Targeting ferroptosis induction via nanomaterials in hepatocellular carcinoma: an insight into mechanism and therapeutic potential

Abstract

Hepatocellular carcinoma (HCC), the most common type of primary liver cancer, continues to be a leading cause of cancer-related deaths worldwide, despite the development of new treatment options, including systemic regimens, locoregional treatments, transplantation, and resection. Significant etiological and genetic heterogeneity, an immunosuppressive tumor microenvironment, dose-limiting toxicities, and high rates of post-treatment recurrence continue to inhibit robust responses. Given the crucial role of the liver in iron metabolism and oxidative homeostasis, ferroptosis-a controlled, iron-dependent cell death triggered by glutathione depletion, GPX4 inactivation, and lipid peroxidation, has become a potential therapeutic vulnerability in HCC. However, the risk of off-target oxidative damage, insufficient intratumoral transport, and the poor solubility and pharmacokinetics of small-molecule ferroptotic inducers limit the therapeutic application of ferroptosis induction.The current review is novel to the best of our knowledge, focusing on a targeted and current synthesis of nanomaterials delivering ferroptotic inducers in order to induce ferroptosis in HCC as a next-generation therapeutic paradigm. We describe how constructed nanoplatforms allow for spatiotemporally controlled ROS formation and iron-catalyzed lipid peroxidation, while also improving tumor-selective accumulation of ferroptosis triggers, extending circulation, and improving stability. In order to transform ferroptotic stress into antitumor immunity, we highlight ferroptotic inducing nanomaterials that co-deliver ferroptosis inducers alongwith chemotherapeutics or photothermal/photodynamic agents, and immunomodulatory designs that invoke innate immune pathways like cGAS–STING. This work establishes ferroptosis-inducing nanomedicine as a new and quickly developing field with the potential to overcome resistance, expand therapeutic windows, and enhance long-term outcomes for patients with HCC.

Graphical abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s12951-026-04711-9.

Highlights

1. Ferroptosis is a mechanistically distinct and therapeutically exploitable feature of HCC.

2. By overcoming the pharmacokinetic and toxicological constraints of small-molecule ferroptosis inducers, nanoparticle-based platforms enable tumor-selective administration.

3. In HCC, ferroptosis is associated with immunological activation via crosstalk between ferroptosis and cGAS-STING signaling, altering the tumor immune microenvironment.

4. Clinically adaptable systems and biomimetic nanocarriers are emerging translational methodologies, making nano-ferroptosis a possible next-generation treatment paradigm for HCC.

5. The combination of ferroptosis-inducing nanoparticles with phototherapy, gas therapy, chemotherapy, and immunotherapy inhibit tumor development and overcomes resistance.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12951-026-04711-9.

Introduction

Hepatocellular carcinoma (HCC), the most common type of primary liver cancer, accounts for approximately 90% of cases worldwide and is a key contributor to cancer related deaths [1]. Even though progresses in locoregional therapy, surgical resection, liver transplantation, and systemic drugs have been made, the overall prognosis for HCC is still depressing, particularly for those with intermediate or advanced stages of HCC [2, 3]. Besides cirrhosis, excessive chronic hepatitis B, alcohol use, and hepatitis C infection, and metabolic disorders like non-alcoholic steatohepatitis and non-alcoholic fatty liver disease, the etiological scale of HCC has many factors and geographically variable [4, 5]. Although Hepatitis C virus-associated (HCV) hepatocarcinogenesis primarily occurs through progressive cirrhosis, chronic hepatitis B virus (HBV) infection increases the risk of HCC by about 12 times and has independent carcinogenic potential, thereby necessitating lifelong monitoring [6, 7]. Other risk factors that influence disease incidence additionally and development across populations, include advanced age, smoking, male sex, obesity, diabetes, metabolic dysfunction-associated steatotic liver disease (MASLD/NASH), genetic vulnerability, and metabolic dysfunction [8, 9]. A mechanistic understanding of risk factors and disease evolution may facilitate the early detection and more effective therapeutic intervention in HCC.

Despite significant advances of the therapeutic background for HCC in recent years, high recurrence rates, therapeutic resistance and treatment-related toxicity continue to limit clinical outcomes. For a small percentage of patients, immune checkpoint inhibitors and systemic agents, specifically tyrosine kinase inhibitors like lenvatinib and sorafenib, limit long-lasting responses. Hypoxia, tumor microenvironment alteration, cancer stem cell morphologies, the activation of compensatory oncogenic pathways, and epigenetic reprogramming are examples of acquired and intrinsic resistance mechanisms that compromise long-term efficacy [10–12]. More than half of patients relapse within five years, even after curative-intent resection or ablation. This is due to aggressive tumor biology that is not well managed by current methods, intrahepatic spread, and remaining microscopic disease [13]. At the same time, toxicity profiles limit the window for treatment. Radiotherapy and locoregional intrusions may worsen hepatic dysfunction in patients with impaired liver, while targeted and immunotherapies often cause adverse effects linked to the cardiovascular system, skin, immune system and gastrointestinal tract [14]. Collectively, these shortcomings highlight the dire need for novel treatment approaches that can decrease toxicity, overcome resistance, and stop relapse [12]. The treatment of HCC remains challenging with constrained progress by hypoxia, resistance involving tumor immune microenvironment and cancer stem cell dynamics. The fact that only a small percentage of patients respond well to current systemic therapies highlights the need for mechanistic, combination, and precision-based therapy approaches in order to address intrinsic and acquired resistance.

Ferroptosis has been a dominant therpeutic pattern. Mechanistically ferroptosis is iron-dependent cell death, different from necrosis and apoptosis, marked by excessive lipid peroxidation and redox imbalance [15, 16]. Intracellular iron accumulation, and glutathione peroxidase 4 (GPX4) inactivation are the causes, which result in the generation of lipid peroxides that are lethal. The critical role of liver in iron metabolism and oxidative homeostasis is established by the notable susceptibility of HCC cells to ferroptotic stress. Preclinical models have shown that inducing ferroptosis can increase the anticancer activity of approved drugs like sorafenib and repurposed drugs like atorvastatin by altering key regulators like system p53, GPX4 and Xc-/SLC7A11, signaling, and iron handling pathways [17, 18]. However, accurate mechanistic management and off-target oxidative damage alleviation continue to be significant obstacles to clinical translation [19]. Moreover, ferroptosis can be integrated with immune signaling pathways and ferroptosis regulatory axes such as NRF2 and FSP1, can be targeted to reverse treatment resistance and improve immunotherapeutic responses [20]. The therapeutic potential of ferroptosis depends on the precise regulation of ferroptotic signaling and the development of effective targeted delivery strategies.

Nanotechnology offers a strong and valuable basis for ferroptosis in HCC therapeutics. Engineered nanomaterial rise the generation of reactive oxygen species (ROS) and lipid peroxidation in tumor tissues while reducing nonspecific supply by improving the stability and intratumoral delivery of drugs such as artesunate. Among the key shortcomings of outdated techniques, ferroptosis inducers’ poor solubility, systemic toxicity, inadequate tumor accumulation and reduced pharmacokinetics, can be addressed by delivery systems based on nanoparticles [21]. Tumor-responsive nanoplatforms, such as pH-sensitive metal–organic frameworks and mesoporous systems, enable the co-administration of ferroptosis inducers with photothermal or chemotherapeutic drugs. This decreases glutathione concentration and inhibits GPX4, promoting ferroptotic signaling in the tumor microenvironment [22]. In addition to direct cytotoxicity, ferroptosis-inducing nanoplatforms can also activate innate immune pathways such as cyclic GMP-AMP synthase-stimulator of interferon genes (cGAS-STING) and cause immunogenic mitochondrial stress. Ferroptosis and antitumor immunity stimulate dendritic cell maturation, enhance T-cell infiltration, and promote macrophage repolarization [23, 24]. Stimuli-responsive nanosystems capable of coordinating ferroptosis induction with immunomodulation may further enhance therapeutic efficacy.

Collectively, HCC treatment using ferroptosis-inducing nanoparticles is a promising and rapidly evolving field of oncologic nanomedicine. The examined nanoplatforms exploit the iron-rich metabolic vulnerabilities and redox dysregulation characteristic of HCC to enhance tumor-selective ferroptosis while minimizing systemic toxicity. The mechanism and advanced nanotechnology of nanoparticle-induced ferroptosis make it a promising next-generation therapy approach to overcome resistance, reduce recurrence, and ultimately treat HCC.

Overview of HCC pathology

Molecular mechanism of HCC pathogenesis including chronic HBV-associated HCC

Hepatocellular carcinoma has varied molecular etiology that is deeply impacted by the type of genotoxic damage and underlying causal variables. The need for etiology-specific treatment approaches is highlighted by the fact that different etiologies, such as metabolic dysfunction-associated steatohepatitis (MASLD/NASH), alcohol-related liver disease, chronic hepatitis B and hepatitis C infection, drive different molecular subtypes and immune backgrounds. The poor druggability of the HCC genome, only around 25% of tumors have actionable modifications, complicates matters further. Effective pharmaceutical targeting of these anomalies is still mostly elusive, despite the fact that almost half of patients possess recurring oncogenic mutations, most frequently in CTNNB1, TP53 or the TERT promoter [25]. Although inhibitors targeting TERT promoter activation and WNT/β-catenin pathway components have been discovered, their clinical efficacy has been limited [26, 27]. So, to overcome the problem of “undruggable” driving events and enable indirect targeting techniques, a better knowledge of how these mutations rearrange downstream signaling networks is essential (Fig. 1A).

Fig. 1.

Mechanism associated with HCC pathology. (A) Hepatitis B/C, alcohol, NASH and aflatoxins are among factors resulting chronic hepatitis, cirrhosis followed by early and advanced HCC. (B) Pharmacological inhibition and alteration of key signaling pathways e.g. TGF-β, Wnt, Notch, Hippo-YAP and Hedgehog in HCC progression. Overexpressed Notch1/Jagged1 signaling system in HCC releases Notch Intracellular Domain (NICD), translocating towards nucleus and contributes to poor differentiation, stemness, tumor development, EMT and metastasis. Dysregulation of YAP and TAZ downstream effectors promoting differentiation, and proliferation in HCC. Overactivated Wnt/β-catenin pathways promotes cancer stem cells self-renewal and drives oncogenic drivers such as, K Ras, Akt, c-Met, LKB1, and Nrf2 to accelerate hepatocarcinogenesis. Dysregulation of hedgehog signaling mediated by Shh/Gli1 promotes tumor cell migration, invasion and hepatocarcinogenesis. Altered expression of TGF-β signaling contributes to the activation of SMAD2/3 which alogwith SMAD 4 contributes to the expression of target genes promoting EMT, metastasis and fibrosis. TGFβR1 inhibitor galunisertib, serves as targeted therapies in HCC

One of the etiologies that best represents a direct virus-driven carcinogenic process is hepatitis B virus-related (HBV-related) HCC. In 80–90% of HCC cases and in around 30% of nearby non-tumorous liver tissue, HBV DNA integration into the host hepatocyte genome is found, suggesting that viral integration is an early carcinogenic event [28]. The carcinogenic potential of viral components is demonstrated by the development of early cirrhosis with steatosis in transgenic mouse models expressing the reverse transcriptase domain of HBV polymerase, with about 10% progressing to HCC [29]. Viral proteins such as HBeAg, HBsAg, HBx and DNA polymerase that are transcribed from integrated HBV DNA or covalently closed circular DNA (cccDNA) actively help in the development of tumors. Among these proteins, HBeAg affects host immune responses and cytokine production, further manipulating liver carcinogenesis [30], whereas HBsAg rises malignant transformation by activating the STAT3-IL-6 cascade [31]. Mainly, HBx is crucial to HBV driven carcinogenesis [32–35], activating signaling pathways and transcription factors such PI3K/AKT, STAT3 and NFκB, which in turn promotes hepatocyte survival, proliferation, and inflammation [36, 37]. Additionally, by modulating glycolysis, mTORC1 and cell cycle progression, C terminally shortened HBx variants have strong carcinogenic concerns in HCC [38]. Nearly 70% of Asian individuals mainly have preS/S and HBx gene regions integrated in HBV DNA and are associated with occult HBV infection (OBI) [39]. These events are related to inactivation of tumor suppressor, genomic instability, TP53 mutations, and telomerase reverse transcriptase (TERT) gene activation [39, 40]. TERT is often targeted by HBV DNA integration, which can activate it through insertional mutagenesis or enhancer hijacking, providing an advantage for early clonal development [41, 42]. Bx can transactivate the hTERT promoter via Sp1 dependent promoter elements. Reporter mapping revealed that Sp1 sites close to the core promoter are necessary to promote HBx to increase Sp1–DNA binding [43]. Additionally, HBx is characterized as a co activator having versatile interaction with transcription factors such as CREB, AP-1, NF-κB, and Sp1, offering several entry points to indirectly increase TERT transcription [44]. Additionally, HBx causes changes in the DNA methylation program, In this way on the signaling layer HBx activates TERT locus-specific methylation alongwith Wnt/β-catenin, MAPK, and PI3K/AKT cascades activation [45]. β-catenin/TCF4 can directly drive hTERT promoter transcription via a critical TCF4 site [46], providing a mechanistic bridge from HBx signaling to TERT upregulation effects. Targeting HBV related genomic alterations and HBx-associated signaling networks might influence TERT pathways for risk stratification, early detection, and precision therapeutic intervention in HBV-associated HCC. Moreover, HBx-mediated signaling, viral-host interaction, and metabolic rewiring provide a novel platform for integrated therapeutic strategies.

Chronic hepatitis B (CHB) inflammation subjects the immune system to dysregulation, through a number of signaling pathways, promoting HCC progression [47, 48]. Among the several immune cell types that mediate chronic inflammation and hasten the development of tumors and fibrosis are HSCs, natural killer T cells, macrophages/monocytes, and CD4⁺T cells mediating chronic inflammation, accelerating the development of fibrosis and tumors. HBV specific CD8⁺T cells and bystander CD8⁺T cells reduce inflammation and hepatotoxicity, but are not virus specific [49, 50]. Similarly, immunosuppressive cell types, such as regulatory T cells, Kupffer cells and myeloid derived suppressor cells (MDSCs), release IL-10 and transforming growth factor-β (TGF-β), reducing antitumor immunity and help in tumor immune evasion [51, 52] (Fig. 1A). Chronic immune-mediated damage leads to increasing fibrosis and cirrhosis, which are known to be precursors to HCC.

Moreover, HCC is characterized by abnormal activation of classical pathways beyond viral oncogenesis. About 20–35% of HCCs have overactivated Wnt/β-catenin pathways, usually due to CTNNB1 gain of function mutations which are common in alcohol-associated, HCV-related, and non-cirrhotic HCC. Similarly, AXIN1 mutations of Wnt/β-catenin pathways are more common in tumors associated with HBV [53–55]. Studies revealed that β-catenin promotes cancer stem cells self-renewal and drives oncogenic drivers such as, K Ras, Akt, c-Met, LKB1, and Nrf2 to accelerate hepatocarcinogenesis [56–62]. Further, downstream targets such cyclin D1, glutamine synthetase, KIF2C, and cMYC of Wnt/β-catenin pathways when activated transcriptionally [63–65], promote tumor growth [66, 67], metastatic potential, mTOR activation, metabolic reprogramming, and resistance to systemic therapies such as sorafenib, lenvatinib, and regorafenib [68–70].

Hippo-YAP signaling pathways is also a vital regulatory axis among other pathways in HCC pathology. Around 5–10% of tumors have YAP gene amplification at chromosome 11q22, while 65–85% of HCC patients have dysregulation of its downstream effectors, YAP and TAZ. Under classic conditions, Hippo signaling promotes apoptosis and restricts proliferation [71–73]. Hyperactivation of YAP/TAZ is linked to aggressive molecular subtypes and a poor prognosis in HCC [74–77]. Although oncogene-driven liver carcinogenesis (such as c-Met/Axin1 models) requires significant oncogenic role of both YAP and TAZ, highlighted by the fact that experimental ablation of upstream Hippo regulators like Nf2 or HPO1/2 results in HCC and related liver malignancies [78, 79]. Overall, YAP/TAZ acts as a transcriptional coactivator that links multiple oncogenic networks. Targeting Hippo-pathway regulators may therefore represent a strategy for modulating metastasis, metabolic reprogramming, and therapeutic resistance.

The Notch signaling system among other pathways contributes to poor differentiation, stemness, tumor development, epithelial-mesenchymal transition (EMT) and metastasis, and is frequently hyperactivated in HCC. In preclinical models, anti-Notch2 antibodies or dual antiplatelet therapy (DAPT) inhibitors promote EMT and cancer stem cell development, while genetic or pharmacologic inhibition of Notch1/Jagged1 signaling inhibits tumor growth [80–87]. These findings suggest that Notch1/Jagged1 and their associated inhibitors could be applied as a differentiation-based therapeutic approach to reduce the aggressiveness of tumors via inhibition of notch pathway.

TGF-β, a tumor suppressor, contributes to prevent proliferation and induce apoptosis in the early stages HCC [88, 89]. Around 40% of HCCs possess somatic mutations, promoting fibrosis, EMT, immune suppression, angiogenesis, and metastatic dissemination in advanced disease. Treatment resistance is also influenced by TGF-β signaling activation through the development of EMT [90], TME regulation, alternate pathway activation, metabolic reprogramming, and the acquisition of stem-like features. Certain patients with advanced HCC have shown manageable safety and potential clinical benefit from targeted therapies, such as the TGFβR1 inhibitor galunisertib, neutralizing antibodies as NIS793, and combination regimens with sorafenib, radiation, or PD-1 inhibition [91–93]. Stage-specific targeting of TGF-β signaling and biomarker-guided patient selection may improve the therapeutic precision in HCC.

Overexpression of Sonic Hedgehog has been exhibited in around 60% of cancers, and more than half of these cases cause downstream components including patched1 (Ptch1), smoothened (Smo), and glioma-associated oncogene homolog 1 (Gli1) to be elevated [94–99]. Hedgehog signaling pathway is abnormally activated in a significant percentage of HCCs. Hepatic fibrosis, HCC tumor cell invasion and migration are all assisted by dysregulated hedgehog signaling, mediated by Shh/Gli1. The potential of Hedgehog signaling as a therapeutic target for HCC is reinforced by the fact that it prevents tumor growth both in vitro and in vivo [100, 101]. The above studies reflect comprehensive role of hedgehog axis in shaping the tumor immune microenvironment and resistance to therapeutic and further emphasizing the integration of this pathway with fibrogenesis and invasive behavior of tumor. These observations support the therapeutic investigation of Hedgehog pathway components, including SMO and Gli, in combination and precision medicine strategies for HCC.

