ferroptosis. Wu et al (120) reported that hypoxia significantly increased the expression level of the DHODH protein, and that knockdown of DHODH under hypoxic conditions did not alter the level of LPO; furthermore, lipid peroxyl radicals significantly increased only when both DHODH and GPX4 were inhibited. These findings illustrate the balancing role of DHODH and GPX4 expression in regulating ferroptosis homeostasis, and suggest that the involvement of DHODH in pyrimidine synthesis may be more conducive to the maintenance of mitochondrial quantity and that DHODH may play a more pronounced role than GPX4 in the scavenging of mitochondrial ROS.

GCH1-BH4 system

GCH1, a rate-limiting enzyme that catalyzes the synthesis of BH4, is a novel oncogene involved in inhibitory ferroptosis (121,122). BH4, a metabolic derivative of GCH1, selectively prevents phospholipid depletion of the two PUFA tails, further leading to lipid remodeling and inhibition of ferroptosis (Fig. 3) (122). Two PUFA tail-containing phospholipids are shielded from oxidative deterioration by BH4 (75). BH4 contributes to the maintenance of cellular redox homeostasis and antioxidant defenses, thereby inhibiting ferroptosis (70). BH4 also has the potential to protect cells from ferroptosis by increasing the levels of reductive CoQ and directly reducing PUFA peroxidation (123). GCH1 not only prevents LPO via the BH4 pathway, but also leads to remodeling of the lipid membrane environment, which increases the abundance of reduced CoQ10 and depletes PUFA-PLs that drive sensitivity to ferroptosis (14). Low expression of GCH1 increases the susceptibility of cells to ferroptosis, whereas high expression of GCH1 increases their resistance (14). Thus, the GCH1-BH4 system can be used as a prognostic indicator of ferroptosis sensitivity (124). Notably, GCH1-BH4 is a regulator of ferroptosis inhibition independent of the GPX4 system (124).

MBOAT1/2-MUFA system

There appears to be metabolic competition between MUFAs and PUFAs, and increased expression of MUFAs may affect the sensitivity of PUFAs to ferroptosis (125). Owing to their lack of bis-allylic positions, MUFAs have low oxidative activity, and can limit LPO and ferroptosis (9,125). Stearoyl-CoA desaturase 1 (SCD1) is an enzyme that produces MUFAs (126). By promoting the synthesis of SCD1-mediated MUFAs, the mitochondrial transporter solute carrier family 25 member 22 contributes to the inhibition of ferroptosis (69,70,105). Saturated fatty acids (SFAs) are converted to MUFAs by SCD1 (105). Increased MUFA production and protection against ferroptosis are two benefits of SCD1 overexpression (9). SCD1 converts SFA to MUFA, which is further converted to MUFA-CoA with the help of ACSL3 (105,127).

Membrane-bound O-acyltransferase (MBOAT) family genes are associated with lipid metabolism and are involved in lipid processing and modification within cell membranes (128). MBOAT1 and MBOAT2, which are transcriptionally induced by estrogenic and androgenic signaling, respectively, are ferroptosis repressors; they selectively transfer MUFAs to phospholipids to impede ferroptosis (129). MUFA-PL is synthesized by MBOAT1 using oleoyl-CoA as a substrate (130). MBOAT2 selectively transfers MUFAs to lysophosphatidylethanolamine, inhibiting ferroptosis through phospholipid remodeling (Fig. 3) (105). Liang et al (105) reported that the inhibition of MBOAT1/2 promoted ferroptosis more effectively than the inhibition of SCD1 did. MBOAT1/2 prevents ferroptosis by reshaping phospholipids in a GPX4- and FSP1-independent manner (105,131). There are >10 family members of MBOATs in humans (131), and their association with ferroptosis requires further research.

