Copper homeostasis and cuproptosis rewire the tumor microenvironment: mechanisms, immune modulation, and therapeutic opportunities

Abstract

Copper, an essential trace element with dual functions in cancer progression, drives tumor growth via oncogenic signaling, metabolic plasticity, and extracellular matrix remodeling. By contrast, copper overload triggers cuproptosis, a form of mitochondrial proteotoxic cell death mediated by the FDX1/LIPT1/DLAT/Fe-S regulatory axis. To date, a unified theoretical framework integrating copper metabolism, tumor microenvironment (TME) remodeling, and antitumor immunity remains lacking. In this review, we reframe the TME as a structured copper ecosystem in which both cellular components and the extracellular matrix are modulated by copper, and propose a contextual copper signaling network, in which copper-mediated tumor cell fate is determined by the labile copper pool, the metabolic state, and tumor cellular heterogeneity. We further delineate a unified causal chain linking cuproptosis-driven immunogenicity and cGAS-STING activation to immune cell activation and PD-L1 modulation. Therapeutically, copper chelation and cuproptosis induction strategies have demonstrated promising efficacy, and combining cuproptosis induction with existing antitumor therapies may reverse therapeutic resistance and enhance treatment efficacy. Future studies need to validate cuproptosis-related signatures as predictive biomarkers for precision oncology and refine copper-targeted therapies to minimize systemic toxicities.

Introduction

Copper (Cu), a redox-active essential trace element, acts as an indispensable cofactor for dozens of enzymes participating in mitochondrial respiration, antioxidant defense, and signal transduction [1]. However, these vital physiological functions are accompanied by inherent cytotoxic risks of copper. Accumulating evidence indicates that dysregulated copper metabolism disrupts cellular homeostasis and contributes to the initiation and progression of multiple diseases, including cancer [2, 3]. Given its high reactivity and toxicity, cells have evolved a tightly coordinated regulatory network that governs copper uptake, intracellular trafficking, utilization, sequestration, and efflux. This network maintains dynamic copper homeostasis in response to metabolic demands and environmental cues, rather than enforcing a fixed intracellular concentration [4]. Disruption of this regulatory network disturbs metal homeostasis and rewires cellular responses to copper-related signaling. The ultimate cell fate is jointly determined by the labile copper pool and cellular metabolic status [4]. Since tumors commonly remodel metal homeostasis and redox equilibrium, dysregulated copper metabolism exerts particularly profound impacts on malignant lesions.

In the human body, disruption of copper homeostasis can elicit a spectrum of pathological consequences. Expansion of the labile copper pool promotes the generation of reactive oxygen species (ROS) and provokes oxidative stress. In parallel, copper can perturb normal protein function by interfering with metal-binding networks and forming aberrant copper–protein interactions [5]. In metabolically active cells with strong dependence on mitochondrial respiration, copper flux that exceeds the cellular buffering capacity can trigger a recently characterized form of regulated cell death (RCD) known as cuproptosis [6]. Elevated copper levels have been reported in tumor cells, tumor tissues, and the circulation of patients across multiple cancer types [7], underscoring the relevance of copper dysregulation in cancer biology. Notably, copper exerts a dual role in tumors: on the one hand, copper facilitates tumor cell proliferation and disease progression via a process defined as “cuproplasia” [8, 9]. On the other hand, in tumor cells relying on mitochondrial oxidative phosphorylation (OXPHOS) with robust tricarboxylic acid (TCA)-cycle activity, copper overload triggers lipoylation-mediated mitochondrial damage and subsequent cuproptosis [10]. These divergent outcomes reflect distinct cellular responses to disrupted copper metabolic homeostasis. Importantly, this balance is highly contingent on metabolic context, which is strongly shaped by the TME.

Tumor immunity refers to the dynamic and multifaceted interactions between tumor cells and the host immune system within the TME, which collectively determine immune surveillance, immune evasion, and therapeutic effects. The TME is dynamic and comprises diverse cell types and non-cellular components such as the extracellular matrix (ECM). The immunosuppressive remodeling of this microenvironment is a central driver of tumor progression, metastasis, and therapeutic resistance across cancer types [11]. Major immune subsets within the TME consist of T lymphocytes, B cells, natural killer (NK) cells, macrophages, dendritic cells (DCs), and myeloid-derived suppressor cells (MDSCs). Their phenotypes and effector functions are strictly controlled by local metabolic and redox signals [12]. These metabolic and redox constraints provide a natural entry point for copper, a redox-active metal, to modulate both tumor and immune cell states within the TME. Accumulating evidence indicates that copper homeostasis represents a critical yet underexplored layer of metabolic regulation within the TME. Alterations in copper distribution and accumulation not only influence tumor cell proliferation and angiogenesis but also play a pivotal role in shaping antitumor immune responses. A pan-cancer single-cell sequencing analysis revealed that highly expressed cuproptosis-related genes were significantly associated with the immunosuppressive TME [13]. This correlation might be attributed, at least in part, to the regulation of immune checkpoint molecule expression in tumor cells by intratumoral copper levels [13]. In addition, copper availability within immune cells can directly affect their activation status and effector functions [14]. Therapeutically, studies have revealed that nano-regulators targeting copper metabolism can enhance the efficacy of cancer immunotherapy, highlighting the substantial potential of copper homeostasis and cuproptosis in overcoming tumor immune evasion and tolerance [15].

Still, a comprehensive review delineating how copper homeostasis and cuproptosis shape the TME and orchestrate antitumor immunity is currently unavailable. In this narrative review, we summarize the underlying mechanisms governing copper homeostasis and its dysregulation. We focus on elucidating the impact of cuproptosis on the TME, especially tumor immunity, from both descriptive and mechanistic perspectives. In addition, we explore the therapeutic landscape of strategies targeting copper metabolism and discuss their translational potential, aiming to integrate existing evidence to identify critical knowledge gaps and provide insights into the mechanisms of cuproptosis and its clinical implications in cancer.

Copper homeostasis in cancer: dysregulation and functional consequences

Copper is an essential trace element for mammals and participates in various critical biochemical reactions, including cellular OXPHOS and the maintenance of redox balance [16]. Dysregulation of copper metabolism can lead to various pathological changes, such as neurological diseases, liver diseases, and hereditary disorders, and is closely associated with the progression of multiple cancers [17], underscoring the importance of maintaining copper homeostasis. Copper homeostasis depends on two layers of regulation. At the systemic level, it involves coordinated absorption, transport, and excretion. At the cellular level, it relies on finely tuned uptake, chaperone-mediated delivery, and compartmentalized storage [4]. In this section, we discuss the regulatory mechanisms of copper homeostasis under physiological conditions and its reprogramming in tumors, providing a framework for the role of copper in tumor initiation and progression (Fig. 1).

Fig. 1.

Copper homeostasis and reprogramming in cancer cells. A Dietary Cu(II) is reduced to Cu(I) in the intestine by STEAPs, and subsequently imported via CTR1. ATOX1 shuttles copper to the Golgi apparatus, where it is loaded onto ATP7A/B for release into the bloodstream. The liver takes up copper, incorporates it into ceruloplasmin, and primarily excretes it via bile. B Tumor cells enhance copper uptake via CTR1/ZnT1. ATOX1, CCS, and COX17 deliver copper to the Golgi apparatus, SOD1, and mitochondria, respectively, supporting oxidative phosphorylation and antioxidant defense. Concurrently, adaptive efflux mechanisms are activated, and excessive copper induces oxidative stress and functional dysregulation

Systemic and intracellular copper metabolism

Copper is present mainly as reduced Cu(I) and oxidized Cu(II) in the human body; systemic and cellular copper homeostasis depends on the coordinated regulation of uptake, intracellular distribution, storage, and excretion. At the systemic level, dietary copper is absorbed by intestinal epithelial cells and transported to the liver. The liver serves as the central hub of whole-body copper balance, either incorporating copper into ceruloplasmin for release into the circulation or eliminating it through ATP7B-mediated biliary excretion [18]. At the cellular level, because cells cannot directly import Cu(II), copper is first reduced to Cu(I) by STEAP-family metalloreductases. It is then transported into cells through copper transporter 1/2 (CTR1/2, encoded by SLC31A1/2) [19, 20]. After entering cells, free copper cannot diffuse randomly and must be transported to distinct subcellular compartments via specific copper chaperones. ATOX1 shuttles copper to the P-type ATPases ATP7A and ATP7B in the Golgi apparatus, which load copper onto ceruloplasmin for secretion [21, 22]. CCS (copper chaperone for SOD1) delivers copper to cytosolic superoxide dismutase 1 (SOD1) to support antioxidant defense [23], whereas COX17 transfers copper to the mitochondrial intermembrane space for the assembly of cytochrome c oxidase (complex IV), a key step linking copper to OXPHOS [24]. Excess intracellular copper is buffered by cysteine-rich metallothioneins [25]. These tightly coordinated processes maintain copper homeostasis under physiological conditions.

Reprogramming of copper homeostasis in cancer cells

Building on the physiological framework above, a defining feature of malignant cells is that they do not maintain copper homeostasis within the narrow physiological range. Instead, they systematically reprogram each node of the copper-handling network, including uptake, chaperone-mediated distribution, utilization, storage, and efflux, to sustain an elevated copper supply. In this section, we discuss the phenomenon of copper-supply upregulation in tumor cells, together with its underlying causes and specific mechanisms.

Compared with normal somatic cells, tumor cells display elevated copper demand and frequently accumulate excessive intracellular copper. Consistent with this, serum copper levels are frequently elevated in cancer patients. Studies have reported that the serum levels of copper, zinc, and iron are significantly increased in patients with oral cancer [26]. Similar findings have been observed in pancreatic ductal adenocarcinoma (PDAC) [27], hepatocellular carcinoma (HCC) [28], renal cell carcinoma (RCC) [29], and hematological malignancies [30]. Overall, high copper levels are associated with poor prognosis in tumor patients [8]. This consistent observation across multiple cancer types suggests that copper overaccumulation is not a random metabolic byproduct but a conserved trait positively selected during malignant evolution.

Mechanistically, copper is crucial for the activation of complex IV, which supports mitochondrial OXPHOS and thus the survival of OXPHOS-dependent tumor cells. Copper deficiency weakens this energy pathway, thereby inhibiting tumor progression [31, 32]. In addition, to maintain an immortalized phenotype and promote invasion and metastasis, tumor cells need to effectively counter oxidative stress, a process in which copper is critically involved [33]. Notably, tumor dependence on copper is also related to the need to remodel the TME. For instance, the copper chaperone protein ATOX1 plays a crucial role in VEGF-mediated angiogenesis [34]. Beyond serving as a static enzymatic cofactor, copper also acts as a direct signaling input for oncogenic pathways, and this dependence has been demonstrated genetically in vivo. In BRAF V600E-driven mouse models, reducing CTR1 levels or introducing MEK1 mutations that disrupt copper binding suppresses MAPK signaling and tumorigenesis [35]. Together, these metabolic dependencies suggest that tumor cells are under persistent evolutionary pressure to acquire copper levels above physiological requirements.

To meet this elevated copper demand while maintaining intracellular homeostasis, tumors employ multiple strategies. Ma et al. found that in cholangiocarcinoma, the SLC31A1 gene that encodes the copper uptake protein CTR1 is significantly upregulated, and its expression is positively correlated with intracellular copper accumulation [36]. CTR1 expression is further responsive to metabolic cues. In non-small cell lung cancer (NSCLC), glucose starvation induces decreased NADPH production and increased ROS levels, which in turn activates AMP-activated protein kinase (AMPK). High-level CTR1 then enhances copper uptake [37]. In a dog osteosarcoma model, Pedro et al. found that ATOX1 expression in tumor cells is higher than that in normal osteoblasts [38]. Bioinformatic analyses also suggest that high ATOX1 expression is associated with poor prognosis in breast cancer (BC) [39]. Similarly, COX17, the copper chaperone that delivers copper to mitochondria, is highly expressed in NSCLC [40]. Another copper chaperone protein, CCS, is also highly expressed in BC, with the highest levels in the Luminal B subtype [41]. Beyond uptake and distribution, tumors also remodel the downstream arms of the network. At the level of storage, the copper-buffering metallothioneins are frequently overexpressed in tumors, and their overexpression is an independent indicator of poor prognosis in melanoma and cholangiocarcinoma, indicating an expanded intracellular copper-sequestration capacity [42, 43]. At the level of efflux, the copper-exporting ATPase ATP7A is upregulated in platinum (Pt)-resistant NSCLC and other carcinomas, where it is an independent prognostic factor and, together with ATP7B, contributes to Pt resistance by exporting both copper and Pt-based drugs [44, 45]. Together, these findings indicate that tumor cells require elevated copper levels to fuel organellar functions, thereby driving proliferation, invasion, and metastasis [46].

Through the above mechanisms, the copper content of most tumor cells is higher than that of the corresponding normal tissues. Copper transporters and chaperones (CTR1, ATOX1, CCS, COX17) have been proposed as prognostic biomarkers [13]. However, copper metabolism is highly heterogeneous across tumor types, with marked differences in copper dependence. Some tumors can still maintain high proliferation and strong invasive capacity under low-copper conditions. Low copper availability also enables tumor cells to escape cuproptosis. For instance, colorectal cancer (CRC) cells downregulate CTR1 expression, leading to decreased copper concentrations in tumor tissues. Intriguingly, CRC cells with copper deficiency exhibit enhanced epithelial–mesenchymal transition (EMT) and invasion [47]. Beyond its role as a copper chaperone, ATOX1 can undergo copper-dependent nuclear translocation and act as a transcriptional regulator. In this capacity, it regulates genes including CCND1, SOD3, and NCF1/p47phox, which govern the cell cycle, oxidative stress, and inflammation. These transcriptional functions are not directly tied to copper metabolism. Therefore, ATOX1 expression and function may vary across tumors depending on its subcellular localization and the upstream and downstream regulators involved [48].

Molecular mechanism of cuproptosis

As discussed above, cancer cells reprogram copper homeostasis to elevate intracellular copper content. However, copper overload also carries severe cellular consequences. In 2019, Tsvetkov and colleagues first reported a copper-dependent form of cell death, which was formally termed cuproptosis in 2022. Cuproptosis is a previously unrecognized form of RCD, mechanistically distinct from apoptosis, necroptosis, and ferroptosis. The copper-enriched phenotype of tumor cells increases their susceptibility to cuproptosis [6, 49] (Fig. 2). In this section, we discuss the molecular mechanisms underlying cuproptosis and its crosstalk with other forms of RCD.

Fig. 2.

Molecular mechanism of cuproptosis. Upon copper overload in tumor cells, Cu(II) enters mitochondria and is reduced to the more reactive Cu(I) by FDX1. In the context of protein lipoylation mediated by LIPT1, Cu(I) directly binds to lipoylated substrates of the TCA cycle such as DLAT, inducing aberrant crosslinking and aggregation of lipoylated proteins, which leads to the collapse of mitochondrial proteostasis. Concurrently, copper accumulation destabilizes and depletes Fe-S cluster proteins, further suppressing mitochondrial respiration and ATP production. The aggregation of lipoylated proteins and the loss of Fe-S clusters jointly drive irreversible proteotoxic stress, ultimately triggering cuproptosis

Core molecular axis of cuproptosis

Cuproptosis is characterized by the direct binding of copper ions to lipoylated enzymes of the TCA cycle, leading to aberrant mitochondrial protein aggregation and disruption of protein homeostasis [6]. Protein lipoylation is a post-translational modification in which lipoic acid is covalently attached to specific lysine residues via an amide bond. This modification is essential for the catalytic activity of multiple mitochondrial enzyme complexes, including the pyruvate dehydrogenase complex and the α-ketoglutarate dehydrogenase complex [50]. The lipoylation process is mediated by lipoic acid metabolism–related enzymes, among which lipoyltransferase 1 (LIPT1) plays a critical role by catalyzing the transfer of lipoic acid to specific mitochondrial target proteins. Consequently, LIPT1 determines the abundance of lipoylated TCA cycle enzymes that serve as direct copper-binding substrates during cuproptosis [51].

Before this mechanism was elucidated, copper-induced cell death was attributed to lipid peroxidation [52]. Cuproptosis is now recognized as a distinct, proteotoxic stress–driven form of cell death, separate from modalities that depend on signaling cascades or oxidative stress. The resulting proteotoxic collapse is irreversible and leads to cell death [53].

Taken together, the core signaling cascade driving cuproptosis can be summarized as the ordered ferredoxin 1 (FDX1)/LIPT1/dihydrolipoamide S-acetyltransferase (DLAT)/Fe-S axis. Upon intracellular copper overload, copper enters the mitochondria, where FDX1 acts at two upstream nodes of this axis. FDX1 reduces Cu(II) to the more reactive Cu(I) and, through its interaction with lipoic acid synthetase (LIAS), also promotes LIPT1-mediated lipoylation of TCA cycle enzymes. Downstream, DLAT represents the terminal lipoylated substrate to which Cu(I) directly binds, triggering aberrant aggregation. In parallel, copper accumulation destabilizes iron–sulfur (Fe-S) cluster proteins, impairing mitochondrial respiration. These two branches, the aggregation of lipoylated DLAT and the loss of Fe-S clusters, are not independent events but converge on a common endpoint, mitochondrial proteotoxic stress, which constitutes the critical execution step of cuproptosis [6]. This axis also accommodates the FDX1-independent mode of cuproptosis recently reported by Lewis and colleagues. In acute myeloid leukemia (AML) cells, inhibition of heme biosynthesis depletes the heme pool required for complex IV assembly, causing complex IV collapse. Notably, this collapse alone is sufficient to initiate lipoylated protein oligomerization and trigger cuproptosis, bypassing the canonical FDX1-mediated copper reduction and lipoylation steps and instead engaging the axis at the Fe-S cluster and respiratory arm [54]. Therefore, the FDX1/LIPT1/DLAT/Fe-S axis provides a unified mechanistic backbone, with each node representing a potential regulatory and therapeutic target.

Crosstalk between cuproptosis and other regulated cell death

Cuproptosis does not operate in isolation but engages in mechanistic crosstalk with other forms of RCD. Beyond triggering cuproptosis, copper broadly participates in multiple RCD modalities, including ferroptosis [55], apoptosis [56], necroptosis [57], pyroptosis [58], and autophagy [59].

Crosstalk between cuproptosis and ferroptosis

Both cuproptosis and ferroptosis are closely related to mitochondrial function, redox and metal ion homeostasis. Hence, they constitute a regulatory network across various tumors, featuring shared upstream triggers, distinct execution pathways, and mutual potentiation effects. On the one hand, dysregulation of the copper and iron metabolic axis may prime ferroptosis by enhancing mitochondrial ROS production, destabilizing Fe-S cluster proteins, and perturbing the cellular supply of GSH and NADPH. On the other hand, the accumulation of lipid peroxidation and membrane damage during ferroptosis can elicit mitochondrial stress that may further sensitize cells to copper overload, thereby lowering the threshold for cuproptosis [55, 60, 61]. Although both pathways are associated with mitochondrial status, their lethal executioners differ fundamentally. Ferroptosis is driven by uncontrolled peroxidation of PUFA-containing phospholipids (PUFA-PL) and collapse of the GPX4 axis [62]. By contrast, cuproptosis depends on copper binding to lipoylated TCA cycle proteins within mitochondria, which provokes protein aggregation and proteotoxic stress accompanied by loss of Fe-S cluster proteins [6, 63]. Accordingly, within the same cell, the metabolic program—including the degree of OXPHOS dependence, protein lipoylation levels, and lipidomic composition—often biases susceptibility toward either ferroptosis or cuproptosis [6]. Building on these insights, an increasing number of studies are exploring co-induction strategies that jointly engage cuproptosis and ferroptosis to achieve synergistic killing. Others modulate the metabolic state to reprogram sensitivity to these RCD routes, providing new combinatorial avenues to overcome therapeutic resistance in cancer [64].

Crosstalk between cuproptosis and apoptosis

Although cuproptosis is mechanistically distinct from apoptosis, copper overload frequently mobilizes apoptotic signaling in parallel, and both programs converge on mitochondria. Oxidative stress is a major bridging node between these two death modalities. The redox cycling between Cu(I) and Cu(II) drives Fenton-like ROS generation, and the resulting redox disequilibrium can initiate the intrinsic apoptotic cascade. In HCC, the disulfiram–copper (DSF/Cu) complex decreases mitochondrial membrane potential and ATP production, activates caspase-9 and caspase-3, and thereby drives ATF3-associated mitochondrial apoptosis [65]. In addition to this mitochondrial route, copper-induced ROS may also act through the endoplasmic reticulum (ER), as copper exposure upregulates CHOP, JNK, and caspase-12 signaling, which promotes ER stress–associated apoptosis [56]. Notably, copper-induced apoptosis is not invariably caspase-dependent. For example, in BC cells, DSF/Cu induces BAK-mediated caspase-independent apoptosis through the mitochondrial-to-nuclear translocation of apoptosis-inducing factor [66]. Apart from redox effects, copper-bound ionophores can also induce caspase inhibition and paraptosis-like cell death in cancer cells [67]. Together, these mechanisms indicate that copper-driven death pathways are heterogeneous. Before cuproptosis was formally defined, copper ionophores had long been classified as ROS-dependent apoptosis inducers, a classification that reflected the historical ambiguity between copper toxicity, oxidative stress, and apoptotic signaling. This interpretation, however, appears to be context-dependent, as antioxidants such as N-acetylcysteine attenuate copper toxicity in some tumor types, including melanoma [68], but fail to suppress copper-induced cellular injury in glioma cells [69].

Crosstalk between cuproptosis and autophagy

Copper is an active regulator of autophagy, and this regulatory network intersects extensively with cuproptosis. Mechanistically, copper can directly bind the autophagy-initiating kinase Unc-51 Like Autophagy Activating Kinase 1 (ULK1), thereby activating the upstream autophagic program. For example, in BRAF V600E-mutant lung adenocarcinoma (LUAD), the interaction between copper and ULK1 maintains autophagic signaling and tumor cell survival. In KRAS-driven lung cancer (LC) models, by contrast, tumor dependence on copper-induced autophagy varies with the inactivated tumor suppressor [70, 71]. The functional consequence of copper-regulated autophagy on cell fate is likewise not uniform. In PDAC, blockade of CTR1-mediated copper uptake instead enhances autophagic flux and renders cells more resistant to death [72]. This suggests that whether copper promotes or suppresses autophagy depends on the genetic context, particularly the specific oncogenic drivers and tumor suppressor losses involved [70]. This duality also extends to copper-based therapeutic contexts. Elesclomol-Cu (ES-Cu) suppresses protective autophagy through the DLAT/mTOR axis, thereby enhancing chemosensitivity [73], whereas DSF/Cu predominantly kills CRC cells through ULK1-dependent autophagic cell death [74]. Copper does not act at a single step of autophagy. Although it activates upstream initiation signaling, copper can also block late-stage autophagic flux by binding to and inhibiting the cysteine protease ATG4B, preventing LC3 delipidation [75]. Overall, whether autophagy functions as a cytoprotective buffer or a pro-death collaborator during copper-induced cell death depends on the genetic background of tumor cells, oncogenic drivers, and their copper-handling state.

Future studies should systematically map the interactions and context dependence of cuproptosis with necroptosis, pyroptosis, and other RCD programs within the TME. Such mapping will support a unified framework for RCD and inform multi-pathway combinatorial intervention and resistance management.

Cuproptosis reshapes the tumor microenvironment

The tumor microenvironment (TME) is the local milieu in which tumor cells reside. It comprises cellular components—including tumor cells, immune cells, cancer-associated fibroblasts (CAFs), cancer-associated adipocytes (CAAs), neural cells, and endothelial cells (ECs)—and non-cellular components such as the ECM, cytokines, and metabolic factors. These components interact with each other and collectively regulate tumor initiation, progression, and response to therapy [12]. For instance, CAFs can influence ECM remodeling and affect tumor drug resistance by secreting extracellular vesicles (EVs) [76]. Copper is enriched in tumor tissues, where it reshapes the TME and modulates tumor immunity through both cuproptosis-dependent and cuproptosis-independent mechanisms. In this section, we first focus on the heterogeneous spatial distribution of copper within the TME, and then discuss the impact of copper and cuproptosis through interactions with the cellular components, including tumor cells, stromal cells, and immune cells.

Tumor microenvironment as a copper ecosystem

Metals, including copper, exhibit stable spatial distribution in tissues, which reflects their distinct biological functions and tightly regulated homeostasis. Metallomics and spatial imaging technologies are important approaches for elucidating the spatial distribution of metal elements [77], including copper in tumors. In this context, the TME can be viewed as a structured copper ecosystem, in which copper is unevenly distributed and exchanged among distinct cellular and non-cellular components.