Finally, epigenetic instability promotes HBV hepatocarcinogenesis through modifying miRNA networks that control a number of carcinogenic signaling cascades including TLR mediated innate immune responses. Noncoding RNAs, particularly miRNAs, which post transcriptionally modify gene expression, are frequently altered by HBV infection [102]. Additionally, aberrant DNA methylation patterns [103], and histone post translational changes are brought on by HBV infection [104] producing stable epigenetic reprogramming that contributes to inflammatory signaling, genomic instability, and malignant transformation (Fig. 1B). Epigenetic reprogramming integrates histone modifications, miRNA dysregulation, DNA methylation, reshaping the expression of genes and immune signaling networks collectively in HBV-associated HCC progression. Host epigenetic machinery and HBx interplay, position miRNAs as key mediators linking viral persistence, inflammatory signaling, and malignant transformation. Moreover, miRNA-based diagnostics, epigenetic-targeted therapies, and integration of epigenetic modulation with antiviral and anticancer treatments can be designed to enhance precision therapy.

The pathophysiology of HCC is caused by abnormal activation of developmental and neoplastic signaling pathways, chronic inflammation, viral integration, and epigenetic remodeling. Besides promoting tumor genesis and progression, convergence of these molecular networks affects therapy response and resistance. Hence, a comprehensive, etiology informed understanding of these interconnected pathways is necessary to develop rational, precision based therapy strategies for HCC.

Tumor immune microenvironment of HCC

Tumor immune microenvironment (TIME) of HCC is a highly dynamic but primarily immunosuppressive network made up of stromal cells, tumor cells, immune populations, and soluble mediators that work together to control the course of the disease and the effectiveness of treatment. Increased infiltration of tumor-infiltrating lymphocytes is repeatedly evidenced in clinical settings to be associated with superior survival outcomes and decreased recurrence [105]. Attenuated tumor progression is specifically linked to the increase of CD8⁺ cytotoxic T lymphocytes (CTLs) [106]. Nevertheless, immunosuppressive mediators such indoleamine 2,3-dioxygenase (IDO), vascular endothelial growth factor (VEGF), interleukin 10 (IL10), hypoxia, and inadequate CD4⁺T cell support sometimes impair their anticancer effectiveness [102]. The tumor suppressive effects of CD4⁺ T lymphocytes are achieved via secreting IFNγ and TNFα [107], however, hepatocarcinogenesis can be aided by their depletion or ROS induced apoptosis [108]. On the other hand, regulatory T cells (Tregs), support immunological tolerance and encourage the spread of metastases [109], with TGFβ being essential for Treg polarization and CD8⁺T cell function inhibition [110]. CTL/Treg is the ratio of effector CTLs to Tregs and is the significant measure of immunological balance in the tumor microenvironment is the ratio of effector CTLs to Tregs [111]. The immunological landscape of HCC is further shaped by myeloid cell populations. TAMs, or tumor associated macrophages that are produced by recruited monocytes as well as indigenous Kupffer cells [112–115], exhibit functional polarization into either protumorigenic M2 or pro-inflammatory M1 phenotypes [116–120]. M2 polarized macrophages are closely linked to a poor prognosis, stimulate CD8⁺T cell death, enable metastasis, and enhance angiogenesis [116, 117, 121]. Similar functional duality is seen in tumor associated neutrophils (TANs), where TGFβ driven N2 subsets promote tumor growth while N1 neutrophils have anticancer effects [122, 123]. TANs increase cancer stemness, attract TAMs and Tregs, and facilitate the creation of neutrophil extracellular traps (NETs), which promote metastasis [124–128]. The TIME in HCC reflects as a very synchronized and adaptive environment, where the balance between immunosuppressive populations (e.g., Tregs, TAMs, TANs) and those of effector cells (e.g., CD8⁺ T cells) dictates progression of tumor and therapeutic responsiveness. Reprogramming the immunosuppressive microenvironment and restoring CTL function may improve the immunotherapeutic responsiveness of HCC.

Further, in HCC, the increase of MDSCs is positively correlated with poor overall survival and tumor burden. Similarly, MDSCs have a wide immunosuppressive effect by suppressing dendritic cells (DCs), T cells, and NK cells and macrophage function [129–136]. Mast cells can either promote tumorigenesis and angiogenesis by releasing growth factors or mediate tumor rejection by secreting proinflammatory cytokines [137–140]. Activated hepatic stellate cells (aHSCs) link chronic liver damage and immune evasion by recruiting MDSCs and Tregs and decreasing CD8⁺T cell activity to contribute to fibrosis and angiogenesis [141–145]. In HCC, higher Cancer associated fibroblasts (CAF) associated markers are consistently linked to a poor prognosis. CAFs secrete chemokines that recruit TANs and MDSCs further amplifying immunosuppression, EMT, and metastatic potential [146–153]. Additionally, innate immune components have important yet context dependent functions. The primary cytotoxic effectors of innate immunity are natural killer (NK) cells [111, 154], lose functional competency over time as HCC advances [155]. NKG2D⁺ NK cells related positive prognosis and that of [156] tumor progression and opposed clinical outcomes are linked with immature NK subsets expressing inhibitory receptors [157–160]. Cytotoxic mediators and antigens presented by B cells aid in antitumor immunity [161, 162]. However, ability of activated B cells to release IL10 via programmed cell death protein 1 (PD1) is emphasized alongwith their relationship with recurrence in advanced HCC. DCs, particularly conventional type 1 DCs (cDC1s), contribute to improved prognoses and are essential for priming cytotoxic responses [163]. Conversely, Tregs and poor results are related with immature or defective plasmacytoid DCs (pDCs) with reduced expression of costimulatory molecules [163–169] (Fig. 2). The TME of HCC exhibits a dynamic imbalance of immunosuppressive cells (MDSCs, Tregs, CAFs) that gradually take over over cytotoxic effectors, suggesting that immune dysfunction is not just a consequence but a driver of disease development. As multiple immune populations cooperatively sustain tumor progression, strategies targeting interconnected MDSC, NK-cell, and CAF signaling networks may provide greater therapeutic benefits than single-pathway interventions.

Fig. 2.

An illustration of the tumor immune microenvironment in HCC. The particular immune response in HCC is mediated by the infiltration of different immune cell subpopulations, regulatory cytokines, and distinct inhibitory signals. IDO, VEGF, IL-10, and hypoxia are among the factors that the HCC tumor cells produce to limit the tumoricidal potential of CTL, whereas TGF-b, IL-10, and the inhibitory receptor NKG2A reduce the tumoricidal potential of NK cells. In order to attract TANs into the tumor stroma, HCC tumor cells also release CXC chemokines, particularly CXCL8. Anti-tumor immunity of Effector T cells is inhibited by IL-10, which is produced when PD-1 signaling of B cells is activated. Many cytokines, including VEGF, IL-b, and IL-6 produced by CAFs and HSCs and VEGF and GM-CSF produced by HCC tumor cells, facilitate the aggregation of MDSCs in the tumor stroma. The cytotoxicity of CTL and NK cells is hampered by the interaction between MDSCs and TAMs, which causes a decrease in IL-6 and IL-12 secretion and an increase in IL-10 secretion. When MDSCs and KCs connect, KCs express PD-L1, which then interacts with PD-1 of T cells to mediate immune evasion. MDSCs that express galectin-9 can attach to TIM-3 on T cells to cause T cell death. Tregs can diminish the antitumor capacity of NK cells by producing IL-2, IL-8, and TGF-b. They can also prevent CTL activation, which is mediated by decreased production of TNF-alpha and IFN-g. DCs can decrease the antitumor response of CTLs by mediating the generation of IL-10 and the lowering of IL-12

Collectively, the interaction of major immunosuppressive mechanisms and antitumor immune components shapes the milieu in which HCC develops, promoting tumor development, immune evasion, and metastasis.

Clinical stages of HCC and its diagnostics

The frequently used framework for clinical stratification of HCC is the Barcelona Clinic Liver Cancer (BCLC) staging system in order to guide prognostic and treatment assessments. To assign stage specific medicines, the BCLC algorithm controls both hepatic reserve and oncologic severity by tumor burden, combining liver function, and patient performance status [170]. HCC is divided into five stages: very early (stage 0), early (stage A), intermediate (stage B), advanced (stage C), and terminal (stage D) (Fig. 1A). For early stages, curative measures like resection, ablation, and transplantation are advised; for intermediate disease, locoregional therapies like transarterial chemoembolization (TACE); for advanced disease, systemic therapy; and for terminal cases, the best supportive care [170, 171]. In suspected cases, the typical arterial phase hyper enhancement and portal/delayed washout on multiphasic CT or MRI establish the noninvasive diagnosis, frequently eliminating the need for biopsy [171]. Early detection through investigation of high risk groups, particularly those with cirrhosis or persistent HBV/HCV infection, utilizing ultrasound with or without serum alpha fetoprotein (AFP) every six months, show better outcome [172]. The Liver Imaging Reporting and Data System (LI-RADS) improves interdisciplinary decision-making and diagnostic consistency by standardizing imaging interpretation [171, 173, 174]. BCLC system integrates liver function and tumor burden effectively, though its strictness may limit decisions of personalized treatment, mainly in borderline or in some cases where patients could get aid from stage migration approaches. The integration of molecular biomarkers, immune profiling, and longitudinal imaging may further improve patient stratification and treatment selection beyond conventional staging criteria.

Present therapeutic strategies for HCC and their limitations

HCC is managed using a multimodal approach that is stage adapted and based on the performance status of patient, hepatic reserve, and tumor burden. Curative intent treatments, such as liver transplantation, percutaneous ablative therapy, and surgical resection, continue to be the mainstay of care for early stage illness and can significantly improve long term survival. Their use is limited to a small proportion of patients, nevertheless, as many arrive with vascular invasion, multifocal tumors, or underlying cirrhosis that make definitive intervention impossible. Additionally, recurrence rates remain vital showing hidden micro metastases even after curative medication, and oncogenic drivers within the sick liver milieu [175, 176].

The therapeutic landscape for progressive HCC has changed dramatically with the introduction of systemic therapies. Though immune checkpoint inhibitors, either alone or in combination, have further extended therapy choices, tyrosine kinase inhibitors (TKIs), such as lenvatinib, sorafenib, cabozantinib and regorafenib have established roles in first and second line conditions. Combination regimens, when compared to TKI monotherapy, such as bevacizumab and atezolizumab, have demonstrated better survival and objective response rates by simultaneously targeting angiogenesis and immune evasion pathways in the tumor microenvironment [177–180]. Two prevalent locoregional therapies for patients with incurable cancers or intermediate stage are radio embolization and transarterial chemoembolization. These treatments shrink tumors while preserving liver function [181, 182]. Though these approaches can slow down the course of the disease and exploit mostly arterial blood supply of HCC, they have little impact on overall survival and often require recurrent treatments due to recurrence and residual disease [183, 184]. Chimeric Antigen Receptor T cell (CART) cell therapy, gene based therapies and nanomedicine platforms, are among of the innovative strategies being actively researched to improve tumor specific medication delivery and get around resistance mechanisms [185, 186].

Despite these developments, there are still significant restrictions in all therapeutic approaches. Recurrence is prevalent even with dynamic management, only a small percentage of patients can benefit from curative approaches as locoregional and systemic treatments seldom achieve long lasting disease control. Moreover, immune checkpoint inhibitors (ICIs), tyrosine kinase inhibitors (TKIs) and usually have low response rates and have associated side effects. Meanwhile, acquired and primary resistance, which are usually caused by tumor heterogeneity, immunosuppressive microenvironmental factors and compromise long term effectiveness. Shortage of donor organs further limits liver transplantation, and many systemic medications show reduced efficacy in patients with HCC. Collectively, these limitations highlight the need for mechanism-based therapeutic strategies that can improve long-term clinical outcomes in HCC.

Insight into ferroptosis, molecular mechanism and its disruption in HCC

Iron metabolism and Fenton reaction, lipid peroxidation, GPX4, and system Xc pathways of ferroptosis

Ferroptosis is a unique type of regulated cell death (RCD) characterized by excessive oxidative stress and iron dependent lipid peroxidation. Ferroptosis varies from necrosis or apoptosis in that it is characterized by an excess of ROS. Redox active Fe²⁺ is crucial to this process because it catalyzes the Fenton reaction, which turns H₂O₂ into reactive hydroxyl radicals. Lipid peroxidation eventually weaken membrane integrity and cause ferroptotic cell death [187, 188]. A key defense against this process is GPX4, which inhibits ferroptosis by converting harmful lipid peroxides to nontoxic lipid alcohols with glutathione (GSH) as a cofactor [189]. Thus, iron availability inside a cell acts as a key factor of ferroptosis susceptibility.

Regulation of iron metabolism guards cellular homeostasis and prevents uncontrolled oxidative damage. Endocytosis, mediated by transferrin receptor 1 (TFR1), ingests Fe³⁺ in circulation. After being reduced to Fe²⁺ inside endosomes by six transmembrane epithelial antigen of the prostate 3 (STEAP3), it is then transported to the cytoplasm through divalent DMT1. Ferritin, ferroportin (FPN), a heteropolymeric complex comprising of ferritin light chain (FTL) and ferritin heavy chain 1 (FTH1) and, or redox mechanisms can all be used to export cytosolic iron and prevent iron overload [190–196]. These processes are post transcriptionally regulated by iron response elements and iron regulatory proteins including IREB2, which affect ferroptosis sensitivity by altering the expression of FTH1, FPN, FTL and TFR1 [193]. Disrupted iron regulation directly affect ferroptotic susceptibility, e.g., exogenous substances like ruscogenin affect FPN and TFR1 expression, increasing the formation of lipid ROS and inducing ferroptosis in pancreatic cancer cells [197, 198]. Autophagic and ferritinophagy, a breakdown process facilitated by Nuclear Receptor Coactivator 4 (NCOA4), is another way through which endogenous regulation takes place improving Fenton chemistry and releasing accumulated Fe²⁺ from ferritin [199]. Similarly substances like dihydroartemisinin increase intracellular iron pools and ROS generation by stimulating ferritinophagy in order to promote ferroptotic death [200].

Polyunsaturated fatty acids (PUFAs) in membrane phospholipids are the chief target of lipid peroxidation, which is the effecting stage of ferroptosis. Long chain polyunsaturated fatty acids (PUFAs) such as, arachidonic acid (AA) and adrenic acid (AdA), are esterified by Acyl-CoA synthetase and long chain family member 4 (ACSL4), to produce AA-CoA and AdA-CoA. Lysophosphatidylcholine acyltransferase 3 (LPCAT3) then integrates these polymers into phosphatidylethanolamine (PE) to form PE-AA and PE-AdA substances that are tremendously susceptible to peroxidation [196, 201, 202]. Ferroptotic lipid peroxidation may also occur independently of LOX activity by nonenzymatic iron-driven radical propagation, but lipoxygenases (LOXs), especially LOX15, can catalyze PUFA oxidation and contribute to ferroptosis [187, 203–206]. Cellular sensitivity to ferroptotic stress is thus largely determined by the makeup of membrane lipids.

A cell uses diverse antioxidant protection mechanisms to fight with iron regulated lipid peroxidation. SLC3A2 and SLC7A11 make up the cystine/glutamate antiporter system Xc⁻, which imports cystine in return for glutamate. This offers the substrate for the synthesis of GSH by γ-glutamylcysteine ligase and glutathione synthetase [207–209]. GSH constrains ferroptosis by reducing phospholipid hydroperoxides (PLOOH) to nontoxic phospholipid alcohols (PLOH), and is a chief cellular antioxidant. GSH is also a vital cofactor for GPX4. Disruption of cystine absorption or GSH reduction compromises GPX4 action, allowing uncontrolled lipid peroxide accumulation. Small molecule inducers including ML210 and RSL3, induce ferroptotic cell death, by directly blocking GPX4, whereas PdPT and FIN56 encourage GPX4 breakdown [210–212]. By reducing coenzyme Q10 (CoQ10) to ubiquinol CoQH₂ in a NADPH dependent manner, FSP4 permits membrane confined radical trapping and inhibits phospholipid peroxidation through CoQ10 regeneration [213, 214]. In mitochondria, GPX4 and dihydroorotate dehydrogenase (DHODH) form a supplementary defensive system. In the inner mitochondrial membrane, DHODH, a crucial enzyme in pyrimidine biosynthesis, converts CoQ10 to CoQH₂, inhibiting mitochondrial lipid peroxidation and maintaining organellar integrity [215–217]. Similarly, the GTP cyclohydrolase1 (GCH1) tetrahydrobiopterin (BH4) pathway promotes ferroptosis resistance via strong lipid antioxidant [218, 219]. Together, ferroptosis is closely regulated by these interconnected iron dependent oxidative and antioxidant processes, and cellular sensitivity to this type of regulated cell death is determined by their dysregulation (Fig. 3).

Fig. 3.

Molecular mechanism of ferroptotic cell death involving lipid peroxidation, Fenton reaction, XC⁻ system and antioxidant pathway. The cysteine glutamate antiporter system allows synthesis of glutathione (GSH), and thus GPX4 to inhibit lipid peroxides and prevent ferroptosis. Lipid metabolism pathways comprizing ACSL4, LPCAT3, and lipoxygenases stimulate the accretion of PUFA and lipid peroxides. Antioxidant systems comprising GCH1-BH4, FSP1-CoQ10, and NRF2 signaling neutralize lipid peroxidation. Mitochondrial metabolism promote ROS production, while ferritinophagy rises the labile iron pool, assisting Fenton reactions and oxidative damage. Dysregulation of these pathways cause oxidative damage of cell membrane and eventually ferroptosis

Disrupted iron homeostasis in HCC

Ferroptosis in HCC is generally regulated by iron homeostasis, where the increase of the intracellular labile iron pool (LIP) aggravates Fenton driven lipid peroxidation and membrane degradation. Mechanistically, PTBP1-NCOA4 axis modulation changes iron metabolism under sorafenib treatment. This iron metabolism in HCC cells is often restarted through increased transferrin-mediated uptake, decreased export and dynamic control of ferritin (FTH1/FTL) and its autophagic degradation via NCOA4 dependent ferritinophagy. The result is release of redox-active Fe²⁺ and increase of ferroptosis sensitivity [15].

Hepcidin induced ferroportin degradation is another way that influences intracellular iron retention via hepcidin-FPN axis, resulting ultimately in tumor ferroptotic susceptibility [220]. In addition, HCC cells create adaptive resistance mechanisms which often needs antioxidant and iron sequestration mechanism mediated by NRF2, increasing ferritin and associated anti ferroptotic mediators. However, sorafenib can cause ferroptosis resistance [221], with activated NRF2, increasing ferroportin and FTH1/FTL expression to inhibit intracellular Fe²⁺ and limit the LIP [222]. Iron homeostasis is a key determinant of ferroptosis susceptibility in HCC. Moreover, lncRNA-mediated control, such as URB1-AS1 induced ferritin aggregation and suppression of NCOA4 mediated ferritinophagy, further restricts Fe²⁺ release and lipid ROS accumulation.

