Biliary tract cancer (BTC) includes a spectrum of invasive malignancies that arise from the bile duct epithelium. It is relatively uncommon but highly heterogeneous, comprising gallbladder cancer (GBC; which arises from the gallbladder or cystic duct) and cholangiocarcinoma (CCA; which arises from the intrahepatic, perihilar or distal biliary tree). BTC accounts for <1% of all human types of cancer and ~3% of digestive system tumors globally (1,2). In recent years, the incidence of BTC has been increasing globally, particularly in Asian countries, where it represents an important health problem (3,4). Globally, adenocarcinoma accounts for ~90% of all BTC cases (5,6). BTC is usually diagnosed at advanced stage. The majority of patients with symptomatic BTC, such as those presenting with biliary obstruction, have incurable tumors with invasion of the surrounding organs, lymph nodes and distant metastases, which results in poor prognoses (2,7). At present, surgery remains the gold standard for treating early-stage tumors; however, only a limited number of patients have this opportunity (8,9). The 5-year overall survival rate of BTC is <20% for patients from the United States of America and Europe (10). Furthermore, despite undergoing curative resection with a negative margin (R0), BTC still have a high recurrence rate of >50% (8,9,11). Gemcitabine-platinum chemotherapy is a recommended first-line treatment for advanced BTC; however, its efficacy is not satisfactory (12). The emergence of targeted therapies and immunotherapies, such as Durvalumab and Pembrolizumab, has revolutionized the treatment paradigm of malignant tumors. Immunotherapy plus chemotherapy has been verified to be effective in treating BTC (13-15). Investigating the key molecular mechanisms and identifying novel therapeutic targets for BTC is imperative.

Cell death can occur either due to unregulated causes or following a regulated process (Fig. 1) (16). Unregulated cell death is the instantaneous demise of cells due to severe physical, chemical or mechanical causes. Unlike unregulated cell death, regulated cell death (RCD) is characterized by organized signaling cascades and specific molecular mechanisms, suggesting that it can be controlled (17). The process of RCD can be influenced, either delayed or accelerated, through pharmacological or genetic interventions. A study by Dixon et al (18) first proposed the notion of 'ferroptosis' in 2012 (19). Ferroptosis is a new type of RCD, differing from necroptosis (Fig. 1A) (20), apoptosis (Fig. 1B) (21), pyroptosis (Fig. 1C) (22) and autophagy (Fig. 1D) (23) in morphology, genetics and biochemistry (24,25). As its name implies, ferroptosis is an iron-dependent mechanism of cell death (26). Ferroptosis is triggered by lipid peroxidation (LPO), which results from an imbalanced cellular metabolism and redox homeostasis. The morphological evidence of ferroptosis includes the presence of small mitochondria with increased mitochondrial membrane densities, reduced or missing mitochondrial crista, outer mitochondrial membrane rupture and a normal nucleus (Fig. 1E) (27,28). Ferroptosis is indicated to serve an important pathophysiological role in the development and occurrence of numerous diseases, such as ischemic heart disease and acute kidney injury, but particularly in cancer (19,29). Non-small cell lung cancer, hepatocellular carcinoma, pancreatic cancer, breast cancer and renal cell carcinoma are reported to be sensitive to ferroptosis (30-34). Additionally, various ferroptosis inducers, such as sorafenib, sulfasalazine and artemisinin, demonstrate a mitigation of tumor progression (35). Due to the issue of drug resistance, there is a growing interest regarding the potential of targeting ferroptosis as a therapeutic strategy for malignancies (36-38).

The present review aimed to provide a comprehensive overview of the signaling pathways and molecular mechanisms of ferroptosis, the research progress on ferroptosis in the occurrence and development of BTC, and the potential application of ferroptosis as a targeted therapy for BTC.

Following the first description of ferroptosis, a form of non-apoptosis and iron-dependent RCD, in 2012, there has been an increasing interest to investigate the process and regulation of ferroptosis (39). Previously, several studies investigated the mechanisms involved in ferroptosis (40-42). The present study aimed to review the core mechanisms of ferroptosis based on three aspects, namely iron metabolism, lipid metabolism and antioxidant defense systems.