Regulatory network of ferroptosis

When the balance between the active and inhibitory systems of ferroptosis is disrupted, lipid peroxides accumulate on the cell membrane and eventually lead to membrane rupture (32). The accumulation of lipid peroxides in the plasma membrane increases plasma membrane tension, thereby activating mechanically sensitive cation channels. The opening of these channels leads to the influx of Ca2+ and Na+, and the efflux of K+. At the same time, the inactivation of the Na+/K+-ATPase enhances the imbalance in ion flux. The loss of ion homeostasis and subsequent changes in transmembrane permeability result in cell rounding and plasma membrane rupture (132). During ferroptosis, nanoscale pores form in the cell membrane. The opening of these pores promotes the influx of Ca2+ and water, leading to osmotic swelling and rupture of the plasma membrane. Ca2+ influx can also activate the endosomal sorting complexes required for transport (ESCRT)-III complex, thereby further affecting the repair capacity of the plasma membrane (132). When the antioxidant system of GPX4 is damaged, PUFA-PL oxidation leads to iron-dependent accumulation of PLOOH, which disrupts the integrity of the lipid bilayer and, subsequently, the cell's defense ability against LPO weakens, resulting in uncontrolled LPO. When the accumulation of lethal LPO exceeds the membrane's antioxidant defense and repair capabilities, related ion pumps or channels lose balance, disrupting ion homeostasis and ultimately leading to membrane rupture and ferroptosis (132,133).

Riegman et al (134) conducted experiments using HT1080, HeLa, HAP1 chronic myeloid leukemia and MCF7 breast cancer cells. The authors demonstrated that ferroptosis is a permeation process, and that ferroptotic cells become rounded and swollen before cell rupture, which may be caused by the opening of nanoscale pores in the plasma membrane. When the plasma membrane pores open, it allows the solute to exchange with the external environment, causing the cells to swell before death. This cellular swelling effect spreads between cell populations in a lipid peroxide- and iron-dependent manner. Pedrera et al (135) validated the factors causing membrane rupture due to ferroptosis in cell models such as NIH-3T3, HT-1080, Mda-157 and H441 cells. The authors demonstrated that the continuous increase of cytoplasmic Ca2+ is one of the markers of ferroptosis before complete cell rupture. Following lipid oxidation, the increase of cytoplasmic Ca2+ and plasma membrane decomposition, and the rise in cytoplasmic Ca2+ also activates the ESCRT-III complex, which is involved in membrane repair. When the continuous increase in cytoplasmic Ca2+ causes the cell to round up and the plasma membrane to completely collapse, the cell's repair mechanisms become overwhelmed, leading to membrane rupture.

In summary, there are various antioxidant systems within the body that work together to alleviate ferroptotic oxidative stress while preserving cellular homeostasis (39). Alterations in the internal environment are highly correlated with ferroptosis (70). When ferroptosis occurs in a single cell, it can rapidly spread to neighboring cells (22). Ferroptosis spreads from one cell to another, but this process is not dependent on rupture of the plasma membrane. Instead, it may be dependent on the release of oxidized lipids from cells with intact plasma membranes, which can then cause ferroptosis in neighboring cells through cytosolic vesicles (14). The dynamic balance between the activation and inhibition systems of ferroptosis ensures the normal physiological function of the organism.

Oxidized lipids, ROS and LPO are the basic elements involved in the occurrence of ferroptosis (32). LPO affects mainly unsaturated fatty acids in cell membranes, and its products include LOOH, malondialdehyde and 4HNE, the levels of which are increased during ferroptosis (22). The metabolic pathways of different organelles in the cell (including mitochondria, ER and peroxisomes) contribute to the regulation of ferroptosis sensitivity (33). Molecules that affect the composition of intracellular lipids may also regulate ferroptosis (35). In conclusion, changes in important markers affecting the metabolic process of ferroptosis promote or inhibit the occurrence of ferroptosis.

Ferroptosis also involves epigenetic modifications such as histone modification, DNA methylation, non-coding RNAs, chromatin remodeling and N6-methyladenosine (m6A) (136). These modifications can dynamically regulate the expression and activity of key molecules involved in ferroptosis. m6A modification alters the mRNA stability of key genes such as SLC7A11, GPX4 and FSP1 through the coordinated action of methyltransferases, demethylases and related proteins (132). The key ferroptosis defense genes GPX4 and FSP1 can be silenced by DNA methylation, which may lead to changes in the expression levels of numerous genes regulating ferroptosis in cancer (136). Ferroptosis can be modified by histones through ubiquitination, acetylation and methylation. Histone modifications can regulate the expression of genes related to ferroptosis. Histone acetylation affects the transcriptional regulation of ferroptosis-related genes and ferroptosis sensitivity. SLC7A11 can be epigenetically activated through the ubiquitination of histones H2A and H2B (132).