Upregulation of antioxidant defense system of ferroptosis in HCC

There is growing evidence that ferroptosis plays a complex role in the development, course, and response to treatment of HCC, and that iron excess is known to be a risk factor for hepatocarcinogenesis [223]. Despite this iron-rich environment, HCC cells often avoid ferroptotic death by strengthening antioxidant defense systems, involving GPX4, GSH and cysteine. In HCC, System Xc⁻ is often overexpressed and neutralizes lipid peroxides, reduces ferroptotic damage by increasing cystine absorption and maintains intracellular GSH production [15]. The transcription of SLC7A11, GPX4, and a wide range of cytoprotective genes, such as NQO1, HO1 and ferritin subunits (FTH1/FTL), is driven by NRF2, a master transcriptional regulator of redox homeostasis in HCC.

Disulfidptosis in addition to ferroptosis regarding overexpression of SLC7A11 and NADPH deprivation is caused by cytoskeletal collapse [224] and is related with metabolic stress e.g. glucose deprivation and aberrant disulfide accoumulation. However ferroptosis is mechanistically dependent on iron metabolism and ROS driven membrane damage, indicating distinct susceptibilities in cancer cells [225]. Redox homeostasis and SLC7A11 link these two processes in response to metabolic limitations. Ferroptosis activation is a prospective therapeutic approach as HCC tumor cells often show high iron consumption and oxidative stress sensitivity. However, metabolic adaptation or the overexpression of antioxidant proteins e.g. SLC7A11 and GPX4 may rise susceptibility to disulfidptosis while decreasing ferroptosis sensitivity. Simultaneous targeting of ferroptosis and disulfidptosis may provide a strategy for overcome therapeutic resistance in HCC [16, 226, 227].

It has been validated that a PI3K-AKT-ABCC5-SLC7A11 signaling cascade keeps increased SLC7A11 expression and inhibits ferroptosis in HCC cells [228]. In HCC, other transcriptional regulators also directly increase the expression of GPX4 e.g. NeuroD1 binds to the GPX4 promoter, raising GPX4 levels and decreasing accumulated level of lipid peroxide [229]. Although GSH depletion is traditionally related to increased ferroptosis, recent studies further emphasize the context dependent role of GSH metabolism by showing that altering GSH homeostasis can change ferroptotic sensitivity and oxidative stress responses. It has been shown that TGFβ1/Smad3 signaling suppresses SLC7A11 activity, which increases lipid peroxidation and makes HCC cells more sensitive to GPX4 inhibition [230]. Together, in HCC, these amendments reduce ferroptosis and rise antioxidant capacity.

Reprograming of lipid metabolism in HCC

The integration and peroxidation of phospholipids containing PUFA within cellular membranes are essential for ferroptosis, and this process is often reprogrammed in HCC to reduce ferroptotic susceptibility. ACSL4, a vital enzyme that converts free PUFAs into membrane phospholipids, is known to be a key factor in ferroptosis sensitivity. Intensified responsiveness to ferroptosis inducers like sorafenib is associated with higher expression of ACSL4 in HCC cells, emphasizing its role as a functional organizer of lipid peroxidation driven cell death [231]. Adaptive downregulation of ACSL4 activity, on the other hand, is a key resistance mechanism. e.g., miR-23a-3p is upregulated in sorafenib resistant HCC cell lines, which directly targets and decreases ACSL4 expression. Consequently reduced PUFA absorption into membrane phospholipids and obstruction in ferroptotic performance occurs [232]. More generally, oxidizable lipid substrate availability is limited by reduced ACSL4 expression and fluctuations in PUFA remodeling enzymes, which confines the accumulated lipid peroxide and damage to ferroptotic membranes [233, 234]. These variations in lipid metabolism point to a slow renovation of the membrane composition in HCC, where inhibition of PUFA metabolism concentrated on ACSL4 acts as a critical defense against ferroptosis and fuels resistance to treatment (Fig. 4).

Fig. 4.

Molecular mechanism and associated disrupted pathways of ferroptosis in HCC. Lipid peroxidation is the primary cause of ferroptosis, and ferroptosis induction may be a useful treatment for HCC as well as associated malignancies. Ferroptosis is mainly caused by system xc⁻, p53, iron metabolism, and lipid peroxidation. The ultimate products of system xc⁻ under the action of peptide peroxidase include glutamate-cysteine ligase (GCL), GSS, and glutathione-induced intracellular lipid peroxidation. Lipid peroxidation promotes iron metabolism characterized by excessive release of iron ions. The primary method of lipid peroxidation is fatty acid-induced lipid peroxidation. Various regulators, including NRF2, p53, and non-coding RNAs, modulate these pathways, while pharmacological agents (e.g., sorafenib, erastin, RSL3, and artesunate) influence ferroptosis sensitivity. The proteins/inhibitors in grey color represent antiferroptotic while those in yellow color represent proferroptotic in HCC

Tumor microenvironment-associated suppression of ferroptosis

The tumor microenvironment (TME) of HCC is often hypoxic and immunosuppressive, which actively inhibits ferroptotic cell death. Tumor provoking immune cells including dendritic cells (DCs), effector T cells, M1 polarized macrophages, and NK cells have a tendency to target and kill cancer cells to prevent tumor. It has been found that HBV infection promotes liver cancer cells to secrete the exosome miR-142-3p, downregulating SLC3A2 expression to induce ferroptosis in M1 macrophages, thereby inducing HCC progression [235]. Therefore, the preservation of antitumor immune cell function is an important consideration in ferroptosis-based therapies. Tumor growth is also promoted by myeloid derived suppressor cells (MDSCs) and regulatory T cells (Tregs), two key cells of immune system. Tumor degradation results from their assistance in removing immune monitoring and destruction. Conversely, the induction of ferroptosis in immunosuppressive cell populations may enhance antitumor responses.

The immunosuppressive microenvironment of HCC is primarily characterized by low immunogenicity. Tumor immunogenicity is the basis of immunotherapy, which refers to the capability of immune system capacity to recognize tumor antigens as foreign and to elicit immunological responses. Tumors with high immunogenicity are often defined as Hot tumors, characterized by the release of pro-inflammatory cytokines and T cell infiltration, elevated PD-L1 expression, and a high tumor mutational burden. Conversely, low immunogenicity tumors are referred to as “cold tumors.” CD8 + T lymphocytes become less sensitive to tumor-related antigens with the HCC progression, demonstrating a drop in tumor immunogenicity and immunotherapy. Ferroptosis can produce a number of DAMPs that aid the presentation of antigens and the activation of antitumor lymphocytes, increasing the susceptibility of “cold tumors” to immunotherapy [236, 237].

By organizing antioxidant and iron regulatory processes in tumor cells, numerous stromal and immune cell populations reduce lipid peroxidation. To support GSH dependent defenses against ferroptosis, Tregs release IL10, which stimulates antioxidant genes like GSTP1, SLC7A11 and activates STAT3 signaling in HCC cells [238]. Similarly, MDSCs release arginase 1, which lowers nitric oxide and reduces arginine, altering redox homeostasis and inhibiting iron dependent oxidative damage [239]. The M2 polarized fraction of TAMs in particular release TGFβ, which increases GPX4 and FPN1 expression in tumor cells through Smad2/3. This promotes iron export and fortifies the GPX4 centered antiferroptotic axis. Cysteine absorption and GSH production are further enhanced when cytokines like IL6 activate STAT3 to upregulate xCT/SLC7A11 [240]. HCC cells are thus protected by the convergence of IL10, IL6, and TGFβ from Tregs, MDSCs, and TAMs on ferroptosis regulators such as, SLC7A11, GPX4, and iron handling proteins. Together, Treg, MDSC, and TAM-derived IL10, IL6, and TGFβ converge on important ferroptosis regulators, such as SLC7A11, GPX4, and iron-handling proteins, to protect HCC cells against ferroptotic harm.

Cancer associated fibroblasts (CAFs) contributes to resistance to ferroptosis. Lactate from CAF is accumulated in the TME and encourages histone lactylation, rewiring the transcription of tumor cells. FTH1 and GCH1 are expressed more frequently when lactate induced H3K18 lactylation upregulates transcription factors such ZFP64, according to recent data. Both approaches reduce lipid ROS accumulation and ferroptosis in HCC. GCH1 promotes the synthesis of BH4, while FTH1 sequesters labile Fe²⁺ into storage complexes [241, 242]. By upregulating the long noncoding RNA DLEU1 or releasing miR-522, CAFs can further suppress ferroptosis and strengthen the antioxidant ability of tumor cells [243, 244].

IFNγ has a dual proferroptotic action by upregulating ACSL4 and downregulating SLC7A11 to prevent cystine uptake, ultimately rising phospholipids containing PUFA and makes cancer cells more sensitive to lipid peroxidation [245, 246]. Neutrophils also increase oxidative stress in tumor cells by transporting granules containing myeloperoxidase, inducing ferroptosis [247]. Furthermore, cytokine signaling might have context dependent effects. For example, it has been demonstrated that TGFβ1 promotes ferroptosis by repressing SLC7A11 through SMAD [230], while, interleukin1β maintains NADPH synthesis and iron sulfur cluster balance by increasing nicotinamide nucleotide transhydrogenase acetylation, which inhibits ferroptosis [248]. These results clearly demonstrate that the TME is a dynamic regulator of ferroptosis in HCC, where cytokine networks, stromal metabolism, and immunosuppressive cells work together to rewire antioxidant defenses, lipid metabolism, and iron management in order to modify ferroptotic exposure (Fig. 5B).

Fig. 5.

Sorafenib resistant and tumor immune microenvironment associated suppression of ferroptosis in HCC. (A) Ferroptosis-related sorafenib resistance mechanisms. Hippo signaling first phosphorylates YAP/TAZ, which NPCs then carry into the nucleus. In order to induce SLC7A11 and rise intracellular GSH levels, YAP/TAZ can either bind to TEAD or activate ATF4 in the nucleus. LCN2 can block the action of transferrin receptor, and when LIFR is deleted, the NF-κB signaling pathway is triggered resulting increased expression of LCN2, which lowers intracellular iron entry and prevents ferroptosis in cells. The expression ferroptosis-related genes, including MT-1G, ABCC5, and GPX4, will rise when the Keap1-Nrf2 system is activated. In HCC cells, MT-1G can stop the process of lipid peroxidation. For a more stable effect, ABCC5 can stabilize the SLC7A11 protein. (B) In CD8 T cells, dendritic cells and M1 macrophages, ferroptosis stimulation alters GPX4 activity, inflammatory mediators (e.g., IL-9, NO), and lipid peroxidation (LPOs), which affects anticancer immunity. In contrast, ferroptosis transforms GPX4-dependent lipid metabolism and affects immune suppression in immunosuppressive M2 macrophages, Tregs, and MDSCs. Ferroptosis induction alters the tumor immunological scene by increasing the release of cytokines and DAMPs resulting “cold tumors” to become “hot tumors”. This enhances the tumor susceptibility to immune checkpoint inhibitor (ICI) therapy and creates a more immunogenic “burning tumor” state

Furthermore, in the liver microenvironment, activated hepatic stellate cells (aHSCs) play a vital role as both targets and regulators of ferroptosis. Hepatocytes and ferroptosis sensitivity of tumor cells is all together shaped by aHSCs through paracrine signaling, even though inducing ferroptosis in these cells can be antifibrotic. However, NRF2 and IL-6/STAT3 signaling pathways improve resistance by upregulating SLC7A11 and GPX4, the components of antioxidant system. aHSC derived TGF-β can also sensitize cells to ferroptosis, also affecting ferritinophagy and iron metabolism, such as the LCN2 axis to control iron availability, a critical element of ferroptosis. Additionally, aHSC driven hypoxia and lipid remodeling might either increase or decrease ferroptosis depending on the situation. Importantly, ferroptosis in hepatocytes may cause cGAS-STING-mediated inflammation, which could aggravate fibrosis. It shows a context dependence of ferroptosis and that it is not always advantageous.

Resistance mechanism of sorafenib-induced ferroptosis in HCC

For advanced HCC, sorafenib and lenvatinib continue to be the first line choice of systemic treatments [3]. Sorafenib, besides its canonical function as a multikinase inhibitor, has been established to induce ferroptosis in a number of solid tumors, such as melanoma, colorectal cancer, and HCC, underscoring ferroptotic cell death as a crucial part of its anticancer action [10, 249]. There is growing evidence that the development of sorafenib resistance in HCC is closely related to controlled cell death (RCD), specifically ferroptosis [250, 251]. Adaptive reduction of ferroptotic signaling can promote treatment resistance, despite the fact that sorafenib induced ferroptosis adds to its lethal efficacy. This highlights the molecular complexity of activity of sorafenib in HCC [252].

Multiple molecular pathways are involved in suppression of ferroptosis to drive drug resistance. The Hippo signaling pathway is a regulator among those pathways driving sorafenib resistance. Phosphorylated YAP/TAZ translocate into the nucleus, where they bind TEAD or activate ATF4, leading to upregulation of SLC7A11 and increased GSH synthesis, thereby reducing ROS accumulation and inhibiting ferroptosis [253]. This antioxidant reprogramming is a key mechanism by which resistant cells evade sorafenib induced lipid peroxidation. Another mechanism involves iron metabolism. Loss of LIFR activates the NF-κB pathway, which upregulates lipocalin 2 (LCN2). LCN2 inhibits function of transferrin receptor, reducing intracellular iron uptake, thus prevents ferroptosis. Since ferroptosis is iron dependent, reduced iron availability directly contributes to resistance. Resistance is further supported by Keap1-Nrf2 signaling cascade that induces antioxidant and ferroptosis suppressing genes such as GPX4, ABCC5 and MT-1G. GPX4 detoxifies lipid peroxides, while Nrf2 driven pathways maintain redox homeostasis and prevent ferroptotic cell death [254]. Specifically, MT-1G inhibits lipid peroxidation in HCC cells, thereby blocking the core execution step of ferroptosis and promoting sorafenib resistance HCC cells. Meanwhile, ABCC5 stabilizes the SLC7A11 protein, enhancing cystine uptake and sustaining GSH production, which further suppresses ferroptosis [228]. These pathways such as Hippo/YAP-TAZ, LIFR, NF-κB, LCN2, and Keap1-Nrf2, join to maintain antioxidant state and thus limiting iron availability, inhibiting lipid peroxidation, thereby enabling tumor cells to escape ferroptosis and develop resistance to sorafenib.

HCC cells are significantly protected against sorafenib induced cytotoxicity when intracellular iron reserves are exhausted with the iron chelator deferoxamine (DFX) [235]. Importantly, DFX does not obstruct sorafenib to access or inhibit its intracellular kinase targets; instead, iron deficiency impedes oxidative stress brought on by sorafenib, which stops the iron dependent lipid peroxidation necessary for ferroptosis [255]. Consistent with these results, Lachaier et al. revealed sorafenib as inducer of ferroptosis in a number of cancer cell lines, supporting the idea that ferroptosis is a conserved mechanism behind its anticancer properties [249]. Together, these results suggest that dysregulated iron homeostasis and antioxidant adaptation are key factors in treatment resistance in HCC, positioning iron availability and ferroptotic competence as critical modulators of sorafenib responsiveness (Fig. 5A).

Nanomaterials-the efficient inducers of ferroptosis in HCC

Various classes of nanomaterials have been used in recent research to induce ferroptotic cell death in HCC (Fig. 6A). Following are some given characteristics which enable them to be utilize in recent research against different diseases.

Fig. 6.

Classification of nanomaterials inducing ferroptosis in HCC and their toxicology. (A) Various organic and inorganic nanomaterials associated with ferroptosis induction alongwith different combination therapeutic drugs in HCC. (B) Physicochemical properties, toxicity mechanism and possible route associated with nanomaterials toxicity

Advantage over ferroptosis-inducing small molecules

Compared to traditional small molecule inducers like RAS selective lethal 3 (RSL3), erastin and sorafenib, nanoparticle (NP) induced ferroptosis therapy has significant advantages, particularly in terms of pharmacokinetics, targeting accuracy, and therapeutic efficacy [256, 257]. Poor water solubility, quick systemic clearance, and dose limiting toxicity have limited the in vivo effectiveness of early ferroptosis inducers, despite their strong anticancer activity in vitro [258, 259]. Optimized delivery systems are necessary e.g. imidazole ketone erastin (IKE), to produce major therapeutic benefits [260]. Ferroptosis inducing agents (FINs) have great therapeutic potential since they can use lipid peroxidation to kill apoptosis resistant cells, particularly in fibrotic diseases and cancer. However, because normal tissues including the liver, kidney and brain, are also susceptible to ferroptotic injury, their lack of cell specificity poses serious safety risks [261]. This leads to a limited therapeutic window and hazards such systemic oxidative stress and organ damage. Toxicity and off-target effects of FINs e.g. sorafenib, delay clinical translation, although preclinical research is promising. Inflammation, inadequate targeting and oxidative damage spread are major obstacles. Above all, FINs are promising but not intrinsically safe. Further, for clinical usage, better targeting and regulated distribution are needed.

The clinical application of many small molecule ferroptosis inducers is limited by their off target toxicity, poor water solubility and rapid systemic clearance; in contrast, nanoparticles recover drug stability, lengthen circulation time, and promote tumor accumulation through the enhanced permeability and retention (EPR) effect [262]. Nanocarrier systems have been shown to solubilize hydrophobic drugs, shield them from early deterioration, lengthen their half-life and rise their bioavailability in circulation [263–267]. For example, a scalable NanoAssemblr microfluidic platform has been used to successfully encapsulate IKE into polyethylene glycol modified poly (lactic-co-glycolic acid) (PEG-PLGA) nanoparticles [268]. PEGylation yields a hydration shell that improves tumor targeting and prolongs systemic circulation by declining clearance by the mononuclear phagocytic system [269]. When combined, these characteristics lead to better tumor accumulation and less off target damage than unformulated small molecules [270, 271]. Co delivery of Nanoparticles and ferroptotic inducers characterizes a well promising approach, mainly in overcoming pharmacokinetic and bioavailability barriers that limit clinical translation. To improve clinical applications, research on targeted specificity, long term safety profiling and scalable manufacturing is required. Overall, combining nanotechnology with ferroptosis based therapies could considerably advance precision oncology while reducing systemic toxicity.

Improved Permeability and Retention in HCC

Nanomaterials with significant concentration, are capable of the enhanced permeability and retention (EPR) effect, promoting better accumulation in tumor tissues with compromised lymphatic drainage and leaky vasculature [270, 271]. Nanoparticles can attain greater intratumoral drug concentrations than free agents where abnormal angiogenesis and sinusoidal capillarization are prevalent in HCC. Targeting ligand surface functionalization enhances cellular uptake and tumor selectivity, for example, IKE nanoparticles encapsulated in PEG-PLGA show improved tumor enrichment through the EPR effect, reducing distribution to healthy abdominal organs and enhancing the therapeutic index [268]. The products of nanoparticle breakdown show changes in retention, potential toxicity and clearance, because cirrhotic livers have altered hepatic structural and function. Sinusoidal fibrosis and capillarization hinder regular absorption of hepatocytes and Kupffer cells, resulting in the buildup and longer retention of breakdown byproducts [269, 272]. These products e.g. polymers and metal ions may degrade oxidative stress and inflammation, particularly in a cirrhotic environment that is already redox imbalanced. Additionally, less effective removal may result from altered activity of macrophage and poor biliary excretion, raising the risk of cirrhosis. Chronic toxicity can also result from reduced biliary excretion and altered macrophage activity. Moreover, stellate cells may be activated via certain degradation products and immunological responses, which could aggravate fibrosis. Safety of nanoparticles in cirrhosis for effective clearance and biocompatibility, is therefore greatly context dependent and requires careful design.