The TME serves pivotal roles in the initiation, progression, metastasis and therapeutic resistance of tumors (72-75). It is a dynamic interacting network that is comprised of diverse cellular components, including malignant cells, immune cells, cancer-associated fibroblasts and endothelial cells, alongside acellular elements such as the extracellular matrix, cytokines and metabolites (76). Ferroptosis engages in a complex bidirectional crosstalk with the TME, particularly through metabolic reprogramming and immunomodulation, which regulates BTC malignancy and treatment responses (77,78).

Ferroptosis in malignant cells modulates antitumor immunity within the TME through immunogenic signaling (79,80). Previous evidence indicates that ferroptotic cancer cells release damage-associated molecular patterns (DAMPs), a hallmark of immunogenic cell death (ICD), which function as endogenous adjuvants to potentiate antitumor immunity (81,82). As well as the classic DAMPs, such as high mobility group protein 1 (83-85), adenosine triphosphate (86,87), calcium reticulum protein (84,86) and the recently identified proteoglycan decorin (88), ferroptotic cancer cells release immunomodulatory cytokines such as C-X-C motif chemokine ligand 1 (CXCL1), tumor necrosis factor (TNF) and interferon (IFN)-β, that contribute to the remodeling of the TME (89). However, whether ferroptosis is a classic form of ICD is still an ongoing scientific debate (89). In addition, the specific role of ferroptotic cancer cells in antitumor immunity is also contradictory. A previous study demonstrates that tumor cell competition for cystine compromises the function of CD8+ T cells within the TME. Cystine deprivation triggers glutamate accumulation, which potentiates CD36-mediated LPO (90). However, ferroptotic cells can activate antitumor immunity by expressing 1-steaoryl-2-15-HpETE-sn-glycero-3-PE. This serves as a signal that mediates phagocytosis by engaging Toll-like receptor 2 on macrophages (91) and enhancing the activities of natural killer cells (92).

Immune cells exhibit differential susceptibility to ferroptosis within the TME, and their functional heterogeneity enables distinct populations to either potentiate or suppress ferroptosis in malignant cells (80). CD8+ T cells exhibit an increased sensitivity to GPX4 inhibition compared with malignant cells, resulting in a premature impairment of antitumor immunity prior to notable tumor cell death (93). However, these effector lymphocytes demonstrate relative resistance to system Xc⁻ inhibitors, such as sulfasalazine, as inhibition of system Xc⁻ did not affect T cell proliferation and antitumor immune responses (94). Activated CD8+ T cells with high expression levels of fatty acid transporter CD36 increases the risk of LPO and ferroptosis (95). Compared with tumor-derived CD8+ T cells, tumor-infiltrating regulatory T cells (Tregs) typically have reduced levels of LPO, suggesting that they are less prone to ferroptosis in the TME (93). However, system Xc⁻ inhibitors rarely impair the viability of Tregs (96). Tumor-associated macrophages (TAMs) are one of the most abundant types of immune cells in the TME and include the proinflammatory antitumor M1 and anti-inflammatory protumor M2 types (97,98). The sensitivity of these two types of cells to ferroptosis is different due to the high levels of inducible nitric oxide synthase and nitric oxide-free radicals in M1 type compared with M2 type TAMs (99). Another group of immune-suppressive cells, namely myeloid-derived suppressor cells (MDSCs), exhibit notable resistance to ferroptosis, which is associated with the upregulation of system Xc⁻ and neutral ceramidase N-acylsphingosine amidohydrolase expression levels (100).

Metabolic reprogramming is a defining hallmark of malignancy, triggering cancer metastasis, aggressiveness and therapeutic resistance (101). Remodeled metabolic environments and immunosuppressive environments enable cancer cells to evade RCD (78). This metabolic reprogramming extends beyond canonical adaptations such as the Warburg effect (in which preferential glycolysis reduces mitochondrial ROS generation), and includes the suppression of pyruvate dehydrogenase, limiting the acetyl-CoA flux for PUFA biosynthesis, which is critical for ferroptosis (102-104). Within the TME, MDSCs increase immunosuppression through IL-6/JAK2/STAT3-mediated SLC7A11 upregulation, which depletes the extracellular cystine pools (105). This metabolic competition induces a GSH deprivation-induced ferroptosis in CD8+ T cells, further inactivating antitumor immunity (90).