MicroRNA (miRNA or miR) regulates ferroptosis by inhibiting mRNA translation or promoting mRNA degradation. miRNA can also influence iron metabolism processes to regulate ferroptosis. Long non-coding RNA (lncRNA) typically competes with miRNA for binding, thereby regulating the interaction between miRNA and its target mRNA, which in turn affects the expression of key genes involved in ferroptosis. LncRNA also interacts with proteins to regulate susceptibility to ferroptosis (132). Han et al (137) demonstrated that the lncRNA LINC00239 can interact with the Kelch domain (NRF2 binding site) of Keap1 to inhibit ferroptosis in colorectal cancer. Tao et al (138) demonstrated that the lncRNA zinc finger NFX1 type-containing 1 antisense RNA 1 promotes ferroptosis in clear cell renal cell carcinoma by regulating the miR-185-5p/SLC25A28 axis. Liu et al (139) demonstrated that miR-193a-5p can target GPX4, and that inhibition of miR-193a-5p or overexpression of GPX4 can reduce lipid ROS and Fe2+ levels in cervical cancer cells, thereby effectively inhibiting ferroptosis. Chen et al (140) demonstrated that epigenetically upregulated NOP2/Sun RNA methyltransferase family member 2 (NSUN2) binds to the mRNA of SLC7A11, catalyzes 5-methylcytosine methylation and then recruits the specific recognition protein Y-box-binding protein 1 to form a stable protein-RNA complex, thereby enhancing the stability of SLC7A11 mRNA and promoting the upregulation of SLC7A11 expression at both the mRNA and protein levels. NSUN2 also makes endometrial cancer cells resistant to ferroptosis by enhancing the stability of SLC7A11 mRNA. Hypoxia plays a crucial role in the resistance and suppression of ferroptosis in cancer cells. Hypoxia inducible factor 1α activation upregulates SLC7A11 and GPX4, thereby enhancing cancer resistance to ferroptosis-based therapies (141). These findings suggest that it is possible to regulate ferroptosis through artificial intervention, which will help develop targeted therapies related to ferroptosis.

Current detection methods for ferroptosis

Both cells and mitochondria undergo morphological changes during ferroptosis, so it is possible to detect ferroptosis by observing these changes (142). Biomarkers of peroxides identified by liquid chromatography-mass spectrometry are the preferred methods for detecting ferroptosis (129). Various small-molecule fluorescent probes are also widely used to detect indicators related to ferroptosis (129). The detection of the Fe2+ content and degree of LPO is an effective method for detecting ferroptosis (142). In addition, the expression of GSH, ROS, GPX4, ACSL4 and other related genes, mRNAs and proteins also contributes to the detection of ferroptosis (142). Ferroptosis and HNC-related genes, such as ferritin heavy chain (FTH1) and Keap1, can not only detect the occurrence of ferroptosis in HNC, but also predict the prognosis of HNC (143). However, monitoring and identifying ferroptosis by detecting a single substance is not accurate (144). The specific markers associated with ferroptosis still need to be further explored.

Ferroptosis and HNC

Role of ferroptosis in carcinogenesis and treatment of HNC

Excessive cell proliferation or suppression of normal CD increases the incidence of cancer (15). Ferroptosis is involved in both health and cancer, and physiological ferroptosis may contribute to the elimination of deleterious cells to preserve tissue homeostasis and development, whereas pathological ferroptosis may be an important contributor to the development of cancer (38). Ferroptosis, a unique iron-dependent form of RCD, has dual roles in cancer, leading to either the proliferation or elimination of cancer cells (145). Restricting the proferroptosis system, and stabilizing and overexpressing the antiferroptosis system, are important mechanisms related to ferroptosis and cancer (9). The role of ferroptosis in carcinogenesis and cancer therapy is influenced not only by oncogenes and tumor suppressors, but also by the tumor microenvironment (TME) (22). The TME is a multifaceted ecosystem where tumor, immune, stromal and other cells coexist, and these cells interact to influence tumor growth and progression (10). Various cytokines secreted by different cells within the TME can promote or inhibit ferroptosis, and the sensitivity of the TME to these cytokines also differs (10). Therefore, ferroptosis in the TME has both immunostimulatory and immunosuppressive regulatory effects on cancer cells (9).