Reduction in systemic toxicity

Application of therapeutic drugs is often accompanied by systemic toxicity and remains a major problem. Noncancerous tissues often experience oxidative stress, resulting nephrotoxicity or hepatotoxicity [269]. Nanocarrier based strategies reduce such side effects by stabilizing therapeutic payloads during circulation and directing them mainly to tumor locations. Strategies of controlled dosage and use of nanocarriers, decrease systemic exposure while preserving effective intratumoral concentrations.

Specific physicochemical properties of nanoparticles directly induce iron metabolism, redox balance, and lipid peroxidation rather than only assisting as drug carriers to induce ferroptosis. Cellular uptake and intracellular distribution is governed by the particle size. Although excessive reduction may increase off-target toxicity [273], however smaller sized nanoparticles increase iron accumulation and ROS production, thereby escalating Fenton reactions and lipid peroxidation. membrane interaction and lysosomal trafficking is regulated by surface charge, where cationic particles promote and iron release and endosomal escape, but may also disrupt membranes non-specifically, highlighting the need for balanced design. Crystallinity directs iron bioavailability, showing that iron oxides that are highly crystalline, sequester Fe(II), while labile iron is released as a result of degradation of less stable structures, producing ROS and induction of ferroptosis [274]. Redox-active compositions such as Fe, nanozymes Cu catalyze continuous ROS production, directly devastating GPX4, GSH and driving ferroptotic signaling [275]. Critically, these properties unite on the three ferroptosis axes: oxidative stress, iron overload, and lipid peroxidation, independent of drug delivery. Future approaches should focus on tunable physicochemical “windows” that maximize tumor-selective ferroptosis while minimizing systemic oxidative injury, alongside standardized nano structure activity relationship (nano-SAR) frameworks for safer clinical translation.

PEG PLGA nanoparticles loaded with IKE have shown tumor selective accumulation and extended circulation to be considerably reduced off target toxicity while maintaining ferroptotic activity. However, further modification in tumor specific targeting and controlled release mechanisms is crucial to maximize clinical benefits and safety.

Collectively, the physicochemical properties of nanoparticles are critical determinants of their ferroptosis-inducing efficacy. The particle size influences tumor penetration, cellular uptake, intracellular trafficking, and iron delivery, thereby affecting lipid peroxidation and ferroptotic cell death. Surface charge regulates membrane interactions, protein adsorption, and endocytosis, thereby influencing biodistribution and intracellular accumulation. Crystallinity governs metal-ion release and catalytic activity, with defect-rich or partially amorphous structures often exhibiting enhanced Fenton reactivity and ROS generation. Likewise, redox-active compositions containing iron, manganese, copper, or mixed-valence metal centers amplify the oxidative stress through catalytic redox cycling. Importantly, these parameters collectively determine intracellular iron availability, ROS production, antioxidant depletion, and lipid peroxidation, which are the key processes underlying ferroptosis. Therefore, the rational optimization of nanoparticle size, surface characteristics, crystallinity, and redox activity is essential for maximizing ferroptosis efficacy while maintaining favorable biocompatibility and safety. To summarize the relationships between these physicochemical properties and ferroptosis-inducing performance, Table 1 highlights the principal mechanistic roles and corresponding ferroptosis-related consequences associated with the key design parameters.

Table 1.

Influence of physicochemical properties of nanoparticles on ferroptosis-inducing performance

Physicochemical parameter	Mechanistic role	Ferroptosis-related impact
Particle size	Tumor penetration and cellular uptake	Intracellular iron accumulation and ferroptosis efficiency
Surface charge	Membrane interaction and endocytosis	Cellular delivery and ROS generation
Crystallinity	Ion release and catalytic activity	Fenton reactivity and lipid peroxidation
Surface area/porosity	Drug loading and catalytic accessibility	Ferroptosis sensitization
Redox activity	ROS amplification and redox cycling	Ferroptotic susceptibility
Surface functionalization	Targeting specificity and intracellular delivery	Tumor selectivity and reduced off-target effects

Controlled Generation of ROS

Iron oxide/iron-salt nanoparticles, Fe-based coordinating frameworks (MOFs/Prussian blue/COFs), lipid or biomembrane carriers, polymer coated nanocarriers, and catalytic nanozymes are examples of nanomaterials that regulate ROS generation. These nanomaterials cause HCC ferroptosis by (i) delivering labile Fe²⁺/Fe³⁺ to catalyze Fenton/Fenton-like conversion of tumor H₂O₂ into •OH under acidic endo/lysosomal pH, (ii) enzymatically “self-supplying” H₂O₂ (e.g., GOx) and cycling Fe²⁺/Fe³⁺ to sustain ROS, and/or (iii) co delivering ferroptosis sensitizers (sorafenib, RSL3, artesunate) that suppress SLC7A11/system Xc⁻ or GPX4 to weaken ROS detoxification [276]. Folate targeted artesunate loaded MSNs elevated ROS and iron while reducing SLC7A11/GPX4 and further slowed HepG2 xenograft growth; and pH dissociable Lipiodol Pickering emulsions co stabilized by CaCO₃ and hemin functioned as selffueling microreactors (LOX/hemin) that produced cytotoxic lipid radicals and inhibited orthotopic N1S1 HCC in rats following embolization [277].

Classification and targeting of nanomaterials-based ferroptosis induction in HCC

Although ferroptosis-inducing nanoplatforms are commonly categorized based on their material composition, their therapeutic performance is ultimately determined by the specific ferroptosis-regulatory mechanisms they engage. From a mechanistic perspective, these nanosystems can enhance ferroptosis through several principal pathways, including intracellular iron supplementation and catalytic ROS generation, disruption of antioxidant defense systems such as the System Xc⁻–GSH–GPX4 axis, modulation of the tumor microenvironment, and synergistic integration with other therapeutic modalities. Therefore, beyond describing individual nanomaterials, the following sections emphasize the distinct ferroptosis-promoting functions and therapeutic implications of each nanoplatform category.

Biomimetic co-delivery of nanoparticles

Biomimetic nanotechnology which combines tumor targeted delivery, immune evasion, and synergistic therapeutic action on a single platform, has become an effective approach to improve ferroptosis based therapy in HCC. To co deliver lenvatinib and FePt nanoparticles (Len/FePt@CMP NPs), one typical method loading poly(lactic-co-glycolic acid) (PLGA) nanoparticles coated with HCC cell membranes is used (Fig. 7). Mechanistically, lenvatinib induces ferroptosis by suppressing GPX4, which is achieved by blocking fibroblast growth factor receptor 4 (FGFR4) signaling. However, FePt nanomaterial increase apoptotic signaling and oxidative stress. The coating of the cancer cell membrane provides immune evasion and homologous targeting, allowing for favored tumor accumulation and extended circulation [278]. This biomimetic design overcomes the drawbacks of monotherapy and reveals a translationally promising approach for improved HCC care by affectedly increasing ferroptosis and apoptosis both in vitro and in vivo.

Fig. 7.

Biomimetic nanoplatform for co delivery of FePt nanoparticles and lenvatinib (Len) in HCC. Len/FePt@CMP is shaped by enclosing FePt and Len in PLGA nanoparticles followed by coating with cancer cell membrane (CCM), which improves tumor targeting and immune evasion. The accretion of nanoparticles in tumor tissue after systemic delivery can be tracked via photoacoustic imaging (PAI) technique. In addition to the role of Len in increasing treatment efficacy, FePt carries out Fenton reactions and results in lipid peroxidation and ROS production. Further GSH depletion and GPX4 inactivation contribute to oxidative stress and thus ferroptosis. Image is reproduced from copyright@ © 2025 Xuan F, et al. Advanced Healthcare Materials published by Wiley-VCH GmbH

Adding to the above paradigm, a bioinspired nanoplatform called Sv@PM-M2p has been created that simultaneously inhibits FSP1 and GPX4, the two important defense axes of ferroptosis. It amplifies lipid peroxidation and alters the immunosuppressive TME in tumor cells. This platform co-encapsulates small-molecule inhibitors (SF and vF) to create Sv@P using a GSH responsive disulfide bond linked PLGA (DS-PLGA) polymer. Sv@PM-M2p is produced by further camouflaging the nanoparticles with separated HCC membranes functionalized with an M2 macrophage binding peptide (M2pep). Both HCC cells and M2 tumor associated macrophages (TAMs) receive cargo from Sv@PM-M2p via homologous membrane targeting and M2pep mediated specificity. Mechanistically, redox triggered release of SF and vF in the GSH-rich intracellular milieu effectively inhibits GPX4 and FSP1, causing ferroptosis in tumor cells and M2 TAMs and eliciting strong immunogenic cell death (ICD) in vitro [279]. According to in vivo research, Sv@PM-M2p alters the immunological TME in allograft HCC models and inhibits tumor growth in cell line derived xenograft (CDX) models. Notably, anti-PDL1 monoclonal antibody combo therapy further inhibits distant metastases and raises long lasting antitumor immunity. Sv@PM-M2p shows better therapeutic efficiency than existing first line treatments in immune humanized patient derived xenograft (PDX) HCC models for a variety of etiologically different HCC subtypes [279]. These biomimetic co-delivery systems represent a next-generation ferroptosis based therapeutic approach that combines immunomodulation, redox-responsive release, dual ferroptosis pathway inhibition, and targeted nanodelivery to eliminate primary and metastatic HCC with the least amount of systemic toxicity.

Iron-based nanoparticles

One important first line treatment for advanced HCC is still sorafenib (Sor) [280], however, poor pharmacokinetics, restricted tumor selectivity, and acquired resistance frequently limit its therapeutic efficacy. To overcome these limitations, delivery techniques based on nanotechnology have been thoroughly investigated to enhance Sor bioavailability and ferroptosis triggering efficacy [281]. These nanomaterial include FeIII based metal organic frameworks (MOFs), which act as effective Sor carriers (Sor@Fe-MOF). They serve as multifunctional tools to combine iron mediated catalysis with drug delivery. Sor@Fe-MOF nanoparticles significantly induce ferroptosis and inhibit tumor development in HCC. These nanostructures increase intratumoral infiltration of CD8⁺ cytotoxic T cells, promoting antitumor immunity beyond direct cytotoxicity, significantly. This underscores their dual ability to induce ferroptosis and modify the immunological milieu. These results highlight the translational potential of Sor@Fe-MOF as a cutting-edge nanotherapeutic strategy for HCC [280]. Mechanistically, iron ions move into the acidic tumor microenvironment by MOF, which are then reduced to Fe²⁺ contributing to Fenton reactions to convert endogenous H₂O₂ into toxic •OH radicals, resulting ROS generation and lipid peroxidation. Concurrently, SLC7A11/system Xc⁻ pathway is inhibited by sorafenib, reducing cystine uptake synthesis of GSH, thereby inactivating GPX4 and weakening the cellular antioxidant defense system.

Likewise, MIL101(Fe) nanomaterial, have shown great potential because of their excellent drug loading efficacy, intrinsic peroxidase like activity, biocompatibility, T2 weighted magnetic resonance imaging capacity and regulated release kinetics. HepG2 cells undergo ferroptosis significantly in response to sor@(Fe)MIL101, as established by elevated levels of malondialdehyde and lipid peroxidation as well as decreased expression of GPX4 and GSH [276]. sor@(Fe)MIL101 capably decreases the growth of tumors in vivo with minimal long term damage, and it increases ferroptotic responses when combined with the tumor penetrating peptide iRGD, emphasizing the significance of improved intratumoral delivery.

Moreover, targeted iron based nanovesicles have been established to increase ferroptosis induced by Sor. For example, transferrin functionalized Fe³⁺ nanovesicles loaded with Sor (sor@TF-Fe³⁺ NVs) selectively target HCC cells that overexpress the TFRC and preferentially concentrate in the liver. Compared to Sor or TF-Fe³⁺ nanovesicles alone, these nanovesicles promote lipid peroxide formation more capably by improving Fe³⁺ absorption and intracellular redox cycling. In HCC mice models, sor@TF-Fe³⁺ NVs thus attain better tumor growth inhibition and longer life [277]. This targeted nanovesicle approach emphasizes on the advantage of biomimetic design in increasing selective drug delivery and enhancing ferroptosis via elevated iron uptake and redox activity.

Besides Sor based systems, various other agents have been used to generate ferroptosis inducing nanotherapeutics. Furthermore, magnetic FeO₄ nanoparticles have been combined with arsenic trioxide (ATO), a strong inducer of ferroptosis whose effects can be reversed by the iron chelator deferoxamine, and disguised with HCC cell membranes to produce AFN@CM. With good biosafety profiles, this biomimetic platform dramatically reduces GPX4 expression, increases intracellular lipid peroxide levels, and improves ferroptosis-mediated tumor suppression in vitro and in vivo [282]. Similarly, GOx/EC@Fe^₃O^₄@CCM nanoplatforms use catalytic pathways to significantly decrease tumor growth by elevating oxidative stress and inducing ferroptosis in HCC cells [283]. Collectively, these nanomaterial based approaches reveal how the rational combination of ferroptosis induction, immune activation, tumor targeting, and iron administration can significantly improve HCC treatment outcomes. Nanoparticles offer an innovative way to improve ferroptosis driven therapeutics by permitting regulated ROS production, improving drug stability, and promising tumor specific accumulation, beyond the limitations of traditional small molecule delivery (Fig. 8A).

Fig. 8.

Iron based nanomaterials and redox driven nanozymes. (A) Schematic illustration of the use of Sor@Fe-MOF for the treatment of HCC. Benzoic acid and ferric chloride hexahydrate alogwith tetrakis makeup FE-MOF. FE-MOF and that of sorafenib collectively produce Sor@Fe-MOF nanoparticles. Following injection, Sor@Fe-MOF NPs may enter the tumor site and prevent tumor growth. Sor@Fe-MOF treatment has been revealed to promote ferroptosis in HCC cells by upregulating ACSL4 and downregulating SLC7A11 and GPX4. Meanwhile, Sor@Fe-MOF has a significant ability to alter the immune microenvironment by increasing CD8 + T cytotoxic cell activation and tumor invasion. Image is reproduced from research article © 2025 The Yuanliang Yan et al. Published by Elsevier Ltd. (B) An illustration of therapeutic and imaging strategy of a redox-driven hybrid nanoenzyme for HCC. In order to improve circulation and targeting, mesoporous organosilica nanoparticles are generated from TEOS/BTES, loaded with iron and glucose oxidase (GOx), and then encapsulated with a tumor cell membrane. Under glucose-rich conditions, iron promotes production of ROS for lipid peroxidation, resulting in depleted GSH and thus ferroptosis. On the other hand GOx catalyzes oxidation of glucose to form acid and H₂O₂. Accumulation of Cystine alongwith NADPH depletion limit the pentose phosphate pathway (PPP) in glucose-deficient conditions, resulting in disulfidptosis. Hence, redox-driven hybrid nanoenzyme promotes anticancer efficacy via ferroptosis or disulfidptosis induction based on metabolic state. Further system’s iron content also make noninvasive tumor monitoring possible in vivo through T2 weighted MRI. Image is reproduced from © 2025 Qiao-Mei Zhou et al. Published by Elsevier Inc

Redox-driven Nanozyme-based activation of ferroptosis

Redox driven nanozyme platforms are a new and very flexible method for precisely inducing ferroptosis in HCC. A GSH responsive, biomimetic hybrid nanoenzyme system (M@GOx/FeHMON) was recently developed. It is made of hollow mesoporous organosilica nanoparticles that are co loaded with GOx and Fe²⁺/Fe³⁺ redox pairs. It is covered with homologous tumor cell membranes to improve tumor-specific targeting and immune evasion [259]. GOx catalyzes the oxidation of glucose to produce H₂O₂ in tumor areas that are high in glucose following transformation into extremely harmful hydroxyl radicals by iron mediated Fenton reactions. Through excessive oxidative stress, this GOx peroxidase cascade increases the generation of intracellular ROS, depletes GSH stores, and upsets lipid redox equilibrium, ultimately leading to ferroptosis [259]. This platform was designed to regulate TME to dynamically control the pathways leading to cell death. Crucially, adding Fe²⁺/Fe²⁺ redox cycling improves catalytic efficiency and maintains ROS production, supporting ferroptotic activity and lipid peroxidation.

On the other hand, glucose shortage alters the nanozyme which is a common incidence in tumor niches with inadequate blood flow. It causes disulfide stress induced cell death (disulfidptosis) and cytoskeletal instability by exacerbating glucose deprivation, which in turn inhibits the formation of NADPH and damages cystine metabolism. Because of its metabolic flexibility, M@GOx/Fe-HMON can disturb redox equilibrium in a variety of tumor areas by selectively inducing ferroptosis or disulfidptosis based on local nutrition availability [259] (Fig. 8B). This approach establishes receptor mediated targeting in inducing ferroptosis and improving therapeutic precision in HCC. Additional studies should explore the pharmacokinetics, scalability, and off-target interactions to confirm clinical safety and reproducibility.

Multifunctional magnetic Fe(3)O(4)-PEI@HA nanoparticles

The ability of multifunctional magnetic nanoplatforms to combine catalytic ROS generation, and targeted delivery and ferroptosis induction in a single system has drawn more interest. Accordingly, Fe₃O₄-PEI@HA nanoparticles with improved stability and tumor targeting potential were produced by effectively synthesizing a Fe^₃O^₄ magnetic nanocube complex modified with polyethyleneimine (PEI) and hyaluronic acid (HA) via thermal decomposition. PEI enhances drug loading efficiency and cellular internalization, while HA functionalization promotes preferred tumor-cell uptake by facilitating selective binding to CD44 receptors, which are often overexpressed in HCC. The subsequent Fe^₃O^₄-PEI@HA-RSL3 nanocubes capably repressed hepatoma cell proliferation by inducing the ferroptosis signaling cascade after loading with the ferroptosis inducer RSL3. Importantly, in vitro research revealed negligible cytotoxicity of these nanoparticles to healthy hepatocytes, emphasizing their role as potential therapeutic agents as tumor selective treatments. Mechanistically, Fe₃O₄-PEI@HA-RSL3 significantly amplified the production of intracellular ROS, which in turn increased ferroptotic cell death and lipid peroxidation. Accordingly, after being treated with nanoparticles, the expression of important ferroptosis related regulators, including lactoferrin, GPX4, FACL4 (ACSL4), and ferritin, was significantly downregulated in a dose dependent manner, indicating a disruption of iron homeostasis and antioxidant defenses [284] (Fig. 9A). Ferroptotic inducers other than RSL3 should be co delivered with Fe₃O₄-PEI@HA both invivo as well as invitro to check the efficacy of ferroptosis awith minimum side effects.

