Lipid metabolism is related closely to ferroptosis, and the accumulation of ROS produced by lipid peroxidation in cells is a central step in ferroptosis involving non-enzymatic (i.e., Fenton) reaction pathways (21–23). PUFAs are crucial components of cell membranes that provide the main substrate for lipid peroxidation in ferroptosis (22–24). Their abundance determines the availability of lipid peroxidation sites, which in turn determines the susceptibility to ferroptosis (25). Under the action of acetyl coenzyme A synthetase long chain family member 4 (ACSL4), PUFAs bind to CoA and are esterified and inserted into the cell membrane by lysophosphatidylcholine acyltransferase 3 (LPCAT3). They are then catalyzed by lipoxygenases (LOXs) and undergo peroxidation at the allylic position to produce and accumulate lipid peroxide, which eventually leads to cell death (26).

Exogenous monounsaturated fatty acids (MUFAs) can replace PUFAs on phospholipids located on the cell membrane or other membrane structures, thereby reducing the sensitivity of these membranes to the deactivation of GPX4 and accumulation of ROS (27, 28). ACSL4 plays an important role in determining the sensitivity of cells (including mesenchymal cancer and ccRCC cells) to ferroptosis, and the reduction of ACSL4 and LPCAT3 expression in cells can reduce the accumulation of substrate for lipid peroxidation, thereby inhibiting ferroptosis (29, 30). In VHL-deficient ccRCC, lipophilic statins have a “synthetic lethal” effect, inhibiting mevalonate pathway biosynthesis (31). Statins reduce the expression of GPX4 by inhibiting HMG-CoA reductase, leading to the production of lipid ROS and induction of ferroptosis (32).

Iron is involved in many physiological and pathological processes (e.g., ferroptosis and ferritin phagocytosis), while iron homeostasis is strictly regulated to prevent harmful effects on organs caused by iron excess or deficiency (33). Different cell types use different strategies to capture various iron subtypes for participation in their intracellular metabolic activities (34). Iron metabolism in cells influences the cells’ susceptibility to ferroptosis (35). The human body contains two iron ions: Fe²⁺ and Fe³⁺ (36). Fe²⁺ is absorbed through the intestinal epithelium via divalent metal ion transporter 1 (DMT1) and then transported to the circulation through ferroportin-1 (FPN1) (37). It is oxidated to Fe³⁺ by ceruloplasmin and binds to transferrin (Tf) in the serum (38). Upon recognition by the transferrin receptor TFRC1 on the cellular membrane, Tf enters the cell through endocytosis. After dissociation with Tf in the endosome, Fe³⁺ is reduced to Fe²⁺ by the metalloreductase six-transmembrane epithelial antigen of prostate 3 (STEAP3). DMT1 then mediates the transport of Fe²⁺ from the endosome to the cell labile iron pool (LIP) (39). Ferritin, which has two subunits – ferritin light chain 1 (FTL1) and ferritin heavy chain 1 (FTH1) – can limit ion participation in redox reactions by forming a complex with iron and proteins (40).

Nuclear receptor coactivator 4 (NCOA4) binds to ferritin and mediates its delivery to autophagosomes (41). When autophagosomes fuse with lysosomes, ferritin is degraded and iron is released. Free Fe²⁺ participates in a series of physiological cell processes, including the generation of ROS through Fenton reactions, which further promotes lipid peroxidation (42). The down-regulation of NCOA4 in cells inhibits ferritin autophagy and ferroptosis, whereas NCOA4 overexpression increases ferritin degradation and promotes ferroptosis, suggesting that NCOA4 is involved in the interaction between autophagy and ferroptosis (43). In addition to participating in Fenton reactions, iron plays roles in other cellular metabolic pathways, such as by being transported to the mitochondria for the synthesis of heme or iron-sulfur (Fe-S) clusters; thus, iron participates in the regulation of iron-mediated cell death in an environment-dependent manner (44).

Ferroptosis-related signature and diverse programmed cell death was used to predict the prognosis and drug sensitivity of various type of cancer (96, 97). Ferroptosis-related genes (FRGs) that are abnormally expressed in ccRCC cells relative to normal cells include CARS1, CHAC1, CD44, STEAP3, AKR1C1, DPP4, SLC7A11, SLC1A5, and NCOA4 (98–101). A survival model for patients with ccRCC based on four of these genes (CD44, DPP4, NCOA4, and SLC7A11) demonstrated excellent prognostic efficacy, and risk scoring was an independent factor for prognosis prediction (102).