Cancer cells undergoing ferroptosis release damage-associated molecular patterns (DAMPs) (146). The role of ferroptosis in cancer promotion or suppression depends on the release of DAMPs and the immune response triggered by ferroptosis in the TME (8). In certain cases, cancer cells stimulated by ferroptosis trigger DAMPs release, which enhances the procarcinogenic response of immunosuppressive cells (9). Impairment of ferroptosis in the TME can trigger immunosuppression, which favors cancer cell proliferation (8). The development and function of immune cells may be hampered by DAMPs generated from ferroptotic cells during LPO, which results in plasma membrane rupture (39). DAMP release from ferroptosis may accelerate tumor growth by maintaining an inflammatory TME (22). However, ferroptosis also regulates the release of DAMPs, immunostimulatory molecules and cytokines by stimulating cancer cells, ultimately leading to enhanced anticancer immunity and cancer suppression (9). In conclusion, the effect of ferroptosis on the immunogenicity of cancer cells depends on the environment in which they are located (10). Thus, DAMP release is a double-edged sword during ferroptosis, and rational induction would be useful for cancer prevention and treatment.

It is well known that tobacco, alcohol and HPV are associated with the occurrence of HNC. Cigarette tar can significantly reduce the expression of SLC7A11 and GPX4, thereby promoting ferroptosis. The mitochondria of tar-treated macrophages exhibit specific ferroptosis morphological characteristics such as contraction, increased membrane density and reduction or disappearance of mitochondrial ridges (147). Alcohol consumption leads to the upregulation of transferrin receptor expression, and also reduces the expression of hepcidin, ultimately resulting in iron homeostasis disorders. Frequent alcohol consumption causes mitochondrial damage and functional disorders, impacting their metabolic capacity and potentially preventing large-scale ion transport. In this case, ferroptosis is activated due to iron overload disorder induced by frequent excessive alcohol consumption, generating a large quantity of iron-dependent ROS (148). HPV can induce oxidative stress by promoting ROS production, further affecting the occurrence of ferroptosis (Fig. 4) (149). HPV can also affect mitochondrial function, alter cystine and GSH metabolism, and make HNSCC cells sensitive to ferroptosis (150).

Figure 4 Ferroptosis in HNC: Mechanisms and treatment. TFR, transferrin receptor; HPV, human papillomavirus; G6PD, glucose-6-phosphate dehydrogenase; FTH1, ferritin heavy chain; DDIT4, DNA damage-inducible transcript 4; ROS, reactive oxygen species; TRIB3, tribbles pseudokinase 3; LPO, lipid peroxidation; ACSL4, acyl coenzyme A synthetase long chain family member 4; GPX4, glutathione peroxidase 4; HNC, head and neck cancer.

Ferroptosis is also involved in the development process of HNC. The inhibitory effect of system Xc-can induce ferroptosis in cisplatin-resistant HNSCC (143). ACSL4 not only promotes the malignant progression of nasopharyngeal carcinoma (NPC), but also induces ferroptosis in NPC cells to enhance radiosensitivity (151). High expression of GPX4 mRNA and protein levels in HNC may be associated with its poor prognosis and radiation resistance (152). As a core regulatory factor of ferroptosis, FTH1 is overexpressed in HNSCC, which is associated with poor prognosis and lymph node metastasis of HNSCC. Knocking down FTH1 can inhibit the metastasis of HNSCC and the epithelial-mesenchymal transition (EMT) process (153). The ferroptosis-related gene DNA damage-inducible transcript 4 is overexpressed in HNSCC, and it can reduce the level of ROS and thereby inhibit ferroptosis, which affects the proliferation, invasion, metastasis, EMT and metabolism of HNSCC cells (Fig. 4) (154).

A deeper understanding of the ferroptosis mechanism in the development of HNC will facilitate the development of relevant biomarkers, construction of risk prediction models, screening of potential therapeutic targets, improvement of prognostic assessment methods, management of high-risk populations and advancement of precision medicine.