Fig. 9.

Magnetic and lactate-fueled nanomaterials inducing ferroptosis in HCC. (A) Fe₃O₄ nanoparticles along with hyaluronic acid (HA), polyethyleneimine (PEI) and GPX4 inhibitor RSL3 are loaded to create Fe₃O₄-PEI@HA-RSL3. After systemic inoculation, nanoparticles aggregate in tumor tissue mediated by HA-mediated targeting and enhance retention and permeability. Reduction in GSH and lipid peroxidation, promote RSL3 production in acidic tumor conditions, inhibiting system Xc⁻ and GPX4 action. Further, Iron ions also increase ROS production along with ferroptotic factors such as lactoferrin, ferritin, and ACSL4 to induce ferroptosis. Image is reproduced from © 2023 Elsevier B.V. All rights reserved. (B) PM@ESL nanoparticles are formed as a result of loading superparamagnetic iron oxide nanoparticles (SPIO), lactate oxidase (LOX), and erastin with a platelet membrane (PM). PM@ESL are accumulated inside the tumor after administration systemically. Lactic acid removal causes hydrogen peroxide (H₂O₂) production, which in turn reacts with Fe²⁺/Fe³⁺ from SPIO to form hydroxyl radicals. PM@ESL nanoparticles in combination with αPD-L1 immunotherapy, showed immunogenic cell death (ICD) and ferroptosis. Copyright © 2025 American Chemical Society

Lactate-fueled nanoparticles in ferroptosis induction

There is increasing evidence that elevated lactic acid level in the HCC microenvironment reduces ferroptosis sensitivity by enhancing metabolic adaptability and redox buffering ability. By co loading superparamagnetic iron oxide nanoparticles (SPIO) with erastin and lactate oxidase (LOX), and a platelet membrane (PM) engineered biomimetic nanoparticle system known as PM@ESL was created to accomplish MRI guided ferroptosis immunotherapy [285]. The capacity of nanoparticles to target tumors and evade the immune system is made possible by the platelet membrane coating promoting their preferred accumulation in tumor tissues. Mechanistically, LOX catalyzes the oxidation of lactate once it has been localized in the tumor microenvironment. This depletes intratumoral lactic acid and interferes with the lactate driven metabolic defense. Although SPIO offers an iron source to improve Fenton reactions and lipid peroxidation, this metabolic reprogramming restores ferroptosis sensitivity. Meanwhile, System Xc⁻ inhibited by erastin which lowers intracellular GSH and impairs GPX4 mediated antioxidant defenses. By a combination of increased iron mediated ROS, lactate depletion, and cystine transport inhibition, PM@ESL NPs significantly increase ferroptotic cell death in HCC (Fig. 9B). However, in depth mechanism is still warranted.

Importantly, SPIO use allows for MRI visibility, facilitating therapy tracking and real time imaging. Lactate modulation may also reduce immunosuppression in the tumor microenvironment, in addition to its direct tumoricidal effects, which combines antitumor immune activation with ferroptosis [285]. Altogether, PM@ESL NPs offer a promising translational methodology for the treatment of HCC by combining ferroptosis induction and immunomodulation in a complex example of metabolically tailored, image-guided nanotherapy.

Metal organic frameworks of nanoparticles

Metal organic frameworks (MOFs) are crystalline materials consisting of metal ions and organic ligands. They have a high porosity, a huge surface area, atomically defined catalytic centers, and a remarkable ability to load drugs [286]. Because of their biodegradability, biosafety, and structural tunability, MOFs have become attractive platforms for tumor diagnostics and treatment [287]. Crucially, by adjusting extrinsic synthesis parameters such as, pH, temperature, and manufacturing techniques [288], and intrinsic factors such as, ligands and metal nodes, their physicochemical properties can be precisely designed, allowing for customized therapeutic performance.

Apatinib (Apa) and Perfluoropentane (PFP) have been co loaded to create an iron based MOF nanoplatform (PFP-Apa-MOF) inducing ferroptosis and microwave ablation (MWA) synergistically in HCC. Acidic lysosomal environment favors iron ion release following cellular absorption, encouraging ferroptosis and lipid peroxidation. PFP alters phases to form microbubbles under microwave irradiation, which allow for improved drug release and real-time ultrasound (US) imaging and accurate ablation zone monitoring. After ablation, apatinib, an antiangiogenic drug, further inhibits tumor recurrence. PFP-Apa-MOF together exhibit strong anticancer efficacy, an efficient induction of ferroptosis, imaging-guided therapeutic capabilities, and acceptable biocompatibility, underscoring the translational promise of MOF based nanomedicine for integrated HCC therapy [289]. Mechanistically, iron released by MOF inhibits glutathione and GPX4 via Fenton reaction, contributing to ROS production and lipid peroxidation. Further PFP-Apa-MOF is non-toxic, indicating its role as a potential therapeutic agent in HCC treatment.

Mesoporous silica nanoparticles (MSNs)

Mesoporous silica nanoparticles (MSNs) have become a prevalent platform in nanomedicine because of their high drug loading capacity, huge surface area, great biocompatibility and variable pore size. However, the lack of intrinsic tumor selectivity in traditional MSNs limits their therapeutic efficacy in HCC. Furthermore, reticuloendothelial system (RES) clearance can decrease nanoparticle bioavailability and prevent tumor formation [290]. Importantly, negatively charged surface of MSNs make functionalization simple, allowing proteins, nucleic acids, medicinal drugs, and targeting ligands to conjugate to create multipurpose, tumor specific delivery systems.

In HCC, artemisinin loaded MSNs with folic acid has been developed to induce ferroptosis. HCC cells have been shown to overexpress folate receptors, so alteration in FA enhances drug absorption and delivery inside tumors. MSN-ART and MSN-ART-FA have been shown to improve the bioavailability of ART and promote ferroptosis significantly.

In vitro results further reveal significant suppression of the migration, invasion, and proliferation in HCC. MSN-ART-FA further inhibited tumor growth in HepG2 mice models, emphasizing the significance of ferroptotic efficacy [21]. Mechanistically the expression of main ferroptotic proteins in ferroptosis and the levels of ROS, malondialdehyde (MDA), Fe2+, and that of GSH were tested. Further studies are needed to explore proper axis confirming this mechanism. Collectively, MSNs offer a flexible and therapeutically promising platform for ferroptosis based therapy in HCC. Hence, MSN-ART-FA offer higher therapeutic potential, tumor specific delivery, and improved drug stability (Fig. 10A).

Fagopyrum dibotrys-derived nanovesicles induce ferroptosis

Fagopyrum dibotrys (Fd), a perennial plant, is generally used in traditional Chinese medicine. It is rich in a variety of bioactive substances, such as flavonoids and polyphenols, linking to anti-inflammatory and anticancer properties [291]. Fagopyrum dibotrys extracts have been shown to have anticancer properties via suppression of inflammatory mediators and apoptosis [292]. Derivative nanovesicles (FdNVs) of Fagopyrum dibotrys have bioinspired nanotherapeutic strategy for hepatocellular cancer.

According to in vitro analyses, both FdNVs and Fagopyrum dibotrys extracts inhibited HepG2 cell migration, invasion, and proliferation in a dose dependent manner. Further ferroptosis induction is a key mechanism behind these anticancer effects of these NVs. FdNV treatment reduces GSH, obvious mitochondrial alterations, ferroptosis, increased intracellular ROS and Fe²⁺ levels. FdNVs downregulated GPX4 and SLC7A11 and raised pro ferroptotic regulators like ALOX15 and p53 [293]. Mechanism in depth is still unclear, however it was revealed that reduction in GSH and that of ROS production might be established through GPX4 and SLC7A11 downregulation as a result of administration of FdNVs derived nanovesicles, which in turn weakened antioxidant defense system of cell to promote ferroptosis. These results represent a viable, biocompatible methodology to ferroptosis based treatment in HCC, combining nanotechnology with conventional herbal medicine to widen the therapeutic perspective.

Lipoprotein (LDL) based nanoparticles

In HCC, lipoprotein based nanoplatforms have become metabolically integrated and biologically compatible methods for inducing ferroptosis. Manganese doped mesoporous silica nanoparticles (MnMSNs) are one strategy that uses redox interactions to efficiently decrease intracellular GSH levels (Fig. 10B). Mechanistically, consuming GSH and blocking system Xc⁻ mediated cystine absorption, these nanoparticles undertake dual ferroptosis induction when further functionalized with folate PEG and loaded with sorafenib (FaPEG-MnMSN@SFB). The results is GSH level is reduced [294, 295]. Endogenous LDLs and exosome mimetic carriers offer a physiologically accurate method of inducing ferroptosis in addition to synthetic silica-based systems. Omega3 PUFA, docosahexaenoic acid (LDL-DHA) reconstituted LDL nanoparticles have shown strong pro ferroptotic action. Incorporating into cellular membranes, DHA, a highly oxidizable lipid substrate, promotes lipid peroxidation. LDL-DHA treatment affectedly increases intracellular lipid ROS, decreases GSH, and inhibits GPX4 activity in liver cancer cells. In vivo, tumor tissues showed decreased GPX4 expression and increased lipid peroxidation, demonstrating successful ferroptosis induction and tumor growth inhibition [296]. These lipoprotein and silica-based nanoplatforms illustrate a metabolically associated strategy to amplify ferroptosis through GSH reduction and increased lipid peroxidation simultaneously. Future directions should emphasize on evaluating long term metabolic interactions, improving targeting specificity, and confirming controlled redox modulation to avoid systemic discrepancy.

Fig. 10.

Mesoporus silica and lipoprotein (LDL) based nanomaterials in HCC. (A) mesoporous silica nanoparticles (MSNs) loaded with ART combined with Folic acid (FA) create MSN-ART-FA nanoparticles. MSN-ART and MSN-ART-FA inside the cell inhibited GPX4 alongwith increased iron ions and inhibits cell proliferation, invasion and migration invitro and significantly rise the bioavailability of ART to induce ferroptosis. Image reproduced from Copyright © 2023, Dengyun Nie et al. (B) Manganese dopped silica nanoparticles (FA-PEG-MnMSNs@FP) in combination with sorafenib inhibit GPX4 antioxidant pathway and induce ferroptosis via ROS production in HCC. Further, low-density lipoprotein nanoparticles, iron-containing Ionic MIL-101(Fe)@SOR, nanocatalyst HKUST-1 and exosomes (Er/Rb@exosome) are designed to induce ferroptosis alongwith increased bioavailability and limited drug degradation

Hypoxia-activated ferroptotic platforms

Under low oxygen circumstances, Hypoxia responsive polymer micelles altered with arachidonic acid (AA) were developed to induce ferroptosis to inhibit HCC development. Amphiphilic polymer AA/ASP-AZO-Fc (AAAF), and FT-IR were created via self-assembly and ^1H-NMR analysis was used to confirm its structure. In vitro examinations verified that curcumin (Cur) nanoparticles AAAF have the ability to release drugs when exposed to hypoxia. Hypoxia improved nanoparticles uptake and significantly increased the antiproliferative effects in HepG2 cells according to recent studies. Both redox modulation and hypoxia responsive release, synergistically make HepG2 cells more susceptible to ferroptosis, whereas AA enrichment provides lipid substrates prone to peroxidation. Collectively, AAAF shows promise as a hypoxia activated nanocarrier for enhancing ferroptotic and anticancer responses in HCC [297]. Mechanistically, in a hypoxic environment AAAF micelles undergo structural cleavage, allowing release of curcumin and improved cellular uptake in HCC cells. Concurrently, arachidonic acid further delivers highly oxidizable lipid substrates, while hypoxia driven reduction of intracellular GSH declines GPX4 mediated antioxidant defense, promoting lipid peroxidation and thus ferroptosis induction.

pH activated nanomaterials

In xenograft models of HCC, ferroptosis and photodynamic treatment (PDT) can be combined in a synergistic manner under multimodal imaging guidance using an in situ pH activatable nanoplatform (SR780@Fe-PAE-GP). The photosensitizer SR780 is first coordinated with Fe³⁺ to generate SR780@Fe in this system. This is then encapsulated in carboxyl functionalized poly(β-amino esters)-poly(ethylene glycol) (PAE10K-PEG5K-COOH), a pH responsive polymer. Active tumor specificity is conferred by surface conjugation with the GPC-3-targeting peptide GP2633, which produces SR780@Fe-PAE-GP nanoparticles that may accumulate in GPC-3 overexpressing HCC cells via enhanced permeability and retention (EPR) as well as endocytose them via receptors. The polymer shell breaks down upon lysosomal acidification, which causes SR780 and Fe²⁺ to dissociate (“off-on” activation). PDT mediated by SR780 is activated by laser irradiation 808 nm, which result in apoptosis and makes photoacoustic (PA) imaging and NIR-II fluorescence possible. Mechanistically, after administration into HCC cells, SR780@Fe-PAE-GP. PDT with NIR808 nm, reduces GSH level that triggers liberated Fe³⁺ to be reduced to Fe²⁺, increasing oxidative stress and hydroxyl radicals via the Fenton reaction. This cascade inhibits GPX4 via GSH reduction, thereby increasing LPO, damages membrane integrity, and eventually ferroptosis induction. Furthermore, the release of Fe²⁺ enhances the signals from T1 weighted MRI, allowing for triple modality imaging (NIR-II/PA/MRI) [298] (Fig. 11).

Fig. 11.

In HCC xenograft/PDX models, SR780@Fe-PAE-GP was developed by loading SR780@Fe with pH-responsive polymers, modified by a glypican-3 (GPC-3) receptor targeting peptide. SR780@Fe-PAE-GP a pH activatable nanoparticle, after its systemic administration, is accumulated in tumor tissue and activated in acidic environment, allowing regulated release of drug. Photodynamic treatment (PDT) increases the production of ROS with NIR808 nm irradiation, depleting GSH, whereas increasing LPO via iron-mediated Fenton reactions, and eventually apoptosis and ferroptosis induction in HCC. GPC-3 receptors facilitate increased tumor selectivity. For real-time tumor monitoring, the system combines multimodal imaging capabilities, such as MRI, NIR-II fluorescence, and photoacoustic (PA) imaging. Copyright © 2022 Elsevier B.V. All rights reserved

Prodrug nanoparticles (PE@PTGA)

Erastin has been capable of inducing ferroptosis in HCC, however its limited bioavailability and poor water solubility inhibit it from being used in clinical settings. In order to solve these issues, A self-assembled prodrug nanoplatform (PE@PTGA) has been created and tested incorporating protoporphyrin IX (PpIX) and erastin within an amphiphilic polymer matrix (PTGA), using orthotopic xenograft model for HCC. PE@PTGA enables PpIX and erastin access to tumor cells. Hyperthermia and ROS are favored by Erastin has been capable of inducing ferroptosis in HCC, however its limited bioavailability and poor water solubility inhibit it from being used in clinical settings. In order to solve these issues, a self-assembled prodrug nanoplatform (PE@PTGA) has been created and tested incorporating protoporphyrin IX (PpIX) and erastin within an amphiphilic polymer matrix (PTGA), using orthotopic xenograft model for HCC. PE@PTGA enables PpIX and erastin access to tumor cells. The mechanism consists of delivery of both erastin and PpIX into tumor cells via PE@PTGA, where erastin inhibits the system Xc^-, resulting in reduced GSH and cystine uptake, following inactivation of GPX4, thereby weakening antioxidant defense of the cell to induce ferroptosis. PpIX produces hyperthermia and ROS when exposed to light, inhibiting the growth of tumor cells directly and increasing oxidative stress [299]. Moreover, PE@PTGA shows activation of ferroptosis and apoptosis related pathways simultaneously, reducing tumor growth. Thus translational potential of PE@PTGA highlighted by its significant biocompatibility and low systemic toxicity promotes ferroptosis-apoptosis treatment approach for HCC. The PE@PTGA nanoplatform exhibits a combination approach, where co-delivery of PpIX alongwith erastin effectively increases oxidative stress and suppressing antioxidant defense system of tumor cells. This dual trigger system, incorporating light-responsive ROS production with inhibition of system Xc^-, reveals a synergistic mechanism that induces ferroptosis and apoptotic pathways simultaneously.

Collectively, the nanoplatforms discussed in this section demonstrate that ferroptosis induction in HCC can be achieved through several complementary mechanistic strategies, rather than through nanomaterial composition alone. Iron-based nanoparticles and MOFs primarily promote ferroptosis through intracellular iron delivery and Fenton-catalyzed ROS amplification, respectively. Nanozyme systems further enhance oxidative stress through enzyme-mimetic catalytic reactions, whereas hypoxia-responsive, pH-responsive, and lactate-modulating platforms increase ferroptosis susceptibility by remodeling the tumor microenvironment. Biomimetic, lipoprotein-based, mesoporous silica, plant-derived nanovesicles, and prodrug nanoplatforms predominantly improve ferroptosis efficacy through targeted delivery, disruption of antioxidant defense pathways, and integration with complementary therapeutic modalities. Importantly, these mechanistic differences are associated with distinct advantages and limitations in terms of ROS generation efficiency, glutathione depletion, tumor selectivity, therapeutic efficacy, and biosafety. Therefore, the future development of ferroptosis nanomedicine should focus on mechanism-guided platform design that balances ferroptosis-induction potency with translational feasibility, manufacturing reproducibility, and long-term biological safety.

To facilitate comparisons among the diverse ferroptosis-inducing nanoplatforms discussed above, Table 2 summarizes their functional classifications, dominant ferroptosis-regulatory mechanisms, tumor microenvironment interactions, therapeutic characteristics, and reported preclinical outcomes in HCC.

Table 2.