At the tissue level, tumors generally show elevated copper compared with normal tissues. However, this increase is not homogeneous but instead manifests as localized and region-specific accumulation. Spatially resolved multiomics analyses revealed pronounced heterogeneity in copper distribution across tumor subregions. Copper-enriched areas showed elevated OXPHOS, consistent with the established role of copper in mitochondrial respiration [78]. At a finer spatial scale, copper distribution is cell-type dependent. Imaging analyses based on laser ablation–inductively coupled plasma–time-of-flight mass spectrometry (LA-ICP-ToF–MS) have shown that copper signals are preferentially enriched in tumor cell–dense regions and neovascular structures, whereas copper levels are lower in regions dominated by ECM components [79].

Copper undergoes trafficking between cellular and acellular TME compartments to sustain regional copper balance. Angiogenesis is a hallmark of tumor progression. Using X-ray fluorescence microscopy (XFM), Finney et al. demonstrated in vitro that spatial redistribution of copper occurs during neovascularization, with copper relocating from the intracellular compartment toward the ECM at the tips of newly forming blood vessels [80]. Interestingly, CAFs can also export copper to the extracellular space and thereby influence tumor progression, primarily through their secretion of lysyl oxidase (LOX). LOX is a canonical cuproenzyme—an enzyme with copper-binding sites whose catalytic activity depends on copper availability. LOX acquires copper in the Golgi apparatus through ATP7A-mediated copper transport [81]. In vitro studies by Lewinska et al. showed that cholangiocarcinoma (CCA) cells took up CAF-derived LOX, which then elevated OXPHOS and stemness and promoted CCA progression [82].

The effects of copper on tumor cells

Cancer cells require elevated copper to sustain multiple oncogenic processes. Beyond driving cuproplasia and proliferation, copper also regulates autophagy, EMT, and genomic instability, thereby shaping the malignant phenotype (Fig. 3).

Fig. 3.

Copper-mediated tumor progression. Tumor cells import Cu(I) via CTR1 and remodel copper homeostasis, thereby activating multiple pro-proliferative signaling axes: Copper delivered through CCS promotes the MEK1/ERK cascade and can also enhance the PI3K/PDK1/AKT and TAK1/IKKβ/NF-κB pathways, driving cuproplasia and metabolic fueling. Copper can also directly regulate p53-ULK1-associated autophagy. Simultaneously, copper participates in networks such as TGF-β/Smad, Wnt/β-catenin, and Hippo/YAP, inducing transcription factors like Snail/Slug and upregulating EMT genes. The Cu(I)/Cu(II) redox cycle generates ROS, and competition for Zn(II) sites interferes with DNA repair proteins, promoting oxidative DNA damage and genomic instability, thereby accelerating tumor progression

Copper-mediated tumor proliferation (cuproplasia)

Cuproplasia has been proposed to describe a phenomenon in which copper availability supports tumor cell proliferation. As an indispensable cofactor of complex IV, copper enhances OXPHOS activity and energy production, providing an essential metabolic basis for sustained tumor growth [83]. Beyond this metabolic role, dysregulated copper homeostasis drives tumor progression through both direct signaling activation and indirect effects such as modulation of protein ubiquitination. Consequently, cuproplasia encompasses both neoplastic proliferation (neoplasia) and non-neoplastic proliferative changes (hyperplasia). Indeed, cuproplasia represents one example of metalloplasia, as other metals, such as iron and zinc, can similarly regulate signaling pathways to promote iron/zinc-dependent cell proliferation [84].

Copper directly interacts with several mitogenic signaling molecules. The MAPK/ERK cascade (RAS/RAF/MEK/ERK) is considered one of the most copper-sensitive pathways [85]. Mechanistically, copper-loaded CCS transfers copper to MEK1, enhancing MEK1 activation and ERK phosphorylation, thereby driving proliferative signaling [86]. In BC, copper enhances NF-κB signaling by directly binding to transforming growth factor-β–activated kinase 1 (TAK1), promoting IκB kinase β (IKKβ)–mediated nuclear translocation of NF-κB [87]. Interestingly, NF-κB in turn represses the copper transporter CTR1 through negative feedback. Pharmacological targeting of NF-κB therefore elevates intracellular copper and activates the MAPK and AKT pathways, which may underlie resistance to NF-κB-targeting inhibitors [87]. In diffuse large B-cell lymphoma (DLBCL), the copper chaperone ATOX1 has also been found to activate the MAPK pathway by regulating copper transport [88]. Copper also activates upstream signaling, including the RTK and BRAF pathways, to promote tumorigenesis [35, 89]. In HCC, copper has been shown to stimulate the PI3K/AKT axis and the PDK1 pathway in a CTR1-dependent manner [90, 91]. Similarly, Cu(II) activates the PI3K/AKT pathway in the p53-mutant BC cell line MDA-MB-231 but not in the p53-wild-type MCF-7 line [92]. This discrepancy may reflect the tumor-suppressive function of p53 in MCF-7 cells. Notably, these effects appear to be at least in part dependent on copper import transporters, which suggests that intracellular copper concentration is a prerequisite for downstream oncogenic signaling.

Copper-mediated autophagy regulation

Autophagy refers to an evolutionarily conserved catabolic process in eukaryotic cells that degrades and recycles damaged cellular components and redundant macromolecules via the lysosomal pathway. It is a core mechanism for maintaining intracellular homeostasis and adapting to various stress conditions [93]. It plays a dual role in cancer, acting as a tumor-suppressive mechanism during tumor initiation while promoting tumor progression in established tumors [94]. In established tumors, copper can sustain tumor cell survival and proliferation by activating pro-survival autophagic signaling. Tsang et al. demonstrated that in BRAF V600E-mutant LUAD, copper directly binds and activates the autophagy-initiating kinase ULK1 [70, 71]. The resulting copper–ULK1 interaction maintains autophagic flux and thereby drives LUAD development and supports tumor cell survival. Through this autophagy-supporting role, copper contributes to the maintenance of the malignant phenotype.

Copper-mediated epithelial-mesenchymal transition

EMT is a phenotypic transition in which epithelial cells lose their polarity and adhesion and acquire the migratory and invasive properties of mesenchymal cells. EMT plays a crucial role in cancer invasion and metastasis [95]. The core mechanisms underlying EMT involve the activation of signaling pathways such as TGF-β/Smad, Wnt/β-catenin, MAPK/ERK, and PI3K/AKT, which subsequently regulate key transcription factors such as Snail, Slug, and Twist. These transcription factors inhibit the expression of epithelial markers such as E-cadherin while promoting the synthesis of mesenchymal markers including N-cadherin and vimentin [96]. Copper directly participates in regulating the EMT process. Lee et al. demonstrated in vitro that copper promotes EMT in CRC cells via the β-catenin pathway by upregulating PLK1 and FOXO3a [97]. In HER2-negative BC, researchers utilized triethylenetetramine (TETA) to chelate intracellular copper for copper depletion. The results showed that long-term copper depletion suppressed EMT through the AKT pathway. This led to increased expression of the epithelial marker E-cadherin, reduced cell migration, and attenuation of TGF-β-induced EMT [98]. Another study in triple-negative breast cancer (TNBC) also confirmed that copper chelators inhibit EMT in TNBC cells through both the canonical TGF-β/SMAD2 pathway and non-canonical MAPK and Hippo/YAP pathways [99]. In addition, a study showed that mitochondrial copper sustains epidermal growth factor (EGF)–induced EMT and the mesenchymal phenotype. However, the anticancer drug metformin can target mitochondrial copper, thereby suppressing copper-mediated pro-tumorigenic effects [100].

Beyond directly regulating EMT-related signaling pathways, copper also acts through the LOX family of copper-containing enzymes. Typically, the LOX family influences the ECM by catalyzing the crosslinking of collagen and elastin [101]. However, emerging evidence indicates that LOX exhibits a transcription factor-like function in anaplastic thyroid cancer (ATC) and BC. Within the nucleus, LOX directly binds and transactivates the promoter of Slug, a key EMT transcription factor, as confirmed by chromatin immunoprecipitation (ChIP) assays. Notably, this activation is independent of its enzymatic activity [102]. A limitation of this study is that it did not identify the specific LOX family member responsible for this effect. In contrast, a separate study in ESCC demonstrated that LOXL2 promotes EMT by activating the focal adhesion kinase (FAK)/Src pathway. This activation leads to increased expression of the mesenchymal marker Snail and decreased expression of the epithelial marker E-cadherin [103].

Copper-mediated genomic instability

Genomic instability, the persistent acquisition of genetic and epigenetic aberrations, is a hallmark of cancer and drives tumor initiation through replication stress, DNA damage, and deficiencies in DNA repair [104]. As an active redox agent, copper drives genomic instability and mutagenesis through two mechanisms. First, Cu(I)/Cu(II) redox cycling promotes hydroxyl radical production via Fenton-like chemistry, which causes oxidative DNA damage [105]. If not properly repaired, such lesions may increase mutagenic potential and genomic instability, and thereby contribute to tumor progression [106]. Second, copper can interfere with the stability of the zinc finger domains in DNA repair proteins through metal-mediated toxicity, such as the Zn(II) coordination structures in PARP-1 and XPA proteins [107, 108]. This interference destabilizes and impairs DNA repair proteins, weakening base excision repair (BER) and nucleotide excision repair (NER). In tumor cells, the resulting decline in repair efficiency promotes mutation accumulation and tumorigenesis [109]. Although existing evidence implicates copper in the modulation of genomic instability, further investigations are required to elucidate the precise underlying mechanisms.

Beyond DNA-damage mechanisms, copper also regulates proteostasis through the ubiquitin–proteasome system (UPS), although its directionality remains controversial. Metal-induced allosteric effects have been proposed to facilitate UPS-mediated degradation of tumor suppressors such as p53 [84, 110]. By contrast, other studies have reported that Cu(II) can directly inhibit 20S proteasome activity [111, 112]. The metal-induced allosteric effects of copper, particularly Cu(II), may contribute to these divergent outcomes, due to the large number of intracellular enzyme systems containing metal-ion binding sites.

Overall, copper can not only directly activate proliferative signaling pathways but also influence tumor progression by modulating autophagy, EMT, and genomic instability. However, the effects of copper depend on the size of the labile copper pool and the tumor genetic background. These contradictory findings indicate that copper has complex regulatory roles in tumor biology and call for further mechanistic studies.

The effects of cuproptosis on tumor cells

Copper exerts a dual role in tumor cells, promoting tumor cell proliferation while also inducing cuproptosis to suppress tumor cell survival. Consequently, a major focus of recent research has been tumor cell adaptation to cuproptosis—specifically, how tumor cells resist cuproptosis to maintain viability or, conversely, what regulatory factors compel them to succumb to this cell death pathway. Elucidating the regulatory mechanisms of molecules involved in this process is important for understanding tumor biology and for developing novel therapeutic strategies. The mechanisms of this regulatory network include regulation of the FDX1/LIPT1/DLAT/Fe-S axis, as well as broader mitochondrial and metabolic pathways (Fig. 4).

Fig. 4.

The effects and regulatory mechanism of cuproptosis in tumors. In tumor cells, cuproptosis induced by copper overload is tightly regulated. Tumors can modulate their sensitivity through various mechanisms, including regulation of the FDX1 axis (SEC14L3–ERK/YY1, p53, CEBPA/B), m6A modification and 3′-end processing (METTL16/METTL3, FMRP, NUDT21, ALKBH5, YTHDF2), non-coding RNAs (circKIAA1797, PVT1, RP11), and ubiquitination and degradation of lipoic acid enzymes LIAS/LIPT1 (TRIM21, XBP1s–MGRN1). Furthermore, pathways such as hypoxia (HIF-1α–PDK1/3), NF-κB, and PI3K/AKT/mTOR are also involved, forming a multi-layered adaptive network that ultimately determines the outcome of cuproptosis

Regulation of cuproptosis through FDX1

As the core driver of cuproptosis, FDX1 has become a major focus of research. Beyond its role in the Cu(II)/Cu(I) redox cycle, recent studies have shown that FDX1 directly interacts with LIAS, a SAM-dependent enzyme that catalyzes protein lipoylation, and thereby promotes lipoylation itself [113]. This finding has two implications for tumor cuproptosis. First, elevated lipoylated proteins provide the substrate pool for cuproptosis. Second, protein lipoylation is essential for several mitochondrial enzyme complexes, including the pyruvate dehydrogenase, α-ketoglutarate dehydrogenase, and branched-chain α-keto acid dehydrogenase complexes, as well as the glycine cleavage system. A baseline level of lipoylation is therefore needed for tumor cell viability, particularly under low-glucose conditions [114, 115]. The human FDX protein family comprises FDX1 and FDX2. Unlike FDX1, which promotes lipoylation, FDX2 is primarily involved in mitochondrial Fe-S cluster biogenesis. It directly interacts with and supplies electrons to proteins within the Fe-S cluster assembly machinery, thereby indirectly contributing to the regulation of cuproptosis [116]. This evidence indicates the central role of FDXs in cuproptosis. Indeed, tumor cells employ various pathways to modulate FDX function, ultimately influencing intracellular cuproptosis susceptibility.

SEC14L3 is a lipid-binding transport protein. Elesclomol-induced copper influx upregulates SEC14L3 through the ERK/YY1/FDX1 axis, enhancing HCC sensitivity to cuproptosis. However, HCC cells appear to have adapted to this axis: SEC14L3 is expressed at low levels in HCC and correlates with poor prognosis [117]. In parallel, tumor cells often downregulate FDX1 expression to evade cuproptosis. Low FDX1 levels are associated with unfavorable prognosis across multiple malignancies, including adrenocortical carcinoma (ACC), RCC, head and neck squamous cell carcinoma (HNSCC), thyroid carcinoma (TC), low-grade glioma (LGG), and LUAD [118]. Zhao et al. discovered that the transcription factor CCAAT/enhancer binding protein alpha (CEBPA) transcriptionally promotes FDX1 expression, thereby increasing LUAD sensitivity to cuproptosis. This provides new insights for LUAD therapies targeting cuproptosis [119]. In contrast, CEBPB, another member of the CEBP family, attenuates cuproptosis. However, this effect is mediated through activation of the PI3K/AKT/mTOR signaling pathway rather than through an FDX1-dependent mechanism [120]. The tumor suppressor p53 exerts potent antitumor effects by regulating various cell death modalities including cuproptosis [121]. Ferredoxin reductase (FDXR) transfers electrons from NADPH to both FDX1 and FDX2, influencing the cuproptosis cascade. In HCC, p53 upregulates both FDXR and FDX1, sensitizing cells to cuproptosis. The combination of the p53 activator CP-31398 and Elesclomol significantly inhibited tumor growth [122]. Furthermore, Liu et al. demonstrated that vasoactive intestinal peptide receptor 1 (VIPR1) enhances FDX1 expression in colon cancer, thereby facilitating cuproptosis, although the detailed molecular mechanisms require further investigation [123].

Copper overload can also modulate FDX1 function at the post-transcriptional and RNA levels. These modifications can act in either direction. In GC, intratumoral copper content correlates positively with global m6A levels. Copper overload induces lactylation of the m6A methyltransferase METTL16, which deposits m6A on FDX1 mRNA to stabilize the transcript and increase FDX1 expression, thereby promoting cuproptosis [124]. Conversely, in HCC, METTL3-mediated m6A modification of FDX1 mRNA suppresses its expression through the m6A “reader” protein FMRP, leading to cuproptosis resistance [125]. Beyond RNA chemical modifications, emerging evidence indicates that alternative mRNA processing also critically regulates FDX1 expression and cuproptosis susceptibility. For example, in esophageal squamous cell carcinoma (ESCC), lactylated NUDT21 remodels the 3’-end processing of FDX1 mRNA, resulting in 3’-UTR lengthening and translational repression, thereby inhibiting cuproptosis [126]. In addition, non-canonical post-transcriptional regulation also contributes to the modulation of cuproptosis. In LC, 8-oxo-7,8-dihydroguanosine (o8G) modification of circKIAA1797 destabilizes its target FDX1 mRNA. The o8G-modified circKIAA1797 also suppresses LIPT1 activity, jointly inhibiting cuproptosis in LC cells [127].

Emerging research highlights the impact of non-coding RNAs, including microRNAs (miRNAs) and long non-coding RNAs (lncRNAs), on tumor cell cuproptosis. Mechanistically, these molecules appear to regulate cuproptosis through two general modes. One mode operates post-transcriptionally at the 3’-untranslated region (3’-UTR), in which miR-144-3p and the lncRNA RP11-199F11.2 suppress FDX1 either indirectly by targeting its upstream activator PUMA or directly by binding to the FDX1 3’-UTR, thereby inhibiting cuproptosis [128, 129]. The other mode occurs at the transcriptional level, where the lncRNA PVT1 increases H3K27ac deposition at the FDX1 promoter, enhances FDX1 transcription, and consequently promotes cuproptosis [130]. In addition to directly targeting FDX1, non-coding RNAs may also act through intermediary regulators. For example, LINC02362 upregulates miR-18a-5p to elevate its downstream target FDX1, thereby promoting cuproptosis in HCC and restoring sensitivity to oxaliplatin [131].

Regulation of cuproptosis through lipoylation pathways

Protein lipoylation constitutes the biochemical foundation of cuproptosis and governs cellular sensitivity to copper overload. This process is strictly orchestrated by lipoylation enzymes, such as LIAS and LIPT1, alongside their substrates, notably DLAT. Post-transcriptional modifications targeting these key lipoylation enzymes serve as pivotal determinants of cell fate. In ESCC, the ubiquitin ligase TRIM21 facilitates the nuclear translocation and K63-linked ubiquitination of the demethylase ALKBH5. Once activated, ALKBH5 removes m6A modifications on the LIAS mRNA, leading to its downregulation and the consequent suppression of cuproptosis [132]. Similarly, in bladder cancer, the m6A reader protein YTHDF2 mediates the degradation of LIPT1 mRNA. Notably, restoring LIPT1 expression disrupts ER homeostasis and potentiates cuproptosis [133]. Furthermore, Gu et al. found that ER stress signaling modulates this process via super-enhancer mechanisms. In LC, spliced XBP1s drives the expression of MGRN1, which mediates the ubiquitin-proteasomal degradation of LIPT1, thereby enabling cuproptosis evasion through the inhibition of lipoylation [134].

As a direct substrate of lipoylation, DLAT is tightly regulated by diverse signaling pathways, and its alteration is closely related to cuproptosis in tumor cells. For instance, in Group 3 medulloblastoma, a subtype associated with poor prognosis, the metabolic enzyme IDH1 upregulates DLAT by maintaining high c-MYC expression, rendering these tumors highly sensitive to copper ionophore-induced cuproptosis [135]. By contrast, in solid tumors, the hypoxia-inducible factor HIF-1α suppresses DLAT expression by activating PDK1/3, conferring resistance to cuproptosis. This represents an adaptive response of tumor cells to hypoxic stress [136]. Additionally, in oral squamous cell carcinoma (OSCC), the circadian protein PER2 promotes AKT degradation by binding to heat shock protein 70 (HSP70), relieving AKT-mediated inhibition of DLAT and restoring cuproptosis sensitivity [137]. In CRC, by contrast, BCL10 downregulates DLAT via the NF-κB pathway, exerting a distinct protective effect [138].

Regulation of cuproptosis in CSCs

The regulation of cuproptosis in cancer stem cells (CSCs) has received increasing attention. Pan-cancer analyses have revealed a significant positive correlation between cuproptosis-related genes and stemness signature across various tumor types. High-stemness subtypes show enhanced cuproptosis resistance, which correlates with poorer survival and altered drug sensitivity [139]. For instance, in LC, a cuproptosis-related stemness gene signature constructed using single-cell and transcriptomic data effectively predicts patient prognosis. High-risk patients exhibit heightened stemness characteristics, lower immune cell infiltration, and poorer survival rates [140]. This association may be attributed to the interaction between cuproptosis and classical oncogenic transcription factors. Molecular features related to cuproptosis are highly enriched in the downstream target genes of c-MYC, suggesting that c-MYC may enhance tumor cell stemness by remodeling copper metabolism and mitochondrial function, thereby supporting cell survival under metabolic stress or death signals [141]. Moreover, the Wnt/β-catenin signaling pathway is often aberrantly activated in CSCs, and β-catenin can transcriptionally induce the expression of copper efflux protein ATP7B, significantly enhancing CSC resistance to cuproptosis [142]. Interestingly, recent studies have shown that glioblastoma stem cells (GSCs) actively regulate copper homeostasis to avoid cuproptosis. Mechanistically, the promoter region of ATP7A shows significant enrichment of H3K27ac, which ensures its high-level transcription. Under circadian regulation, ATP7A dynamically facilitates intracellular copper efflux, thereby limiting copper accumulation and effectively inhibiting cuproptosis [143]. Furthermore, the metabolic plasticity of CSCs also contributes to their resistance to cuproptosis. Specifically, CSCs adapt their metabolic dependencies, shifting between glycolysis and OXPHOS depending on TME conditions. Glycolysis is preferentially used under hypoxia [144]. These findings provide important insights into the mechanisms underlying resistance to cuproptosis in CSCs. However, further mechanistic studies are still required to delineate the complex regulatory network.

Regulation of cuproptosis through other mechanisms

The intracellular bioavailability of copper ions is another critical determinant of cuproptosis. Tumor cells tightly control this balance through coordinated regulation of copper influx, efflux, and chaperone proteins. In addition to the Wnt/β-catenin–ATP7B axis in CSCs [142] and the NF-κB–CTR1 negative feedback loop [87] discussed above, several additional regulatory mechanisms have recently been identified. For instance, TRIM21 regulates copper uptake through a mechanism distinct from its role in LIAS modulation. TRIM21 catalyzes non-proteolytic ubiquitination of the transcription factor inhibitor of DNA binding 1 (ID1), thereby releasing the transcription factor TCF12, which subsequently upregulates the copper transporter SLC31A1 to promote copper influx and cell death [145]. In addition, aberrant splicing of mitochondrial copper chaperones is another mechanism of resistance to cuproptosis. In gastric cancer peritoneal metastasis (GCPM), the splicing factor PTBP3 induces exon skipping of COX11, generating a functionally defective isoform that impairs copper delivery to mitochondria, consequently inhibiting cuproptosis [146].

Furthermore, regulation at the gene expression level has also been shown to modulate cuproptosis. Wang et al. revealed that extrachromosomal circular DNA (eccDNA)–mediated amplification of KIF3C downregulates MUC20 expression in multiple myeloma, thereby impairing MUC20-driven cuproptosis through a MET–CDKN2A–dependent axis and contributing to proteasome inhibitor resistance [147]. In LC cells, the RNA-binding protein YTHDC2 recognizes and binds the m6A modification on SLC31A1 mRNA, promoting SLC31A1 expression and copper uptake. This increases cuproptosis susceptibility and reverses acquired resistance to EGFR tyrosine kinase inhibitors (EGFR-TKIs) [148].

In summary, copper has dual effects on tumor cells: it drives malignant phenotypes such as EMT and genomic instability while also inducing cuproptosis under copper overload. Tumor cells balance these opposing effects through multilayered regulation of copper homeostasis, mitochondrial metabolism, and key cuproptosis determinants, which shapes their survival and progression (Table 1).

Table 1.