According to FRG expression profiles, patients can be divided into groups with different degrees of immune infiltration and PD-L1 expression. The CARS level correlates positively with PD-L1 expression and may induce the trans-sulfur pathway, leading to cysteine deprivation and ferroptosis inhibition (103). Based on the mRNA expression profiles of 57 FRGs, a ferroptosis scoring system reflecting different prognoses and degrees of immune cell infiltration for ccRCC can be created to predict the sensitivity of patients to different treatments, illustrating that ferroptosis may involve multiple complex tumor microenvironments (98). However, many FRGs identified by bioinformatics remain to be verified experimentally. The FRGs identified by bioinformatics or by experiments in ccRCC are shown in Table 2 .

By binding with KEAP1, DPP9 promotes the stability of NRF2 and overcomes oxidative stress, and the accumulation of DPP9 can lead to the hyperactivation of the NRF2 pathway and inhibit ferroptosis, thereby driving ccRCC development and sorafenib resistance (68). Targeting the Nrf2/Heme oxygenase-1 pathway can significantly increase oxidative stress in RCC and promote apoptosis and death of cancer cells (104), while Heme oxygenase-1 has a dual role in ferroptosis induction (105). The farnesyl-diphosphate farnesyltransferase gene has the potential to be used as a biomarker for the ccRCC diagnosis and to treat ccRCC by regulating ferroptosis (105).

Long non-coding RNAs (lncRNAs) are longer than 200 nucleotides, lack protein translation ability, and are expressed specifically in some tumor tissues (106). They regulate gene expression through transcription, translation, protein modification, and RNA–protein complexing (107). In tumor cells, lncRNAs can act as oncogenes or tumor suppressor genes by regulating various signaling pathways (108). LncRNAs act as biomarkers of or therapeutic targets for cancer based on their specific signaling of certain cell states (109–112). Recent studies have shown that the lncRNA 00312/miR-34a-5p/ASS1 axis reduces proliferation and invasion and promotes apoptosis in RCC (113), and that the SNHG12/SP1/CDCA3 ferroptosis-related long non-coding RNA (FRIncRNA) axis promotes RCC progression and sunitinib resistance, thereby serving as a new target for the treatment of sunitinib-resistant RCC (114).

The expression patterns of FRIncRNAs can be used to predict prognoses, and a risk model constructed using patient-specific FRlncRNAs has shown good diagnostic and prognostic value (115). The p53 signaling pathway, TNF-mediated signaling pathways, cytokine-cytokine receptor interaction, and helper T-cell immune response signaling pathways are associated with high-risk FRlncRNAs (116). The expression characteristics of lncRNAs are related to the immune infiltration of ccRCC, and immune checkpoint gene expression differs significantly between low-risk and high-risk groups, which may be related to patient sensitivity to immune and targeted therapies (117).

As a key regulator of ferroptosis, SLC7A11 is up-regulated in ccRCC, chRCC, and pRCC, and its overexpression is associated with their poor prognosis and immune cell infiltration (118). SLC7A11, HMOX1, and MT1G were the 3 FRGs associated with the prognosis of adrenocortical carcinoma (ACC), KICH, KIRC and KIRP, while the SLC7A11 is a prognostic risk factor for above 4 different renal tumors, and high expression of MT1G increases the prognostic risk of ACC, ccRCC and chRCC (119).

GCLC, HSBP1 and ACSF2 are independent prognostic FRGs in pRCC (120, 121). The FRGs features composed of CDKN1A, MIOX, PSAT1 and RRM2 are independent prognostic indicators of pRCC, and have a good predictive ability for the prognosis of pRCC patients (122).

For chRCC, targetable pathways for innate lymphoid cell/IL-15 and cysteine homeostasis/ferroptosis have recently been discovered (123). TFRC and SLC7A11 are highly expressed FRGs in chRCC tissues and can predict the clinicopathological features and prognosis of chRCC (124). Gamma-glutamyltransferase 1 (GGT1), a key enzyme in glutathione homeostasis, is significantly down-regulated in chRCC compared with normal kidneys, while overexpression of GGT1 can inhibit the chRCC cells proliferation, and inhibit the cystine uptake and GSH levels (125).