To resist ferroptosis, cancer cells use lipid metabolic reprogramming as a defense strategy via two primary mechanisms: i) PL membrane restructuring via the activation of the sterol regulatory element-binding protein-1-SCD1 axis, replacing oxidation-vulnerable PUFAs with MUFAs (106); and ii) mevalonate pathway-derived antioxidant production, such as isopentenyl pyrophosphate from cholesterol metabolism, which neutralizes lipid peroxides (107,108). Iron homeostasis is also inhibited through TFR1-mediated uptake and ferritin heavy chain (FTH)-dependent storage, which reduces the levels of cytotoxic free Fe²+ (109,110). Furthermore, gut microbiota-derived metabolites, such as short-chain fatty acids (such as butyrate) and tryptophan metabolites (such as indole-acrylic acid), also affect the TME and influence the resistance of cancer cells to ferroptosis (111,112).

Epigenetic regulation is a heritable mechanism that regulates the expression of genes through chemical modifications without altering the DNA sequence. It primarily involves DNA methylation, histone modifications, non-coding RNA (ncRNA)-mediated regulation and RNA methylation (113).

Several key genes are dynamically controlled by DNA methylation. For example, in normal tissues, the core enzyme GPX4, which is essential for eliminating lipid peroxides, has notably reduced expression levels due to transcriptional repression caused by the hypermethylation of CpG islands within its promoter region compared with cancer tissues. This suppression increases cellular susceptibility to ferroptosis (114-117). By contrast, the promoter of the system Xc⁻ light chain subunit SLC7A11 is frequently hypomethylated, leading to its elevated expression levels (118,119). This enhances cystine uptake and GSH synthesis, which inhibits ferroptosis. Furthermore, hypermethylation of the FSP1 promoter results in a reduction to its expression levels, which sensitizes cells to ferroptosis (120,121).

RNA methylation, the most abundant post-transcriptional modification in eukaryotic mRNA, also contributes to ferroptosis regulation. AlkB homolog 5 regulates glutamate-cysteine ligase modifier subunit (GCLM) mRNA levels though N⁶-methyladenosine (m⁶A) modification of GCLM and YTH m⁶A RNA binding protein-mediated decay of GCLM, which regulates ferroptosis (122). Additionally, methyltransferase-like protein 16 (METTL16) serves a role in ferroptosis and tumorigenesis by catalyzing the m⁶A modification of Sentrin/small ubiquitin-like modifier-specific protease 3 mRNA and regulating the stability of lactotransferrin (123).

Histone modifications serve an important role in shaping chromatin structure. It exerts precise transcriptional control over ferroptosis-associated genes, such as SLC7A11 and ACSL3, and impacts gene expression and cancer development (124-126). p53 promotes the nuclear translocation of a histone H2B monoubiquitylation (H2Bub1) deubiquitinase, ubiquitin-specific protease 7 (USP7), which negatively regulates H2Bub1 levels. The interaction between p53 and USP7 reduces the occupancy of H2Bub1 at the regulatory region of SLC7A11, leading to transcriptional repression of SLC7A11and enhancing the ferroptosis induction sensitivity of cells (125). Histone acetylation and methylation also demonstrate bidirectional regulatory effects on ferroptosis (126-129).

ncRNAs regulate the expression of ferroptosis-associated genes via post-transcriptional regulation, forming protein-RNA interaction networks. Specifically, microRNA (miR)-137/miR-9 and miR-15a-5p promote ferroptosis by targeting SLC7A11 and GPX4 mRNA, respectively (130,131), whereas miR-27a and miR-4717 exert anti-ferroptosis effects by targeting ACSL4 and NCOA4 (132,133). Long ncRNAs (lncRNAs) also participate in ferroptosis regulation by acting as protein decoys, chromatin modifiers or miRNA sponges (134,135).

Ferroptosis induction has emerged as a novel strategy to overcome tumor therapy resistance (35,36). However, tumor cells establish complex ferroptosis defense networks by remodeling immune-metabolic environments, notably elevating the threshold for ferroptosis initiation (78). The development of this resistance limits the clinical application of ferroptosis-inducing therapies (78). However, a series of innovative strategies have been developed, providing a theoretical foundation for designing effective treatments to circumvent resistance. Recently identified, reversible palmitoylation modification of GPX4 is a key regulator of ferroptosis. The palmitoylation inhibitor 2-bromopalmitate notably enhances the antitumor efficacy of ferroptosis inducers by inhibiting GPX4 palmitoylation (136). Furthermore, targeted protein degradation represents an emerging therapeutic paradigm. The GPX4-targeted autophagy-targeting chimera (GPX4-AUTAC), based on selective autophagic degradation, induces ferroptosis and demonstrates antitumor activity in vitro, in vivo and in patient-derived organoids (137). In addition, proteolysis-targeting chimeras that target GPX4/DHODH have also been designed (138,139). The use of combination strategies that target immune checkpoints alongside ferroptosis inducers are also gaining attention (140). Furthermore, the development of novel nanodrug delivery systems offers additional approaches to overcome chemoresistance, such as high-density lipoprotein nanoparticles with dual metabolic disruption (selenium deprivation and LPO) (141) and antibody-targeted nanoplatforms (such as Fe-MOF@Erastin@ Herceptin), which deliver ferroptosis inducers (such as erastin) specifically to tumor cells (142).