Ferroptosis and treatment of HNC

Theoretically, inhibiting either pathway at the onset of ferroptosis could affect ferroptosis. However, multiple pathways influence the occurrence of ferroptosis, and only by inhibiting the 'culprits' of pathological ferroptosis can the incidence of ferroptosis be effectively reduced. This is a key point for exploring therapeutic targets related to ferroptosis. Ferroptosis occurs through multiple pathways, which also means that a variety of targeted drug therapies related to ferroptosis can be developed. The key points that hinder the activation or inhibition of ferroptosis may provide a theoretical basis for the study of targeted drugs related to HNC.

HNC is susceptible to ferroptosis induction, particularly in subpopulations that exhibit drug resistance, and inducing ferroptosis is a viable treatment for HNC (37,155). Compared with normal cells, cancer cells are more dependent on iron and promote ferroptosis by increasing iron uptake, decreasing iron storage, limiting iron efflux and effectively killing cancer cells (24). These findings may contribute to the development of novel therapeutic options for HNC associated with ferroptosis. Iron metabolic pathways and ferritin phagocytosis are two key factors that regulate ferroptosis (15). Excess iron increases the risk of HNC, and promotes invasion and metastasis. Regulating iron metabolism can increase the sensitivity of cancerous tissues (15,156); thus, interfering with iron absorption and metabolism may be a way to treat HNC. Lee et al (157) reported that the cytosolic iron chaperone poly(rC)-binding protein 1 (PCBP1) negatively regulates ferroptosis by inhibiting lipid synthesis and peroxidation, and that excessive iron accumulation leads to mitochondrial dysfunction in PCBP1-inhibited cells. Knocking down PCBP1 may be a promising approach to promote ferroptosis in HNC cells. Roh et al (158) demonstrated that, by inducing iron-dependent, ROS-accumulated ferroptosis, artesunate specifically kills HNC cells but not normal cells, and that inhibition of the NRF2-antioxidant response element (ARE) pathway reverses resistance to ferroptosis in drug-resistant HNC cells and enhances artesunate sensitivity. Shin et al (159) reported that activation of the NRF2-ARE pathway contributes to the resistance of HNC cells to GPX4 inhibition, and that inhibition of this pathway reverses the resistance of HNC to ferroptosis.

GPX4 constitutes the strongest defense against ferroptosis (11). Cystine promotes both GSH and GPX4 synthesis (160). The inhibition of glutamine hydrolysis blocks cystine deprivation-induced ferroptosis in HNC cells (35). Shin et al (161) reported that cystine deprivation led to increased CD and induced ferroptosis in HNC cell lines, and that silencing of the dihydrolipoamide dehydrogenase gene reduced LPO and Fe2+ accumulation, which in turn inhibited ferroptosis. Therefore, exploring treatment options for HNC associated with ferroptosis is also important.

Ferroptosis sensitivity and HNC microenvironment

Ferroptosis is associated with immune activity in the TME. In addition to affecting the number and function of immune cells themselves, ferroptosis can also influence macrophage polarization to regulate the immune microenvironment. In addition, DAMPs released by ferroptotic cells themselves, once recognized by immune cells, can trigger inflammatory responses in the TME (162). Changes in the metabolism of HNC cells lead to elevated levels of PUFAs in the lipid membrane, thereby making HNC cells sensitive to LPO toxicity closely related to ferroptosis (46). Downregulation of GPX4 expression renders cells more sensitive to ferroptosis (163). Epigenetic modifications of ferroptosis-related genes, such as DNA methylation and histone modifications, can also affect the susceptibility of HNC cells to ferroptosis (164). Chung et al (162) demonstrated that inflammation and immune-related features are closely related to ferroptosis in HNSCC. Their findings indicate that ferroptosis inducers modulate the TME to inhibit the growth of HNSCC and make anti-PD-L1 antibodies sensitive to the treatment of HNSCC. Lee et al (165) demonstrated that E-cadherin expression in HNC can regulate susceptibility to ferroptosis, and this regulatory effect is influenced by cell density. Additionally, miR-200 family inhibitors, 5-azacytidine and cadherin-1 can also regulate the sensitivity of HNC cells to ferroptosis. Cai et al (166) demonstrated that the combination of triptolide and erastin can reduce the tumorigenicity of NPC and inhibit the expression of SLC7A11, increasing the sensitivity to ferroptosis. The aforementioned examples illustrate that there are numerous factors influencing ferroptosis in the HNC microenvironment. Artificial intervention in these influencing factors can induce or inhibit the occurrence of ferroptosis, thereby further aiding in the treatment of HNC.