Functional classification, ferroptosis-regulatory mechanisms, and preclinical characteristics of representative nanoplatforms for HCC therapy

Nanoplatform	Functional class	Primary ferroptosis driver	Antioxidant pathway affected	TME-responsive feature	Additional therapeutic function	Preclinical therapeutic outcome	Safety
Sor@Fe-MOF	Iron-supplying catalytic platform	Iron delivery and Fenton catalysis	SLC7A11/GPX4	Immune activation	Sorafenib delivery	Tumor suppression	Good
sor@(Fe)MIL101	Iron-supplying catalytic platform	Iron-mediated lipid peroxidation	GPX4	Not reported	iRGD-assisted delivery	Tumor inhibition	Good
M@GOx/Fe-HMON	Redox-driven nanozyme platform	GOx–Fenton cascade	GSH depletion	Metabolic stress	Ferroptosis/disulfidptosis	Marked efficacy	Favorable
Fe₃O₄-PEI@HA-RSL3	Antioxidant-disrupting platform	ROS generation	GPX4 inhibition	Not reported	CD44 targeting	Tumor inhibition	Favorable
PM@ESL	TME-responsive platform	Lactate depletion and iron-mediated ROS	System Xc⁻ inhibition	Lactate remodeling	MRI-guided immunotherapy	Potent efficacy	Favorable
MSN-ART-FA	Targeted ferroptosis platform	Artemisinin-mediated ROS generation	GPX4 regulation	Not reported	Folate targeting	Tumor inhibition	Good
FdNVs	Plant-derived bioactive platform	ROS accumulation	GPX4/SLC7A11 suppression	Not reported	Natural bioactive cargo	Antitumor activity	Excellent
FaPEG-MnMSN@SFB	Antioxidant-disrupting platform	GSH depletion	System Xc⁻/GPX4	Not reported	Folate targeting	Ferroptosis induction	Favorable
AAAF micelles	Hypoxia-responsive platform	Lipid peroxidation	GSH depletion	Hypoxia activation	Curcumin delivery	Enhanced efficacy	Good
SR780@Fe-PAE-GP	pH-responsive platform	Iron release and PDT-enhanced ferroptosis	GPX4 suppression	Acidic pH activation	Multimodal imaging	Tumor suppression	Good
PE@PTGA	Multifunctional prodrug platform	System Xc⁻ inhibition and ROS amplification	GPX4 suppression	Not reported	Ferroptosis–apoptosis synergy	Tumor inhibition	Favorable

From a translational perspective, ferroptosis-inducing nanoplatforms have distinct advantages and limitations. Biomimetic nanoparticles, plant-derived nanovesicles, and lipoprotein-based systems generally offer superior biocompatibility and reduced immunogenicity. In contrast, iron-based nanoparticles, nanozymes, and metal–organic frameworks often achieve stronger ferroptosis-inducing efficacy but may face challenges related to biodegradation, metal-associated toxicity, and manufacturing reproducibility. Multifunctional platforms that integrate ferroptosis with immunotherapy, phototherapy, or imaging can further enhance therapeutic performance, but they also increase production and regulatory complexity. Therefore, successful clinical translation will require a balance between ferroptosis-induction potency and safety, scalability, reproducibility, and regulatory feasibility.

The cGAS-STING Axis: A positive feedback regulator of ferroptosis and anti-tumor immunity

Mechanistic interface between cGAS-STING activation and Ferroptosis in HCC

In order to create next generation treatment approaches for HCC, it is necessary to clarify the molecular interaction between ferroptosis and cyclic GMP-AMP synthase-stimulator of interferon genes (cGAS-STING) signaling. Since the liver is naturally iron rich and iron metabolism is frequently dysregulated in hepatic cancers, ferroptosis an iron dependent type of cell death caused by catastrophic lipid peroxidation and GSH depletion, has special significance in HCC. According to new research, ferroptotic stress may increase the release of oxidized DNA fragments and damage associated molecular patterns (DAMPs), which may activate the cGAS, STING pathway and improve type I interferon signaling and antitumor immunity. On the other hand, STING activation might make tumor cells more susceptible to ferroptotic death by intensifying oxidative stress and inflammatory signals [24]. Hence ferroptosis-STING coupled platform is positioned as a prospective immunometabolic therapeutic pathway in HCC because of this reciprocal interaction. Combining innate immune activation and iron dependent lipid peroxidation may result in direct tumor cell death as well as the remodeling of the immunosuppressive tumor microenvironment, which would improve immunotherapy response and immune surveillance (Fig. 12).

Fig. 12.

Activation of cGAS-STING platform as a prospective immunometabolic therapeutic pathway in HCC. (A) Ferroptotic stress increases the release of oxidized DNA fragments and damage associated molecular patterns (DAMPs), activating cGAS-STING pathway. On the other hand ferroptosis is stimulated in cancer cells when the cGAS-STING pathway is activated. Activated STING signaling enhance NCOA4-dependent ferritinophagy, which rise free iron (Fe²⁺) and lipid peroxidation and thus ferroptosis. Ferroptotic cell death is further fueled by Mfn1/2-mediated mitochondrial fusion and GPX4 degradation, which simultaneously increase ROS levels. DNA fragments on ferroptosis, trigger cGAS, producing cGAMP, which in turn triggers STING and type I interferon (IFN) signaling, creating a positive feedback loop that intensifies ferroptosis. (B) ZMRPF is an ion-coordinated and self-assembled nanoparticle which initiates ferroptosis by inducing high levels of lipid ROS, leading to mitochondrial stress and the release of endogenous mtDNA. This mtDNA activates the cGAS-STING cascade via Mn²⁺ ions released from ZMRPF. ZMRPF uptake by HCC is facilitated when it is loaded with folic acid (FA). Later ZMRPF is translocated to acidic lysosomes where dissociation of its ionic coordination takes place, releasing Mn²⁺ and RSL3 into the cytoplasm. RSL3 induces ferroptosis generating cytotoxic lipid peroxides following mitochondrial damage and mtDNA release. Meanwhile, Mn²⁺ increases cGAS sensing of mtDNA, promoting cGAS-STING pathway activation and immune cell maturation alongwith production of IFN-β. IFN-β causes DC maturation and the release of damage-associated molecular patterns (CRT, HMGB1). Enhanced immune activation results in increased response of CD4⁺, CD8⁺, and M1 macrophages as well as reduced Tregs and M2 macrophages, thereby stimulating pro-inflammatory cytokine release supporting antitumor immunity. © 2026 Wiley-VCH GmbH

cGAS-STING-ferroptosis signaling in HCC tumor microenvironment

Ferroptosis and the cyclic GMP-AMP synthase, cGAS-STING pathway interact functionally in HCC through intricate, two-way feedback mechanisms that affect antitumor immunity and tumor growth. Mechanistically, in HCC cells, ferroptotic stress is characterized by high lipid peroxidation and mitochondrial dysfunction, and might overexpress TBK1/IRF3 signaling. As a result of ferroptosis induced oxidative damage to the mitochondria, mitochondrial DNA is translocated into the cytosol, where it acts as a strong cGAS activator. cGAMP is synthesized by cGAS upon cytosolic DNA recognition. This production activates STING and initiates downstream TBK1/IRF3 signaling. The generation of pro inflammatory cytokines and type-I interferons changes the tumor immunological milieu, encourages dendritic cell maturation, and cytotoxic T cells to infiltratration [23, 24].

Consequently, ferroptosis can activate cGAS-STING, which in turn initiates innate immune signaling [23]. Ferroptosis driven mitochondrial DNA translocation may convert immunologically “cold” HCC tumors into more inflammatory and immune responsive state.

However, despite these immunostimulatory effects, ferroptosis does not universally promote antitumor immunity and may exert immunosuppressive effects under specific conditions. Excessive or poorly controlled ferroptotic stress can impair effector T-cell survival and function, induce ferroptotic loss of activated CD8⁺ T cells, alter macrophage polarization, and promote the accumulation of immunosuppressive myeloid cell populations in the TME. Persistent lipid peroxidation and oxidative stress may further sustain chronic inflammatory signaling that supports tumor progression rather than immune elimination. Therefore, the immunological consequences of ferroptosis are highly context-dependent and influenced by the intensity, duration, and cellular targets of ferroptotic stress in the tumor microenvironment. Consequently, the therapeutic success of ferroptosis-based nanomedicine is likely to depend on maximizing tumor-selective ferroptosis while preserving the beneficial antitumor immune responses. Targeting the ferroptosis–cGAS-STING signaling axis therefore represents a promising strategy to enhance immunotherapy and control tumor progression in HCC.

Nanoparticles-based targeting of STING-ferroptosis pathway in HCC

ZIF-8-MOF a novel biomimetic nanoplatform that could activate ferroptosis via cGAS-STING pathway. In addition to encouraging DC maturation and M2 TAM polarization toward M1 TAMs, this cascade also improved CD4⁺ and CD8⁺T cell infiltration at the tumor site, fully activating the immune response, exhibiting more anti HCC activity [23, 24] (Fig. 12). Likewise, the FABP5 targeting nanocarrier co-delivery technology presents a novel framework for improving clinical therapy approaches and a promising method to increase RFA efficacy in HCC via ferroptosis. Mechanistically, coactivation of cGAS-STING and ferroptosis axis reports three main types of resistances. First, disabling the antioxidant defense axis of GPX4 and NRF2 overcomes chemoresistance resulting rise in lipid peroxidation and making tumor cells more susceptible to oxidative death. Secondly, this co-activation pathways reduces immunoresistance by remodeling the immunosuppressive tumor microenvironment, stimulating cytotoxic T cell infiltration and dendritic cell maturation, and triggering STING dependent type I interferon signaling. Third, it overcomes metabolic resistance by disrupting adaptive pathways including HIF-1α, and STAT3, which normally allow tumors to survive in inflammatory and hypoxic environments [24].

The merging of naommunotherapy, and multi-omic profiling offers opportunities for developing next generation HCC therapeutics capable of breaking down the intricate resistance networks that define this aggressive malignancy. With first-in-human trials expected in 2025–2027 and possible regulatory approval trajectories extending to 2030.

Other biological advances in chemo-immuno-photothermal therapy

Integration of c-GAS-STING axis with nanoparticle-induced ferroptosis and chemo-immunotherapy

Combining ferroptosis inducing nanomaterials alongwith traditional chemotherapeutic agents is one of the most effective methods for overcoming treatment resistance in HCC. Combining some other chemotherapeutic agents such as apatinib, doxorubicin, or sorafenib with ferroptosis activators can activate multiple cytotoxic pathways increasing tumor cell death and decreasing systemic toxicity. Ferroptosis and chemotherapy synergy can be supported mechanistically by focusing on metabolic and oncogenic signaling pathways. Mechanistically, SAMD1 suppression leads to the inhibition of Hippo pathway, and reduced expression of NUAK2. Additionally, SAMD1 inhibition also upregulates ITIH5 via H3K4me3 demethylation at its promoter region [300], further modulating Hippo signaling and increasing the sensitivity of HCC cells to chemotherapeutic agents. Example of a novel biomimetic nanoplatform integrating chemotherapy and immunotherapy is PZ@M-T which is created as result of intergration of Zn²⁺ and 2-methylimidazole (2-MIM) to make ZIF-8 Metal-organic framework (MOF). Tannic acid (TA) alongwith ZIF-8 generates hollow TA-ZIF-8 which is subsequently loaded with the chemotherapeutic drug PPII, yielding PPII/ZIF-8 (PZ). PZ Surface was coated with MSCm to create PZ@M aiding enhanced tumor accumulation. Upon intravenous administration, PZ@M-T precisely targets tumor mitochondria, releasing PPII and induce ferroptosis in tumor cells, causing damage to mitochondria, disrupting antioxidant system, and activation of cGAS-STING signaling pathway. Activated cGAS-STING pathway initiates immunomodulatory cascade, promoting DC maturation and subsequent T-cell activation, as well as polarization of tumor-associated macrophages (TAMs) from the pro-tumoral M2 phenotype to the anti-tumoral M1 phenotype. These processes potentiate systemic antitumor immunity.

FePt nanoparticles encapsulated in PLGA carriers coated with cancer cell membranes (Len/FePt@CMP NPs) have shown strong ferroptosis-apoptosis synergy. Their immune evasion and homologous targeting capabilities greatly enhance intratumoral delivery and therapeutic response in vitro and in vivo [278]. Similarly, iron based metal-organic frameworks, such Sor@Fe-MOF, enhance the anticancer efficacy of sorafenib by remodeling the tumor immunological milieu and promoting ferroptosis through iron-mediated lipid peroxidation [301]. The advantages of combination techniques are further demonstrated by other MOF-derived systems. Another example reveals that MIL-100@Apa@MPN has better antiproliferative effects than apatinib alone in HepG2 cells, inhibiting tumor development in vivo with lowest systemic toxicity [302]. Similarly, MIL101(Fe)@sor nanoparticles capably promote ferroptosis induction and serve as therapeutic potential agent in HCC models, particularly when given in combination with iRGD [276]. Using urea fueled enzymatic propulsion, urease driven nanomotor platform (CUNMs + Sor) allows magnetic targeting. It comprises sorafenib loaded with porous magnetic silica nanoparticles alongwith altered cyclodextrin-urease hybrids. This design improves tumor penetration of material under acidic (pH < 6.5) environment, and attains pH responsive drug release. Mechanistically, this combination suppresses tumor growth effectively by blocking the GSH/SLC7A11/GPX4, inducing ferroptosis [303]. These results collectively show that combining nanoparticle mediated chemotherapeutic agents offer a comprehensive strategy to HCC therapy.

Ferroptosis-based nanoparticles and immunotherapy

Combining immunotherapy with ferroptosis inducing nanomaterials, is a potent way to overcome immune resistance in HCC. Notably, it has been established that upregulating FSP1 counteracts sorafenib induced ferroptosis in GPX4 dependent way. This is associated with a worse clinical prognosis and a greater infiltration of immunosuppressive TAMs. Biomimetic nanoparticles co-loaded with Sor and the FSP1 inhibitor viFSP1, have been developed to target tumor cells and TAMs simultaneously to address this dual resistance mechanism. This dual target nanoplatform supports antigen presentation and increases cytotoxic CD8⁺ T cell infiltration in preclinical HCC models by increasing ferroptosis in both malignant cells and immunosuppressive macrophages. Crucially, anti PDL1 antibody combination therapy further inhibits metastasis and tumor recurrence, proving that immune checkpoint suppression can sensitize HCC due to ferroptosis driven tumor microenvironment modification. Thus, this dual target strategy based on nanoparticles creates a synergistic ferroptosis-immunotherapy paradigm and offers a viable way to combat immune evasion and SF resistance in HCC [279]. In addition to this approach, carrier free nanoprodrugs have been created to combine chemo-immunotherapy and ferroptosis induction. In HCC, FANPs, which self-assemble from a binary prodrug made of 3,19-isopropyl andrographolide (IPADE) and 5-fluorouracil (5-FU) connected by an ester bond, have strong anticancer activity [304]. Compared to its parent chemical andrographolide (ADE), IPADE exhibits higher cytotoxicity, increases CD8⁺ T cell infiltration, and triggers ferroptosis. With a remarkable tumor growth inhibition rate of 97.6% in murine HCC models, FANPs significantly boost adaptive immune responses when paired with a PD-1/PD-L1 small molecule inhibitor.

Nanomaterials can induce cell mediated, humoral, and immunological responses depending on the form, size, and surface chemistry [305]. Intrinsic immunogenicity of nanoparticles may increase antigen production and thus adaptive immune responses, contributing to more effective immunotherapies and vaccines [306]. Unchecked immunogenicity can cause immunotoxicity causing ROS generation, inflammasome activation and tissue damage [307]. Cytokine pathways and the complement system are also activated by nanomaterials, increasing systemic immune responses and hypersensitivity. Some nanoparticles cause immunosuppression, affecting B and T cell activity and high risk of infections [305]. Since nanomaterials can either suppress or stimulate systemic immunity based on their biodistribution and physicochemical properties, however immunogenicity is a vital factor representing nanomedicine safety and design.

Nanomaterial induced ferroptosis can either result compensatory PD-L1 overexpression via stress/innate immune signaling or lower PD-L1 by rewiring hypoxia. Since ferroptosis activators cause HMGB1 to release DAMP, and HMGB1 can promote PD-L1 transcription that is dependent on transcription factors such as TBK1, IRF3, and NF-κB, resulting in a DAMP checkpoint feedback loop [236, 308]. It was shown that Fe/Cu ion induced ferroptosis and Fc-SS-Fe/Cu induced mild PTT downregulate PDL-1 expression by inhibiting HIF-1α, and improving anti-PD-L1 efficacy [309] (Fig. 13a). Ferroptosis can both increase immunogenicity and induce PD-L1 (surface/exosomal), thus checkpoint blockade is often mechanistically justified. However, the direction and magnitude of PD-L1 change remain nanoplatform and context-specific, and are often not quantitatively resolved.

Fig. 13.

Combining nanomaterials with chemotherapy, immunotherapy, gas therapy and photothermal therapy to induce ferroptosis in HCC. (A) Example of a novel biomimetic nanoplatform integrating chemotherapy and immunotherapy is PZ@M-T which is created as result of intergration of Zn²⁺ and 2-methylimidazole (2-MIM) to make ZIF-8 Metal-organic framework (MOF). Tannic acid (TA) alongwith ZIF-8 generates hollow TA-ZIF-8 which is subsequently loaded with the chemotherapeutic drug PPII, yielding PPII/ZIF-8 (PZ). PZ Surface was coated with MSCm to create PZ@M aiding enhanced tumor accumulation. Upon intravenous administration, PZ@M-T precisely targets tumor mitochondria, releasing PPII and induce ferroptosis in tumor cells, causing damage to mitochondria, disrupting antioxidant system, and activation of cGAS-STING signaling pathway. Activated cGAS-STING pathway initiates immunomodulatory cascade, promoting DC maturation and subsequent T-cell activation, as well as polarization of tumor-associated macrophages (TAMs) from the pro-tumoral M2 phenotype to the anti-tumoral M1 phenotype. These processes potentiate systemic antitumor immunity. Image reproduced from Copyright © 2025. Hulinyue Peng et al. (B) Lenvatinib, PLGA-PEG-MAL, and IR780 assembly generates PP@LVN/IR780 nanoparticles. These NPs are further altered by cRGD to form PRP@LVN/IR780 for enhanced tumor targeting. Irradiation at 808 nm causes inhibition of VEGF/FGF pathway with lenvatinib, mediated by photothermal/photodynamic effects via IR780 to cause oxidative stress and ferroptosis in tumor cells. Moreover, this combination treatment in addition to the significant damage to tumor cells, allows for multimodal imaging-guided detection, such as photothermal imaging (PTI), fluorescence imaging (FLI), and photoacoustic imaging (PAI). Image reproduced from Copyright © 2025 American Chemical Society. (C) Combining CDT/SDT/gas treatment to increase the synergistic ferroptosis against orthotopic HCC model. The CO releasing chemical CORM-401 was encapsulated in bimetallic compound FM to create a multimodal therapeutic nanoplatform (FM/C). SP94-peptide was injected into the macrophage membrane to provide dual targeting capacity toward HCC. Besides promoting CDT to produce •OH, multi enzyme activity of FM enhanced SDT performance to produce O_1/2 by reducing hypoxia, which in turn activated CO release. GSH depletion and elevated ROS levels resulted toxic lipid peroxidation (LPO) accumulated thus induced ferroptosis in orthotopic HCC model. Image reproduced from © 2025 Wen Meng et al. Advanced Science published by Wiley-VCH GmbH

It is interesting to note that type I IFN via the cGAS–STING pathway can encourage PD-L1 induction if ferroptotic damage causes cytosolic DNA sensing [310]. Through the JAK/STAT3/p38 pathway, type I IFN can increase PD-L1 on immune cells [311]. However, direct connection along the nanoparticle-ferroptosis-STING-PD-L1 axis is still unclear in basic nanomedicine research and requires more investigation.