Regulatory mechanism of cuproptosis in tumor cells

Regulators	Cancer type	Mechanism	Cuproptosis	Ref.
FDX1	Pan-cancer	Interacts with LIAS to promote protein lipoylation	↑	[113]
FDX2	Pan-cancer	Mitochondrial Fe-S cluster biogenesis	↑	[116]
SEC14L3	HCC	ERK/YY1/FDX1 axis	↑	[117]
FDX1 downregulation	Pan-cancer	Downregulation to evade cell death	↓	[118]
CEBPA	LUAD	Transcriptionally promotes FDX1 expression	↑	[119]
CEBPB	Pan-cancer	PI3K/AKT/mTOR signaling pathway	↓	[120]
p53/FDXR	HCC	Upregulates FDXR and FDX1 expression	↑	[122]
VIPR1	Colon cancer	Enhances FDX1 expression	↑	[123]
METTL16 (Lactylated)	GC	m6A modification stabilizes FDX1 mRNA	↑	[124]
METTL3/FMRP	HCC	m6A modification suppresses FDX1 expression	↓	[125]
NUDT21/CPSF6	ESCC	3’-UTR lengthening and translational repression	↓	[126]
circKIAA1797	LC	Decreases FDX1 mRNA stability and inhibits LIPT1	↓	[127]
miR-144-3p	CRC	Inhibits PUMA function to activate FDX1	↓	[128]
lncRNA RP11	HGSOC	Binds to 3’-UTR of FDX1 mRNA to inhibit FDX1	↓	[129]
lncRNA PVT1	CRC	Enhances FDX1 promoter H3K27ac deposition	↑	[130]
LINC02362	HCC	miR-18a-5p/FDX1 axis; restores drug sensitivity	↑	[131]
TRIM21/ALKBH5	ESCC	ALKBH5 removes m6A on LIAS mRNA	↓	[132]
YTHDF2	Bladder cancer	Mediates degradation of LIPT1 mRNA	↓	[133]
XBP1s	LC	XBP1s/MGRN1 axis induces ubiquitin-proteasomal degradation of LIPT1	↓	[134]
IDH1	Medulloblastoma	IDH1 sustains high c-MYC expression to upregulate DLAT, increasing cuproptosis sensitivity	↑	[135]
HIF-1α	Solid tumors	PDK1/3 axis; suppresses DLAT expression	↓	[136]
PER2	OSCC	PER2/HSP70/AKT axis restores DLAT level	↑	[137]
BCL10	CRC	NF-κB pathway; downregulates DLAT	↓	[138]
c-Myc	Pan-cancer/LC	Remodeling copper metabolism & stemness	↓	[141]
Wnt/β-catenin	CSCs	Transcriptionally upregulates copper exporter ATP7B	↓	[142]
ATP7A	GSCs	H3K27ac enrichment; facilitates copper efflux	↓	[143]
NF-κB	BC	Feedback loop; represses copper importer CTR1	↓	[87]
TRIM21	Pan-cancer	TRIM21/TCF12 axis upregulates copper transporter SLC31A1	↑	[145]
PTBP3	GCPM	PTBP3/COX11 impairs mitochondrial copper delivery (splicing)	↓	[146]
eccDNA	Multiple myeloma	eccDNA/KIF3C inhibits MUC20 and MET–CDKN2A axis	↓	[147]
YTHDC2	LC	m6A modification; promotes SLC31A1 expression	↑	[148]

Contextual copper signaling network

As discussed above, copper exerts dual biological effects on tumor cells: it can promote proliferation, activate autophagy, induce EMT, and exacerbate genomic instability, while excessive copper can trigger tumor cell death via activation of the cuproptosis pathway. This bidirectional biological effect is associated with multiple dimensions, including copper ion concentration, cellular metabolic state, and cell heterogeneity. Accordingly, in this section, we construct a conceptual model to elucidate how these factors influence cellular responses to copper and guide therapeutic intervention (Fig. 5).

Fig. 5.

The contextual copper signaling network. The functional outcome of copper in tumor cells is dictated by three converging dimensions. The labile copper pool acts bidirectionally, with basal levels sustaining physiological function, elevated levels supporting cuproplasia, and supra-threshold accumulation triggering cuproptosis. The threshold is movable, lowered by OXPHOS dependence and raised by glycolytic shift. Cellular heterogeneity further tunes susceptibility through molecular and microenvironmental rewiring (e.g., AKT1/FDX1 axis and HIF-1α/DLAT axis). These dimensions guide therapy, with chelators favored when copper acts as a pro-progression cofactor, and ionophores or nanomedicines favored for OXPHOS-dependent tumors with high FDX1 activity and abundant lipoylated substrate

The first core dimension is the copper concentration, which here refers specifically to the bioactive labile copper pool within the cell rather than the total cellular copper content. When the labile copper pool remains at a low level, copper functions primarily as a catalytic or allosteric cofactor, sustaining proliferative and autophagic signaling and thereby mediating its pro-proliferative effect [84]. As the labile copper pool progressively accumulates, copper gradually initiates mitochondrial injury programs and ultimately elicits proteotoxic stress [6]. Critically, the threshold separating these two regulatory regimes is not a fixed concentration but a movable boundary: the threshold decreases when cells rely on oxidative phosphorylation, whereas a shift toward aerobic glycolysis raises it. This boundary is jointly governed by the metabolic state together with glutathione, metallothioneins, and copper transporters [149]. Because intracellular copper levels in tumor cells often operate near the upper boundary of their copper-tolerance range, and because their proliferative signaling has been at least partially rewired into copper-dependent pathways, tumors exhibit bidirectional sensitivity to copper perturbation: copper deficiency collapses their proliferative capacity, whereas copper excess triggers cuproptosis, while normal cells retain a wider tolerance range for equivalent fluctuations [7]. Notably, from a histological perspective, the vast majority of tumor tissues possess the capacity to concentrate copper [150]. Therefore, the discussion of copper concentration in this section is framed in terms of relative levels against a copper-rich tissue background. In other words, both the copper level that supports copper-driven proliferation and that which induces cuproptosis already lie above the normal physiological range.

The second core dimension is the cellular metabolic state, specifically the relative reliance on mitochondrial OXPHOS versus aerobic glycolysis. This dimension determines the position at which the threshold described above is set. Because the execution of cuproptosis depends on the TCA cycle and the electron transport chain, tumor subsets that predominantly utilize aerobic glycolysis rather than OXPHOS generally display reduced susceptibility to cuproptosis, whereas cells that retain or depend on OXPHOS are more prone to undergo it [91, 151]. Mechanistically, this difference is governed by FDX1 activity and the level of mitochondrial protein lipoylation, the two metabolically coupled and cuproptosis-indispensable variables. Lu et al. demonstrated that upregulation of FDX1 induces mitochondrial uncoupling and sensitizes gastric cancer cells to cuproptosis, as evidenced by enhanced degradation of Fe-S cluster proteins [152]. Conversely, a metabolic shift toward glycolysis contracts the lipoylated-substrate pool and thereby confers resistance to cuproptosis. For instance, Liao et al. found that wild-type p53 transcriptionally activates the circular RNA circFRMD4A, which suppresses pyruvate kinase PKM2 and drives a metabolic shift from glycolysis toward the TCA cycle; this metabolic reprogramming enhances sensitivity to cuproptosis [153]. Pharmacological interventions such as 4-octyl itaconate (4-OI), which inhibits aerobic glycolysis by targeting GAPDH, can similarly sensitize cells to cuproptosis [154]. The metabolic dimension thus determines the precise point at which a cell crosses from copper-supported proliferation into cuproptosis.

The third core dimension is cellular heterogeneity. Copper concentration, metabolic state, FDX1 activity, and lipoylation status do not vary at random but are assembled into recognizable configurations by tumor type, molecular subtype, and microenvironmental niche. Therefore, the response of a cancer to copper perturbation is, in principle, foreseeable from its molecular features rather than discoverable only empirically. At the molecular-subtype level, Sun et al. reported that although copper is significantly elevated in TNBC tissues relative to adjacent normal tissue, TNBC cells nonetheless resist cuproptosis, a phenotype associated with worse clinical outcomes. Mechanistically, AKT1 phosphorylates and inactivates FDX1, while the same signaling axis upregulates glycolysis-related genes, further reinforcing a glycolytic state unfavorable for cuproptosis [155]. In addition, the hypoxic state confers greater resistance to cuproptosis. Mechanistically, HIF-1α activates PDK1/3 and downregulates the DLAT subunit of lipoylated TCA cycle enzymes. Given that DLAT serves as the direct copper-binding substrate, its loss attenuates cuproptosis, which reflects an adaptive response of tumor cells to the hypoxic environment [136]. Collectively, tumor cell heterogeneity, encompassing both differences in cell type and heterogeneity in cell state, shapes the susceptibility to cuproptosis.

By integrating these three dimensions, we propose the contextual copper signaling network as a conceptual model. In this model, the functional outcome of copper does not depend on any single variable but on the convergence of labile copper pool concentration, metabolic state, and cellular heterogeneity. The principal value of this model lies in transforming such context-dependence into a basis for therapeutic decision-making, thereby framing therapeutic intervention as a fourth actionable dimension of the model. Copper-directed therapy can be pursued from two opposing directions, and these three dimensions jointly indicate which direction a given tumor is amenable to. The first direction exploits copper excess to force cells across the cuproptosis threshold, with representative strategies including copper ionophores and copper-based nanodelivery systems. This strategy is predicted to be effective when a tumor occupies an OXPHOS-dependent metabolic state and retains high FDX1 activity together with an abundant lipoylated substrate [122, 156]. The second direction instead restricts copper availability mainly through copper chelators, so as to suppress the copper-dependent signaling that sustains proliferation, autophagy, EMT, and angiogenesis. This direction is rational when copper within a tumor functions chiefly as a pro-proliferative cofactor [157]. The model therefore reframes the choice between these two therapeutic directions as a decision that can be read from a tumor’s measurable metabolic and molecular features. In this section, we provide guidance for therapeutic intervention only through the dimensions discussed above, whereas the specific therapeutic strategies will be discussed in detail in the therapeutic strategies section.

Copper and cuproptosis in stromal cells

In the TME, stromal cells constitute an indispensable cellular compartment. These stromal populations encompass CAFs, vascular-associated cells including ECs and pericytes, CAAs, mesenchymal stem cells (MSCs), as well as other tissue-specific resident supportive cell types [158]. Beyond paracrine interactions with tumor cells, stromal cells actively remodel the ECM by secreting structural proteins (e.g., collagen, fibronectin) and matrix-remodeling enzymes such as matrix metalloproteinases (MMPs). Through these activities, they establish the physical, biochemical, and metabolic landscape of the tumor niche [159, 160].

Emerging evidence has revealed that stromal cells within the TME display varying degrees of copper dependency, suggesting that copper homeostasis is a crucial regulator of stromal-tumor crosstalk. Copper serves not only as an essential cofactor for ECM-modifying enzymes—including the LOX family, which catalyzes collagen crosslinking and ECM stiffening—but also as a modulator of stromal cell phenotype and function [161, 162]. In parallel, copper-induced cuproptosis has been implicated in stromal cell dynamics within the TME [163].

The stromal landscape is highly heterogeneous, but current research on copper metabolism and cuproptosis has focused predominantly on CAFs, owing to their abundance and central role in ECM remodeling. Studies on copper-mediated CAA-driven remodeling of the TME are also expanding rapidly. By contrast, the roles of copper in other stromal populations, including MSCs and ECs, remain less well characterized. This section first reviews the well-established copper-dependent mechanisms in CAFs, then summarizes emerging findings in other stromal populations (Fig. 6).

Fig. 6.

Cuproptosis in stromal cells and ECM remodeling. In the TME, stromal cells remodel the ECM through copper-dependent mechanisms and influence tumor fate. CAFs secrete LOX/LOXL, which can be loaded into EVs and anchored to collagen fibers via integrin α2β1, POSTN, FN, and BMP-1, catalyzing collagen crosslinking and leading to ECM stiffening. The resulting mechanical forces activate FAK and promote nuclear translocation of YAP/TEAD, upregulating EMT-related genes while altering OXPHOS and sensitivity to cuproptosis. The downregulation of miR-148b-3p in CAF-derived exosomes relieves its inhibition of ATP7A, enhancing copper efflux and helping tumor cells evade cuproptosis. CAAs support tumor OXPHOS by releasing FFA via FOXO1–cAMP-mediated lipolysis and participate in LOX-driven pre-metastatic niche formation. Copper can also promote MSC recruitment via the SDF-1/CXCR4 axis

Copper and cuproptosis in CAFs regulating ECM remodeling

CAFs are the predominant stromal cell population in many solid tumors, particularly in those characterized by dense fibrotic stromal reactions such as PDAC. As the principal architects of the ECM, CAFs adopt an activated phenotype tightly linked to copper-dependent mechanisms governing matrix synthesis, crosslinking, and stiffening [164, 165]. The relationship between copper homeostasis and CAF function is multifaceted, with the LOX family acting as canonical cuproenzymes that link copper metabolism to CAF-mediated ECM remodeling.

It has been demonstrated that both LOX and LOXL family enzymes are highly expressed in CAFs, and this upregulation correlates with poor prognosis in cancer patients [166, 167]. CAF-derived LOX and LOXL enzymes directly promote tumorigenesis and progression through both paracrine and exosome-mediated mechanisms. In GC, CAF-secreted LOXL2 significantly enhances tumor cell invasion and migration through activation of the FAK pathway in recipient cancer cells [168]. A similar paracrine axis is observed in OSCC, where LOXL2 is packaged into CAF-derived EVs and internalized by tumor cells to exert pro-tumorigenic effects [169]. Extending these findings, Lewinska et al. demonstrated that CCA cells actively internalize and accumulate inflammatory CAF-derived LOX, which promotes cancer stem cell properties. Furthermore, internalized LOX upregulates OXPHOS pathways via mitochondrial transcription factor A, thereby driving metabolic reprogramming in CCA cells [82].

During the progression of solid tumors, the physical properties of the ECM undergo extensive remodeling, among which matrix stiffening represents a central hallmark of desmoplastic tumors. As the most abundant cellular component of the tumor stroma, CAFs constitute the primary driving force underlying this physical remodeling process [170]. Beyond being internalized by tumor cells, CAF-derived LOX and LOXL enzymes can also anchor within the ECM, catalyzing covalent crosslinking of matrix components and increasing matrix stiffness.

EVs released by CAFs contribute to this stiffening, which in turn modulates tumor cell phenotype through biomechanical signaling. For instance, in OSCC, EVs released by OSCC-associated fibroblasts are enriched in the mature, enzymatically active form of αLOX rather than its precursor. On the vesicle surface, αLOX assembles with periostin (POSTN), fibronectin (FN), and bone morphogenetic protein-1 (BMP-1) into a functional complex that promotes vesicle binding to the ECM and induces collagen crosslinking [169, 171]. Unlike conventional paracrine signaling, these LOX-enriched EVs preferentially associate with collagen fibers in the ECM rather than being internalized by recipient cells. This directional binding is mediated by integrin α2β1 expressed on the EV surface. Blockade of integrin α2β1 activity using neutralizing antibodies significantly reduces EV deposition on collagen fibers, thereby inhibiting local collagen crosslinking and matrix stiffening [169].

Intriguingly, LOX-mediated matrix remodeling by CAFs is not confined to the primary tumor site. During the pre-metastatic stage, soluble LOX secreted by CAFs can reach distant organs such as the lung and bone prior to tumor cell arrival, where it induces local collagen crosslinking and fibrosis. In salivary adenoid cystic carcinoma (SACC), CAF-derived LOX activates pulmonary metastasis-associated myofibroblasts (MAFs) at distant sites. This activation increases collagen I deposition and tissue stiffness in the lung. The resulting microenvironment supports colonization and proliferation of disseminated circulating tumor cells (CTCs) [172]. Once ECM stiffness is elevated through CAF-mediated remodeling, tumor cells are subjected to sustained biomechanical stimulation. This stimulus promotes phosphorylation of FAK, which drives nuclear translocation of YAP, the core effector of the Hippo pathway. Within the nucleus, YAP associates with TEAD family transcription factors to drive expression of mesenchymal markers. This drives EMT and enhances the migratory, invasive, and stem-like capacities of cancer cells [169, 173].

CAF-mediated ECM remodeling via LOX also exerts profound effects on tumor resistance. Indeed, therapeutic strategies based on the delivery of copper ionophores such as Elesclomol or cuproptosis-inducing nanomaterials represent emerging tumor treatment modalities built upon the mechanistic foundations of cuproptosis [174]. However, the dense stromal architecture produced by CAF-mediated, LOX-dependent ECM remodeling forms a physical barrier that limits the penetration and accumulation of these agents in deep tumor regions. To overcome this physical obstacle, recent investigations have proposed a strategy of “CAF exhaustion” mediated by near-infrared (NIR) light-triggered nanogenerators. This approach uses photodynamic ROS generation to ablate CAFs and degrade the ECM, removing the penetration barrier. Elesclomol-containing nanocomposites can then precisely trigger cuproptosis in tumor cells and achieve antitumor efficacy [175].

Apart from directly acting on tumor cells or indirectly regulating their functions through ECM remodeling, CAFs can modulate tumor cell susceptibility to cuproptosis through multiple mechanisms. In OSCC, CAF-derived exosomes exhibit reduced levels of miR-148b-3p, a negative regulator of the copper efflux pump ATP7A, leading to upregulation of ATP7A expression in recipient tumor cells. This enhances efflux of free intracellular copper and alleviates copper overload–induced proteotoxic stress. As a result, cancer cells evade cuproptosis while retaining their malignant phenotype [176]. Moreover, arecoline, a carcinogen strongly associated with OSCC pathogenesis, not only promotes CAF proliferation and inhibits apoptosis but also correlates with substantial changes in cuproptosis-related gene expression profiles, as shown by single-cell RNA sequencing and bioinformatic analyses. These findings suggest that arecoline may reshape the copper metabolic landscape within the TME by modulating CAF activation status, consequently influencing tumor cell function [176].

Beyond their architectural role in ECM remodeling, CAFs are increasingly recognized as drivers of immune evasion in the TME. Jia et al. concluded that, alongside ECM remodeling, metabolic support, and the promotion of EMT and stemness, CAFs can induce an immunosuppressive TME and contribute to therapeutic resistance [177]. Studies on CAF metabolic reprogramming further indicate that CAFs and their reprogrammed metabolism cooperate with infiltrating immune cells to establish an immunosuppressive, chronically inflamed niche and to support pre-metastatic niche formation [178]. Functionally, these activities encompass the recruitment and maintenance of immunosuppressive populations such as MDSCs, Tregs, and M2-like macrophages together with restraint of effector T and NK cells, and the dense LOX-crosslinked matrix discussed above reinforces this state by physically excluding cytotoxic lymphocytes, a configuration typical of the immune-excluded phenotype [179]. Because the canonical cuproenzyme LOX is central to CAF-driven matrix stiffening, we propose that copper availability may indirectly sustain this immune-excluding architecture, although whether copper homeostasis within CAFs directly governs their immunosuppressive output has not yet been established and warrants dedicated study.

In summary, copper tightly links CAF activation, LOX-dependent ECM remodeling, and tumor progression. By harnessing cuproenzymes, CAFs orchestrate structural and mechanical evolution of the ECM, ultimately establishing a fibrotic barrier. This barrier not only contributes to the promotion of malignant phenotypes but also physically impedes the delivery of cuproptosis-inducing therapies. In addition, CAFs modulate cuproptosis in tumor cells, underscoring their protective role within the TME. However, although the influence of CAFs on tumor cell cuproptosis is becoming clearer, whether CAFs themselves are susceptible to cuproptosis and the regulatory mechanisms involved remain largely unexplored. Future studies should therefore examine the susceptibility of CAFs themselves to cuproptosis. Targeting stromal copper metabolic axes may help dismantle the physical barriers of solid tumors and enhance the efficacy of copper-based anticancer therapies.

Copper and cuproptosis in CAAs

Adipocytes represent a critical stromal component of the TME, and accumulating evidence has delineated their pivotal role in tumor progression. CAAs promote malignant phenotypes in tumor cells by secreting a broad spectrum of bioactive mediators, including chemokines, lipid metabolites, and exosomes [158, 180].

In BC, Chen et al. delineated a mechanistic framework in which obesity-associated adipocytes drive metastatic progression by rewiring tumor cell copper homeostasis through a noncoding RNA–mediated regulatory axis. Mechanistically, CAA-derived secretory factors suppress the expression of circCNIH4 in BC cells, thereby inhibiting its function as a molecular sponge binding to miR-135b. The accumulated miR-135b directly targets and destabilizes the mRNA of the tumor-suppressive transcription factor FOXO1. Critically, FOXO1 loss compromises transcriptional control over the copper exporter ATP7A, leading to intracellular copper dyshomeostasis. This copper imbalance acts as a pro-metastatic signal that induces EMT, thereby enhancing tumor cell migratory and invasive capacities [181].

Copper also functions as a key endogenous signaling molecule that regulates adipocyte metabolic programs. Copper directly binds to and inhibits phosphodiesterase 3B (PDE3B), resulting in sustained elevation of intracellular cAMP levels and activation of lipolytic cascades. This process promotes triglyceride hydrolysis and the release of free fatty acids (FFAs) into the TME. During metastasis, tumor cells preferentially take up these FFAs via lipid transporters such as CD36. The FFAs then serve as substrates for mitochondrial OXPHOS, supporting the energy demands of rapid proliferation and dissemination [182]. In addition to metabolic regulation, CAA-derived copper-dependent enzymes contribute to metastatic niche remodeling. In lipid-rich microenvironments such as the bone marrow, ECM-derived signals stimulate CAAs to secrete LOX, resulting in increased matrix crosslinking and stiffness. This biomechanical reprogramming promotes mesenchymal–epithelial transition (MET) in disseminated BC cells, a step critical for metastatic colonization and outgrowth at secondary sites [183].

From a clinical perspective, the principal systemic copper carrier ceruloplasmin has recently been identified as an adipokine. Under obese conditions, adipose tissue contributes approximately 22% of circulating ceruloplasmin, increasing systemic copper transport capacity. Elevated ceruloplasmin levels positively correlate with tumor angiogenesis, TME remodeling, and disease progression, which supports its potential as a biomarker for monitoring obesity-associated cancer progression [184]. Given the central role of CAA-mediated copper signaling in tumor progression, copper-targeting strategies for inducing cuproptosis have attracted increasing interest. By exploiting CAA-driven upregulation of CD36 on tumor cells, nanotechnology-based “fatty acid camouflage” systems have been developed to selectively deliver copper ionophores into cancer cells, inducing cuproptosis and achieving robust antitumor efficacy [185].

Copper and cuproptosis in other stromal cells

Beyond CAFs and CAAs, MSCs are increasingly recognized as active participants in copper-mediated TME remodeling. For instance, MSCs exhibit a distinctively high tolerance to copper stress, often surviving concentrations exceeding 250 µM through the dynamic upregulation of the efflux pump ATP7B [186]. In the copper-rich TME, this survival advantage allows MSCs to be selectively recruited via the copper-activated SDF-1/CXCR4 signaling axis [187]. Furthermore, it has been demonstrated that copper can effectively direct the differentiation and homing of MSCs within the microenvironment [162]. Beyond their copper-dependent recruitment, MSCs form an important immunoregulatory compartment of the TME. Tumor-associated MSCs interact with tumor cells and other stromal components and have been shown to participate actively in TME immunosurveillance and to contribute to an immunosuppressive microenvironment through extensive communication with immune cells [188, 189]. Given that the copper-rich niche selectively favors MSC survival and SDF-1/CXCR4-driven enrichment, MSC accumulation may not only expand the stroma but also reinforce local immunosuppression, although direct evidence linking copper handling in MSCs to their immunomodulatory output is currently lacking.

ECs are the principal effectors of tumor angiogenesis, undergoing proliferation, directional migration, and lumen formation in response to pro-angiogenic factors such as vascular endothelial growth factor (VEGF) and angiopoietin-2 (Ang-2) [190]. The pro-angiogenic role of copper is now well established. Copper promotes angiogenesis primarily by modulating the hypoxia-responsive HIF-1α/VEGF axis [191]. Mechanistically, upon VEGF stimulation, the copper-transporting ATPase ATP7A translocates from the trans-Golgi network to the plasma membrane and interacts with VEGF receptor 2 (VEGFR2). This interaction prevents autophagic degradation of VEGFR2 and enhances the angiogenic capacity of ECs [192]. In parallel, under hypoxia or inflammatory cytokine stimulation, ROS oxidatively inactivate SENP1, thereby leading to accumulation of SUMOylated ATOX1. This facilitates NF-κB nuclear translocation and enhances the transcriptional activity of ATOX1 itself, which together upregulate key pro-angiogenic genes including IL-8, RANTES, and VCAM-1 [193]. Given the well-recognized copper addiction of tumors, copper enrichment in tumor cells may further sensitize ECs to pro-angiogenic stimuli, establishing a self-reinforcing feedback loop that sustains neovascularization within the TME. In addition to driving angiogenesis, the tumor endothelium influences antitumor immunity by regulating the trafficking and survival of immune effectors, a feature well documented in tumor vascular immunology, in which an anergic, dysfunctional endothelium limits cytotoxic lymphocyte adhesion and extravasation and thereby promotes T-cell exclusion. Because copper is a key driver of the HIF-1α/VEGF axis, copper-fueled aberrant angiogenesis may concurrently establish a poorly infiltrated, immunologically cold vasculature, whereas restricting copper could in principle favor vascular normalization and improved immune-cell trafficking. Stromal cells, including endothelial cells, also participate in the neogenesis of tertiary lymphoid structures through chemokine-mediated recruitment of lymphocytes, and the presence of mature tertiary lymphoid structures correlates with improved prognosis and immunotherapy responsiveness [194]. How copper homeostasis within these stromal niches shapes tertiary lymphoid structure formation remains an open and clinically relevant question.

In summary, copper and cuproptosis regulate stromal–tumor interactions, mediating TME remodeling through effects on tumor cell function, metabolism, and the ECM. Nearly all stromal cell types contribute to this process. CAFs and CAAs are the most extensively characterized, owing to their abundance and central role in ECM remodeling, whereas MSCs and ECs, though less studied in the copper context, are not passive bystanders but active regulators of the immune microenvironment, suppressing effector lymphocytes and gating immune-cell infiltration, respectively. Their roles in copper-dependent TME remodeling therefore warrant closer attention.