ccRCC is the most common type of kidney tumor, and traditional treatment methods for it include surgical resection, chemotherapy, and radiation therapy (126). As ccRCC is often associated with VHL gene mutations, targeted drug therapies have show greater efficiency and safety than traditional therapies which offer new hope for kidney cancer patients (127). However, VHL deletion alone is usually not sufficient to drive tumor development; the co-deletion of VHL and one (or more) of the common and important drivers of ccRCC (polybrominated 1 gene, BRCA1-associated protein 1, and SET domain 2) is often required (128, 129). In addition to the VHL-HIF-VEGF/vascular endothelial growth factor receptor (VEGFR) pathway, VHL mutations affect signaling regulation in the mTOR and epidermal growth factor receptor (EGFR) pathways, promoting tumor growth (130). HIF signaling changes the extracellular matrix, regulates the tumor immune response, increases angiogenesis, and affects the tumor microenvironment, altering the metabolism of tumor and stromal cells and increasing tumor invasiveness and metastasis (131). Traditional radiation treatment and chemotherapy are almost completely ineffective for RCC, and the development of targeted drugs against VEGFR and its ligand VEGF is considered to be the primary therapeutic intervention required for patients with advanced RCC (132). However, there is increasing evidence that induction of ferroptosis can reverse tumor chemotherapy resistance (133). Bevacizumab was the first anti-angiogenic drug used in patients with RCC, followed by VEGFR inhibitors such as sunitinib, sorafenib, pazopanib, cabozantinib, and axitinib (101). Beyond that, targeted therapies for DNA damage repair, tumor microenviroment remodeling and tumor cell metabolism reprogramming show potential (134, 135). In March 2021, the FDA approved tivozanib, a potent selective VEGFR-TKI, for the third - and fourth-line treatment of relapsed refractory RCC (136). ccRCC often involves the reprogramming of glucose and lipid metabolism and the tricarboxylic acid cycle, as well as the alteration of tryptophan, arginine, and glutamine metabolism, all of which may provide potential biomarkers or treatment strategies (137). However, resistance to anti-angiogenic therapy often develops in the clinical treatment of ccRCC due to the crosstalk between VEGFR and other tyrosine kinases or downstream pathways; the use multi-target (e.g., HIF) inhibitors and combination strategies carries more potential for the treatment of ccRCC (138).

Crosstalk between different cell death pathways may also be related to resistance to targeted therapy, and the induction of different types of cell death, such as ferroptosis, through targeted therapy has become a new strategy in kidney cancer treatment (139). The metabolic reprogramming properties of ccRCC give it cysteine- and GPX4-dependent sensitivity to ferroptosis, and cysteine deprivation and GPX4 inhibition can induce cell death (51, 80). In addition, some commonly used targeted drugs have been found to induce tumor ferroptosis. For example, sorafenib, used commonly in the treatment of liver and kidney cancers, has been reported to have a similar ferroptosis induction effect for RCC cells to that of erastin (140). Disulfiram/copper can synergistically inhibit the RCC cells growth and induce ferroptosis with sorafenib (141). The accumulation of DPP9 leads to the overactivation of NRF2 and its transcriptional target SLC7A11, which inhibits ferroptosis and induces the resistance of ccRCC cells to sorafenib (142). In addition to sorafenib, sunitinib and glycogen synthase 1 inhibitors can produce synthetic lethal effects in RCC cells (143). Inhibition of fibroblast growth factor receptor 4 significantly reduces glutathione synthesis, leading to excessive ROS production and unstable iron pool accumulation, which triggers ferroptosis to overcome drug resistance of HER2-positive breast cancer (144). Everolimus, an mTOR signaling pathway inhibitor typically used in tumor-targeted therapy, inhibits the mTOR–3EBP4 axis, reduces GPX4 expression, and acts synergistically with RSL3 or erastin to induce ccRCC cell ferroptosis (145).

Activation of ferroptosis by IL-6 is may provide a new strategy for the treatment of advanced RCC and new ideas for TKIs resistance (146). Lysine acetyltransferase 7 inhibits ccRCC by regulating cell cycle and ferroptosis sensitivity (147). As a key enzyme in the tricarboxylic acid cycle, MDH2 play a non-metabolic role in ccRCC progression by reducing FSP1 protein levels and increasing the ccRCC’s sensitivity to ferroptosis (148). SETD2 catalyzes the interaction of histone H3 lysine 36 trimethylation (H3K36me3) with the ferrochelatase promoter, and knockdown of SETD2 significantly increases erastin sensitivity and promotes ferroptosis in ccRCC (149). With the progress of clinical trials such as KEYNOTE-426, the combined use of immune checkpoint and protein tyrosine kinase inhibitors is now one of the standard treatment options for most patients with metastatic RCC (150).

Immune therapies, particularly the use of immune checkpoint inhibitors (e.g., anti-PD-1, anti-PD-L1, and CTLA-4 antibodies; CAR-T cell adoptive immunotherapy; and cancer vaccines) are used commonly and represent significant advances in the treatment of malignant tumors (151).