Nanoparticle induced ferroptosis and gas therapy

Ferroptosis inducing nanomaterial in combination with gas therapy has created new opportunities for multimodal treatment of HCC. To improve immune evasion, and tumor selectivity, FeMoO₄ nanoenzyme, carbon monoxide (CO)-releasing compounds, and SP94 modified macrophage membrane coating are integrated into an HCC targeted biomimetic nanoplatform. Mechanistically, the FeMoO₄ nanoenzyme promotes Fenton reactions, consequently producing hydroxyl radicals and increased lipid peroxidation. GSH deficiency also affects redox homeostasis, making tumor cells more susceptible to ferroptosis. The release of CO further alters the tumor microenvironment, which may improve oxidative stress mediated cytotoxicity with lower hypoxia. This platform has strong anticancer efficacy when exposed to ultrasonic radiation, indicating an altered tumor microenvironment and inhibited HCC tumor growth [312]. Following this idea, dual functional nanobubble systems that combine ferroptosis and sonodynamic treatment (SDT) have established distinguished synergistic benefits in vitro [313]. Collectively, ferrtoptosis inducing nanoparticle-gas therapy techniques offer better tumor selectivity, therapeutic efficacy, microenvironment manipulation, as a suitable multimodal paradigm for the treatment of HCC (Fig. 13c).

Nanoparticle induced ferroptosis and photothermal therapy

Combination of ferroptosis inducing nanomaterial alongwith photothermal therapy (PTT) has been discovered as an effective synergistic approach for HCC (Fig. 13). Biocompatible SQ890@Fe nanoparticles that can react to the TME are produced by encapsulation of SQ890@Fe in a GSH sensitive polymer (PLGA-SS-mPEG). Near infrared (NIR) dye SQ890, which serves as both a photothermal converter and an iron-chelating agent. Activation of nanoparticles in high GSH conditions promotes photothermal conversion and accelerates the Fenton process, producing hydroxyl radicals thus inducing lipid peroxidation and ferroptosis. GSH depletion weakens antioxidant system resulting more susceptibility of tumor cells to ferroptotic death. ROS by ferroptosis notably reduce the expression of heat shock protein (HSP), which reduces tumor thermotolerance and increases PTT efficacy. This reciprocal reinforcement of PTT and ferrotherapy highlights the potential of SQ890@Fe nanoparticles as a smart nanomedicine platform for better anticancer efficacy with better safety profiles [314].

Furthermore, cyclic RGD peptide (cRGD) utilizing PLGA-PEG polymer was created for integrin targeted administration in HCC. Together with ferroptosis induction, this approach enables combination photodynamic treatment and PTT by co-encapsulating lenvatinib (LVN) and the NIR photosensitizer IR780. After both active and passive tumor targeting, the nanoparticles hydrolyze in the TME due to acid, releasing IR780 and LVN. With IR780-mediated PDT/PTT, LVN-induced ferroptosis and apoptosis are synergistically enhanced with 808 nm laser irradiation, leading to substantial tumor suppression in both in vitro and in vivo models with negligible off-target damage. Additionally, by combining fluorescence imaging, photoacoustic imaging, and photothermal imaging, this nanoplatform enables it to visualize therapy reactions in real time [315] (Fig. 13).

The risk of thermal damage to normal liver tissue is reduced but not eliminated. It is reduced because both ferroptosis and PTT are designed to restrict photothermal activity to the tumor by using tumor biased nanoparticle accumulation and TME activation. For example, GSH responsive SQ890@Fe release and acid responsive cRGD-PLGA-PEG/LVN-IR780 release should increase heat generation mainly where acidity, glutathione, and tumor uptake are highest. Relevant risk is the laser itself. The 808 nm wavelength lies in the NIR-I window, which is commonly used for PTT but has lower penetration and more scattering than longer NIR-II wavelengths. A study clearly shows that NIR-I penetration is relatively limited, and 808 nm has a lower maximum permitted exposure than NIR-II lasers [316]. In practice, many 808 nm photothermal systems are designed to push tumor temperatures toward the ablation range, which is therapeutically useful but narrows the safety margin for the surrounding liver if heating is not spatially restricted. Ferroptosis alongwith PTT have a better safety profile than non-targeted PTT, because ferroptosis lowers the heat requirement and PTT can enhance ferroptosis, allowing synergistic killing rather than relying on heat alone. The highest risk scenarios would be superficial or infiltrative tumors, cirrhotic livers, broad irradiation fields, and off-target nanoparticle uptake by Kupffer cells or nearby parenchyma [317, 318]. Overall, the thermal risk appears manageable but clinically non trivial, and it should be interpreted as a design dependent safety issue rather than assumed absent.

Preclinical assessment of nanomaterials inducing ferroptosis

Invitro models, Invivo subcutaneous and orthotopic models of HCC and ferroptosis

In vitro, in vivo subcutaneous and orthotopic tumor models are necessary for analyzing the preclinical validation of nanotherapeutics based ferroptosis in HCC. Ferroptosis inducing nanoparticles have been shown to have strong anticancer effect by inhibiting tumor development in animal models and triggering lipid peroxidation cascades in cultured HCC cells.

Biomimetic FeO₄ iron oxide nanoparticles enveloped with HCC cell membranes (AFN@CM) might improve tumor targeting and immune evasion, arsenic trioxide (ATO) is best example of such combination. Ferroptotic cell death was confirmed in vitro by AFN@CM treatment, which significantly increased intracellular Fe²⁺ and ROS levels, inhibited GPX4 production, and increased lipid peroxidation. Importantly, the iron dependent specificity of the ferroptosis pathway was confirmed when the effects were abolished by the iron chelator deferoxamine [282]. AFN@CM nanoparticles successfully suppressed tumor development and showed increased tumor accumulation in vivo due to homologous membrane targeting in mice with HCC. Both orthotopic and subcutaneous xenograft models established reduced systemic toxicity and significant tumor suppression when compared to ATO treatment alone, indicating higher biosafety. Orthotopic models, liver specific vascularization, showing metabolic context, and immunological interactions that affect ferroptosis sensitivity, provide physiologically relevant tumor microenvironment [282]. Table 3 presents a comprehensive overview of experimental designs and results.

Table 3.

Ferroptosis inducing nanomaterails and their preclinical status

NPs type	Nanoparticle design	Drug	Ferroptosis mechanism	In vitro	In vivo	Findings	Ref.
Biomimetic NPs	Len/FePt@CMP	Lenvatinib	FGFR4 axis blocking, GPX4 inhibition	HCCLM3 and HepG2 cells	BALB/c nude mice	Induced ferroptosis, appoptosis	[278]
	Sor@PM-M2p	Sorafenib	FSP1 and GPX4 inhibition, lipid peroxidation ↑	patient derived xenograft, Huh-7, Hep3B	HCC mice	Induced ferroptosis	[279]
Iron based NPs	Sor@Fe-MOF	Sorafenib	CD8⁺ increased	Hep3B, HUVEC	Male athymic nude mice	Induced ferroptosis	[281]
	sor@(Fe)MIL101	Sorafenib	GSH, GPX4 levels reduced	HepG2 cells	5 week old Kun Ming mice	Induced ferroptosis	[276]
	sor@TF-Fe³⁺	Sorafenib	SLC7A11 inhibition	HepG2, Huh7, 7701, 7703	8 live cancer patients tissue samples, female BALB/c nude mice	Induced ferroptosis	[277]
	GOx/EC@Fe₃O₄@CCM	GOx	GSH, GPX4 levels reduced	Huh7, Hepa1-6	BALB/c nude mice	Induced ferroptosis	[283]
	AFN@CM	Arsenic trioxide (ATO)	GPX4 reduced	HepG2	mice	Induced ferroptosis	[282]
Nanozyme based	M@GOx/FeHMON	Glucoseoxidase (GOx)	GPX4 reduced, elevated ROS	H22 murine HCC cells	subcutaneous and orthotopic tumor bearing mouse models	Induced ferroptosis	[259]
Magnetic based NPs	Fe₃O₄-PEI@HA-RSL3	Magnetic nanocube Fe3O4	GPX4, ACSL4 downregulated, ROS↑, low expression of Lactoferrin, FACL 4, GPX 4, Ferritin	Hepatoma cells	--	Induced ferroptosis	[284]
Lactate fueled	PM@ESL	Lactate oxidase	GSH, GPX4 low	--	Tumor mice	Induced ferroptosis	[285]
Metal organic	PFP-Apa-MOF	Apatinib (Apa)	Elevated ROS	HepG2 cells	Female BALB/c nude mice	Induced ferroptosis	[286, 287, 289]
Mesoporous silica NPs	MSN-ART-FA	Artesunate	GPX4 depletion	HepG2	female BALB/cmice models	Induced ferroptosis	[21]
Fagopyrum dibotrys NPs	FdNVs	Fagopyrum dibotrys	GPX4, SLC7A11 downregulation	HepG2 cell inhibition	Male BALB/c nude mice	Induced ferroptosis	[291–293]
Lipoprotein based NPs	FaPEG-MnMSN@SFB	Sorafenib	GSH, GPX4 inhibited	HepG2,	Male SD rats	Induced ferroptosis	[294–296]
Hypoxia activated NPs	AA/ASP-AZO-Fc (AAAF)	ASP from Angelica root	GSSH, GSH depletion	HepG2 cells	GSH, GPX4 inhibited	Induced ferroptosis	[297]
PH activated	SR780@Fe-PAE-GP	SR780	GSH, GPX4 inhibited, ↑LPO	HCC cells	Nude mice	Induced ferroptosis	[298]
Prodrug NPs	PE@PTGA	Erastin	↑ROS	HCC cells	mice	Induced ferroptosis	[299]

Pharmacokinetics and biodistribution

The unique pharmacokinetic (PK) and biodistribution (BD) features of nanoparticle based ferroptosis treatments in HCC support their superiority over traditional small molecule inducers. Nanocarrier systems, such as iron oxide cores, polymeric micelles, and mesoporous silica frameworks, extend circulation time, improve tumor localization, controlled tissue distribution, in contrast to free agents like erastin or RSL3. Such characteristics are important to achieve prolonged iron dependent lipid peroxidation in vivo.

Because many nanomaterials have a liver dominant biodistribution, effective and safe treatment depends on maximizing tumor to normal liver partitioning and guaranteeing sufficient intratumoral penetration. By combining ferroptosis specific pharmacodynamic biomarkers e.g., lipid peroxidation products, GPX4 suppression, GSH depletion, with advanced quantitative BD/PK techniques e.g. radiotracing, inductively coupled plasma mass spectrometry (ICP-MS), translational development can be accelerated and exposure response relationships established [319, 320].

When compared to free drug formulations, biomimetic FeO₄ nanoparticles loaded with arsenic trioxide covered within tumor cell membranes, exhibit improved blood retention and prolonged systemic circulation because of immune evasion and decreased renal clearance, thus improving bioavailability. Ferroptosis based nanotherapy is controlled by multilayer PK dynamics. Systemic circulation and mononuclear phagocyte system clearance impact liver and spleen exposure; and microenvironment responsive drug release determines whether adequate intracellular iron-mediated oxidative stress is attained.

Clinical translation status

Existing nanoparticles induced ferroptosis-based clinical therapies of HCC

There are currently no approved routine clinical ferroptosis therapies based on nanomaterials specific to HCC as of February 27, 2026, and the majority of ferroptosis oriented nanoplatforms are still in the preclinical stage. Two primary approaches, however, are showing early signs of clinical translation: (i) first-in-human trials with ferroptosis specific nanomaterials and (ii) the repurposing of clinically proven hepatic nanoparticles that interact with the biology of iron and ROS. The formulation is expressly positioned as a ferroptosis based anticancer treatment, and its safety and pharmacokinetics are being evaluated in advanced solid tumors using intratumoral carbon nanoparticle loaded Fe(II) (CNSI-Fe(II), which is being evaluated with ClinicalTrials.gov identifier NCT06048367 [321]. Despite not being specific to HCC, this experiment is extremely pertinent to liver cancers that can be delivered intratumorally and is one of the initial stages toward the therapeutic application of ferroptosis nanotherapy [322].

Iron oxide nanoparticles, including ferumoxytol and superparamagnetic iron oxide nanoparticles (SPIONs), are also being studied for their immunomodulatory and macrophage-polarizing properties in irradiated hepatic tissue. They are already used in liver oncology practice, specifically for MRI-guided radiotherapy planning. Although these uses are not officially classified as ferroptosis treatments, they offer clinically verified exposure to nanomaterials in the context of HCC and could be the starting point for next ferroptosis-sensitizing techniques [322].

Additionally, locoregional particle platforms drug-eluting bead transarterial chemoembolization, or DEB-TACE, are well established in the treatment of HCC. These microsphere based systems allow for tumor-focused drug retention, ischemia, and oxidative stress within hepatic tumors, despite not being marketed as ferroptosis-inducing treatments. This provides a feasible and scalable way to incorporate ferroptosis-inducing payloads into current clinical workflows [323, 324].

NCT06048367, the first-in-human Phase I, single-arm, open-label trial, is evaluating intratumoral CNSI-Fe, which is projected to cause ferroptosis through lipid peroxidation and iron mediated oxidative stress and. In order to mechanistically evaluate ferroptosis activity, it uses pharmacodynamic indicators, such as malondialdehyde (MDA), H2O₂, GSH and GPX4, and a characteristic 3 + 3 dose escalation design with a 21 day dose limiting toxicity (DLT) window. In early dose escalation data, there was only one DLT at 90 mg which indorsed advancement to higher dose cohorts following safety evaluation. Preliminary findings from 19 patients point to favorable tolerability with minor local side effects and no significant systemic toxicity noted. Likewise, the early efficacy signals are uncertain but intriguing, with disease control rate (DCR) ~ 84% and overall response rate (ORR) ~ 10.5%, including infrequent full responses. However, diverse tumor types, the lack of recognized survival endpoints (e.g., PFS/OS), and small sample size limit interpretation of the study. All things measured, the trial offers early proof of concept for ferroptosis based treatment; nevertheless, still larger and controlled studies are required for validation [325]. Clinical translational analysis at nanomaterial based ferroptosis was shown in Table 4.

Table 4.

Clinical translational analysis at nanomaterial based ferroptosis

Category	Details
Trial Design	Phase I, first-in-human, single-arm, open-label study
Therapeutic Strategy	Intratumoral CNSI-Fe inducing ferroptosis via iron-mediated oxidative stress and lipid peroxidation
Dose Escalation	Standard 3 + 3 design; dose levels: 30, 60, 90 mg (Fe²⁺); escalation 120–150 mg
DLT Window	21 days
Population	Adults (18–80 years) with advanced solid tumors, refractory or no standard treatment
Key Exclusion Criteria	Iron metabolism disorders (e.g., thalassemia, G6PD deficiency), uncontrolled comorbidities, recent anticancer therapy,
Pharmacodynamics (PD)	Ferroptosis indicating biomarkers: H₂O₂, GSH, MDA, GPX4
Pharmacokinetics (PK)	Serum iron levels post-dose
Enrollment	19 patients
Timeline	Start: Nov 2022; Primary completion: Sept 2024; Final completion: Feb 2025
Safety Findings	Only 1 DLT at 90 mg; no major systemic toxicity reported, mostly mild local AEs (nausea, pain)
Treatment Plan	Intratumoral injection Day 1; monitoring on Days 7, 14, 21; optional second cycle if tolerated
Follow-up	Up to 28 days after last dose
Limitations	Small sample size, heterogeneous tumors, lack of PFS/OS and detailed response duration
Risk Profile	Potential risks: liver toxicity, hematologic effects, iron overload (based on ICF, not confirmed incidence)
Efficacy Signals	ORR ~ 10.5%, DCR ~ 84%, rare complete responses
Overall Interpretation	Early proof-of-concept for ferroptosis-based therapy; requires larger controlled trials for validation

Number of individuals and eligibility criteria

The trial enrolls adult around 18–80 years with solid tumors at advance stage, who lack effective treatment options or have failed standard therapies. Main exclusions criteria involves patients with disorders related to iron metabolism e.g. G6PD deficiency, thalassemia, recent anticancer treatments, and uncontrolled comorbidities such as hypertension.

Endpoints, response and PD plan

One of the key safety endpoints include the incidence of dose limiting toxicities (DLTs). Efficacy estimation is carried out via response evaluation criteria in solid tumors version 1.1 (RECIST v1.1) with confirming imaging examinations, including disease control response (DCR, ORR) and overall response rate. Strong sampling on early basis following dosage in pharmacokinetics was used to measure serum iron levels. Tumor tissues were used in pharmacodynamic analyses to assess lipid peroxidation indicators like GPX4 and MDA markers of ferroptosis. A standard 3 + 3 dose escalation is the design of trail including cohorts (30, 60, 90 mg Fe²⁺) and it could be expanded to 6 patients in case of DLTs. The DLT evaluation period is 21 days, with possible adjustment of dose based on toxicity.

Treatment plan and follow-up

After screening (-28 to -days), patients receive intratumoral injection on day 1, with safety monitoring on Days 7, 14, and 21. A second treatment cycle may be given if tolerated, followed by safety follow-up up to 28 days after the last dose.

Enrollment status and key dates

The trial started from November 2022, with primary completion in September 2024 and final completion in February 2025 and enrolled 19 patients. The results were posted in 2025. Though results submission is confirmed, detailed result tables are not fully accessible.

Safety signals and DLTs

DLTs were defined using CTCAE v5.0, including severe hematologic and ≥ grade 3 non-hematologic toxicities. Only one DLT was observed at the 90 mg dose, allowing escalation to higher doses. Sponsor data suggest good tolerability, with mainly mild local adverse effects and no major systemic toxicity, though this is based on non-peer-reviewed summaries. Patient still highlight potential risks such as hematologic effects, iron overload and liver toxicity.

Preliminary efficacy and protocol evolution

Diverse tumor forms were included in the trial, indicating a heterogeneous population characteristic of phase I trials. Better efficacy of the drug was revealed in early outcome data, with an overall response rate (ORR) of about 10.5%, and a disease control rate (DCR) of about 84%. However, vital results such as comprehensive response durations and PFS/OS and were not revealed. The protocol adjustments (V3.1/3.2) which came out in March 2024, were based on the results of the safety assessment, mainly focusing on the single DLT at 90 mg. These changes made it possible to keep escalating the dose to greater levels such as 120–150 mg). There is no availability of complete public comparisons of protocol.