Copper and cuproptosis in immune cells

Immune cells are indispensable core components of the TME. The immune cells within the TME include T cells, B cells, NK cells, tumor-associated macrophages (TAMs), and neutrophils, among others. These cells, through their complex interactions with tumor cells, stromal cells, and non-cellular components, collectively form the tumor immune microenvironment (TIME), which can both mobilize effector cells to execute tumor cytotoxicity and, conversely, promote immune evasion through immunosuppressive polarization and cytokine secretion [195]. Increasing evidence suggests that copper and cuproptosis are closely involved in shaping the TIME. However, their roles are far more nuanced than a uniform supplementation or depletion would suggest. Different immune cell subsets maintain distinct copper homeostasis, respond differently to changes in copper availability within the TME, and compete with tumor and stromal cells for a limited pool of labile copper [196]. Herein, we summarize how copper and cuproptosis shape immune cells within the TME from three perspectives: their regulatory effects on immune cell function, the heterogeneous copper distribution across immune subsets, and the intercellular competition for copper that arises within the TIME (Fig. 7).

Fig. 7.

Cuproptosis modulates immune cell function and tumor immunity. Copper homeostasis and cuproptosis jointly shape the TIME. In innate immunity, copper promotes TAM polarization toward the M2 phenotype by inhibiting IL-1β/IL-6 production in a CTR1-dependent manner. In contrast, agents like CuET or copper-based nanoplatforms can release high doses of copper, activate cholesterol oxidase–related ROS signaling, induce macrophage migration and promote M1 reprogramming, while also elevating the effector molecule perforin in NK cells. In adaptive immunity, copper deficiency reduces IL-2 activity and inhibits T cell proliferation, whereas copper overload leads to ROS accumulation, inhibits FOXO/FOXM1, and results in cytoskeletal collapse and apoptosis. Intracellular copper in tumor cells can also upregulate PD-L1 through transcriptional activation via EGFR/STAT3 and the TAK1–IKKβ–NF-κB pathway. Conversely, inducing cuproptosis can inhibit the Wnt/β-catenin pathway to reduce PD-L1, while concurrently triggering ICD with release of DAMPs (HMGB1, ATP, and calreticulin) and mitochondrial DNA leakage that activates the cGAS-STING signaling to drive IFN-β production. These signals collectively promote DC maturation and M1 macrophage polarization while suppressing Treg infiltration, reshaping the TIME toward antitumor immunity

Copper and cuproptosis modulate immune cell function

Copper exerts pervasive regulatory effects on the function of both innate and adaptive immune cells within the TME. In the innate compartment, macrophages, NK cells, and DCs are particularly responsive to copper-mediated modulation, with macrophages representing the most extensively characterized target. TAMs undergo stimulus-dependent polarization into pro-inflammatory M1 macrophages or anti-inflammatory M2 macrophages, thereby jointly shaping tumor immunity [197]. Accumulating evidence has demonstrated that copper suppresses the production of pro-inflammatory cytokines such as IL-1β and IL-6 in macrophages, and promotes M2 polarization [198, 199]. Notably, Xu et al. combined single-cell RNA sequencing with in vitro experiments to reveal that the copper uptake protein CTR1 is also critically involved in M2 macrophage polarization [200]. Importantly, the immunomodulatory effects of copper are highly dose-dependent. For instance, Lu et al. reported that diethyldithiocarbamate-containing Cu(II) (CuET) could release high doses of copper into the BC microenvironment, which then induced macrophage migration and increased the proportion of M1 macrophages, thereby enhancing antitumor immunity [201]. Beyond CuET, copper-based nanoplatforms have similarly been shown to reprogram TAMs toward pro-inflammatory phenotypes through copper-redox signaling and cholesterol oxidase–mediated cholesterol metabolism interference [202]. In addition, the tissue-resident macrophages of distinct ontogeny and self-renewal capacity occupy specialized niches across organs and can exert either pro-tumorigenic or anti-tumorigenic effects via phagocytosis and the secretion of cytokines, chemokines, and growth factors that modulate the adaptive immune response, adding a further layer of macrophage heterogeneity relevant to copper-mediated modulation [203].

NK cells are innate lymphocytes that mediate MHC-unrestricted cytotoxicity against tumor cells, and their effector function is frequently blunted within the immunosuppressive TME, motivating a range of NK-directed therapeutic strategies including adoptive transfer, CAR-NK cells, and bispecific or trispecific engagers [204, 205]. The essential role of copper in supporting the cytotoxic function of NK cells has been recognized for decades [206]. In the context of tumor immunity, delivery of copper into the CRC immune microenvironment using CuET significantly enhanced the expression of the NK cell effector molecule perforin, thereby potentiating NK cell–mediated antitumor activity [207]. NK abundance and function can decline in an age-dependent manner, and in aged mice a reduction in NK infiltration impairs the recruitment and activation of CD103-positive dendritic cells, weakening tumor-antigen-specific CD8⁺ T-cell responses and contributing to resistance to PD-L1 therapy, whereas adoptive transfer of NK cells from young mice remodels the TME and reverses this resistance [208]. This identifies the NK compartment as an additional and still underexploited effector arm that copper-targeting strategies may help engage. DCs constitute a unique class of antigen-presenting cells that process and present antigens to adaptive lymphocytes, thereby serving as a crucial functional bridge between innate and adaptive immunity [209]. Interestingly, in BC, chelation of copper promotes the maturation and activation of the human plasmacytoid dendritic cell line CAL-1, as evidenced by the significantly increased expression of CD80, CD86, and CD83 [210].

MDSCs are a heterogeneous population of immature myeloid cells and constitute a principal barrier to effective antitumor immunity, mediating immune escape mainly by suppressing T-cell activity and helping to constitute an immunosuppressive microenvironment, for which they are now an actively pursued therapeutic target [211, 212]. The effector mechanisms include arginase-1 and inducible nitric oxide synthase (iNOS), among others. Within this framework, copper adds a metabolically encoded layer of suppression. Baumann et al. demonstrated that semicarbazide-sensitive amine oxidase (SSAO), a copper-containing amine oxidase, is enriched in MDSCs and catalyzes the synthesis of methylglyoxal. This metabolite is subsequently transferred directly to CD8⁺ T cells through cell–cell contact, inducing metabolic paralysis and loss of effector function in T cells [213]. This mechanism reveals that copper not only supports the intrinsic functions of immune cells as a metabolic cofactor and signaling mediator, but can also become an instrument of active tumor-mediated immunosuppression through copper-enzyme-mediated intercellular metabolic transfer.

Recent studies have revealed that copper shapes the recruitment of tumor-associated neutrophils (TANs) into the TME. In PDAC, neutrophils are selectively recruited via a copper-dependent CCL2/CCR2 chemotactic axis to form a pro-metastatic TAN-2 subpopulation. Blockade of this axis suppresses such pathological recruitment while restoring CD8⁺ T cell infiltration [214]. This finding highlights copper as an important metabolic factor governing TAN recruitment patterns within the TME.

Adaptive immunity is a highly specific immune response mediated by T and B lymphocytes, capable of recognizing tumor-associated antigens and establishing immunological memory, and it plays a central role in tumor immune surveillance and responses to immunotherapy [215]. T cells are particularly dependent on tightly regulated copper homeostasis and are highly sensitive to fluctuations in copper levels. Copper deficiency impairs DNA synthesis in activated T cells by limiting IL-2 activity, thereby restricting T cell clonal expansion and weakening immune responses [216]. A recent study in zebrafish demonstrated that copper overload also disrupts effector T cell homeostasis, leading to impaired T cell proliferation and increased apoptosis of T cells and their progenitors. Mechanistically, copper excess induces ROS accumulation while suppressing FOXO transcription factors, which normally drive the expression of antioxidant defense genes, and concurrently inhibits the FOXM1–cytoskeleton axis. Together, these effects drive activated T cell apoptosis and proliferative defects [217]. This dose-dependent regulation further exhibits lineage specificity at the level of T-cell subsets. Zhang et al. demonstrated that CTR1, by regulating copper transport in Treg cells, maintains mitochondrial OXPHOS and the NAD⁺/NADH balance, while stabilizing histone acetylation at core functional gene loci. Notably, both in vitro T-cell receptor (TCR) stimulation and human autoimmune disease settings increase the intracellular labile copper pool in Treg cells, revealing an activation state-dependent copper uptake program in Tregs and its supportive role in peripheral immune tolerance [218]. In contrast, Noyer et al., through in vivo genetic screening, found that CTR1 is required for pathogenic Th17 (pTh17) cell differentiation and IL-17A production, whereas Th1 differentiation and function remain unaffected. This subset specificity further indicates that copper metabolic dependence varies between T cell subsets [219]. Taken together, these findings suggest that the effect of copper on T-cell function is not a simple binary switch between activation and inhibition, but rather a finely tuned program shaped by subset identity and functional context.

The roles of copper or cuproptosis in B cells remain relatively underexplored. Indeed, emerging evidence suggests that most studies do not investigate copper-mediated immune regulation in isolated immune cell systems, but instead focus on how copper-containing nanomedicines modulate immune cell functions. At the single-cell and transcriptomic levels within the TME, several studies have correlated the expression of cuproptosis-related genes (CRGs) with immune cell infiltration patterns and macrophage polarization states [220, 221]. However, there are currently no direct reports of immune cells undergoing cuproptosis themselves. Nonetheless, the regulatory effects of copper on immune cell function and the crosstalk between immune cells and tumor cell–intrinsic cuproptosis represent a rapidly evolving area of research, with a central focus on how copper homeostasis remodeling shapes tumor immunity within the broader framework of the TIME.

Copper distribution heterogeneity across immune cell subsets

The baseline copper levels and copper distribution vary among different immune cell types. Myeloid cells, particularly those of the mononuclear phagocyte system, possess substantial copper reserves, in contrast to T or B cells. For instance, inflammatory macrophages harbor measurable Cu(II) stores within mitochondria. This mitochondrial copper pool is highly active and chemically reactive, capable of sustaining NAD⁺ levels to drive metabolic and epigenetic programs toward a pro-inflammatory state [14]. These differences in copper content are closely associated with the expression levels of the copper transporter CTR1. Pan-cancer bioinformatic analyses have revealed that CTR1 expression is elevated in multiple tumor tissues, and its expression positively correlates with the infiltration of macrophages and myeloid cells, while negatively correlating with CD8⁺ T cells, NK CD56^bright cells, and plasmacytoid DCs [222]. These findings are further validated in the Database of Immune Cell Expression, Expression Quantitative Trait Loci (eQTLs), and Epigenomics (DICE) (https://dice-database.org/) [223].

Beyond cell-type heterogeneity, copper distribution within immune cells is also shaped by subcellular architecture. The copper pool in inflammatory macrophages described above is localized within mitochondria, consistent with the dependence of their pro-inflammatory function on active mitochondrial energy metabolism [14]. Beyond mitochondria, copper in myeloid cells is also redistributed toward the phagolysosomal system upon innate immune activation. In IFN-γ–stimulated macrophages, ATP7A relocates from the trans-Golgi network to phagosome-associated vesicles, delivering copper into the phagosomal lumen to support immune activity [224]. Furthermore, recent work has demonstrated that ATP7A-mediated copper delivery to the lysosome is required not only for phagolysosome formation and acidification, but also for downstream lysosomal ROS generation and chemokine secretion in both neutrophils and macrophages, mechanistically through ATP7A interactions with LAMP1/CTSB and activation of the Rab5/Rab7 cascade [225]. These compartmentalized copper fluxes are integrated at the Golgi apparatus, where ATP7A and ATP7B load copper onto cuproenzymes traversing the secretory pathway and direct their post-Golgi sorting to distinct destinations, including the plasma membrane, secretory vesicles, and phagosomes, depending on cellular state [226]. The Golgi thus serves as a hub linking intracellular copper handling to cuproenzyme secretion. This state-coupled redirection of copper allows immune activation to rapidly reorganize subcellular copper allocation—within minutes to hours—in response to evolving functional demands.

However, the resolution of evidence regarding subcellular copper distribution varies across immune cell types: while the mitochondrial and phagolysosomal copper pools have been well characterized in macrophages and are increasingly being studied in T cells, the subcellular localization of copper in NK cells, B cells, and DCs remains poorly defined. For instance, although copper supports perforin expression and NK cell cytotoxicity [207], direct evidence for copper enrichment within cytotoxic granules has yet to be established. Filling these gaps requires the systematic application of subcellular copper imaging approaches to immune cell populations isolated from the TME, including X-ray fluorescence microscopy and reactive copper-selective fluorescent probes combined with organelle-specific labeling.

Intercellular competition for copper within the TIME

Whether tumor cells and immune subsets actively compete for the limited labile copper pool within the TME remains a central unresolved question. An analogous mechanism has been described for iron: in patients with bone metastases, tumor cells hijack a specialized VCAM1⁺CD163⁺CCR3⁺ macrophage population that normally delivers iron to erythroblasts, rerouting this iron to support iron-dependent metabolic and redox processes that drive tumor proliferation and hypoxic adaptation [227]. This supports the idea that metal competition between the tumor and immune compartments is a real biological phenomenon, which naturally raises the same question for copper. Emerging evidence suggests that copper undergoes competitive allocation in the TIME, not only between tumor and immune compartments, but also among immune subsets themselves.

Direct evidence comes from neuroblastoma, in which tumor cells sequester copper from infiltrating neutrophils, depriving them of cofactors required for effector function. The clinically approved copper chelator TETA reverses this suppression, restoring antibody-dependent cellular cytotoxicity (ADCC) and enhancing anti-GD2 antibody therapy [228]. Competition also operates among immune subsets, with copper-avid suppressive populations acting as a metabolic sink. For instance, the copper-dependent enzyme SSAO catalyzes methylglyoxal synthesis, and the methylglyoxal is transferred from MDSCs to CD8⁺ T cells through cell–cell contact and induces metabolic paralysis [213]. Together, these examples show that copper within the TIME is not freely shared but actively partitioned, with functional consequences for antitumor immunity.

On the basis of the foregoing subsections, we propose an integrated molecular interaction network through which copper homeostasis and cuproptosis collectively reshape the TME. (1) Copper is heterogeneously partitioned across tumor cells, neovasculature, ECM, and stromal and immune compartments, establishing the TME as a structured copper ecosystem in which intercompartmental copper redistribution continuously occurs. (2) Within tumor cells, the labile copper pool acts as a bidirectional input. The relatively low concentration of copper sustains cuproplasia, autophagy, EMT, and genomic instability through oncogenic cascades. Upon copper overload, the FDX1/LIPT1/DLAT/Fe-S axis is engaged to execute cuproptosis, and the threshold between these two regimes is jointly dictated by labile copper levels, the relative reliance on OXPHOS versus glycolysis, and intrinsic cellular heterogeneity. (3) Within the stromal compartment, copper-dependent LOX/LOXL enzymes link CAFs and CAAs to ECM crosslinking and downstream mechanotransduction. Stromal cells further modulate tumor cell cuproptosis sensitivity and provide lipid-based metabolic support, while copper concurrently amplifies EC-driven angiogenesis and MSC recruitment into the niche. (4) Within the immune cells, copper is unevenly distributed across immune subsets, with myeloid cells maintaining both mitochondrial and ATP7A-directed phagolysosomal copper pools. Functionally, copper polarizes TAMs, regulates NK perforin and DC maturation, and dose-dependently controls T-cell expansion through IL-2 and the ROS/FOXO/FOXM1 axis. (5) Across these compartments, copper undergoes competitive allocation within the TIME, in which tumor cells sequester copper from infiltrating neutrophils, and MDSC-expressed SSAO transfers methylglyoxal to paralyze CD8⁺ T cells. Together, these five interconnected layers position copper as a unifying molecular hub linking the tumor, stromal, and immune compartments of the TME, reframing the microenvironment as an integrated copper-regulated ecosystem in which the balance between copper-driven progression and cuproptosis dictates tumor fate.

The effects of cuproptosis on tumor immunity

Copper homeostasis and cuproptosis act as key modulators of tumor immunity, with bidirectional impacts on immune escape and antitumor immune reactions. On one hand, cuproptosis-induced immunogenic cell death (ICD) directly activates antitumor immunity through tumor-cell release of damage-associated molecular patterns (DAMPs) and engagement of the cGAS-STING pathway, thereby promoting DC maturation, eliciting CD8⁺ T-cell antitumor function, and driving M1 polarization of macrophages. On the other hand, copper overload in viable tumor cells reinforces immunosuppression by modulating immune checkpoint molecules, particularly PD-L1, through transcriptional activation and post-translational stabilization, thereby promoting T-cell exhaustion and immune evasion (Fig. 7). In this section, we systematically delineate and contrast these dual roles, integrating these mechanisms within a unified molecular interaction network.

Cuproptosis triggers immunogenic cell death

The immunogenic potential of regulated cell death mainly depends on its ability to release DAMPs, which activate innate immune sensing and trigger downstream adaptive immunity. The three canonical hallmarks of ICD comprise surface-exposed calreticulin (an “eat-me” signal), extracellular adenosine triphosphate (ATP, a “find-me” chemoattractant acting through P2Y2 and P2X7 receptors), and released high-mobility group box 1 (HMGB1), which binds TLR4 and RAGE [229]. Multiple lines of evidence consistently indicate that cuproptosis represents a bona fide ICD inducer.

Mechanistically, cuproptosis exhibits the canonical features of ICD. The direct consequence of cuproptosis is severe mitochondrial proteotoxic stress and dysfunction, which propagate to the ER and activate the unfolded protein response (UPR). The downstream PERK/eIF2α/CHOP signaling axis subsequently drives the translocation of calreticulin from the ER lumen to the plasma membrane, where it serves as a recognition marker for phagocytosis by conventional type 1 dendritic cells (cDC1). In CRC cells, cuproptosis upregulates the expression of CHOP, eIF2α, PERK, and GADD34, whereas the antioxidant N-acetylcysteine (NAC) scavenges copper-induced ROS and concurrently suppresses ER stress and the membrane localization of calreticulin, thereby mechanistically linking mitochondrial proteotoxic stress to ER-mediated calreticulin trafficking [230]. Consistent with this mechanism, surface exposure of calreticulin has likewise been documented across multiple copper-based regimens, including DSF/Cu in HCC cells [231], the ES-Cu nanoplatform in 4T1 BC cells [232], and Cu-MOFs in fibrosarcoma cells [15]. Concurrently, extracellular ATP secretion accompanied by intracellular ATP depletion has been observed in parallel with calreticulin translocation [231, 233].

Cuproptosis is likewise accompanied by HMGB1 release. However, this process is mechanistically distinct from the passive HMGB1 efflux that characterizes ICD induced by oxaliplatin, anthracyclines, and ionizing radiation. Copper accumulation provokes energy metabolism dysfunction, which in turn activates AMPK. Activated AMPK then phosphorylates HMGB1 and facilitates its release. Pharmacological inhibition of AMPK simultaneously blocks both cuproptosis execution and HMGB1 release, whereas HMGB1-deficient cells undergoing cuproptosis fail to elicit RAGE-dependent inflammatory cytokine secretion [234]. More direct evidence has been provided by confocal microscopy, which captured calreticulin membrane translocation accompanied by the activation of CD8⁺ T cells within tumor-draining lymph nodes [235]. Furthermore, an in vitro comparative analysis of twelve RCD modalities in CRC suggested that cuproptosis has pronounced ICD-inducing capacity, with cuproptosis intensity, DAMP release, and antigen-presenting cell maturation closely correlated [230].

Collectively, cuproptosis constitutes a highly immunogenic mode of RCD, capable of triggering ICD in tumor cells and thereby exerting profound effects on tumor immunity.

Cuproptosis activates the cGAS-STING pathway

The cGAS-STING axis is a central cytosolic DNA-sensing pathway in which cGAS recognizes cytosolic DNA and activates STING, driving IRF3- and NF-κB-dependent production of type I interferons and pro-inflammatory cytokines, and it has become a pivotal, therapeutically tractable bridge between innate sensing and adaptive antitumor immunity, with recognized dual antitumor and immunosuppressive roles [236, 237]. Cuproptosis engages precisely this axis. Beyond inducing ICD, cuproptosis also activates the cGAS-STING innate immune signaling axis, thereby driving the transcriptional induction of type I IFNs. Using the copper complex Cu-DPPZ-Py⁺ and an ES-Cu nanosystem as model platforms, Zhu et al. demonstrated that cuproptosis triggers mitochondrial permeability transition pore (mPTP)-dependent leakage of mitochondrial DNA (mtDNA), whereas the apoptosis-inducing control compound Cu-DPPZ-Ph did not [238]. The released mtDNA activates cGAS to generate 2′3’-cGAMP, which in turn engages STING and drives TBK1/IRF3-dependent IFN-β production. Notably, both cuproptosis inhibitors and mPTP inhibitors abolished mtDNA release, mechanistically anchoring this signaling cascade to the cuproptosis–mitochondrial axis [238]. In addition, DSF/Cu generates substantial ROS that damage both mitochondrial and nuclear DNA, producing abundant cytosolic double-stranded DNA (dsDNA) fragments that potently activate the cGAS-STING pathway. In tumor-bearing mouse models, this activation enhanced the efficacy of anti-PD-1 immunotherapy [239].

Cuproptosis regulates immune cell function

Cuproptosis intrinsically generates ROS, induces ICD, and activates the cGAS-STING pathway, and these mechanisms collectively underlie its broad regulatory effects on immune cell function. Specifically, cuproptosis promotes DC maturation, activates CD8⁺ T cells, modulates the proportion of Tregs, and drives M1 polarization of macrophages, thereby concurrently potentiating both innate and adaptive antitumor immunity.

Cuproptosis promotes DC maturation

DC maturation is hallmarked by elevated expression of antigen-presenting and costimulatory molecules, including MHC class I/II, CD80, and CD86. Among these, MHC class I/II present processed antigens to the TCR, while CD80 and CD86 provide costimulatory signals, together promoting T-cell activation. Cuproptosis has been shown to promote DC maturation through multiple complementary mechanisms. For instance, HMGB1 released during cuproptosis is recognized by TLR4/RAGE on the DC surface, engaging the downstream MyD88/NF-κB axis to drive the maturation transcriptional program [240]. In parallel, tumor cells undergoing cuproptosis activate the cGAS-STING pathway to secrete type I IFNs, which subsequently upregulate MHC class I/II and CD80/CD86 on DCs, completing their functional maturation [241]. Luo et al. provided further quantitative evidence in a 4T1 BC-bearing model, demonstrating that cuproptosis induction increased the intratumoral proportion of mature DCs by approximately 8-fold relative to control, and elevated CD4⁺ and CD8⁺ T-cell infiltration by approximately 5.5- and 6.5-fold, respectively, effectively reversing the immunosuppressive TME [242]. Building on this foundation, Lien et al. engineered an N-doped carbon dots/mesoporous silica nanoparticle platform (CMS) for ES-Cu delivery. In this system, tumor-cell cuproptosis is coupled with active DC capture and retention within the TME through the pH-responsive cationic surface of CMS. Tumor antigens released in situ are subsequently directed to retained DCs by the platform, enabling sustained antigen exposure and phenotypic reprogramming. This study provided spatially resolved evidence that the DC retention/reprogramming/activation cascade within the TME can be precisely initiated by cuproptosis [243].

Cuproptosis regulates T-cell function and promotes adaptive antitumor immunity

Following cuproptosis-driven DC maturation, antitumor immunity is principally executed by the adaptive compartment, in which T cells, particularly CD8⁺ T cells, serve as the most direct and central effectors. Cuproptosis first enhances T-cell chemotaxis and infiltration. In clear cell renal cell carcinoma (ccRCC), it has been demonstrated that cuproptosis transcriptionally upregulates HLA-DRA in a dose-dependent manner through ROS generation. Elevated HLA-DRA in turn promotes the expression of the chemokines CCL5, CXCL9, and CXCL10 within the TIME, augmenting CD4⁺ and CD8⁺ T-cell infiltration and suppressing the development of a pro-tumor microenvironment [244]. Beyond infiltration, cuproptosis further potentiates T-cell effector function. In microsatellite-stable colorectal cancer (MSS-CRC), in vitro studies have demonstrated that co-culture of ES-Cu–treated cuproptotic tumor cells with CD8⁺ T cells markedly enhances T-cell immune activity while concurrently reducing surface PD-1 expression [245]. In parallel with augmented effector T-cell infiltration and function, cuproptosis suppresses the Treg compartment. In a murine BC model, Guo et al. constructed an “in situ vaccine” through intratumoral injection of DSF/Cu combined with local radiotherapy and observed concomitant reductions in Tregs and MDSCs alongside a significant increase in CD8⁺ T-cell numbers [246]. Consistent observations have also been reported in LC [247] and CRC [248]. These findings appear to contrast with the previously discussed role of copper in maintaining Treg function [218]. However, this apparent discrepancy can be reconciled by recognizing that cuproptosis-driven remodeling of the TIME indirectly restricts Treg infiltration, rather than perturbing intrinsic Treg copper metabolism. Collectively, cuproptosis in tumor cells enhances both T-cell infiltration and effector function while suppressing Treg infiltration, jointly potentiating adaptive antitumor immunity and reversing the immunosuppressive TME.