Ferroptosis is demonstrated to have therapeutic potential in various types of cancer, including hepatocellular carcinoma, lung carcinoma, lymphoma, pancreatic ductal carcinoma and renal cell carcinoma (30-34). BTC is an uncommon type of gastrointestinal malignancy that has a poor prognosis. In the early stages of disease, surgical resection with negative margins can effectively treat BTC (8,9). For unresectable and metastatic disease, gemcitabine and cisplatin chemotherapies plus immunotherapy (such as with Durvalumab or Pembrolizumab) are the only preferred regimens approved by the National Comprehensive Cancer Network (143,144). With the increasing number of patients with BTC, investigating novel targets and alternative options is crucial. Ferroptosis, which is considered to be one of the most promising potential antitumor methods, can affect the occurrence and development of BTC by regulating intracellular iron and ROS levels, providing new treatment options for patients with BTC (70,124,145). The following section primarily concentrates on the advancements in ferroptosis-associated research in BTC, including potential therapeutic targets, treatment options and associated mechanisms (Table I; Fig. 3).

TP53 is acknowledged as a classic tumor suppressor gene, which serves important roles in controlling the cell cycle, cell proliferation and cell death (146,147). TP53 serves a role in controlling the susceptibility to ferroptosis via both transcription-dependent and transcription-independent mechanisms (148). Shank-associated RH domain interacting protein (SHARPIN) is a crucial part of the complex responsible for activating the linear ubiquitin chain, which inhibits ferroptosis through the p53/SLC7A11/GPX4 signaling pathway and promotes cell proliferation in CCA. Silencing the SHARPIN gene results in the suppression of p53 ubiquitination and degradation, as well as a decreased expression of SLC7A11, GPX4, superoxide dismutase (SOD)-1 and SOD-2. Targeting SHARPIN may serve as a potential strategy benefiting CCA treatment (149). Human hydroxysteroid dehydrogenase-like 2 (HSDL2) is also a regulator of cancer progression and lipid metabolism. Knocking down HSDL2 promotes CCA progression by inhibiting ferroptosis through the p53/SLC7A11 axis (150). As a member of Runt-domain family, Runt-related transcription factor 3 (RUNX3) has a reduced expression level in GBC. This downregulation, mediated by promoter DNA hypermethylation, is notably associated with adverse outcomes in patients with GBC. RUNX3 activates the transcription of inhibitor of growth protein 1, which leads to the suppression of SLC7A11 through a p53-dependent pathway and the induction of ferroptosis (151). Golgi phosphoprotein 3 also promotes CCA malignancy by inhibiting ferroptosis through SLC7A11 upregulation (152). Additionally, TP53-induced glycolysis and apoptosis regulator (TIGAR) is associated with adverse outcomes and ferroptosis resistance in patients with intrahepatic CCA (iCCA). TIGAR encodes an enzyme responsible for regulating glycolysis and scavenging ROS. Knockdown of TIGAR reduced the expression of GPX4, a key inhibitor of ferroptosis (153). Furthermore, combining TIGAR knockdown with cisplatin treatment synergistically induces a notable increase in ferroptosis (153).