Immunotherapy associated with ferroptosis in HNC

When HNC invasion and metastasis occur, drug resistance often occurs (164). Ferroptosis is associated with the immunologically active state of HNSCC (162). Immunotherapy has clear advantages in the anticancer process (167). Ferroptosis recruits and activates immune cells at the site of cancer, which contributes to anticancer immunotherapy (12). Targeting ferroptosis-associated metabolism in cancer may increase the effectiveness of cancer immunotherapy (168). Loss of immune function and promotion of leukocyte subpopulation death are two immunological characteristics of ferroptosis (144). The proferroptotic activity of the immune system has the potential to reduce cancer incidence and mortality (169). Immunotherapy resistance may be overcome by inducing ferroptosis (170). Yuan et al (171) reported that Keap1 mutations are relatively common in HNSCC, and that a mutation in or lack of Keap1 inhibits ferroptosis in HNSCC by increasing NAD(P)H quinone dehydrogenase 1 (NQO1) levels through NRF2 activation. The authors also reported that NQO1 is a key target in mediating ferroptosis resistance in Keap1-deficient tumors. Lee et al (165) reported that ferroptosis susceptibility in HNC cells can be epigenetically reprogrammed by EMT, which helps combat ferroptosis resistance.

Ferroptosis induces impaired anticancer immunity in immune-stimulating cells such as natural killer and B cells (10). The number and function of immune cells can be affected by ferroptosis (162). Ferroptosis in cancer cells can be triggered by T-cell-derived signal transduction molecules (17). CD8+ T cells and neutrophils promote cancer cell ferroptosis by secreting interferon γ (IFN-γ) and transferring myeloperoxidase-containing particles, followed by the release of immunostimulatory signals from ferroptotic cancer cells, which promotes the maturation of dendritic cells (DCs), the activation of M1-polarized macrophages, and the enhancement of intratumoral T-cell infiltration and activity (10). IFN-γ released by CD8+ T cells downregulates the expression of SLC3A2 and SLC7A11, and inhibits cystine uptake by cancer cells (Fig. 5) (168). The inhibition of ferroptosis in CD8+ T cells enhances their anticancer function in vivo (172). However, IFN-γ alone does not directly trigger ferroptosis (56).

Figure 5 Regulation of ferroptosis by immune cells in HNC. IFN-γ, interferon γ; ACSL4, acyl coenzyme A synthetase long chain family member 4; GPX4, glutathione peroxidase 4; DC, dendritic cell; NK, natural killer; Treg, regulatory T; HNC, head and neck cancer.

The survival and proliferation of cancer cells may be related to the transport activity of system Xc- (106). Intracellular GSH levels may be increased by upregulation of system Xc-expression, leading to cisplatin resistance in HNC cells (164). Cisplatin is a commonly used chemotherapeutic agent for the treatment of HNSCC, but it has a high rate of resistance (24). CD8+ T cells can block system Xc-via IFN-γ, leading to a decrease in cystine and GSH, which contributes to an increase in the sensitivity of cancer cells to cisplatin chemotherapy (142). Lee et al (173) reported that inhibition of glutaredoxin 5 (GLRX5) promotes ferroptosis in HNC cells and enhances the sensitivity of HNC cells to sulfasalazine (an inhibitor of system Xc-). Their study suggested that inhibition of GLRX5 overcomes chemotherapy resistance in HNC by promoting ferroptosis.

Immune cells in the TME can be affected by ferroptosis through the activation of DCs (174). The differentiation and maturation of DCs can be affected by ferroptotic cancer cells (175), which can release certain phospho-LPO products that impede DC maturation, which in turn affects the ability of DCs to cross-present antigens (10). The different effects of ferroptotic cancer cells and DCs are related to the stage of ferroptosis; for example, in the early stage of ferroptosis, cancer cells trigger the maturation of DCs. However, in the later stages of ferroptosis, their ability to promote DC maturation is diminished (10). In addition, ferroptosis is closely associated with a wide range of immune cells, including macrophages, fibroblasts, and B, plasma and mast cells. Therefore, a comprehensive exploration of its immune role in HNC could aid in the development of immunotherapies related to HNC. Ferroptosis involves a variety of immune cells in the pathogenesis of HNC, and their coexistence affects the proliferation and progression of HNC cells. The association between ferroptosis and immune cells in HNC is currently not well understood. Therefore, a deeper understanding of the function of ferroptosis in different immune cells is needed.