Cuproptosis regulates macrophage polarization

Following cuproptosis-mediated ICD induction and cGAS-STING activation, TAMs within the TME undergo rapid repolarization from the immunosuppressive M2 toward the antitumor M1 phenotype. Mechanistically, tumor cell–derived ecto-CRT, ATP, and HMGB1 are recognized by LRP1, P2X7/P2Y2, and TLR4/RAGE on TAMs, respectively, activating the MyD88–NF-κB axis and the NLRP3 inflammasome, which together promote pro-inflammatory M1-like polarization [249]. In addition, type I IFNs downstream of cGAS-STING engage the IFNAR–JAK–STAT1 axis in TAMs to upregulate IRF1 and IRF8 while suppressing the STAT6/PPARγ-driven M2 transcriptional program [240]. Concurrently, the oxidative burst triggered by copper-mediated Fenton-like reactions, superimposed on cuproptosis-driven perturbation of lactate metabolism, shifts TAM metabolic preference from fatty acid oxidation toward a hybrid glycolysis–OXPHOS state [240]. Hou et al. further dissected this regulation using the copper chelator tetrathiomolybdate (TTM) in DSF/Cu-treated macrophages, demonstrating that CTR1-mediated copper influx alone is sufficient to upregulate CD86, TNF-α, and iNOS while downregulating CD206, TGF-β, and IL-10, with all phenotypic changes reversed by TTM. This indicates that TAM repolarization is directly driven by copper-induced events rather than being a bystander effect [250]. Functionally, Gao et al. quantitatively documented the microenvironment-remodeling capacity of cuproptosis in a PDAC model, reporting that intratumoral M2 proportions decreased from 50.7 to 18.4% following cuproptosis induction [251], and consistent M2-to-M1 repolarization has been reproduced in melanoma [252].

Cuproptosis modulates immune checkpoints and synergizes with ICI therapy

Immune checkpoints are key molecular regulators of immune system activity. Canonical immune checkpoint molecules include programmed cell death-1 (PD-1) and cytotoxic T-lymphocyte–associated protein 4 (CTLA-4) expressed on T cells, as well as programmed death-ligand 1 (PD-L1) expressed on tumor cells and other immunosuppressive cell types. Under physiological conditions, immune checkpoints restrain immune cells, particularly T cells, from mounting attacks against normal tissues [253]. However, under conditions of elevated intracellular copper, tumor cells can modulate the expression of immune checkpoint molecules on their own surface or in surrounding cells, thereby attenuating T-cell–mediated cytotoxicity and facilitating immune evasion. For instance, in glioblastoma (GBM), Florida et al. found that high expression of the copper transporter CTR1 promotes copper uptake, and CTR1 expression is positively correlated with PD-L1 levels [254]. At the molecular level, copper regulates PD-L1 through multiple mechanisms. First, copper accumulation enhances the phosphorylation and activation of the oncogenic kinase EGFR and the transcription factor STAT3. Activated STAT3 directly binds to the PD-L1 promoter, thereby increasing its transcriptional activity. Second, copper ions appear to influence the post-translational stability of PD-L1. Experimental evidence indicates that depletion of intracellular copper promotes PD-L1 ubiquitination, leading to accelerated proteasomal degradation [254]. Similarly, in BC, intracellular copper directly binds to TAK1, resulting in activation of the NF-κB signaling pathway and subsequent TNF-α-induced PD-L1 expression [87].

By contrast, cuproptosis in tumor cells decreases immune checkpoint expression. In MSS-CRC, Zeng et al. demonstrated through both in vitro and in vivo experiments that cuproptosis suppresses Wnt signaling activation, leading to enhanced β-catenin degradation and impaired nuclear translocation, ultimately resulting in reduced PD-L1 expression in tumor cells [245]. This regulatory pattern has been further observed across diverse cancer types and is mechanistically supported by multiple signaling axes. In GBM, Chen et al. reported that a copper-coordination nanoplatform induces cuproptosis and activates the AMPK pathway, which in turn promotes the ubiquitin-mediated degradation of PD-L1 [255]. Li et al. further extended this regulatory axis to both CRC and TNBC models, where a mitochondria-targeted Cu/Ti nanoparticle simultaneously triggered cuproptosis and downregulated PD-L1 expression, thereby promoting intratumoral infiltration and activation of cytotoxic T cells [256]. Similarly, researchers have found that in osteosarcoma (OS), induction of tumor cell cuproptosis can also reduce PD-L1 expression. Based on this finding, they combined a PD-L1 monoclonal antibody with ES-Cu, which not only induced cuproptosis in OS cells but also downregulated tumor PD-L1 expression, thereby enhancing CAR-T cell infiltration within the TME [257]. Collectively, cuproptosis can induce ICD, activate the cGAS-STING signaling pathway, and enhance antitumor immune cell functions, thereby promoting the conversion of tumors from an immunologically “cold” to a “hot” phenotype [258]. Although cuproptosis downregulates PD-L1 expression across multiple tumor types, which might theoretically reduce the target available for anti-PD-L1 blockade, this intrinsic effect is outweighed by concurrent multi-layered remodeling of the TIME. In particular, the establishment of an inflammatory milieu characterized by macrophage M1 polarization, reduced Treg infiltration, and expanded effector T-cell infiltration provides a favorable context for ICI-based therapy. Consistently, preclinical studies have shown that combining cuproptosis induction with ICIs significantly enhances antitumor efficacy, indicating a synergistic—rather than merely additive—effect. We discuss these combination strategies in the combined therapy section.

On the basis of the foregoing discussion, we propose an integrated molecular network from cuproptosis to antitumor immunity: (1) mitochondrial proteotoxic stress and mPTP-mediated leakage during cuproptosis are the upstream triggers of downstream immune events. (2) cuproptosis directly triggers ICD and activates the cGAS-STING axis, eliciting the release of canonical DAMPs. (3) within the immune compartment, these signals promote DC maturation, drive M1 macrophage polarization, and enhance both CD8⁺ T-cell infiltration and effector function while suppressing Treg accumulation, thereby reshaping the TIME into an immune-promoting milieu. (4) cuproptosis modulates PD-L1 expression and thus influences the effector arm of adaptive immunity, providing a mechanistic rationale for combining cuproptosis induction with ICI therapy. Together, these four causal layers position cuproptosis as a unified upstream node for orchestrating antitumor immunity and provide the mechanistic foundation for the copper-based therapeutic strategies and combination regimens in tumor management.

Therapeutic targeting of copper metabolism and cuproptosis

To date, therapeutic interventions targeting dysregulated copper metabolism and cuproptosis have diversified into several modalities, including copper ionophores, copper chelators, and nanotechnology-based copper regulatory platforms. Importantly, modulation of copper homeostasis can be integrated with immunotherapeutic and metabolic approaches to potentiate synergistic antitumor efficacy [84]. Conceptually, these modalities can be grouped along two complementary strategic axes including cuproptosis induction (elevating intracellular copper beyond the cuproptosis threshold) and copper depletion (sequestering labile copper to disable copper-driven oncogenic processes). The two strategies differ systematically in mechanism, supported indications, response-predicting tumor states, resistance mechanisms, and pivotal clinical signals (Table 2). Accordingly, we begin this section by summarizing candidate predictive biomarkers that may guide future precision-oncology applications before discussing each therapeutic modality and its combination strategies in turn (Fig. 8).

Table 2.

Comparison between cuproptosis-inducing strategies and copper depletion strategies

Feature	Cuproptosis induction (ionophores/Cu-nanoplatforms)	Copper depletion (chelators)
Representative agents	Elesclomol, DSF/Cu (CuET), 8-HQ derivatives, Copper-delivery nanoplatforms	TTM, TETA, D-penicillamine
Mechanism	Mitochondrial copper overload/FDX1/LIPT1/DLAT aggregation [6]/proteotoxic cuproptosis; ROS and NPL4 aggregation [259]	Sequestration of labile copper/loss of cuproenzyme activity (cytochrome c oxidase, LOX, SOD1)/impaired OXPHOS, angiogenesis, ECM remodeling, EMT [260]
Indications	OXPHOS-dependent cancer and CSC–enriched tumors: GBM (DSF/Cu) [261–263]; PDAC [264], Group 3 medulloblastoma [135], melanoma (ES-Cu) [264, 265]	Highly metastatic/angiogenesis-driven tumors: TNBC (TTM) [266, 267]; HER2-negative BC (TETA) [98]
Resistance mechanisms	GSH upregulation via Nrf2/GCLC [268]; ATP7A/B-mediated copper efflux; HIF-1α–driven DLAT downregulation [136]; glycolytic switch (Warburg); MGRN1-mediated LIPT1 degradation [134]; AKT1-phosphorylated FDX1 [155]	Compensatory CTR1/ATOX1 upregulation; rebound angiogenesis after withdrawal; metabolic adaptation away from copper-dependent OXPHOS [266, 267]
On-target/off-target toxicities	Mitochondrial toxicity in OXPHOS-reliant normal tissues (cardiomyocytes, hepatocytes, neurons); DSF-related neurotoxicity; bone marrow suppression in combination [261, 263]	Copper-deficiency myeloneuropathy on prolonged use; reversible neutropenia (TTM); sideroblastic anemia; hepatotoxicity [269]
Pivotal clinical trials	NCT00522834; NCT02678975; NCT02715609	NCT00195091

Fig. 8.

Therapeutic targeting of copper homeostasis and cuproptosis. Targeting copper homeostasis and cuproptosis represents a promising therapeutic strategy against tumors. On one hand, copper loading/ionophores and nanodelivery platforms can be utilized to elevate intracellular copper levels in tumors and induce cuproptosis. On the other hand, copper chelation can be employed to deplete copper and inhibit copper-dependent oncogenic processes. These strategies can be combined with chemotherapy, radiotherapy, immunotherapy, metabolic interventions, and energy-triggered therapies. By optimizing delivery efficiency and therapeutic synergy, such integrated approaches can enhance overall antitumor efficacy and reduce the risk of drug resistance

Cuproptosis-related predictive biomarkers for precision cancer therapy

The clinical translation of copper-targeted therapeutics fundamentally depends on the identification of robust predictive biomarkers for patient stratification and precision oncology. Although copper-targeted therapy has shown potent antitumor activity in preclinical models, the heterogeneity of clinical responses necessitates molecular signatures that can accurately predict cuproptosis sensitivity [270]. Accordingly, we propose an integrated biomarker framework encompassing cuproptosis-execution biomarkers, including FDX1, DLAT, protein lipoylation, and copper levels.

Within the FDX1/LIPT1/DLAT/Fe-S mechanistic axis of cuproptosis, FDX1 acts upstream of this axis and is currently the most extensively investigated biomarker. However, tissue-level validation and multi-cohort bioinformatic analyses have revealed substantial heterogeneity in FDX1 expression across tumor types. In solid tumors that retain mitochondrial OXPHOS dependence, such as LGG, high FDX1 expression is associated with increased immune checkpoint expression and poorer overall survival (OS) [271]. By contrast, in ccRCC, FDX1 is downregulated, and its loss instead predicts shorter survival and is independently associated with higher tumor immune dysfunction and exclusion (TIDE) scores, enhanced immune exhaustion, and resistance to immunotherapy [272, 273]. Similarly, in HCC, post-transcriptional suppression of FDX1 by miR-3130-5p accelerates tumor growth and predicts unfavorable clinical outcomes [274]. This paradox may be mechanistically explained by the broader biological functions of FDX1 beyond cuproptosis. In adrenal and gonadal tissues, FDX1 participates in steroidogenesis by transferring electrons to mitochondrial cytochrome P450 enzymes. In addition, FDX1 cooperates with FDX2 in heme A biosynthesis, thereby supporting complex IV assembly and OXPHOS [275]. Therefore, FDX1 expression primarily reflects the metabolic profile of a tumor. In tumors with a high OXPHOS burden, elevated FDX1 expression is biologically plausible and may more reliably indicate susceptibility to cuproptosis-inducing strategies [276]. However, in tumors with active steroidogenic programs, such as ACC, FDX1 expression may partly reflect steroid biosynthetic activity rather than cuproptosis sensitivity. These observations indicate that FDX1 should not be interpreted as an isolated predictive biomarker without consideration of tumor-specific biological context. Instead, FDX1 should be evaluated together with downstream markers such as DLAT and DLST, metabolic phenotype, and tissue origin.

DLAT acts downstream in the cuproptosis pathway and may therefore represent a more proximal and mechanistically informative predictor of cuproptosis susceptibility than FDX1. Pan-cancer analyses have shown that DLAT expression is positively associated with poor prognosis across multiple tumor types [277]. In GC, DLAT is upregulated and positively correlated with Th1/Th2 infiltration and the BTN2A1 immune checkpoint [278]. In PDAC and glioma, tumor cells with high DLAT expression also exhibit stronger proliferative, migratory, invasive, and drug-resistant phenotypes [279, 280].

Mitochondrial protein lipoylation represents a more direct biomarker for predicting cuproptosis, because lipoylated proteins are the immediate substrates for cuproptosis execution. In AML, a seven-gene lipoylation-cuproptosis regulatory model integrating FDX1, DLAT, PDHB, PDHA1, dihydrolipoamide dehydrogenase (DLD), LIAS, and LIPT1 can classify patients into three immune phenotype-coupled subtypes: immune-desert, immune-inflamed, and immune-excluded. The model also predicts differential responses to combined cuproptosis-immunotherapy strategies [281]. In PDAC, a four-gene cuproptosis-risk score consisting of LIPT1, LIAS, PDP1, and GCSH has been validated as an independent predictor of immune checkpoint blockade efficacy. This study further confirmed in clinical samples that LIPT1 is significantly downregulated in PDAC, and that in vitro silencing of LIPT1 promotes pancreatic cancer cell proliferation, migration, and invasion [282]. Notably, in preclinical models of LUAD, XBP1s-MGRN1-mediated degradation of LIPT1 protein has been identified as a core mechanism of acquired resistance to ES-Cu [134], possibly because LIPT1 degradation reduces cellular sensitivity to cuproptosis.

Compared with gene- and protein-level markers, tissue and circulating copper levels represent the most direct and longitudinally monitorable cuproptosis-related readouts. Several large-scale population studies have demonstrated that elevated serum copper is an independent prognostic factor in NSCLC [283], HCC [284], and BC [285], with high copper levels correlating with poorer outcomes. Notably, the Cu/Zn ratio appears to be more sensitive than copper alone. As an emerging biomarker, copper isotope fractionation (δ⁶⁵Cu) may provide independent information beyond conventional copper concentration. Multi-collector ICP-MS analyses have shown that serum δ⁶⁵Cu is significantly lower in patients with BC and OC than in healthy controls, indicating relative enrichment of ⁶³Cu in serum, whereas tumor tissues are correspondingly isotopically heavier. Mechanistically, this has been attributed to redistribution of copper into sulfur-rich metallothioneins, altered tumor lactate metabolism, and changes in copper transporter activity [286, 287]. Because changes in δ⁶⁵Cu may precede the elevation of existing tumor markers, such as CA-125, this marker has particular potential for early detection and pharmacodynamic monitoring of copper-targeted therapy. However, the current level of clinical evidence remains limited to cross-sectional studies, and prospective cohort validation is still required.

From the perspective of clinical implementation, serum copper levels and the Cu/Zn ratio may have clear advantages in predicting responses to copper-chelating agents, because tumors with greater copper overload are more likely to depend on copper availability and may therefore derive greater benefit from copper depletion. By contrast, predicting responses to cuproptosis-inducing therapies depends more heavily on the integrity of the downstream FDX1/LIPT1/DLAT machinery and requires integrated consideration of tumor metabolic status and tumor-type-specific differences.

Copper ionophores

Copper ionophores are a class of small-molecule compounds capable of traversing biological membrane barriers and mediating the directed transport of extracellular copper ions into cells. By facilitating the intracellular delivery of excess copper, frequently with preferential accumulation in mitochondria, these agents induce oxidative stress and trigger cuproptosis [288]. In addition to the well-characterized copper ionophores Elesclomol and DSF, this category also encompasses several emerging small-molecule compounds, including bis(thiosemicarbazone) analogues [289], diacetyl-bis(N4-methylthiosemicarbazone), and glyoxal-bis(N4-methylthiosemicarbazone) [290]. In this section, we discuss the antitumor activities of these copper ionophores, while also exploring the underlying mechanisms of resistance and potential strategies for therapeutic reversal (Table 3).

Table 3.

Antitumor effects of copper ionophores

Compound	Cancer type	Method	Mechanism	Ref.
Elesclomol (ES-Cu)	Uveal melanoma	In vitro	Activates Hippo pathway, promotes YAP phosphorylation, and inhibits SNAI2	[292]
	Advanced BC	Both in vitro and in vivo	Reduces FOXP3 expression; modulates FOXP3–β-catenin axis	[293]
	PDAC	Both in vitro and in vivo	Selectively eliminates CSCs and induces mitochondrial DLAT aggregation	[294]
	KRAS-mutant LC	Both in vitro and in vivo	OSTM1-mediated ATP7A degradation, leading to mitochondrial copper overload	[295]
	G3 Medulloblastoma	Both in vitro and in vivo	IDH1/c-MYC axis upregulates DLAT, increasing metabolic liability	[135]
	HCC	Both in vitro and in vivo	UCHL3 stabilizes PKM2, promoting glycolytic flux toward pyruvate production	[296]
	LC	Both in vitro and in vivo	p53/circFRMD4A axis suppresses PKM2, shifting metabolism to TCA cycle	[153]
DSF/Cu (CuET)	A549 LC	In vitro	Triggers ROS elevation, cell cycle arrest, and apoptosis	[201]
	Osteosarcoma	Both in vitro and in vivo	ROS–JNK activation and NF-κB/Nrf2 mediated antioxidant suppression	[301]
	AML	Both in vitro and in vivo	ROS–JNK activation and antioxidant suppression	[302]
	Lymphoid malignancies	Both in vitro and in vivo	ROS–JNK activation and Nrf2 inhibition	[303]
Clioquinol	BC	In vitro	Inhibits proteasome and induces apoptotic cell death	[310]
	PC	In vitro	Lysosomal damage triggered Bid cleavage mediates apoptosis	[311]
	PC	In vitro	Targets XIAP/IAPs axis to induce apoptosis	[312]

Elesclomol

Elesclomol is a prototypical copper ionophore belonging to the lipophilic bis(thiosemicarbazone) class of compounds. It forms a stable complex with Cu(II), and its antitumor mechanism has been intensively investigated over the past decades. Prior to the formal introduction and conceptualization of “cuproptosis”, Elesclomol was primarily regarded as a redox-active agent that exerts cytotoxicity by disrupting intracellular oxidative homeostasis [68]. Mechanistically, Elesclomol chelates Cu(II) in the extracellular milieu to form a 1:1 complex, which is subsequently internalized. Once inside the cell, Cu(II) is reduced to Cu(I), thereby triggering Fenton-like reactions and leading to excessive accumulation of ROS. This profound oxidative stress has been shown to activate the mitochondrial apoptotic pathway, including caspase-3 cleavage, ultimately resulting in apoptosis [291]. In vitro studies further demonstrated that the ROS scavenger N-acetylcysteine markedly attenuates the cytotoxic effects of Elesclomol, underscoring the central role of oxidative stress in its mechanism of action [68]. Beyond apoptosis induction, Elesclomol-mediated ROS accumulation has been implicated in the modulation of oncogenic signaling pathways. In uveal melanoma, elevated ROS levels selectively activate the Hippo pathway, promoting YAP phosphorylation and inhibiting its nuclear translocation. This leads to downregulation of SNAI2 expression and consequent suppression of tumor invasiveness [292]. In advanced BC, Elesclomol has also been reported to reduce FOXP3 expression, thereby restraining tumor progression through modulation of the FOXP3–β-catenin signaling axis [293]. Importantly, the pharmacological activity of Elesclomol is highly dependent on the availability of copper ions. At the systemic level, depletion of serum copper, which lowers extracellular copper levels, largely abolishes its cytotoxic effects, whereas copper supplementation markedly enhances the sensitivity of cancer cells to Elesclomol. At the cellular level, chelation of intracellular copper with glutathione similarly attenuates the toxicity of Elesclomol [268]. Collectively, these findings have positioned ES-Cu as a promising therapeutic strategy. Functionally, ES-Cu has been shown to reduce mitochondrial spare respiratory capacity without significantly affecting basal respiration or ATP production, suggesting a selective inhibition of TCA cycle–associated enzymes rather than an immediate disruption of the electron transport chain [270].

Following the identification of the cuproptosis pathway, preclinical investigations of Elesclomol as an anticancer agent have regained considerable momentum. Its central mechanism of action is now understood to involve the elevation of intracellular copper levels, thereby triggering cuproptosis. Notably, conclusions derived from studies on cuproptosis do not contradict the conventional paradigm of oxidative stress–mediated cytotoxicity. Indeed, oxidative stress appears to be a downstream consequence of ES-Cu–induced cuproptosis in tumor cells [6, 49]. PDAC is a solid malignancy enriched in CSCs, and it has been shown to be particularly susceptible to copper-dependent targeting strategies. Yu et al. reported that ES-Cu increases intracellular copper concentrations in PDAC cells by approximately 2–4-fold, selectively eliminating CSCs and inducing aggregation of DLAT within mitochondria, which is a characteristic molecular hallmark of cuproptosis [294]. In KRAS-mutant LC, Zhao et al. demonstrated that ES-Cu upregulates the E3 ubiquitin ligase OSTM1, thereby promoting intracellular copper accumulation and activating cuproptosis. Mechanistically, OSTM1 interacts with the IRGY motif of ATP7A, facilitating its ubiquitin-mediated degradation. This process disrupts copper delivery to Golgi-resident cuproenzymes and impairs subsequent copper efflux, ultimately leading to aberrant copper accumulation within mitochondria and other subcellular compartments [295]. The highly aggressive Group 3 (G3) medulloblastoma subtype exhibits marked sensitivity to ES-Cu–induced cuproptosis. This vulnerability is attributable to elevated c-MYC expression in G3 tumors. Through the IDH1/c-MYC signaling axis, c-MYC upregulates DLAT expression, creating a metabolic liability that substantially enhances the cytotoxic efficacy of ES-Cu [135]. In HCC, the deubiquitinase UCHL3 mediates the deubiquitination of pyruvate kinase M2 (PKM2), thereby enhancing its protein stability and enzymatic activity in cancer cells. This modification promotes glycolytic flux toward pyruvate production, ultimately potentiating the effect of ES-Cu [296]. Similarly, p53 transcriptionally activates circFRMD4A expression, which in turn suppresses PKM2 activity. This metabolic reprogramming shifts tumor cell metabolism toward enhanced TCA cycle activity, thereby augmenting the cuproptotic response elicited by ES-Cu [153].

Although ES-Cu has demonstrated considerable antitumor potential, its clinical efficacy remains constrained by multiple resistance mechanisms. Tumor cells can acquire resistance to ES-Cu through complex molecular adaptations. Therefore, elucidating the underlying mechanisms and developing reversal strategies are critical for improving therapeutic outcomes. Emerging evidence indicates that prolonged exposure to ES-Cu activates HIF-1α and Wnt/β-catenin signaling pathways, resulting in the enrichment of CD44⁺/CD133⁺ stem-like tumor cell populations and the eventual development of therapeutic resistance [136, 142]. In PDAC, ES-Cu has also been shown to stabilize the transcription factor Nrf2, thereby upregulating the expression of the rate-limiting enzymes involved in GSH synthesis, namely glutamate–cysteine ligase modifier subunit (GCLM) and glutamate–cysteine ligase catalytic subunit (GCLC). Elevated intracellular GSH levels can chelate copper ions and consequently suppress ES-Cu–mediated cuproptosis [268]. Furthermore, in LUAD, activation of the ER UPR leads to the accumulation of spliced X-box binding protein 1 (XBP1s), which promotes super-enhancer formation and enhances the interaction between MGRN1 and its target promoters. MGRN1 subsequently facilitates the ubiquitin-mediated degradation of LIPT1, a key regulator of cuproptosis, thereby attenuating ES-Cu–induced copper-dependent cell death [134]. Notably, degradation of LIPT1 suppresses TCA cycle activity. To compensate for energy demands, LUAD cells undergo XBP1s-driven glycolytic reprogramming to sustain metabolic homeostasis [134].

To overcome resistance to Elesclomol, several strategies have been explored. In HIF-1α–mediated resistance models, co-administration of ES-Cu with the HIF-1α inhibitor PX-478 enabled low-dose ES-Cu to achieve antitumor efficacy comparable to high-dose monotherapy, while significantly reducing hepatotoxicity [136]. In addition, intricate molecular crosstalk exists between ferroptosis and cuproptosis. Wang et al. reported that ferroptosis inducers such as sorafenib and erastin enhance the susceptibility of HCC cells to cuproptosis, and their combination with ES-Cu exerts synergistic antitumor effects [61]. In castration-resistant prostate cancer (CRPC), enzalutamide increases tumor cell dependence on mitochondrial metabolism and elevates the expression of lipoylated proteins. Given the strong reliance of cuproptosis on mitochondrial metabolic activity, the combination of enzalutamide and ES-Cu results in markedly enhanced copper-dependent cytotoxicity in both in vitro and in vivo models [297].