The transcription factor Nrf2 controls the expression of genes involved in counteracting oxidative and electrophilic stresses, which serves a role in regulating the cellular antioxidant response (154,155). A study by Huang et al (156) reports that transcription factor activating enhancer-binding protein 2 α (TFAP2A) may serve a role as a regulator of ferroptosis in GBC via the Nrf2 signaling pathway. GBC cells exhibit elevated levels of TFAP2A, compared with non-tumorigenic human intrahepatic bile duct cells (H69), and knocking down TFAP2A lead to a decrease in GBC cell proliferation, migration and invasion, as well as a decreased expression of oxidative stress-associated genes, such as heme oxygenase 1 and Nrf2. A study by Zheng et al (157) demonstrates high methyltransferase-like 3 expression levels in cisplatin-resistant iCCA cells compared with parental cells. This upregulation enhances m⁶A modifications, leading to the inhibition of ferroptosis and a resistance to cisplatin through the stabilization of Nrf2 mRNA and an increase in the Nrf2 protein expression levels. Additionally, overexpression of METTL16 in patients with CCA is associated with poor prognosis. Mechanistically, METTL16 increases the m⁶A modifications of activating transcription factor 4 (ATF4) mRNA, which increases the expression of ATF4 and subsequently results in the suppression of ferroptosis (158).

At present, GPX4 is recognized as the only enzyme with the ability to directly reduce complex phospholipid hydroperoxide (159). Therefore, GPX4, which is responsible for the efficient removal of phospholipid hydroperoxides, is critical for cell survival (160). When GPX4 fails to effectively remove PLOOHs, there is an increase in LPO and ferroptosis (57,161). GPX4 is also one of the important adverse prognostic indicators in iCCA (162). A study by Lei et al (163) demonstrates that JUND enhances the transcription of linc00976, which promotes the development of CCA and prevents ferroptosis by regulating the miR-3202/GPX4 axis. F-box protein 31 (FBXO31) stimulates the ubiquitination process of GPX4 and consequently promotes the degradation of the GPX4 proteasome, which enhances the occurrence of ferroptosis. Additionally, FBXO31-upregulated cancer stem cell-like cells present enhanced sensitivity to cisplatin compared with control cells (164). In addition, lncRNA paired box 8-antisense RNA 1 binds to p62 and activates Nrf2, which promotes GPX4 transcription and stabilizes GPX4 mRNA by interacting with insulin-like growth factor 2 mRNA binding protein 3. This inhibits ferroptosis and promotes resistance to gemcitabine and cisplatin. Treatment with JKE-1674 (1 μM), a GPX4 inhibitor, combined with gemcitabine (5 μM) and cisplatin (10 μM) exhibits an improved antitumor potential in preclinical models of both subcutaneous tumor and orthotopic mice models, compared with gemcitabine and cisplatin therapy (165).

Previous studies reveal ACSL4 promotes ferroptosis by catalyzing the esterification of long-chain PUFAs into PUFA-CoA, which serve as substrates for LPO (166,167). Data from the Gene Expression Omnibus and The Cancer Genome Atlas databases demonstrates that there is a higher level of ACSL4 in CCA compared with normal adjacent tissues. Additionally, ACSL4 levels are associated with the prognosis of patients with CCA as well as the immune infiltration in CCA (168). Targeting the ACSL4 pathway may be an anti-cancer approach. A study by Liao et al (169) reveals that, via ACSL4-dependent lipid reprogramming, IFNγ from cytotoxic T lymphocytes and AA from the TME induce tumor cell ferroptosis. Furthermore, cancer cells that derive from migration exhibit a higher ferroptosis sensitivity and PUFA lipid contents compared with primary-tumor-derived cells. This supports the possibility of targeting ferroptosis in the treatment of cancer (170). Specific and targeted inhibitors of ACSL4 with anti-ferroptosis function such as AS-252424 have been screened and identified (171). In GBC, sirtuin 3 has tumor suppressive effects through the induction of the expression of ACSL4 and AKT-dependent ferroptosis, and inhibition of the epithelial-mesenchymal transition (172). Another member of the acyl-CoA synthetase long chain family, ACSL3, synthesizes MUFAs that suppress PUFAs peroxidation, which confers ferroptosis protection. Furthermore, ACSL3 is an unfavorable prognostic biomarker in patients with CCA and a mediator of ferroptosis resistance in CCA cells. Therefore, ACSL3 may be a potential therapeutic target (173).

Furthermore, several other factors potentially regulate ferroptosis sensitivity. For example, circular (circ)forkhead box protein P1 and Homo sapiens_circ_0050900 affect ferroptosis by interacting with OTU domain-containing protein 4 and regulating SLC3A2, respectively (174,175). TFR modulates the ferroptosis sensitivity of cells by regulating intracellular iron levels. Additionally, TFR is also an adverse prognostic factor in patients with iCCA (176). The regulatory importance of the E26 transformation-specific variant 4-ALY/REF export factor-pyruvate kinase M2 and Aldo-keto reductase family 1 member C1-cytochrome P450 family 1 subfamily B member 1-cAMP axes also warrant attention (177,178). Additionally, gut microbiotas inhibit ferroptosis in iCCA by altering glutamine metabolism through the regulation of the activin receptor-like kinase 5/NADPH oxidase 1 axis (179).