In addition to the traditional perforin–granzyme and Fas–Fas ligand cell death induction pathways, interferon-γ secreted by CD8+ T cells activated in anti–PD-L1 therapy can down-regulate System X_c ^– expression to reduce cystine uptake and induce ferroptosis (152). The release of damage-associated molecular patterns by tumor cells undergoing ferroptosis also mediates antitumor immunity, and immune therapy can synergize with targeted therapy or radiotherapy through ferroptosis (153, 154). USP8 inhibitor combined with ferroptosis inducer can significantly inhibit tumor growth and promote the infiltration of CD8+T cells in the tumor, and further improve the therapeutic effect of PD-1/PD-L1 blocking antibodies (155). Activation of ferroptosis and inhibition of myeloid suppressor cells (MDSC) combined with anti-PD-1 monoclonal antibody can greatly improve the survival rate of liver tumor mouse models and reduce liver metastasis of other tumors in situ (156). Intermittent deprivation of methionine in the diet significantly enhances tumor sensitivity to ferroptosis and collaborates with PD-1 antibodies to inhibit tumor progression and enhance T-cell-mediated anti-tumor immune response (157). However, ferroptosis may be a double-edged sword for tumor cells, which can enhance not only the function of anti-tumor immune cells but also the tumor-promoting response of immunosuppressive cells under certain circumstances (158). Thus, ferroptosis interacts with the immune system, and the combined use of immune and ferroptosis-targeting therapy for RCC is anticipated in the future.

PD-1 is highly expressed in infiltrating lymphocytes in various tumors, and PD-1 ligands are often upregulated on the surfaces of tumor cells and inhibit the local antitumor immune response mediated by T cells (159). PD-L1 was found to be expressed in 24.2% of RCC tumors, and the level of PD-L1 expression correlated positively with the risk of patient death (160). Recently, the combined use of targeted drugs and immune therapy has yielded favorable outcomes, and combined anti-angiogenic therapy and anti-PD-1/PD-L1 pathway targeting has been shown to improve patients’ clinical outcomes (161). By promoting dendritic cell differentiation, reducing Treg cell aggregation, increasing cytotoxic T-cell abundance, and increasing tumor susceptibility to immune-mediated cytotoxicity, targeted therapy and tumor immunity are believed to interact with each other (162–165). In addition to the use of TKIs combined with PD-1/PD-L1 inhibitors, dual immunotherapies (ipilimumab with nebuliumab and atrilizumab with bevacizumab) are currently used for the first-line treatment of advanced renal cancer (166). RCC-associated drugs, targets, and tumor microenvironment associations are summarized in Figure 2 .

Nanoparticles (NPs) are a new form of cancer treatment delivery that overcome some limitations of traditional anti-cancer drugs and can induce various cancer cells to undergo ferroptosis through various means (167, 168). Due to the unique biophysical and chemical properties of metal-based nanomaterials after entering tumor cells, they can also induce ferroptosis and enhance the ability of catalytic therapy (169). Iron oxide (Fe₃O₄) NPs produce toxic heat effects and induce ferroptosis via ACSBG1, an acyl-CoA synthetase that acts as a key pro-ferroptotic factor in this tumor-specific process (170). MIL-101 (Fe)@RSL3 NPs deliver iron ions and RSL3 to ccRCC cells, which generate large amounts of ROS through Fenton reactions, leading to the accumulation of LOX, and as GPX4 activity is inhibited by RSL3, this combination induces cell death by ferroptosis (171). Although nanotherapy has many advantages, some problems remain: the complexity and heterogeneity of tumor biology lead to uncertainty about its efficacy, as the interaction of NPs with this biology is not yet clear, and challenges related to NP cost, chemical processes, manufacture, and control persist (172, 173). Furthermore, current research is mostly limited to animal models, and gap between research and clinical application remains (172, 173).

Natural compounds are used widely in the treatment of various cancers, including RCC (174). Luteolin, a natural flavonoid found widely in fruits and vegetables, promotes ferroptosis via GSH depletion and ROS accumulation through Fenton reactions (175). Icariside II upregulates miR-324-3p and inhibits GPX4 expression to induce ferroptosis in ccRCC (176). Artemisinin inhibits resistant ccRCC growth by downregulating GSH and GPX4 expression (177). Curcumin increases ADAMTS18 gene expression and restores ccRCC sensitivity to sunitinib, and this regulation can be reversed by ferroptosis inhibitors, indicating that curcumin mediates ferroptosis in ccRCC through ADAMTS18 (178). Ursolic acid and sorafenib can synergically inhibit various tumor, which may be related to the induction of Mcl-1-related apoptosis and SLC7A11-dependent ferroptosis (179). Erianin can induce ferroptosis of RCC stem cells by promoting N6-methyladenosine modification of ALOX12/P53 mRNA, thus may play a role in RCC treatment (180). Sanguinarine reduces the expression of GPX4 and increases the expression of ACSL4, thereby inducing the ferroptosis of RCC cells (181). Ginsenoside Rh4 can increase the sensitivity of RCC cells to ferroptosis by inhibiting the expression of NRF2 signaling pathway and antioxidant enzymes (182). Overall, the natural compounds associated with ferroptosis provide numerous new structural bases for the treatment of RCC. Various compounds and genes targeting ferroptosis in RCC are summarized in Figure 3 .