DSF

DSF is a classical anti-alcoholism drug approved by the United States Food and Drug Administration (FDA). Unlike Elesclomol, DSF is not a canonical metal ionophore, but rather a small-molecule compound containing a dithiocarbamate moiety. Upon administration, DSF is rapidly metabolized in vivo to diethyldithiocarbamate (DDC), a metabolite with strong metal-chelating capacity. DDC readily coordinates with Cu(II) to form stable complexes and can be further oxidized to generate bis(diethyldithiocarbamate)-copper (CuET). In the context of copper-dependent anticancer activity, DSF alone exhibits relatively limited cytotoxicity across multiple tumor types, whereas CuET is widely considered the principal bioactive species mediating its antitumor effects [298–300]. Mechanistically, CuET is a lipophilic complex that facilitates transmembrane copper transport and increases intracellular copper burden, thereby profoundly disrupting copper homeostasis and creating a permissive environment for subsequent mitochondrial dysfunction and the induction of cuproptosis [201].

DSF/Cu (CuET) exerts anticancer effects by perturbing endogenous antioxidant systems and inducing oxidative stress. In A549 LC cells, the complex readily penetrates the plasma membrane and accumulates intracellularly, triggering a marked elevation in ROS, which leads to cell cycle arrest and apoptosis [201]. Similar observations have been reported in osteosarcoma [301], AML [302], and lymphoid malignancies [303], where CuET activates the ROS–JNK signaling axis while suppressing key antioxidant and inflammatory regulators such as Nrf2 and NF-κB, and partially depletes cancer stem-like cell populations. Although tumor cells may counteract oxidative stress through compensatory upregulation of Nrf2, pharmacological inhibition of Nrf2 markedly enhances CuET sensitivity and synergizes with sorafenib to suppress tumor growth [303]. The excessive oxidative stress induced by CuET is likely linked to the activation of multiple cell death modalities, including apoptosis, ferroptosis, and ICD [304–307]. At the level of proteostasis, the work by Skrott et al. provided an additional mechanistic framework for the antitumor activity of CuET. They demonstrated that CuET directly binds to NPL4 in cells and induces its aggregation, thereby disrupting the functional interaction between NPL4 and the segregase p97/VCP. As a consequence, p97/VCP fails to efficiently process ubiquitinated substrates, leading to the accumulation of polyubiquitinated and misfolded proteins, the induction of proteotoxic stress, and ultimately cell death [259]. Notably, recent studies have further shown that CuET promotes cuproptosis-associated phenotypes in HCC and OC models, including the depletion of iron–sulfur cluster proteins. The subsequent analysis underscores the copper dependency of DSF-mediated antitumor activity, as neither DSF nor copper alone produces robust antitumor effects [308, 309].

Notably, in addition to Elesclomol and DSF, several other copper ionophores have been reported to exert antitumor activity, including 8-hydroxyquinoline (8-HQ) and its derivatives such as 7-iodo-5-chloro-8-hydroxyquinoline (CQ). Current studies have not systematically examined their capacity to induce cuproptosis. Instead, the anticancer effects of these compounds have largely been attributed to mechanisms such as proteasome inhibition [310], or activation of apoptotic pathways [311, 312]. Nevertheless, given their pronounced ability to remodel intracellular copper homeostasis, these ionophores may possess substantial potential to trigger cuproptosis. Future studies based on rigorous preclinical models are urgently needed to clarify the precise role of these copper ionophores in cuproptosis induction and to evaluate their translational potential in cancer therapy.

Clinical translation of copper ionophores

Clinical translation of copper ionophores has primarily focused on evaluating their therapeutic efficacy in combination with chemotherapy or other treatment modalities. However, despite robust preclinical activity, the clinical translation of copper ionophores has thus far yielded inconsistent outcomes.

A phase II clinical trial (NCT00084214, n = 80) demonstrated that weekly Elesclomol plus paclitaxel was superior to paclitaxel monotherapy in patients with advanced melanoma. Specifically, the combination therapy doubled the median progression-free survival (PFS) compared with monotherapy, with a favorable trend in overall survival (OS) [264]. This prompted the subsequent double-blind phase III SYMMETRY trial (NCT00522834, n = 651). However, SYMMETRY failed to confirm this benefit, showing no significant improvement in PFS (HR 0.89, P = 0.23), and was terminated early because of excess deaths in the combination arm, predominantly among patients with elevated baseline lactate dehydrogenase (LDH) levels [265].

Clinical studies of DSF/Cu have similarly provided important translational evidence. In newly diagnosed GBM, the phase I/II trial (NCT02715609, n = 33) established a maximum tolerated dose of DSF at 375 mg/day and a recommended phase II dose of 250 mg/day in combination with radiotherapy and temozolomide (TMZ), with dose-limiting toxicity probabilities of 10% (95% CI, 3–29%) and 21% (95% CI, 7–42%), respectively [261]. However, after a median follow-up of 26 months, no significant differences in OS or PFS were observed between IDH-mutant and NF1-mutant cohorts, and the only signal of benefit was observed in three patients with BRAF mutations who experienced prolonged remission. These findings suggest that DSF/Cu confers limited clinical benefit in most patients, although BRAF-mutant GBM may warrant further investigation [261]. In the recurrent setting, the open-label single-arm phase II trial (NCT03034135, n = 23, 21 evaluable) in TMZ-resistant GBM reported an objective response rate of 0% and a clinical benefit rate of only 14%, with a median PFS of 1.7 months and a median OS of 7.1 months, although tolerability was acceptable, as reflected by a low rate of dose-limiting toxicity (4%) [262]. More importantly, the randomized open-label phase II/III DIRECT trial (NCT02678975, n = 88), evaluating DSF/Cu in combination with alkylating chemotherapy in recurrent GBM, failed to meet its primary endpoint (6-month survival: 44% vs 62% for chemotherapy alone), and demonstrated numerically inferior median OS (5.5 vs 8.2 months) and PFS (2.3 vs 2.6 months). The combination was associated with significantly higher rates of grade ≥ 3 adverse events (34% vs 11%, P = 0.02) and serious adverse events (41% vs 16%, P = 0.02), with 24% of patients discontinuing DSF due to toxicity. Collectively, these findings indicate that DSF/Cu should not be recommended for recurrent GBM outside clinical trials [263].

Notably, clinical trials of DSF/Cu have focused predominantly on GBM, reflecting the intrinsic pharmacokinetic properties of DSF. DSF has favorable lipophilicity that allows effective penetration of the blood–brain barrier, giving it a pharmacokinetic advantage in GBM and other central nervous system malignancies [313]. In other solid tumors, a phase I clinical trial (NCT05210374) is currently recruiting patients to evaluate the safety of a triple-combination regimen comprising DSF, copper gluconate, and liposomal doxorubicin in patients with recurrent or treatment-refractory sarcomas (Table 4).

Table 4.

Clinical trials for copper ionophores

NCT number	Phases and status	Conditions	Regimen	Primary endpoint	Result	Ref.
NCT00084214	II, n = 80; Completed	Advanced melanoma	Elesclomol and Paclitaxel vs paclitaxel	PFS	Met (median PFS 112 vs 56 days, HR 0.583, P = 0.035)	[264]
NCT00522834	III, n = 651; Terminated	Advanced melanoma	Elesclomol and Paclitaxel vs paclitaxel	PFS	Not met (HR 0.89, P = 0.23); terminated early	[265]
NCT02715609	I/II, n = 33; Completed	Newly diagnosed GBM	DSF/Cu and TMZ	Safety/Maximum tolerated dose	Maximum tolerated dose 375 mg/d; Recommended phase 2 Dose 250 mg/d	[261]
NCT03034135	II, n = 21; Completed	TMZ-resistant GBM	DSF/Cu and TMZ	ORR	ORR 0%; median PFS 1.7 months; median OS 7.1 months	[262]
NCT02678975	II/III, n = 88; Completed	Recurrent GBM	DSF/Cu and Alkylating chemotherapy vs Alkylating chemotherapy	6-month survival	Not met (44% vs 62%, P = 0.10); median OS 5.5 vs 8.2 months	[263]
NCT05210374	I, Recruiting	Recurrent or refractory sarcomas	DSF/Cu and Liposomal doxorubicin	Safety	Recruiting	NA

Limitations of copper ionophores

Although preclinical results have been encouraging, the translational potential of copper ionophores, as clinical trials have shown, remains limited due to interlinked issues of strategy-specific toxicity, resistance, and the patient-selection dilemma. With respect to dose-limiting toxicity, the DIRECT trial documented a grade ≥ 3 adverse event rate of 34% with DSF/Cu plus alkylating chemotherapy in recurrent GBM, with 24% of patients discontinuing DSF on account of hepatotoxicity and peripheral neuropathy [263]. In addition, the broader experience across off-label oncology applications of DSF has identified poor oral bioavailability, rapid hepatic conversion to inactive metabolites, and a narrow therapeutic window as recurrent barriers to achieving effective intratumoral CuET concentrations without prohibitive systemic exposure [314]. Recent formulation efforts have begun to address these pharmacokinetic deficits through bioavailability-enhanced nano-self-emulsifying systems and implantable controlled-release platforms, although none has yet reached clinical use [315, 316]. Elesclomol illustrates a related but distinct translational pattern. Although the phase II melanoma trial supported its combination with paclitaxel, the subsequent phase III SYMMETRY trial was terminated early because patients with elevated baseline LDH derived no benefit and exhibited excess mortality [265], illustrating how an inadequately prespecified biomarker can convert a mechanistically rational therapy into a futile or unsafe one in unselected populations. Resistance mechanisms compound these limitations. Tumors evade ionophore-induced cuproptosis through GSH-mediated copper sequestration driven by NFE2L2/Nrf2 activation [268], HIF-1α–driven enrichment of stem-like populations, AKT1-mediated FDX1 phosphorylation [155], and broader metabolic rewiring toward glycolysis that desensitizes cells to copper-induced mitochondrial proteotoxic stress [317]. Cuproptosis can also be co-opted as a downstream effector of resistance to non-copper therapies, as exemplified by DLD-mediated rewiring of cuproptosis in endocrine-resistant ER-positive BC [318]. Finally, prospective patient-selection strategies remain the most clinically consequential gap, and the SYMMETRY experience exemplifies the cost of advancing to phase III without a prespecified enrichment biomarker. Beyond LDH, several biologically rational stratifiers, including mitochondrial OXPHOS dependence, FDX1/DLAT/LIPT1 expression, and the lipoylation status of TCA-related enzymes, have been nominated through retrospective cuproptosis-related gene-signature analyses [319]. The tentative subgroup signals such as the prolonged remission observed in BRAF-mutant GBM await confirmation in molecularly stratified cohorts [261]. Therefore, patient-selection strategies that integrate predictive biomarkers such as the cuproptosis axis and metabolic status should be established before clinical trial initiation to improve safety and reliability of outcomes.

Copper chelators

Copper chelators are another important class of anticancer agents. Distinct from the mechanism of copper ionophores, copper chelators exert their effects by directly sequestering intracellular copper ions, thereby suppressing copper-dependent tumor progression. A variety of copper chelators have been investigated to date, including tetrathiomolybdate (TTM), trientine (triethylenetetramine, TETA), and D-penicillamine (D-PA). Notably, the clinical application of D-PA has been largely restricted to the treatment of Wilson’s disease [320], whereas both TTM and TETA have advanced into clinical trials for cancer therapy [321]. This section focuses on recent preclinical advances and translational progress of TTM and TETA in oncology (Table 5).

Table 5.

The antitumor effect and clinical translation of copper chelators

Compound	Cancer type	Method	Mechanism	NCT number	Ref.
TTM	TNBC	In vivo	Targets SOX2/OCT4+ subpopulation; depletes copper to impair OXPHOS and activates AMPK while antagonizing mTORC1 signaling	NA	[31]
	BC (Lung metastasis)	In vivo	Remodels pre-metastatic niche by attenuating pulmonary collagen deposition and augmenting CD4+ T cell infiltration	NA	[322]
	TNBC	Both in vitro and in vivo	Suppresses cuproplasia via the STEAP3/CDK16/JAK1 axis	NA	[8]
	Luminal-A BC	Both in vitro and in vivo	Disrupts copper signaling by modulating Na/K-ATPase subunit ATP1B1, induces apoptosis, and impairs invasive growth	NA	[323]
	Solid tumors	Both in vitro and in vivo	Inhibits copper-dependent tumor angiogenesis and exhibits synergistic effects with lenvatinib	NA	[324]
	Advanced solid tumors	Clinical trial	Phase I (n = 18) safety validation; reduces serum copper to 20% of baseline	Not Found	[269]
	Advanced solid tumors	Clinical trial	Phase I (n = 18) safety validation: starting dose of 300 mg/day for phase II trials	Not Found	[326]
	BC	Clinical trial	Phase II (n = 16); reduces intracellular copper and circulating EPC counts to promote tumor dormancy	NCT00195091	[266, 267]
	Advanced Cancer	Clinical trial	Phase I/II (recruiting); evaluates safety and efficacy in combination with capecitabine and immunotherapy	NCT06134375	NA
TETA	HER2-negative BC	Both in vitro and in vivo	Long-term exposure suppresses migratory potential and invasiveness via sustained AKT activation and STAT3 signaling inhibition	NA	[98]

The antitumor mechanisms and clinical translation of copper chelators

TTM is a molybdenum(VI) sulfide complex composed of tetrahedral tetrathiomolybdate anions and ammonium ions. Its pronounced thiophilicity and soft coordination environment endow it with exceptional copper-chelating capacity. In the context of TNBC, Ramchandani et al. identified a highly metastatic SOX2/OCT4 + tumor cell subpopulation. Oral administration of TTM effectively targets this subset by depleting intracellular copper, which triggers mitochondrial cristae disorganization and impairs complex IV function, thereby suppressing OXPHOS. This metabolic disruption—characterized by ATP depletion and resultant AMP accumulation—activates the AMPK signaling pathway while antagonizing mTORC1, ultimately suppressing TNBC metastasis [31]. Furthermore, TTM facilitates the remodeling of the pre-metastatic niche. In vivo experiments have demonstrated that TTM treatment significantly attenuates pulmonary collagen deposition and augments CD4⁺ T cell infiltration, thereby hindering the colonization of circulating BC cells in the lung [322]. Regarding oncogenic proliferation, copper ions drive TNBC progression via the STEAP3/CDK16/JAK1 axis, whereas TTM has been shown to robustly suppress this cuproplasia in both in vitro and in vivo models [8]. In Luminal-A BC, Hancock et al. identified COMMD3 as a pivotal gene suppressing aggressive proliferation. Its deficiency disrupts copper signaling by modulating the Na/K-ATPase subunit ATP1B1, which enhances cellular dependence on copper homeostasis. Consequently, TTM-mediated copper chelation induces apoptosis and impairs the invasive growth of COMMD3-deficient cells [323]. Notably, TTM also inhibits copper-dependent tumor angiogenesis, exhibiting a potent synergistic effect when combined with lenvatinib [324]. TETA is a tetradentate ligand whose four nitrogen atoms can chelate Cu(II) ions through coordination interactions to form stable five-membered chelate rings [325]. In HER2-negative BC, studies have shown that short-term and long-term exposure to TETA exert differential effects on tumor invasiveness. Specifically, short-term exposure enhances the invasive capacity of BC cells, whereas long-term exposure suppresses their migratory potential. These effects may be associated with the distinct consequences of transient versus sustained activation of AKT kinase, as well as the inhibition of STAT3 signaling [98].

Clinically, the safety profile of TTM was assessed in a phase I clinical trial (n = 18). The study employed three dose escalation levels of 90, 105, and 120 mg/day, and when serum copper levels decreased to 20% of baseline, no additional toxicities were observed across dose cohorts, with five of six patients achieving stable disease [269]. Another phase I clinical trial (n = 18) in patients with advanced solid tumors similarly supported the favorable tolerability of TTM and recommended a starting dose of 300 mg/day for phase II trials [326]. A phase II clinical trial (NCT00195091, n = 16) showed that TTM significantly reduced intracellular copper in BC patients, with the effect particularly pronounced in TNBC. In addition, TTM treatment reduced circulating endothelial progenitor cell (EPC) counts, thereby promoting tumor dormancy and preventing recurrence [266, 267]. An ongoing phase I/II trial (NCT06134375) is evaluating the safety and efficacy of TTM combined with capecitabine and immunotherapy.

Limitations of copper chelators

Although copper chelators have shown a more favorable safety record than ionophores in early-phase oncology trials, they also face obstacles in clinical translation. Sustained systemic copper depletion can cause the dose-limiting toxicities of copper deficiency, including macrocytic anemia, neutropenia, and progressive myeloneuropathy [327]. All of these systemic toxicities constrain prolonged dose escalation of TTM and TETA and necessitate strict serum-copper or ceruloplasmin monitoring during chronic administration. From a resistance standpoint, the antitumor activity of copper chelation hinges on whether tumor cells remain copper-dependent, yet several copper-handling adaptations can blunt this dependence, such as mobilization of metallothionein-buffered intracellular copper stores and copper redistribution among the cytosolic and mitochondrial copper chaperones ATOX1, CCS, and COX17 [42]. Together, these mechanisms can sustain proliferation and copper-dependent angiogenic signaling despite systemic copper depletion [317]. The patient-selection challenge is equally pronounced. Serum copper is an imperfect pharmacodynamic surrogate because it primarily reflects ceruloplasmin-bound copper rather than the labile pool that drives tumor signaling, and no validated predictive biomarker has been clinically qualified. Exploratory work has nominated intratumoral ATOX1 expression as a candidate that may identify patients more likely to benefit from TTM in high-risk BC [328], but such markers require prospective validation in adequately powered trials before they can guide patient stratification.

Nanoplatforms inducing cuproptosis for tumor therapy

Recent advances in nanotechnology have provided engineered platforms for the precise modulation of copper metabolism. Nanoregulators, by virtue of their tunable physicochemical properties, enable tumor-targeted delivery and microenvironment-responsive release of copper ions, thereby achieving localized remodeling of copper homeostasis within the tumor [329]. Elevation of intracellular copper ion levels can induce cuproptosis, thereby exerting antitumor effects. Mechanistically, there are two major categories of cuproptosis-inducing therapeutic strategies (Table 6): (1) disrupting copper homeostasis through nanoplatforms to induce copper overload; and (2) enhancing cuproptosis-mediated tumor-killing effects by concomitantly inducing other forms of RCD.

Table 6.

Nanoplatforms inducing cuproptosis for tumor therapy

Nanoplatform	Cancer type	Key physicochemical feature	Effect	Ref.
ES-Cu/DNAzyme/ZnO	GC	ZnO as Zn(II) reservoir activates ATP7B DNAzyme; concurrent GSH depletion	ATP7B silencing suppresses Cu efflux; cuproptosis induction	[330]
MC@BSA	Not specified	BSA template confers colloidal stability and tumor accumulation	Cu efflux inhibition; PKM2/HIF-1α/DLAT axis downregulation	[331]
Cu-RNP	Not specified	CRISPR-Cas9 RNP co-delivery with copper	Concurrent BMAL1 silencing and copper overload	[332, 333]
RNP@Cu2O@SPF	CRC	Dual pH/GSH-responsive organosilica shell	Tumor-confined Cu release; ferroptosis and cuproptosis	[334]
CuO2@G5-BS/TF	Not specified	BS-PAMAM surface chemistry targets CA IX; self-supplied H2O2	MRI-guided ferroptosis and cuproptosis	[335]
Cu1.8S/MIL-88B	Not specified	Surface charge engineering for mitochondrial targeting; Fe co-supply	Mitochondria-directed cuproptosis with ferroptosis	[336]
Fe/Cu-MOF (MetaCell)	Not specified	Porous coordination architecture; thermosensitive liposome in live neutrophils	Tumor-homing ferroptosis and cuproptosis	[338]
Cu2(PO4)(OH) NPs	CRC	H2S-triggered phase transformation reduces particle size	Enhanced uptake; pyroptosis and cuproptosis	[339]
Hollow Ca/Cu platform (D@HCC-CuTH)	Not specified	Hollow CaCO3 scaffold enables DSF loading and Ca(II) release	Mitochondrial Ca overload; cuproptosis, paraptosis, apoptosis	[340]

Disrupting copper homeostasis to drive copper overload

According to the contextual copper signaling network discussed above, the most direct strategy to induce cuproptosis is to elevate the intracellular labile copper pool beyond the proteotoxic threshold, which necessitates nanoplatforms endowed with design features capable of regulating copper delivery, promoting copper retention, and dismantling metabolic buffering systems. The nano–nucleic acid platform reported by Zhang et al. is characterized by the functional synergy between ZnO and ES-Cu. Specifically, ZnO is engineered as a zinc ion reservoir, whose released Zn(II) activates the catalytic core of an ATP7B-targeting DNAzyme, thereby silencing ATP7B and suppressing copper efflux. Meanwhile, ZnO depletes intracellular GSH, preserving the bioactivity of the delivered ES-Cu, ultimately inducing cuproptosis in GC cells [330]. In the manganese–copper bovine serum albumin nanocomposite MC@BSA, the decisive physicochemical feature is the albumin template itself, which confers colloidal stability and tumor-accumulating capability on the nanoplatform, thereby enabling efficient delivery of the Mn–Cu core to tumor sites. Subsequently, the released metal ions not only inhibit copper efflux but also downregulate the PKM2/HIF-1α/DLAT axis, thereby enhancing cuproptosis through increased copper burden [331]. The copper-loaded CRISPR-Cas9 ribonucleoprotein carrier (Cu-RNP) exemplifies the capability of nanoplatforms to integrate gene-editing functionality. Its structural design enables the co-delivery of a BMAL1-targeting ribonucleoprotein, such that circadian rhythm disruption and copper overload are simultaneously induced by a single nanoparticle, thereby potentiating the antitumor cytotoxic effect [332, 333].

Inducing cuproptosis alongside other regulated cell death

Because cuproptosis, ferroptosis, pyroptosis, and paraptosis are mechanistically associated with mitochondrial function and metal homeostasis, a class of nanoplatforms has been designed to activate two intrinsic cell death programs through a single material. The central design question is what physicochemical properties allow the particles to drive both death programs concurrently. In the folate-modified, organosilica-coated CRISPR-Cas9 ribonucleoprotein nanomedicine RNP@Cu₂O@SPF developed by Wu et al., the core design feature is the dual environmental responsiveness of the coating layer. This design enables the organosilica shell and Cu₂O core to undergo specific degradation within the acidic and glutathione-enriched TME, thereby confining copper release to the tumor region. The released Cu(I) can not only drive Fenton-like reactions but also be partially oxidized to Cu(II), accompanied by GSH depletion, thereby sequentially triggering ferroptosis and cuproptosis. This process represents a cascade reaction jointly driven by pH- and redox-responsive mechanisms [334]. The dendrimer–metal–phenolic nanonetwork CuO₂@G5-BS/TF highlights another structure-dependent design principle, namely molecular targeting conferred by surface chemistry. In this system, CuO₂ is loaded onto 4-carboxybenzenesulfonamide (BS)-modified polyamidoamine (PAMAM) dendrimers and crosslinked with iron ions and tannic acid (TA). Since BS functions as a carbonic anhydrase IX (CA IX) inhibitor, it mediates the binding of CuO₂@G5-BS/TF to CA IX, thereby promoting targeted accumulation within tumor cells. The subsequent decomposition of CuO₂ self-supplies hydrogen peroxide, which continuously supports ferroptosis and cuproptosis under magnetic resonance guidance [335]. In the platform constructed by Liu et al., the decisive feature is subcellular targeting achieved through surface charge engineering. Functionalized Cu_1.8S nanodots can be selectively delivered to mitochondria. Given that cuproptosis strictly depends on copper accumulation within mitochondria, this targeting design is mechanistically indispensable. Meanwhile, anchoring the nanodots onto the iron-based framework MIL-88B provides an additional iron supply, thereby synergistically inducing ferroptosis [336]. Copper-based metal–organic frameworks (Cu-MOFs) are structurally well suited to fulfill this dual role, as their porous coordination architecture enables them to serve simultaneously as nanocarriers and copper donors [337]. In the bimetallic Fe/Cu-MOF, a thermosensitive liposomal membrane was further internalized into live neutrophils. This design makes drug release temperature-responsive, while biodistribution depends on the intrinsic tumor-homing property of the cellular carrier, thereby jointly achieving the synergistic activation of ferroptosis and cuproptosis at the tumor site [338].