Ferroptosis serves a role in the pathogenesis, progression and treatment resistance of BTC. Furthermore, accumulating evidence demonstrates that ferroptosis-associated indicators are notably associated with patient prognosis and represent potential prognostic biomarkers (Table II). Techniques such as RNA sequencing and immunohistochemistry reveal elevated expression of ferroptosis-associated markers, including ACSL3/4, SLC7A11 and GPX4, in BTC tumor tissues compared with normal adjacent tissues, which is notably associated with poorer clinical outcomes (168,173,176,180-183). By contrast, isocitrate dehydrogenase 1 (IDH1) mutations are associated with a more favorable prognosis in BTC (183,184). In addition, other key ferroptosis-associated markers, such as LPCAT3 and FSP1, are also associated with prognosis in various solid tumors including ovarian cancer and lung adenocarcinoma; however, their clinical validation in BTC requires further investigation (185-192). Taken together, these ferroptosis-associated indicators may serve as novel prognostic markers for BTC, potentially offering a new perspective for precise prognosis assessment and individualized treatment strategies. Furthermore, integrating these indicators to establish multi-parameter prognostic models may enhance the accuracy of existing BTC prognostic assessment systems.

In addition to conventional GPX4 inhibitors and system Xc⁻ inhibitors, numerous other compounds have the potential to induce and notably contribute to the ferroptosis process in BTC. A number of these compounds can function independently, while others require co-administration with traditional chemotherapies, inducers and technologies.

Erastin, an inducer of ferroptosis, specifically inhibits system Xc⁻ (18). A study by Su et al (193) demonstrates that IDH1 mutations suppress CCA progression through erastin-induced ferroptosis. Erastin-treated IDH1 mutation cells have increased numbers of propidium iodide-positive cells, increased levels of lipid ROS and decreased cell viability compared with that in erastin-treated wild type cells. However, IDH1 mutation-targeted inhibitors, such as ivosidenib (AG-120), demonstrates notable clinical efficacy in CCA and are, at present, incorporated into second-line therapy. Therefore, the feasibility of combining ferroptosis inducers with IDH1-targeted agents requires further investigation (12,194). Furthermore, erastin and lenvatinib combined with photodynamic therapy (PDT) can increase ROS levels and inhibit antioxidant systems intracellularly, thereby enhancing the induction of ferroptosis in CCA cells (195).

Traditional Chinese medicine (TCM) is currently recognized as a viable complementary treatment in malignancies (196). Accumulating evidence demonstrates that TCM can notably enhance the chemosensitivity of a patient, potentiate the antitumor efficacy of therapeutic agents and mitigate treatment-associated adverse effects in cancer management (197,198). Quercetin (QE), a flavonoid in flowers, stems and leaves of various plants, possesses the ability to impede the progression of breast and colon cancer by interrupting the cell cycle (199,200). Previous studies demonstrate that QE induces ferroptosis in iCCA cells by inhibiting the NF-κB signaling pathway, which inhibits iCCA cell invasion (201,202). Isoliquiritigenin (ISL), also known as 2',4',4-trihydroxychalcone, is also a type of flavonoid and is extracted from the root of the liquorice plant. ISL induces apoptosis and inhibits proliferation in tumors (203,204). A study by Wang et al (205) reveals that ISL triggers ferroptosis in GBC by activating the p62-Kelch-like ECH-associated protein 1-Nrf2-haem oxygenase-1 (HMOX1) signaling pathway and reducing the expression of GPX4. Silencing HMOX1 or increasing GPX4 expression levels decreases the susceptibility of GBC cells to ISL-induced ferroptosis and enhances the survival of GBC cells (205). Another naturally occurring triterpenoid compound named liquidambaric acid (LCD) also exhibits notable therapeutic potential. LCD can bind to and inhibit signal transducing adaptor molecule binding protein like 1, which stabilizes Nrf2 through de-ubiquitination, and notably enhances CCA cell proliferation and migration while impeding ferroptosis (206). Wu-Mei-Wan, a long-utilized TCM formula, demonstrates potential efficacy as a second-line GBC therapy. It enhances gemcitabine chemosensitivity and induces ferroptosis through phosphorylated (p)-STAT3-mediated transcriptional regulation of ferroptosis-associated targets, downregulating GPX4, HIF-1α and FTH1, while upregulating ACSL4 (207).