Ferroptosis combined with conventional therapies in HNC

Promoting ferroptosis may increase the fragility of ferroptosis, which is beneficial for targeted cancer treatment (11). Selective manipulation of ferroptosis to destroy mutated cancer cells is beneficial for cancer treatment (34). The treatment of cancer should not only consider the selective eradication of cancer cells but also minimizing damage to normal cells (9). RCD is an ingenious way to specifically target cancer cells and enhance the efficacy of drug-induced CD while simultaneously reducing adverse effects on normal cells (9). Inducing ferroptosis to target cancer sites can kill cancer cells while keeping normal cells intact (161,176). Inhibition of ferroptosis prevents cancer development, inhibits cancer progression and attenuates the side effects of conventional therapies for cancer (9). Different forms of conventional cancer therapies can trigger ferroptosis. Therefore, the use of ferroptosis induced by these therapies could further increase the effectiveness of cancer treatment. This approach has been shown to have synergistic effects and be well tolerated in preclinical models (11). The combination of ferroptosis inducers and conventional therapies has great potential in cancer treatment (11). Targeting ferroptosis may provide new treatment options for tumors that are difficult to treat with conventional therapies (11).

Radiation can trigger ferroptosis by stimulating LPO (177). Ferroptosis may also lead to normal tissue damage caused by radiation therapy (10). Ferroptosis inducers may also synergize with chemotherapeutic agents to inhibit cancer cell proliferation in HNC (8). Drugs that target ferroptosis can improve the efficacy of radiation, chemotherapy and immunotherapy for cancer treatment (8). Therefore, the combination of ferroptosis with conventional treatment has the potential to increase the efficacy of treatment for HNC while reducing the adverse effects of conventional therapies.

Role of ferroptosis in the prognosis of HNC

Currently, despite notable advances in targeted drug therapy and tumor immunotherapy for HNC, the prediction of HNC prognosis remains inadequate. Seeking new prognostic markers for HNC remains one of the research directions that need to be explored. Effective prognostic assessment can reduce the probability of recurrence, metastasis and deterioration of HNC. Individualized treatment plans and effective prognosis assessment tests are conducive to promoting the development of the field of precision oncology. Detecting the intermediate metabolites of the ferroptosis promotion or inhibition system after the treatment of HNC is helpful for evaluating the prognosis of HNC. Chen et al (178) demonstrated that tribbles pseudokinase 3 (TRIB3) can reduce the levels of Fe2+ and cellular LPO, thereby inhibiting ferroptosis in HNSCC cells. Targeting TRIB3 to increase ferroptosis in HNSCC cells may effectively inhibit cancer progression, suggesting that TRIB3 is a potential prognostic marker and therapeutic target for HNSCC. Jia et al (179) constructed a prognostic model for ferroptosis in HNSCC. Their findings indicate that heat shock protein family A member 5 is associated with poor prognosis in patients with HNSCC, and high levels of autophagy-related gene 5 expression may be linked to low overall survival in patients with HNSCC. High expression of glucose-6-phosphate dehydrogenase in HNSCC may be associated with lymph node metastasis and prognosis, and aurora kinase A may serve as a therapeutic target and prognostic marker related to ferroptosis in HNSCC (Fig. 4). These findings suggest that ferroptosis plays an important role in the prognosis of HNSCC.