Pyroptosis can also be synergistically activated through similar mechanisms. Studies on Cu₂(PO₄)(OH) nanoparticles (NPs) in CRC have shown that stimulus-triggered physical morphological transformation itself can function as a key mechanism. In the H₂S-enriched TME, these NPs undergo in situ transformation into smaller CuS particles, and the reduced particle size enhances cellular uptake efficiency. Subsequently, the released Cu(II) can not only cleave gasdermin D through Fenton reactions to trigger pyroptosis but also downregulate ATP7A to enhance copper loading [339]. In the hollow calcium–copper bimetallic platform, the key feature is the hollow calcium carbonate scaffold. This structure not only provides loading space for DSF but also decomposes under acidic conditions to release calcium ions, inducing mitochondrial calcium overload. Consequently, copper-induced DLAT oligomerization and calcium-induced ER stress converge within the same system, synergistically executing cuproptosis, paraptosis, and apoptosis [340].

Collectively, whether through copper overload induction or concurrent induction of other forms of RCD, cuproptosis-inducing nanocarriers exhibit coherent physicochemical design principles. The first principle is stimulus-responsive release. By recognizing TME signals such as acidic pH, elevated glutathione levels, or H₂S enrichment, these platforms can spatiotemporally regulate copper release [334, 335, 339]. The second principle is that particle size and surface chemical properties dominate subcellular fate. Since cuproptosis strictly depends on the delivery of copper to mitochondria, small-sized particles that ensure cellular uptake, together with specific surface charges or ligands that enable mitochondrial or tumor-cell targeting, constitute a potential design criterion. The in situ size reduction of Cu₂(PO₄)(OH) and mitochondria-directed Cu_1.8S nanodots exemplify this principle [336, 339]. Another design principle is the catalytic advantage of bimetallic components. Copper–iron bimetallic platforms can simultaneously drive ferroptosis and cuproptosis because, compared with monometallic systems, bimetallic systems exhibit stronger hydrogen peroxide adsorption capacity and can reduce the energy barrier of peroxidase-like reactions. This bimetallic enhancement of catalytic activity can be interpreted and predicted through density functional theory calculations [336, 338]. Therefore, the unifying core principle is that the efficacy of cuproptosis-inducing nanoplatforms is not achieved merely by their copper-loading function, but is also determined by their structural configuration and physicochemical properties.

Limitations of copper-targeted nanoplatforms

The clinical translation of cuproptosis-inducing nanoplatforms confronts a distinct set of constraints arising from their fate as engineered constructs in vivo. In terms of in vivo metabolism, adsorption of serum proteins forms a corona on circulating particles that alters their hydrodynamic size, surface chemistry, and targeting fidelity, while stimulus-responsive coatings may degrade and release copper prematurely outside the tumor compartment [341]. Toxicity arises chiefly from off-target accumulation of copper-loaded particles in normal organs, with hepatic enzyme elevation and pulmonary oxidative stress from non-specific Cu(I)/Cu(II) release as the principal concerns [342]. These agents are primarily eliminated via sequestration by Kupffer cells and splenic macrophages within the reticuloendothelial system. Such biological retention traps inorganic copper cargo for prolonged periods and interferes with heterogeneous tumor targeting mediated by the EPR effect in human malignancies [343, 344]. In terms of therapeutic resistance, multiple intratumoral obstacles can suppress cuproptosis even when NPs successfully reach tumor sites. Abundant GSH reserves and hypoxic TME neutralize released copper ions, while compact ECM and high interstitial fluid pressure hinder their deep penetration into tumor tissues [344]. Difficulties in patient screening for these nanotherapies are highly consistent with those seen in copper ionophores. To date, no phase I clinical trial targeting cuproptosis via nanoplatforms has been finished, and no validated companion diagnostic biomarkers have been established either.

Copper-targeted strategies in combination with other therapeutic strategies

As a monotherapy, cuproptosis induction or copper overload alone may encounter limitations such as therapeutic resistance and insufficient tumor eradication, underscoring the necessity for rational combination strategies. Emerging preclinical and clinical evidence has indicated that integrating copper-based therapy strategies with established therapeutic modalities can achieve synergistic antitumor effects through complementary mechanisms of action, thereby enhancing overall treatment efficacy. This section delineates recent advances in cuproptosis-based combination strategies including chemotherapy, radiotherapy, immunotherapy, and metabolic therapy, and other novel therapy strategies.

Copper-targeted strategies in combination with chemotherapy

Chemotherapy, which employs cytotoxic agents to eliminate tumor cells and suppress tumor progression and metastasis by disrupting cell proliferation, division, or DNA integrity, is a cornerstone of systemic antitumor therapy and remains widely used in multimodal cancer treatment [345]. However, its clinical efficacy is frequently constrained by two principal limitations: (1) the inherently restricted cytotoxic potency of single-agent chemotherapy, which is insufficient to eradicate highly heterogeneous tumor cell populations; (2) the propensity of tumor cells to develop acquired drug resistance under therapeutic pressure, ultimately leading to treatment failure and disease progression [346]. The combination of copper-targeted therapeutic strategies with chemotherapy offers a promising approach to address both of these challenges, grounded in the pivotal role of intracellular copper homeostasis in regulating tumor cell survival, modulating chemotherapeutic responses, and driving resistance mechanisms. Copper ionophores such as DSF and Elesclomol, as well as copper-containing nanoregulators, can independently induce tumor cell death through cuproptosis, oxidative stress, and proteasome inhibition, thereby complementing the mechanisms of conventional chemotherapeutics and generating synergistic antitumor effects [6]. In addition, aberrant copper transport pathways have been identified across multiple solid tumor types, where they impair the intracellular uptake and distribution of chemotherapeutic agents and thus mediate drug resistance. A representative example is CTR1, which mediates both the uptake of extracellular copper ions and the cellular internalization of Pt-based drugs. Accordingly, it has been demonstrated in many cancer types that downregulation of CTR1 expression or disruption of its membrane localization is closely associated with Pt resistance [347–349]. These findings collectively suggest that modulation of copper homeostasis may restore chemosensitivity in resistant tumors. In this section, we systematically review the preclinical and clinical progress of these combination strategies from two perspectives: synergistic antitumor effects and reversal of chemoresistance. These strategies include the conventional combined use of copper ionophores or chelators with chemotherapeutic drugs, as well as nanoplatforms that co-deliver copper-related agents and chemotherapy payloads (Table 7).

Table 7.

Copper-targeted strategies in combination with chemotherapy

Combination strategy	Cancer type	Effect	Mechanism	Ref.
Elesclomol and TMZ	GBM	Synergistic efficacy	Cuproptosis and DNA damage synergy	[69]
Cs-Gn-Cu and DOX	NSCLC	Augmented activity	ROS generation and oxidative DNA damage	[350]
DOX, CPT, and Copper (NPs)	TNBC	Synergistic efficacy	Dual apoptosis and cuproptosis induction	[351]
Carfilzomib and CuET	GBM	Prolonged survival and reduced resistance	Proteostasis disruption and cuproptosis induction	[352]
DDP-loaded Cu-MOFs	CRC	Enhanced cytotoxicity and stemness suppression	Chemotherapy combined with cuproptosis and stemness inhibition	[353]
CuDT (CuO2-DOX) in HA	BC	Promoted killing	Oxidative stress and cuproptosis	[354]
Copper chelator and PTX	TNBC	Reduced toxicity	Copper depletion and PTX cytotoxicity	[355]
Baicalein and ES-Cu	Cervical Cancer	Reverse DDP resistance	AKT-mediated enhancement of cuproptosis	[356]
ES-Cu and DTX	DTX resistant PC	Enhance DTX sensitivity	DLAT/mTOR inhibition and G2/M arrest	[73]
Elesclomol and CuS NPs	PDAC	Reverse GEM resistance	Targeting CSCs via cuproptosis	[294]
DOX, Hydralazine, and DSF	BC	Reverse DOX resistance	Efflux inhibition and CSC depletion	[357]
DSF and DTX	BC	Reverse DTX resistance	Induced autophagic cell death	[358]
DSF and DDP	NSCLC	Enhance DDP sensitivity	Formation of potent Pt(DDTC)3+ complex	[359]
DSF, DDP, and ALDH inhibitor	AT/RT	Promoted efficacy	ALDH inhibition and synergistic toxicity	[360]
DSF and DTX ([DD]NpH-T)	DTX resistant BC	Reverse DTX resistance	pH-responsive release and deep penetration	[361]
HFn, Regorafenib, and Cu(II)	GBM	Enhanced bioavailability	TfR1-targeted blood–brain barrier penetration	[362]
bCCM (Cu-nanocarrier) and Pt	Pt resistant HCC	Reverse Pt resistance	GSH depletion and ATP7B downregulation	[363]
PTX-DSF co-crystal and Cyt C	MDR Tumors	Reverse PTX resistance	Co-crystal delivery to MDR cells	[364]

The synergistic antitumor effects achieved by combining copper-targeted strategies with chemotherapy are fundamentally rooted in the additive and complementary cytotoxic effects of the two modalities. In addition to the tumor-killing activity exerted by conventional chemotherapeutic agents, copper-based therapeutics contribute to tumor cell elimination by inducing cuproptosis, thereby enhancing overall antitumor efficacy. At the level of direct drug combination, Buccarelli et al. demonstrated in a GBM model that the co-administration of Elesclomol and TMZ potentiated cytotoxicity against both GBM cells and GBM stem cells, leveraging Elesclomol-induced cuproptosis in concert with TMZ-mediated DNA damage [69]. Building on this design, nanomaterial-based co-delivery platforms have emerged as a particularly compelling approach owing to their higher targeting precision and delivery efficiency. In NSCLC, Pandit et al. fabricated a guanidinylated chitosan-copper complex (Cs-Gn-Cu) by chelating copper to guanidine moiety-modified low-molecular-weight chitosan. This polymeric construct induced intracellular ROS generation, which further exacerbated doxorubicin (DOX)-induced oxidative DNA damage and lowered the apoptotic threshold, thereby synergistically augmenting the anticancer activity of DOX [350]. Wang et al. engineered NPs co-loaded with DOX, camptothecin (CPT), and copper, which demonstrated potent synergistic antitumor efficacy in TNBC by concurrently inducing apoptosis via DOX and CPT while triggering cuproptosis through copper overload [351]. This co-delivery logic extends naturally to other chemotherapeutic partners. Ding et al. addressed the dual obstacle of the blood–brain barrier and GBM-cell resistance by co-loading the proteasome inhibitor carfilzomib with CuET into Y-shaped pHA-VAP-modified nanodiscs. The two agents disrupt protein homeostasis at complementary stages, jointly inducing apoptosis and cuproptosis while reducing carfilzomib resistance in glioma stem cells and prolonging survival in tumor-bearing mice [352]. A comparable framework operates in CRC, where cisplatin-loaded Cu-MOFs exert chemotherapeutic cytotoxicity while the released copper simultaneously mediates cuproptosis and downregulates the cancer-stemness proteins ZEB1 and c-MYC, coupling direct tumor killing to suppression of the stem-like populations that underlie chemoresistance [353]. In BC, Qian et al. encapsulated CuO₂-doxorubicin nanodots (CuDT) within hyaluronic acid, achieving a marked elevation in intracellular oxidative stress and synergistically promoting cuproptosis in conjunction with doxorubicin [354]. In contrast to the copper-elevating strategies described above, Hao et al. adopted a copper-depleting approach by conjugating a copper chelator to a paclitaxel (PTX) prodrug via a disulfide bond, which was subsequently encapsulated in distearoylphosphoethanolamine-PEG2000 nanocarriers. This formulation exhibited greater cytotoxicity against TNBC than the free drug, while significantly reducing systemic toxicity [355].

Beyond synergistic cytotoxicity, the strategic modulation of copper homeostasis offers a potent tool to overcome established chemoresistance. ES-Cu has demonstrated chemosensitizing potential across multiple resistance models. In DDP-resistant cervical cancer, the combination of baicalein with ES-Cu reversed DDP resistance through AKT-mediated enhancement of cuproptosis [356]. Furthermore, ES-Cu suppressed autophagy via the DLAT/mTOR pathway, inducing G2/M cell cycle arrest that inhibited tumor cell proliferation and enhanced sensitivity to docetaxel (DTX) [73]. In PDAC, both the co-administration of Elesclomol with CuS NPs and the direct encapsulation of Elesclomol within CuS NPs enhanced cytotoxicity against PDAC cells and CSCs. Furthermore, in vivo studies further confirmed that this strategy attenuated gemcitabine resistance in PDAC [294]. DSF similarly represents an important agent for chemosensitization and resistance reversal. BC cells can develop resistance to DOX, DTX, and related agents through drug efflux mechanisms and CSC enrichment. Lafi et al. established a triple-combination regimen comprising DOX, hydralazine, and DSF, which significantly impaired cell proliferation and viability in both wild-type and DOX-resistant BC cell lines in vitro, while reducing the effective DOX dose required to eliminate BC cells [357]. Given that chemotherapy-induced protective autophagy is a key resistance mechanism, Swetha et al. combined DSF with DTX, exploiting DSF-induced autophagic cell death to successfully reverse DTX resistance in BC [358]. In DDP-resistant NSCLC, Yuan et al. identified DSF as a potent DDP sensitizer. The two agents react to form Pt(DDTC)₃⁺, which has broader-spectrum and more potent anticancer activity compared with DDP alone [359]. Analogous findings were reported in atypical teratoid/rhabdoid tumor (AT/RT), a malignant central nervous system neoplasm, where Jangra et al. demonstrated that DSF combined with DDP and an aldehyde dehydrogenase (ALDH) inhibitor significantly potentiated the antitumor efficacy of DDP [360]. At the level of nanoregulators, rational delivery system design can further amplify these resistance-reversing effects. Swetha et al. engineered a pH-responsive NP system ([DD]NpH-T) co-loaded with DTX and DSF at a fixed molar ratio, surface-modified with histidine-conjugated star-shaped PLGA and TPGS. In the acidic TME, [DD]NpH-T achieved stimuli-responsive drug release, enhanced penetration into three-dimensional tumor spheroids, and effectively reversed DTX resistance [361]. To address the challenges of poor bioavailability and limited blood–brain barrier permeability, Jia et al. constructed a targeted nanoregulator by integrating human H-ferritin (HFn), regorafenib, and Cu(II). By exploiting the high-affinity interaction between HFn and transferrin receptor 1 (TfR1), this nanoregulator achieved selective accumulation within GBM, thereby overcoming the pharmacokinetic limitations of regorafenib monotherapy [362]. In addition, the polyphenol-copper nanocarrier bCCM demonstrated the capacity to target mitochondria and deplete intracellular GSH, thereby blocking the formation of GSH–Cu/Pt complexes and enhancing the bioavailability of both copper and Pt-based drugs. Concurrently, bCCM downregulated ATP7B expression to suppress metallic drug efflux, synergistically contributing to chemoresistance reversal [363]. These combination strategies also hold considerable promise for multidrug-resistant (MDR) tumors. Employing a cocrystal@protein strategy, Zou et al. anchored PTX–DSF co-crystal nanobodies onto cytochrome C (Cyt C), achieving significant reversal of paclitaxel resistance in both in vitro and in vivo settings [364].

Copper-targeted strategies in combination with radiotherapy

Radiotherapy exerts antitumor effects mainly through ionizing radiation–induced DNA damage and oxidative stress, yet its efficacy is often compromised by radioresistance and tumor hypoxia [365]. Given the central role of copper in redox homeostasis and mitochondrial metabolism, targeting copper homeostasis can amplify radiation-induced oxidative damage and impair DNA repair. Intriguingly, emerging evidence suggests that radiotherapy can induce cuproptosis in radiosensitive tumor cells. Mechanistically, ionizing radiation upregulates the expression of the copper transporter CTR1 while concurrently depleting intracellular GSH, two complementary events that cooperatively drive copper accumulation and ultimately trigger cuproptosis, as evidenced by marked reductions in lipoylated proteins and Fe-S cluster proteins [366]. Importantly, RNA sequencing analysis of radioresistant tumor cells revealed downregulation of BTB and CNC homology 1 (BACH1), which in turn suppressed the expression of the copper-sequestering metallothioneins MT1E and MT1X, thereby attenuating intracellular copper buffering capacity and inhibiting cuproptosis [366]. Collectively, these findings suggest that radiotherapy itself can sensitize tumor cells to cuproptosis. However, this sensitivity is diminished in radioresistant tumor cells, highlighting the promising potential of combining radiotherapy with cuproptosis-inducing therapies for overcoming radiotherapy resistance (Table 8).

Table 8.

Copper-targeted strategies in combination with radiotherapy

Combination strategy	Cancer type	Effect	Mechanism	Ref.
Radiotherapy (RT) alone	Radiotherapy resistant cancers	Self-sensitization	Upregulates CTR1 and depletes GSH to trigger cuproptosis	[366]
RT combination with DSF and Cu(II) D-gluconate	ESCC	Enhance radiosensitivity	Suppress PI3K/AKT pathway in TICs	[367]
RT combined with GLS2 inhibition and copper	ESCC	Radiosensitization	GLS2 suppression plus copper downregulates LIAS and DLST, impairing the α-KGDH complex and TCA cycle to promote cuproptosis-linked radiosensitization	[368]
RT combination with Cu-polyoxometalate (NPs)	Radiotherapy resistant cancers	Reverse radioresistance	Radiation-triggered Cu release to induce cuproptosis	[369]
RT combination with CuO and Elesclomol (Liposomes)	Radiotherapy resistant cancers	Enhance radiosensitivity	Controlled co-delivery and cuproptosis induction	[370]
RT combination with Cu-doped Polydopamine (PC NPs)	Radiotherapy resistant cancers	Synergistic killing	GSH depletion and Cu-mediated cuproptosis	[371]
RT combination with DSF	Osteosarcoma	Enhance radiosensitivity	Induction of apoptosis and survival inhibition	[372]

Building on the heightened sensitivity of tumor cells to cuproptosis following radiotherapy, accumulating evidence suggests that copper-based agents serve as radiosensitizers to potentiate antitumor efficacy. In ESCC, ALDH1-positive tumor tissues harbor an enriched population of tumor-initiating cells (TICs), which confer resistance to conventional radiotherapy. In vivo studies demonstrated that systemic administration of DSF combined with Cu(II) D-gluconate suppressed the PI3K/AKT signaling pathway and restored radiosensitivity in TICs [367]. In a mechanistically complementary ESCC study, Jing et al. found that knockdown of glutaminase 2 (GLS2) suppressed proliferation and enhanced radiosensitivity, and that GLS2 suppression cooperated with copper to downregulate LIAS and dihydrolipoamide S-succinyltransferase, lowering α-ketoglutarate dehydrogenase complex activity and obstructing the TCA cycle to potentiate radiotherapy [368]. Notably, in cells with intact GLS2 expression, exogenous copper still augmented radiosensitivity but without triggering cuproptosis, indicating that the cuproptosis-linked component of this radiosensitization depends on the combined suppression of GLS2 and copper exposure. Beyond small-molecule copper complexes, nanomaterial-based strategies have emerged as versatile platforms for copper-mediated radiosensitization. Liao et al. engineered a copper-containing nanoregulator polyoxometalate that, upon exposure to ionizing radiation, released copper ions to induce cuproptosis and reverse radioresistance [369]. In a complementary approach, CuO and Elesclomol were co-encapsulated within nanoliposomes to achieve controlled co-delivery, and both in vitro and in vivo experiments showed that this combinatorial system significantly enhanced tumor radiosensitivity [370]. Further expanding this paradigm, Xin et al. developed a biomimetic nanoplatform by doping copper into polydopamine NPs (PC NPs). In this system, radiation-generated ROS depleted intracellular GSH while the released copper synergistically promoted cuproptosis; the combined regimen achieved a tumor growth inhibition rate of 93.0% in vivo [371]. In osteosarcoma, concurrent treatment with DSF and radiotherapy markedly suppressed cell survival, proliferation, and migration, while DSF additionally induced apoptosis in osteosarcoma cells, supporting its potential as a promising radiosensitizing agent in this malignancy [372].

Copper-targeted strategies in combination with immunotherapy

Cancer immunotherapy has emerged as a transformative therapeutic modality, encompassing immune checkpoint inhibitors (ICIs), adoptive cell therapy, cancer vaccines, and modulators of the tumor immune microenvironment, among which ICIs—particularly those targeting the PD-1/PD-L1 axis—are the most widely utilized [373]. The integration of copper-targeting strategies with immunotherapy is supported by a strong mechanistic rationale. Mechanistically, cuproptosis, a mitochondria-associated form of RCD, can trigger ICD, thereby promoting antitumor immune activation. In addition, copper dyshomeostasis or copper overload induced by the copper ionophore DSF has been shown to upregulate PD-L1 expression in tumor cells, highlighting the therapeutic relevance of combining copper-based interventions with ICIs [374]. Accordingly, recent studies have increasingly focused on integrating cuproptosis-inducing strategies with immune checkpoint blockade to enhance antitumor immunity and overcome immunotherapy resistance (Table 9).

Table 9.

Copper-targeted strategies in combination with immunotherapy

Combination strategy	Cancer type	Effect	Mechanism	Ref.
CuO/ES and αPD-1	Melanoma	Enhanced killing	Enhanced immune cell infiltration and ICD induction	[375]
NP@ESCu and αPD-1	General	Synergistic killing	ROS-triggered cuproptosis and TIME reprogramming	[376]
Pt@PCN-Cu and αPD-L1	PDAC	Synergistic killing	Glycolysis impairment and PD-L1 upregulation	[377]
Cu-PN NPs and αPD-1	CRC	Reverse immune evasion	DC maturation and CD8+ T-cell activation	[378]
CMS and αPD-1	LC	Extended survival	Antigen release via magnetic-field cuproptosis	[243]
Cu-DBCO/CL	General	Metastasis suppression	Cholesterol oxidase mediated PD-1/TIM-3 downregulation	[379]
DOX@MINPs-TRF/ChO	GBM	Sustained killing	TRF targeting, ICD induction and cholesterol oxidase mediated PD-1 downregulation	[380]
CussOMEp and αPD-1	BC	Metastasis suppression	ATP7A inhibition and T-cell proliferation via ICD	[381]
CMO/MMF hydrogel and αPD-1	BC	Enhanced infiltration	cGAS-STING activation and DNA release	[382]
CuO@Gd2O3 and LOX and RCC membrane	RCC	Enhanced immunotherapy	Lactate depletion, cuproptosis, PD-L1 downregulation	[383]
HMSN-Met@HA-CuO2	General	Sustained immune activation	ROS/RNS amplification, ICD induction	[384]
Cu9S8 and siATP7A	BC	Tumor suppression and anti-metastasis	ATP7A silencing, copper accumulation, immune activation	[385]
DSF/IONC@Au/MSN-TA NPs	General	Synergistic metalloimmunotherapy	Cuproptosis/ferroptosis and cGAS-STING activation	[386]
CuX-P (T-cell membrane) and αPD-L1	TNBC	Synergistic killing	PD-L1 positive feedback loop and targeted delivery	[387]
Cu(CQ)2 and αPD-1	CC	Synergistic killing	Scheduling dependent and transient sensitization	[388]

Increasing evidence has highlighted a pivotal role for cuproptosis induction in enhancing immune infiltration and sustaining the efficacy of immune checkpoint blockade. In vivo, Lu et al. engineered a nanoparticulate delivery system co-encapsulating CuO and ES, which not only suppressed melanoma growth in mice but also improved immune cell infiltration. Notably, its combination with PD-1 antibodies (αPD-1) resulted in markedly augmented antitumor efficacy [375]. Consistently, Guo et al. developed a ROS-responsive nanoplatform (NP@ESCu) to co-deliver Elesclomol and copper, enabling ROS-triggered cuproptosis and TIME reprogramming. Notably, NP@ESCu synergized with αPD-1 to enhance antitumor efficacy, representing an early example of combining cuproptosis-inducing nanomedicine with immune checkpoint blockade [376]. Furthermore, the researchers constructed a nanozyme, Pt@PCN-Cu, which efficiently induces cuproptosis by stabilizing intracellular copper accumulation. Mechanistically, this nanozyme promotes the dissociation of hexokinase 2 (HK2) from mitochondria, thereby impairing glycolysis and upregulating PD-L1 expression. In a PDAC model, the combination of Pt@PCN-Cu with αPD-L1 not only significantly suppressed tumor growth but also effectively remodeled the immunosuppressive TIME, offering a synergistic metabolic-immunotherapeutic strategy for refractory PDAC [377]. Similarly, Weng et al. developed a copper–propranolol nano-platform (Cu-PN NPs), which remodeled the TIME in a CT26 CRC model by alleviating T-cell exhaustion, promoting DC maturation, and enhancing CD8⁺ T-cell activation, thereby further potentiating tumor suppression when combined with αPD-1 [378]. Lien et al. reported a pH-responsive nanoplatform (CMS) that preferentially accumulated in tumors via charge reversal in the acidic microenvironment and, upon alternating magnetic field stimulation, amplified cuproptosis and antigen release, thereby initiating robust antitumor immunity. When combined with PD-1 blockade, CMS significantly inhibited lung metastasis and prolonged mouse survival beyond 60 days [243]. In parallel, Liu et al. described a single-site nanozyme (Cu-DBCO/CL) that amplified ROS generation through cascade catalysis, triggering both cuproptosis and ferroptosis. Concurrently, the release of LOX-IN-3 remodeled the ECM to facilitate CD8⁺ T-cell infiltration, while cholesterol depletion mediated by cholesterol oxidase (CHO) downregulated immune checkpoint molecules, including PD-1 and TIM-3, thereby restoring T-cell cytotoxicity and significantly suppressing tumor growth and metastasis [379]. Zhao et al. constructed a biomimetic nanocomposite (DOX@MINPs-TRF/CHO) that achieved TRF-mediated tumor targeting in a GBM model, followed by intracellular release of DOX and Cu(II), synergistically inducing cuproptosis and apoptosis and enhancing ICD. Meanwhile, selective binding to cholesterol reprogrammed cholesterol metabolism and downregulated PD-1/PD-L1 expression, thereby sustaining checkpoint blockade efficacy [380].