At present, chemoimmunotherapy is the first-line treatment option for BTC due to the durvalumab plus gemcitabine and cisplatin in advanced BTC and KEYNOTE-966 studies (143,144). A study by Zhu et al (208) reveals that, in AKT-hyperactivated iCCA, the p-AKT-p-phosphoenolpyruvate carboxykinase 1 (PCK1)-p-lactate dehydrogenase A-SPRINGlac axis is a driver of ferroptosis resistance. This combines p-PCK1-mediated glycolytic activation and mevalonate flux reprogramming. However, simvastatin treatment effectively reverses this resistance, underscoring its therapeutic potential for improving the chemoimmunotherapy efficacy through ferroptosis sensitivity (208).

In addition, metabolic products such as lithocholic acid (LCA) are reagents involved in inhibiting tumorigenesis. A study by Li et al (209) reveals that LCA induces ferroptosis in GBC by suppressing glutaminase-mediated glutamine metabolism, which may be a tumor-suppressive mechanism with therapeutic potential.

PDT is a treatment modality that selectively destroys tumor tissue through the light activation of photosensitizers and release of ROS (210). The tumor-selective destruction capacity of PDT minimizes damage to healthy tissues, which suggests it may be a potential antitumor therapeutic strategy (211,212). The first-generation photosensitizer-HiPorfin-mediated PDT promotes the apoptosis and ferroptosis of CCA cells through the activation of the p53/SLC7A11/GPX4 signaling pathway (213). Furthermore, the second-generation photosensitizer-Hypericin-mediated PDT induces ferroptosis in CCA by inhibiting the AKT/mTORC1/GPX4 signaling pathway (214). These findings provide possible insights into individualized precision therapy for CCA. Furthermore, combination strategies represent novel therapeutic paradigms for BTC. Surufatinib (SUR), a multi-targeted kinase inhibitor, has comparable clinical efficacy in patients with advanced CCA compared with regorafenib, which is one of the recommend regimens used for subsequent-line therapy of BTC (12,215). Mechanistically, both SUR and PDT induce ferroptosis through the upregulation of ACSL4 and suppression of GPX4, and their effect is synergistically enhanced during combination treatment (215). In addition, an integrated nanotherapeutic platform known as CMArg@ Lip has been successfully developed for combined PDT and gas therapy. This platform encapsulates the photosensitizer, NO gas-generating agent and Nrf2 inhibitor with ROS-responsive liposomes, which enables it to effectively address the challenges of tumor hypoxia and the antioxidant microenvironment (216).

Ferroptosis, an iron-dependent, LPO-driven form of RCD, is a research focus in BTC therapeutics. Accumulating evidence demonstrates marked vulnerability of BTC cells to ferroptosis, unveiling novel avenues for targeted treatment. Despite promising advances, challenges persist. While core ferroptosis pathways, such as the GPX4-GSH axis and system Xc⁻ regulation, are delineated, the integrated regulatory network, particularly crosstalk with metabolic reprogramming and TME interactions, is yet to be fully elucidated. At present, clinically applicable biomarkers for predicting ferroptosis sensitivity are lacking, which hampers patient stratification. Furthermore, current ferroptosis inducers, such as erastin analogs and RSL3 derivatives, have limitations in tumor specificity, systemic toxicity and pharmacokinetic profiles.

Future studies should prioritize the following: i) Systematic investigations of ferroptosis synergism with immune checkpoint inhibitors and molecular-targeted agents, using multiomics approaches to decipher resistance mechanisms; ii) development of integrated biomarker panels incorporating genetic, epigenetic and microenvironmental features for precision patient stratification; iii) therapeutic validation through physiologically relevant platforms, including patient derived organoids, orthotopic BTC models and humanized mouse systems simulating tumor-stroma crosstalk; and iv) engineering of tumor-targeted nano-formulations, such as the CMArg@Lip platform, to enhance the tumor-targeted delivery of ferroptosis inducers while mitigating off-target effects.

In summary, bridging mechanistic insights from laboratory studies with rationally designed clinical trials incorporating biomarker-driven enrollment may highlight ferroptosis modulation as a potnetial therapeutic paradigm for BTC.