Conclusions and outlook

Ferroptosis is a double-edged sword. On the one hand, it can stop the spread of disease; on the other hand, uncontrolled ferroptosis can lead to uncontrolled cellular damage as well as inadequate immune responses. Therefore, careful consideration is needed to determine the need to inhibit or activate ferroptosis. Careful evaluation of treatment strategies is needed to determine the optimal duration of treatment, therapeutic dose and indications during the application of ferroptosis to HNC treatment. However, a series of issues, such as whether ferroptosis produces drug resistance and/or off-target effects, and how to improve selectivity in the treatment of HNC, are worthy of careful consideration. Since ferroptosis plays a dual role in the occurrence and development of cancer, how to utilize its advantages and avoid its disadvantages is one of the key factors for exploring its application in the treatment of HNC. Specifically, a series of problems needs to be solved through continuous research. For example, it may be possible to develop specific ferroptosis inhibitors that can effectively treat HNC, and selectively induce ferroptosis during HNC treatment. Whether combining ferroptosis-related HNC treatments with existing HNC treatments may produce improved outcomes, as well as how the adverse effects of ferroptosis in the treatment of HNC can be avoided, are questions to be addressed by future research. It may be possible to develop predictive biomarkers to accurately predict the response of HNC to the induction of ferroptosis.

To address current challenges, future research should prioritize the following directions: i) Drug resistance and off-target effects: Systematic studies are needed to determine whether ferroptosis inducers or inhibitors induce resistance or off-target effects. High-throughput screening approaches may identify compensatory pathways, while nanoparticle-mediated delivery may enhance selectivity; ii) combination therapy: The synergistic effects of ferroptosis modulators with conventional therapies (such as radiotherapy, chemotherapy and immunotherapy) should be evaluated in preclinical models. For example, ferroptosis inducers may overcome cisplatin resistance by depleting GSH in HNC, but this effect needs to be validated in vivo; iii) predictive biomarker development: Lipidomics and transcriptomics can identify biomarkers (such as ACSL4 expression levels and LPO levels). Clinical validation of these biomarkers is critical for translational applications; and iv) homeostasis restoration: Studying the interaction between ferroptosis and cellular homeostasis (such as autophagy and ERS) may reveal new drug targets to mitigate adverse effects.

Future research priorities should include the following: i) Strengthening mechanistic studies to elucidate the regulatory pathways of ferroptosis in HNC subtypes; ii) preclinical trials combining ferroptosis modulators with standard treatments; and iii) standardized detection methods for quantifying ferroptosis in clinical samples. By addressing those points, ferroptosis modulation has the potential to develop into a precision medicine tool for HNC. However, a deeper understanding of the role of ferroptosis in HNC needs to be obtained in future studies. A comprehensive understanding of the role of ferroptosis in HNC induction and inhibition can help improve strategies for HNC treatment. With respect to the potential dual regulatory role of ferroptosis in HNC, future studies should aim to shed light on this emerging area. However, further research is still needed to answer remaining questions related to ferroptosis. For example, GPX4 is a steward gene and its absence is irreversible. Therefore, antiferroptosis drugs can temporarily inhibit CD and ferroptosis may still occur once these drugs are discontinued (21). This is still a challenge that needs to be addressed. It may be possible to improve the current understanding of diseases while using ferroptosis to control their progression. Clarifying the association between cell homeostasis and ferroptosis may contribute to the development of drugs that inhibit CD or new ways of restoring cell homeostasis. Clarifying the mechanism and application of ferroptosis allows us to utilize or engineer ferroptosis to selectively kill unwanted and abnormal cells to treat diseases, which is a promising option. The diseases and mechanisms associated with ferroptosis are still being explored, and more in-depth studies may lead to a clearer understanding of the association between ferroptosis and HNC, and the development of highly effective therapeutic regimens related to ferroptosis.

Availability of data and materials

Not applicable.

Authors' contributions

CK wrote this manuscript, conducted the literature search and participated in the conceptualization of the study, methodology, formal analysis and writing-review and editing. XL contributed to formal analysis and writing-review and editing. XW was involved in formal analysis and writing-review & editing. Data authentication is not applicable. All of the authors have read and agreed to the final version of the manuscript.

Ethics approval and consent to participate

Not applicable.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Acknowledgments

The figures were created using PowerPoint (2019 version; Microsoft Corp.).

Funding

This work was supported by grants from the Natural Science Foundation of China (grant no. 82260475), Major Scientific Research Project on Science and Technology Innovation in Gansu Province Health and Wellness Sector (grant no. GSWSZD2024-02), Key Research and Development Program of Gansu Provincial Science and Technology Department-International Cooperation Field Projects (grant no. 25YFWA028) and Lanzhou City Science and Technology Development Guiding Program Project (grant no. 2024-9-16).

References