Production and scalability at market level

Currently, manufacturing complexity and chemistry, manufacturing, and controls (CMC) preparation are more important factors limiting the large scale translation of ferroptosis inducing nanomedicines for HCC than biological feasibility. Though limited ferroptosis inducing nanomaterials entering human trials are still in the early dose-escalation phases, such as current intratumoral carbon-nanoparticle Fe(II) formulation assessed in NCT06048367, where scalable production workflows and good manufacturing practice (GMP) processes are being optimized before commercialization. The majority of HCC targeted nano-ferroptotic platforms are still in the preclinical stage. From an industrial point of view, scaling depends on regulatory compliance, reproducibility, batch to batch consistency, and sterility assurance, all of which are especially difficult for complex nanostructures that contain iron, catalytic components, or biomimetic coatings. The transition of nanoparticle platforms from laboratory scale fabrication to commercial manufacturing is being aided by the encouraging growth of specialized contract development and manufacturing organizations (CDMOs). The use of high throughput production technologies indicate continuous flow synthesis, microfluidics, controlled extrusion, and nanocrystallization. Many ferroptotic inducing nanomaterials are designed using lipid based or hybrid nanocarriers for cancer therapeutics [325], however market analyses indicate a remarkable rise in the CDMO capacity of lipid based nanoparticles.

A formal cost analysis for NCT06048367 has not been publicly disclosed, which is typical for early phase industry sponsored trials due to commercial confidentiality. From a patient view, early phase oncology studies can create extensive financial burden, as over half of patients indicate out of pocket expenses above $1,000 per month, mainly for travel and housing [326]. Same challenges are likely present due to its specialized single site design, although not exclusive to this trial. NCT06048367 seems to be a small trial, which means it will probably cost less overall than later phase studies. However, each patient’s expenditures may go up due to its high mechanistic complexity (ferroptosis biomarkers, nanomedicine platform). Likewise, the absence of efficacy endpoints (PFS/OS) limits the evaluation of cost effectiveness.

Challenges for clinical nanomedicine in HCC

Significant biological and translational challenges still exist regarding the clinical application of ferroptosis inducing nanodrugs despite promising preclinical results for HCC. The key challenge is tumor heterogeneity, as ferroptosis susceptibility varies greatly depending on metabolic phenotype, genetic background, and tumor microenvironment. Emerging evidence suggests that HCC subtypes characterized by elevated iron metabolism, increased lipid peroxidation, and greater dependence on oxidative metabolic pathways are more susceptible to ferroptotic cell death. In contrast, tumors exhibiting enhanced antioxidant defense mechanisms, including increased GPX4, SLC7A11, and NRF2 activities, may exhibit relative ferroptosis resistance. Furthermore, differences in hypoxia, nutrient availability, and immune composition within the tumor microenvironment can further influence ferroptosis responsiveness. These findings underscore the importance of biomarker-guided patient stratification and personalized ferroptosis-based nanotherapeutic strategies for HCC. Consistent with this heterogeneity, the regulatory networks controlling iron metabolism, antioxidant defenses, and lipid remodeling vary substantially among HCC subtypes, reducing the predictability of treatment response and complicating patient stratification [327].

Moreover, there are important physiological limitations on nanocarrier delivery. The rapid clearance by the mononuclear phagocyte system, limited intratumoral penetration, and preferential sequestration in the liver and spleen may lead to reduced accumulation of therapeutic agents within malignant hepatocytes and increased off-target toxicity [328]. Precise regulation of drug release kinetics, catalytic iron activity, and redox responsiveness is necessary to activate ferroptosis without causing systemic oxidative damage or unintentional immune activation [329]. Translational barriers continue to exist at the manufacturing and regulatory levels in addition to biological ones. Multifunctional nanomedicines must be produced on a large scale with batch reproducibility, strict quality control, and detailed safety assessment [330]. The regulatory pathways for complex ferroptosis-inducing nanomedicines require rigorous evaluation of pharmacokinetics, biodistribution, long-term toxicity, and risk–benefit profiles before integration into the routine clinical management of HCC.

Future perspectives of commercial development of clinical drugs

Advances in nanotechnology engineering, precision oncology and tumor biology are expected to speed up the commercial development of ferroptosis inducing nanomedicines for HCC. In order to improve alter TME, tumor specific accumulation, and combined ferroptosis with immunotherapy, chemotherapy, or phototherapy, scientists have been logically developing novel next generation nanodrug delivery systems (NDDSs). These techniques offer a potential framework for enhancing therapeutic efficacy, while preventing systemic toxicity to increase the chance of clinical translation in the upcoming years [331]. Current preclinical researches illustrate promise for biomimetic and targeted nanoplatforms that co-deliver synergistic medicines, targeting peptides or immune modulators, alongwith ferroptosis inducers. Reducing off-target effects, maximizing intratumoral specificity and overcoming resistance mechanisms, are the goals of such approaches. These novel nanotechnologies show great translational potential and commercial capability if issues regarding scalable production, long term safety validation, and regulatory compliance can be resolved. While ferroptosis based nanodrugs have not yet been approved for HCC, the development of responsive, immune compatible, and precision targeted nanocarriers, along with a growing understanding of ferroptosis biology, should hasten the transition from bench to bedside. The next generation of economically viable nanotherapeutics for HCC may be defined by biomimetic and microenvironment responsive systems, rational combination techniques, and substantial antitumor activity demonstrated in preclinical models [332].

AI-driven nanoparticle designs for HCC

Ferroptosis based nanotherapy for HCC is being advanced by the revolutionary technique of artificial intelligence (AI)-guided nanoparticle engineering. To optimize tumor accumulation while reducing reticuloendothelial system (RES) sequestration and off-target oxidative damage, machine learning (ML) models can systematically optimize nanoparticle size, surface chemistry, drug loading capacity, and release kinetics by substituting data driven optimization for empirical, trial and error formulation strategies. According to recent reviews, ML frameworks combine biological outputs like toxicity profiles, protein corona formation, cellular uptake, and pharmacokinetic/biodistribution (PK/BD) data with physicochemical descriptors to predict nanoparticle performance and effectively navigate formulation space, which speeds up translational development and makes patient-tailored nanomedicine design possible [333]. At the same time, prediction models based on machine learning are being created to stratify tumors based on prognostic risk and ferroptosis susceptibility. These models are derived from ferroptosis related gene signatures in HCC. The patients who are most likely to be benefited from ferroptosis inducing techniques, such as GPX4/GSH inhibitors or iron catalytic nanoplatforms, may be identified using these computational algorithms as companion diagnostics [334]. Precision nano ferroptosis therapy for HCC appears to be a viable short term option when AI-optimized nanoparticle engineering is combined with AI-driven patient selection (“who will respond”) to determine “what to deliver and how” [335]. AI guided technique has transformed revolutionarily the experimental nanomedicine design to rational data driven precision therapy, significantly increasing the effectiveness and customization. Combining patient classification techniques with AI-based nanoparticle assembly, offers a robust platform for next-generation ferroptosis based precision oncology in HCC.

Toxicity, biosafety, and nanomaterial-host relationships

Overload of iron and hepatotoxicity

Iron administration or iron catalyzed ROS formation are frequently used by ferroptosis inducing nanomedicines for HCC to increase lipid peroxidation. However, the liver is inherently susceptibility to iron mediated damage because it is the primary organ for systemic iron metabolism and nanoparticle elimination. The oxidative stress, inflammatory cascades, and lipid homeostasis disturbances caused by excess hepatic iron all culminate in lipid peroxidation pathways linked to liver injury [336]. Systemically delivered nanomaterials have a tendency to aggregate especially in the liver because of their diffusion by Kupffer cells and the mononuclear phagocytic system. Such accumulation can cause inflammatory signaling, oxidative stress, mitochondrial dysfunction, and fibrogenic responses, according to analyses of nanomaterial induced hepatotoxicity. These processes may work with iron driven ferroptosis to degenerate off-target harm [337]. Therefore, a good therapeutic window and reduced risk of iron overload-associated hepatotoxicity, depend on spatiotemporal control of iron release, careful dose optimization, and detailed biosafety evaluation, even though iron based nanoplatforms have therapeutic potential for HCC (Fig. 6B).

Risk of off-target ferroptosis

Ferroptosis based nanotherapy for HCC increases serious safety concerns due to the possibility of off-target ferroptotic cell death in immunological and stromal compartments, as well as non-malignant liver cells such as hepatocytes and cholangiocytes [338]. Ferroptosis is a controlled cell death linked to a variety of liver disorders and drug induced liver injury, instead of being tumor specific. It occurs when inflammatory signaling and iron dependent lipid peroxidation combine to degenerate tissue damage. Nanoparticle based delivery systems often accumulate in the liver and spleen because of sequestration by the mononuclear phagocyte system (MPS), despite of their intended improvement of tumor selectivity and reduction of systemic exposure. Absorption of catalytic elements like Cu, Fe, Mn, peroxides, or GPX4, which raises the likelihood of lipid peroxidation in healthy tissues via local redox stress impairment [337]. Thus, careful dosage of catalytic metal, tumor specific activation mechanisms, and thorough assessment of redox sensitive toxicity profiles are essential to reduce off-target ferroptosis.

Considerations for long-term biosafety

Therapeutics based on nanomaterials are evaluated for long term biosafety utilizing a risk based paradigm from a translational and regulatory viewpoint. Regulatory bodies stress that impurity profiles, surface chemistry, physicochemical stability, particle size distribution, biodegradability, and in vivo fate are all important factors that affect safety profiles which are very product specific. Sponsors must therefore demonstrate both short and long term sustainability by presenting preclinical safety data, repeatable production controls, and rigorous characterization. Biodistribution, persistence in hepatic and reticuloendothelial tissues, immunogenicity, possible buildup of catalytic metals, and delayed oxidative or fibrotic effects are all carefully assessed for ferroptosis-inducing nanomedicines in HCC. To maintain a favorable risk-benefit balance, as well as to enable regulatory approval and clinical adoption, it will be imperative to establish uniform quality control criteria and long term toxicological monitoring [339].

Certain nano carriers such as PLGA, many lipids break down into familiar metabolites, while others like, PEG, gold, many carbon frameworks persist and increase the risk of ion-mediated oxidative stress, lysosomal storage, and innate immune activation, particularly in the liver. Although quantitative cirrhosis-specific kinetics data is limited in databases [340], however altered hepatic architecture, kupffer cell activity, and hepatobiliary clearance pathways may increase retention and toxicity concerns. He et al. 2024 revealed that up to ~ 99% of phagocytic clearance of administered nanoparticles cause metabolite toxicity by concentrating particles and breakdown products in the liver. In polymeric PLGA systems, ester hydrolysis is accelerated in acidic macrophage lysosomes, produces lactic and glycolic acids that enter central metabolism and disappear as CO₂ and water [341]. Since PEG is non-biodegradable, risk switches from “metabolic toxicity” to tissue/lysosomal accumulation and immunology for PEGylated carriers.

Immune landscape shows that anti-PEG antibodies can enrich in the protein corona, increasing the uptake of macrophages, increasing clearance and minimizing adverse effects [342]. Enzymatic lipid processing and hydrolysis of engineered biodegradable linkers dominate lipid nanoparticle (LNP) degradation. Recent research emphasizes designing ionizable lipids with degradable motifs e.g., ester linkages to improve clearance, but disease specific (cirrhosis) component kinetics are hardly described in clinical pharmacology literature [343].

Similarly, CuO nanoparticles have been shown to lower GSH/SOD/CAT and increase MDA in the liver cells, suggesting ion driven lipid peroxidation as a primary toxicity axis of soluble metal metabolites [344, 345]. Inorganic/metal oxide nanomaterial degradation often involves phagolysosomal dissolution “Trojan horse” including endocytosis, acidic lysosomes, increased solubility, intracellular ion overload bypassing transporter homeostasis, and ROS chemistry and biomolecule damage from redox-active ions [346]. In steatotic hepatic cells, iron oxide nanoparticles can cause lysosome linked stress e.g., cathepsin release and cell death pathways. Iron NPs show uptake with partial dissolution and genotoxicity signals at higher exposures.

The core of gold nanoparticles is poorly degradable and can exhibit long term hepatic persistence. Some studies show increased gold levels of liver/spleen over time, suggesting limited clearance, while others show surface chemistry can shift towards hepatobiliary elimination versus Kupffer/LSEC sequestration, suggesting the risk of metabolites is often residual retention rather than chemical catabolites [347, 348]. Similarly, peroxidases and macrophage oxidant systems can biodegrade carbon based nanomaterials into oxidized pieces, but degradation is generally incomplete, leaving possible carbon residues over time. Kupffer cells can dominate capture of PEGylated liposomal therapeutics and create a sequence of capturing, payload release and hepatocyte accumulation that prolongs metabolism/excretion in the fibrosis and cirrhosis context. Fibrosis and cirrhosis further adds sinusoidal capillarization and remodeling of ECM that significantly increase retention. However, chronic accumulation kinetics of nanoparticle metabolites specific to cirrhosis, remains unspecified in most public studies.

It is shown that “degradation” can turn carriers into immunostimulatory signals. Metabolite retention interacts with innate immunity show that nanoparticles can activate the NLRP3 inflammasome via lysosomal damage and release of cathepsin using cGAS-STING signaling pathway.

Conclusion

Hepatocellular carcinoma is still a major health concern worldwide because of its limited responses to current systemic and locoregional treatments, high recurrence rates, and therapeutic resistance. Ferroptosis, an iron dependent, lipid peroxidation driven mechanism of controlled cell death, has emerged as a game changing approach for addressing the redox and metabolic deficiencies characteristic of HCC. As liver is essential for iron metabolism and oxidative balance, ferroptosis induction provides a tumor selective and biologically sound treatment approach. Clinical implementation needs careful control of iron supply, increase of ROS, and activation specific to tumors. To inhibit off-target damage, clinical implementation needs careful control of iron supply, and activation specific to tumors. By offering adaptable platforms that can be used for tumor targeted administration, controlled ferroptosis activation, immunological modulation, and multimodal therapy integration, nanotechnology has significantly progressive this field. Redox responsive nanozymes, biomimetic membrane coated systems, hypoxia activated micelles, iron based metal-organic frameworks, and theranostic nanoplatforms, are just a few examples of the nanomaterials that have shown strong preclinical efficacy in inducing ferroptosis, altering the TME, and working in concert with immunotherapy, chemotherapy, phototherapy, and gas therapy. Importantly, the combination of ferroptosis and cGAS-STING signaling provides a dual cytotoxic immunomodulatory example by connecting metabolic cell death to innate immune activation.

Ferroptosis oriented nanotherapies are still mostly in the preclinical stage, but early translational initiatives, such as clinically proven hepatic nanoparticle platforms and ferroptosis explicit nanomaterials in clinical trials, indicating a promising way toward clinical application. Future developments will depend on on combining AI assisted nanocarrier design with biomarker driven patient stratification, as well as optimizing pharmacokinetics, biosafety, biodistribution, and scalable manufacturing. Collectively, nanomaterial induced ferroptosis, is a deliberately important and rapidly emerging area of HCC treatment that may assist patients overcome resistance, become more sensitive to immunotherapy, and ultimately better long term consequences.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

AA

Arachidonic acid

ACSL4

acyl-CoA synthetase long-chain family member 4

AFP

Alpha-fetoprotein

ATO

Arsenic trioxide

BCLC

Barcelona Clinic Liver Cancer staging

BD

Biodistribution

CAF

Cancer-associated fibroblast

cDC1

Conventional type 1 dendritic cell

cGAMP

cyclic GMP–AMP

cGAS

cyclic GMP–AMP synthase

CHB

Chronic hepatitis B

CTL

Cytotoxic T lymphocyte

DEB-TACE

Drug-eluting bead transarterial chemoembolization

DFX

Deferoxamine

DHODH

Dihydroorotate dehydrogenase

DMT1

Divalent metal transporter 1

DNA/mtDNA

DNA and mitochondrial DNA

EPR

Enhanced permeability and retention

FPN

Ferroportin

FSP1

Ferroptosis suppressor protein 1

FTH1/FTL

Ferritin heavy/light chain

GCH1

GTP cyclohydrolase 1

GOx

Glucose oxidase

GPX4

Glutathione peroxidase 4

GSH

Glutathione

HBV/HCV

Hepatitis B virus/hepatitis C virus

HCC

Hepatocellular carcinoma

HIF-1α

Hypoxia-inducible factor 1 alpha

HSC

Hepatic stellate cell

HSP

Heat shock protein

ICD

Immunogenic cell death

ICI

Immune checkpoint inhibitor

ICP-MS

Inductively coupled plasma mass spectrometry

IDO

Indoleamine 2,3-dioxygenase

iRGD

Tumor-penetrating peptide iRGD

IRE/IREB2

Iron response element/iron regulatory protein 2

IRF3

Interferon regulatory factor 3

LIP

Labile iron pool

LI-RADS

Liver Imaging Reporting and Data System

LOX/LOX15

Lipoxygenase/15-lipoxygenase

LOX

Lactate oxidase

LPCAT3

Lysophosphatidylcholine acyltransferase 3

LPO

Lipid peroxidation/lipid peroxides

MASLD

Metabolic dysfunction–associated steatotic liver disease

MDA

Malondialdehyde

MDSC

Myeloid-derived suppressor cell

MOF

Metal–organic framework

MSN

Mesoporous silica nanoparticle

NAFLD/NASH

Non-alcoholic fatty liver disease/non-alcoholic steatohepatitis

NCOA4

Nuclear receptor coactivator 4

NF-κB

Nuclear factor kappa B

NK/NKT

Natural killer/natural killer T cell

NQO1

NAD(P)H quinone dehydrogenase 1

NRF2

Nuclear factor erythroid 2–related factor 2

PAI

Photoacoustic imaging

PD

Pharmacodynamics

PDX

Patient-derived xenograft

PEG

Poly(ethylene glycol)

PEI

Polyethyleneimine

PK

Pharmacokinetics

PLGA

poly(lactic-co-glycolic acid)

POD

Peroxidase-like activity

PTT/PDT

Photothermal therapy/photodynamic therapy\

RCD

Regulated cell death

RES/MPS

Reticuloendothelial system/mononuclear phagocyte system

ROS

Reactive oxygen species

SLC7A11/xCT

System Xc⁻ light chain subunit

SPIO/SPION

Superparamagnetic iron oxide nanoparticle

STAT3

Signal transducer and activator of transcription 3

STEAP3

Six-transmembrane epithelial antigen of prostate 3

STING

Stimulator of interferon genes

TAM/TAN

tumor-associated macrophage/tumor-associated neutrophil

TBK1

TANK-binding kinase 1

TFR1/TFRC

Transferrin receptor 1

TGF-β

Transforming growth factor beta

TIME/TME

Tumor immune microenvironment/tumor microenvironment

TKI

Tyrosine kinase inhibitor

VEGF

Vascular endothelial growth factor

WNT

Wingless/Int signaling pathway

Author contributions

Yang Song.Xiaohan Qu.:Conceptualization, Methodology, Writing–original draft; Writing–review & editing.Shasha Yu. Haishan Zhang.:Data curation, Formal analysis; Writing–review & editing.Fengshou Chen. Yizi Wang.:Supervision, Writing–review & editing.All authors reviewed and approved the final manuscript.

Funding

No financial support was provided for this work.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Competing interests

The authors declare no competing interests.