In addition, intrinsic immunosuppression and delivery barriers present in certain tumor types pose further challenges. To address this, researchers developed a copper-based nanozyme (CussOMEp) loaded with the ATP7A inhibitor omeprazole, which induced cuproptosis and promoted T-cell proliferation and DC maturation via ICD, and significantly suppressed contralateral and pulmonary metastases in BC models when combined with αPD-1 [381]. Furthermore, Jiang et al. designed a puncturable tumor implant to overcome the immunosuppressive microenvironment and vascular basement membrane barriers in BC. This implant employed a photocrosslinked GelMA hydrogel to co-deliver CMO NPs and MMF, enabling MMP-responsive local release. This platform simultaneously induced cuproptosis and ICD, promoted mitochondrial DNA release into the cytosol, and activated the cGAS-STING pathway in synergy with Mn(II), thereby enhancing CD8⁺ T-cell infiltration and pro-inflammatory cytokine production. In parallel, it modulated PD-L1/PD-1 expression to further improve αPD-1 therapeutic efficacy [382].

Building on this approach, a range of inorganic copper nanomaterials has likewise been engineered to convert cuproptosis into an immunological advantage. Working from the principle that cuproptosis-derived ICD can be amplified by relieving metabolic immunosuppression, Xu et al. designed an RCC-targeting platform (CGYL) in which an RCC cell-membrane coating directs lactate-oxidase-decorated CuO@Gd₂O₃ particles to tumor cells. In this system, intratumoral lactate is consumed to fuel chemodynamic therapy, and released copper induces cuproptosis and downregulates PD-L1 expression. Consequently, the combined metabolic and immunogenic remodeling supports antitumor immunotherapy [383]. A related strategy targets the antioxidant barrier directly. The HMSN-Met@HA-CuO₂ platform amplifies intracellular ROS through copper peroxide, while metformin-derived nitric oxide converts these into longer-lived ROS, producing sustained immunogenic damage that drives an antitumor immune response [384]. Copper efflux can also be exploited to strengthen this synergy. Guan et al. developed a Cu₉S₈-based platform co-delivering siATP7A, in which silencing of the copper exporter both deepens cuproptosis and activates immune responses that suppress tumor growth and metastasis [385]. Extending this toward innate immune signaling, Xu et al. constructed DSF/IONC@Au/MSN-TA NPs by doping copper and manganese into iron oxide nanomaterials and co-loading DSF into MSN. Beyond driving cuproptosis and ferroptosis, this platform activates the cGAS-STING pathway through immunogenic cuproptosis, delivering synergistic metalloimmunotherapy [386].

Unlike nanoparticulate systems that directly induce cuproptosis, bio-inspired strategies have also been developed to enhance the synergy between copper-based therapies and ICIs. For instance, MXene nanosheets loaded with CuET(II) were innovatively coated with membranes derived from PD-1–overexpressing T cells to construct a biomimetic nanoformulation termed CuX-P. The key feature of this system is its ability to camouflage as PD-1–expressing T cells, thereby enabling specific recognition of PD-L1–overexpressing tumor cells and subsequent internalization. Following uptake, CuX-P exerts potent antitumor activity through the induction of cuproptosis. Notably, the CuET(II) complex further upregulates PD-L1 expression in tumor cells, which in turn enhances their capacity to internalize CuX-P NPs present in the TME, thereby establishing a positive feedback loop. Importantly, this platform demonstrated robust antitumor efficacy in a murine model of TNBC [387].

Importantly, the synergy between copper-based strategies and ICIs is not universal and may depend on treatment scheduling. Although in vitro studies revealed that Cu(DDC)₂ and Cu(CQ)₂ specifically activated cuproptosis and immune-related pathways, in vivo experiments revealed that short-term administration of Cu(CQ)₂ transiently enhanced PD-1 blockade in CT26 models, whereas prolonged treatment induced immunosuppression. Moreover, antagonistic effects were observed in MC38 models. These findings suggest that tumor-intrinsic copper handling characteristics and the chemical properties of copper ionophores jointly determine therapeutic outcomes, and copper isotopic signatures may serve as predictive biomarkers for copper-based combination immunotherapy [388].

Copper-targeted strategies in combination with metabolic therapy

Tumor metabolic therapy represents an emerging anticancer strategy targeting metabolic reprogramming in malignancies. To meet the demands of rapid proliferation, cancer cells characteristically exhibit enhanced glycolysis, aberrant mitochondrial metabolism, and dysregulated lipid metabolism, displaying metabolic profiles distinct from those of normal cells. This therapeutic approach seeks to selectively disrupt tumor-specific metabolic pathways, thereby depriving cells of energy and biosynthetic substrates essential for growth, with the potential to overcome therapeutic resistance [389]. As a mitochondria-dependent form of cell death, cuproptosis is intimately coupled with tumor metabolic reprogramming, suggesting potential synergistic effects between cuproptosis induction and metabolic intervention. In this section, we examine the therapeutic implications of combining copper-targeted strategies with metabolic therapy in cancer treatment. Overall, this synergy manifests through three principal mechanisms: (1) suppression of glycolysis coupled with enhancement of TCA cycle/mitochondrial respiration to reverse cuproptosis resistance; (2) restriction of tumor energy supply synergizing with cuproptosis to exert antitumor effects; and (3) coordinated regulation of auxiliary metabolic pathways including lipid, choline, lactate, and amino acid metabolism to further potentiate tumor cuproptosis (Table 10).

Table 10.

Copper-targeted strategies in combination with metabolic therapy

Combination strategy	Cancer type	Effect	Mechanism	Ref.
p53 agonists and Elesclomol	BC/CRC	Enhance cuproptosis sensitivity	Shifts metabolism from glycolysis to TCA cycle	[153]
PROTAC proteasome sensitizers	LC	Enhance cuproptosis sensitivity	p53 modulation to increase cuproptosis sensitivity	[391]
Cu(II) and APR-246 (CuF16@246)	Mut-p53 tumors	Synergistic killing	Restores p53; suppresses glycolysis via TIGAR/GLS2	[392]
4-OI and Elesclomol	CRC	Enhance cuproptosis sensitivity	GAPDH inhibition to reduce glycolysis	[154]
2-DG/3-BP and Elesclomol	BC	Synergistic killing	Direct inhibition of key glycolytic nodes	[393]
Cu@CoNC(O) NPs	Hypoxic tumors	Enhance cuproptosis sensitivity	Alleviates hypoxia and shifts glycolysis to OXPHOS	[394]
Cu-Ox@HA (LDHA inhibitor)	NSCLC	Enhance cuproptosis sensitivity	Inhibits LDH-A and shifts glycolysis to OXPHOS	[395]
MHRC@Cu NPs	General	Synergistic killing	Shifts glycolysis to OXPHOS	[396]
Cu(L1)2Cl complex	Metastatic TNBC	Synergistic killing	ATP depletion and inactivates copper efflux (ATP7A)	[397]
Cu–O/Phloretin nanozyme	General	Synergistic killing	Blocks glucose uptake and induces copper deposition	[398]
Orlistat (ORL) @ Cu-MOF	OSCC	Synergistic killing	Induces cuproptosis and inhibits fatty acid metabolism	[399]
CuET and Lenvatinib	HCC	Reverse TKI resistance	FOXO6/CHKα axis and blocks choline metabolic reprogramming	[400]
Syrosingopine (Syr) @ mPDA@CP	General	Synchronous killing	Inhibits lactate efflux, inactivates ATP7B and induces Cu/Fe overload	[401]
NCT-503 (Serine inhibitor) @ NPs	General	Enhance cuproptosis sensitivity	Suppresses GSH production and disrupts redox imbalance	[402]
Gabapentin (Gp) @ CMSNs	PCa	Synergistic killing	Inhibits BCAT1; Dual cuproptosis and starvation therapy	[403]

Targeting metabolic reprogramming in cancer cells can reverse cuproptosis resistance by suppressing glycolysis while augmenting TCA cycle activity, thereby amplifying copper-mediated tumor cytotoxicity. A single-cell sequencing study encompassing multiple cancer types (BRCA, CRC, LUAD, PAAD, PRAD, and STAD) systematically constructed a single-cell transcriptomic atlas of metabolic heterogeneity, revealing metabolic disparities among distinct tumor cell populations. Notably, this investigation demonstrated that the glycolysis-dominant metabolic landscape in CRC significantly correlated with Elesclomol resistance, suggesting that metabolic intervention may circumvent ES-Cu resistance [390]. Mechanistically, p53 orchestrates the metabolic shift from glycolysis toward the TCA cycle in tumor cells. Studies have suggested that this transition may be mediated through the CircFRMD4A/PKM2 axis. In vivo, combining p53 agonists with Elesclomol induced metabolic reprogramming and significantly suppressed tumor growth in BC and CRC models [153]. In LC, researchers employed proteolysis-targeting chimera (PROTAC) technology to engineer proteasome sensitizers that enhanced A549 cell susceptibility to cuproptosis through p53 modulation [391]. Furthermore, Ma and colleagues designed an acid-responsive nano-coordination polymer CuF16@246, co-encapsulating Cu(II) and the p53 activator eprenetapopt (APR-246) within a perfluorodecanedioic acid coordination framework. Upon acidic TME-triggered release, this system synergistically deploys both components: Cu(II) directly induces DLAT oligomerization and Fe-S cluster protein depletion, while APR-246 restores mutant p53 to its wild-type conformation, upregulating metabolic targets including TIGAR and GLS2, thereby suppressing glycolysis and enhancing TCA cycle flux to substantially amplify cuproptosis. Both in vitro and in vivo studies showed greater tumor inhibition by this dual-functional platform in mut-p53-positive tumors compared with monotherapy, without eliciting appreciable toxicity [392].

Beyond p53-mediated glycolysis-to-TCA reprogramming, direct inhibition of key glycolytic nodes or alternative mechanisms promoting the metabolic shift from glycolysis-dependent toward mitochondrial respiration-dependent metabolism represent viable strategies for enhancing cuproptosis sensitivity. 4-Octyl itaconate (4-OI) has been shown to reduce glycolytic flux through GAPDH inhibition. When combined with Elesclomol, 4-OI significantly promotes ES-induced tumor cuproptosis [154]. Another investigation employed the glycolytic inhibitors 2-deoxy-D-glucose or 3-bromopyruvate in conjunction with Elesclomol exposure to evaluate combinatorial efficacy, demonstrating that both strategies enhanced antiproliferative and cytotoxic effects in BC cells [393]. From an engineering perspective, researchers developed a copper-loaded cobalt single-atom nanozyme (Cu@CoNC(O)) possessing both catalase-like activity and copper ion delivery capability. This nanozyme alleviates TME hypoxia, reprogramming tumor cell metabolism from glycolysis toward mitochondrial OXPHOS, thereby restoring TCA cycle activity and enhancing cuproptosis sensitivity. Notably, it induces significant cuproptosis and antitumor effects even under hypoxic conditions [394]. Similarly, the coordination polymer Cu-Ox@HA encapsulated within hyaluronic acid blocks key glycolytic steps by releasing oxidants that inhibit lactate dehydrogenase A. Both in vitro and in vivo experiments demonstrated that this strategy drives metabolic transition from glycolysis toward OXPHOS while enhancing cuproptosis sensitivity [395]. Additionally, the synthetically engineered peptide-conjugated probe (MHRC) promotes the metabolic shift of glycolysis-dependent tumor cells toward mitochondrial respiration. MHRC@Cu NPs co-loaded with MHRC and copper(II) release both molecules in the acidic TME, enhancing copper(II)-induced tumor cuproptosis and achieving 96% tumor growth inhibition [396].

Beyond enhancing TCA flux through metabolic reprogramming, certain strategies disrupt cellular reducing capacity and restrict energy supply, synergizing with cuproptosis to produce therapeutic effects. Chaudhary et al. designed a copper(II) dipyridophenazine complex (Cu(L1)₂Cl) to deplete GSH and enhance ROS generation, thereby compromising mitochondrial function and suppressing ATP production. ATP depletion further impairs ATP7A function, causing massive intracellular Cu(I) accumulation that triggers DLAT oligomerization and LIAS depletion, ultimately synergistically driving cuproptosis [397]. Zhang et al. developed a GSH-responsive in situ synthesis strategy that transforms carrier-adsorbed copper ions into oxygen-anchored copper single-atom nanozymes (Cu–O₂/Cu–O₄) through carrier deformation. This process not only promotes copper ion dissociation and GSH consumption, thereby triggering cytoplasmic copper deposition and initiating cuproptosis, but also blocks glucose uptake via the released glucose transport inhibitor phloretin (Ph), thereby jointly mediating antitumor effects [398].

Beyond targeting glucose metabolism, increasing studies extend investigation to lipid, choline, lactate, and amino acid metabolism to evaluate their roles in enhancing cuproptosis-inducing cancer therapy. One study constructed orlistat (ORL)-loaded Cu-MOF nanomedicine (ORL@Cu-MOF), achieving responsive release upon stimulation by elevated glutathione in the TME, simultaneously inducing cuproptosis while inhibiting fatty acid metabolism. In an OSCC lymph node metastasis model, this platform effectively triggered cuproptosis and blocked tumor lipid metabolic reprogramming, significantly enhancing antitumor efficacy [399]. Regarding choline metabolism targeting, Wu and colleagues discovered that CuET modulates HCC cell sensitivity to lenvatinib through FOXO6 targeting. Integrated transcriptomic and metabolomic analyses revealed that CuET induces autophagy through FOXO6 targeting and downregulates choline kinase CHKα expression, thereby suppressing AKT pathway activation and blocking choline metabolic reprogramming, ultimately enhancing lenvatinib efficacy in HCC. Both in vitro and in vivo experiments confirmed significant synergistic antitumor effects of CuET combined with lenvatinib against HCC, providing a potential strategy for overcoming TKI resistance by targeting choline metabolic pathways [400]. In lactate metabolism modulation, researchers constructed a syrosingopine (Syr)-loaded nanodelivery system (Syr@mPDA@CP), utilizing mesoporous polydopamine as a carrier for co-delivery of Syr and CuO₂. Syr inhibits lactate efflux, leading to intracellular lactate accumulation and cellular acidification. This acidic environment promotes ferritin FTH1 dissociation with endogenous iron release while suppressing glycolysis and reducing ATP levels, consequently inactivating the copper efflux protein ATP7B. This system synergistically amplifies intracellular copper/iron accumulation, ultimately synchronously activating both cuproptosis and ferroptosis pathways [401]. Furthermore, Ma et al. investigated the impact of amino acid metabolism on tumor cuproptosis and constructed a nanoregulator containing the serine metabolism inhibitor NCT-503. NCT-503 suppresses GSH production from the serine metabolic pathway, causing redox imbalance and mitochondrial dysfunction, ultimately enhancing copper-induced cuproptosis [402]. Inhibition of the catabolism of branched-chain amino acids (BCAAs)—leucine, isoleucine, and valine—may affect tumor cell metabolic reprogramming and proliferative capacity. Based on this premise, Shi and colleagues developed a copper ionophore-functionalized mesoporous silica nanoplatform (XQ/Gp@CMSNs), achieving PCa-targeted delivery through XQ-2d aptamer modification and responsive release of gabapentin (Gp) and Cu(II) in the acidic TME. Gp blocks BCAA catabolism by inhibiting branched-chain amino acid transaminase 1 (BCAT1), while Cu(II) induces mitochondrial copper overload. Their synergy drives DLAT oligomerization, FDX1 downregulation, and TCA cycle inhibition, achieving dual enhancement of cuproptosis and tumor starvation therapy [403].

Copper-targeted strategies in combination with novel therapeutic strategies

The elucidation of cuproptosis has provided conceptual avenues for the advancement of sonodynamic therapy (SDT). SDT is a non-invasive therapeutic modality that employs ultrasound to activate sonosensitizers, thereby generating ROS in situ at tumor sites to exert cytotoxic effects on tumor cells. Owing to its exceptional tissue penetration depth, SDT is particularly well-suited for the treatment of deep-seated tumors. Chlorin e6 (Ce6) represents one of the most widely utilized sonosensitizers [404]. One study designed a carrier-free NP system (Ce6@Cu NPs) fabricated through the direct self-assembly of Cu(II) and the sonosensitizer Ce6. In vitro experiments using the GBM U87MG cell line demonstrated that ultrasound-triggered Ce6 activation facilitated Cu(II) release, which was subsequently reduced to Cu(I) by the elevated intracellular GSH levels in tumor cells, thereby triggering cuproptosis [405]. Building upon this foundation, surface functionalization of this nanostructure with d-α-tocopheryl polyethylene glycol succinate (TPGS) activates macropinocytosis in tumor cells, further augmenting the intracellular accumulation of copper nanomaterials [406]. Chen et al. coated macrophage membranes onto copper-doped zeolitic imidazolate framework-8 (ZIF-8) co-loaded with perfluorocarbon and the sonosensitizer Ce6, achieving enhanced tumor cell uptake and eliciting cuproptosis [407].

Photothermal therapy (PTT) is particularly compatible with cuproptosis, as many copper nanomaterials are themselves efficient near-infrared absorbers. Zhou et al. fabricated an Au-nanorod-cored, Cu-doped, DSF-loaded mesoporous-silica platform (Au@MSN-Cu/PEG/DSF) in which NIR irradiation triggers on-demand release of copper and DSF. The DSF chelates copper in situ to form cytotoxic CuET, and photothermal heating acts in synergy with the resulting cuproptosis [408]. On the same principle, loading the ES-Cu complex onto triphenylphosphine-chitosan-modified MoS₂ nanosheets (EsCu@TCM) couples mitochondria-directed copper delivery with efficient NIR photothermal conversion, the photothermal effect additionally suppressing ATP7A to enhance copper retention [409]. Photothermal enhancement can also be paired with ferroptosis. The CuFeTe₂ nanosheets release Fe(II) and Cu(II) under acidic conditions to drive ferroptosis and cuproptosis, while NIR-II photothermal activation degrades HSP70, lowers ATP, and accelerates ROS generation, amplifying both pathways [410].

Photodynamic therapy (PDT) employs light-activated photosensitizers to generate ROS, inducing oxidative damage and tumor cell death. Given that both PDT and cuproptosis converge on mitochondrial dysfunction and redox dysregulation, their combination represents a rational therapeutic strategy [411]. Copper-based nanomaterials can serve dual functions as photosensitizers or photosensitizer carriers while simultaneously delivering copper payloads, thereby synergistically amplifying oxidative stress and cuproptosis-mediated tumor cytotoxicity [412]. Ce6-based bifunctional nanoplatforms have demonstrated this synergistic effect. Chen et al. co-assembled Ce6 with mitochondria-targeting triphenylphosphine (TPP), thymopentin (TP5), and copper into NPs. TPP mediated the mitochondrial accumulation of NPs, and upon light irradiation, these NPs released ROS and copper to induce tumor cell cuproptosis [255]. Hu et al. constructed a Cu₂O@SiO₂-Ce6 (CSC) nanoplatform in which a cuprous oxide core was encapsulated within a silica shell and covalently linked to the Ce6 photosensitizer. Following tumor accumulation via the enhanced permeability and retention (EPR) effect, the acidic microenvironment triggered massive copper ion release, inducing mitochondrial cuproptosis through promotion of lipoylated TCA cycle protein aggregation and Fe-S cluster protein depletion. Concurrently, photoactivated Ce6 generated cytotoxic ROS that actively disrupted tumor membrane structures. This dual-mechanism strategy significantly enhanced tumor cell killing while reducing normal tissue toxicity, providing insights into overcoming PDT resistance [413]. Beyond Ce6-based NPs, TME-responsive platforms have also been developed. CJS-Cu NPs represent a TME-responsive copper-based nanoplatform for photo-induced cuproptosis synergistic therapy. Upon co-activation by intratumoral H₂O₂ and light irradiation, this system mediated Cu(II) release, which was subsequently reduced to Cu(I) by superoxide radicals (∙O₂⁻) to specifically trigger tumor cell cuproptosis. Moreover, these nanoregulators exhibited high tumor-targeting efficiency and low systemic toxicity while inhibiting tumor metastasis [414]. In a related TME-responsive design, Ning et al. developed a platelet-vesicle system co-loading Cu₂O with a type-I aggregation-induced emission photosensitizer. The photodynamic activation depletes intracellular GSH and suppresses copper efflux, thereby amplifying cuproptosis and suppressing BC pulmonary metastasis [415]. To minimize the potential toxicity of chemical photosensitizers, one study employed the plant extract aloe emodin (AE) as a photosensitizer, loading copper and surface-modifying with PEG-DSPE-FA to construct NPs for delivery to NSCLC. Results demonstrated that NPs accumulated in NSCLC tissues and released Cu and AE, thereby inducing tumor cell cuproptosis and enhancing PDT efficacy [416].

Furthermore, PDT has been integrated with immunomodulatory strategies to construct multimodal combination therapies. For instance, under blue light irradiation, cells upregulate major histocompatibility complex class I (MHC-I) expression via the NF-κB-SUSD6 axis. Based on this finding, Yang et al. isolated MHC-I-enriched melanoma cell membranes and used them to load copper and ferrihydrite, constructing M-Cu@Fh. In a mouse melanoma model, this system was demonstrated to not only induce cuproptosis and ferroptosis but also promote DC and T cell maturation [417]. Liang et al. constructed copper-coordinated nanoassemblies (CCNAs) based on photosensitizer-chemotherapy prodrugs, achieving multiple synergies of apoptosis-cuproptosis-immunotherapy. This system self-assembled zinc phthalocyanine (ZnPc)-doxorubicin (DOX) prodrugs with the indoleamine 2,3-dioxygenase inhibitor 1-MT via Cu(II) coordination. Under near-infrared irradiation, the photodynamic effect of ZnPc generated ROS and triggered DOX release to induce apoptosis, while Cu(II) not only enhanced the photodynamic process by catalyzing oxygen generation but also promoted toxic mitochondrial protein aggregation to trigger cuproptosis. The synergistic cuproptosis-apoptosis effect elicited ICD, which, in combination with 1-MT-mediated reversal of the immunosuppressive TME, achieved substantial regression of primary tumors and 83% growth inhibition of distant tumors [418] (Table 11).

Table 11.

Copper-targeted strategies in combination with SDT, PTT, and PDT

Combination strategy	Cancer type	Effect	Mechanism	Ref.
Cu(II) and Ce6 (SDT)	GBM	Triggered cuproptosis	Ultrasound triggers Ce6 to release Cu(II) and GSH reduces Cu(II) to Cu(I) to induce cuproptosis	[405]
TPGS-functionalized Cu-Ce6 (SDT)	GBM	Enhanced Cu accumulation	TPGS activates macropinocytosis, increasing intracellular uptake of copper nanomaterials	[406]
Cu-doped ZIF-8 and Ce6 (SDT)	Not specified	Enhanced Cu uptake and cuproptosis	Macrophage membrane coating improves targeting and ultrasound activates ROS and copper release	[407]
Au@MSN-Cu/PEG/DSF (PTT)	Not specified	Synergistic killing	NIR triggers Cu/DSF release; in situ CuET formation and photothermal heating synergize with cuproptosis	[408]
EsCu@TCM (PTT)	Not specified	Enhanced cuproptosis	Mitochondria-directed Cu delivery with NIR photothermal conversion; photothermal effect suppresses ATP7A	[409]
CuFeTe2 nanosheets (PTT)	TNBC	Amplified cuproptosis/ferroptosis	Acidic Fe(II)/Cu(II) release; NIR-II photothermal enhancement	[410]
Cu and Ce6 and TPP (PDT)	Not specified	Triggered cuproptosis	TPP mediates mitochondrial targeting and light triggers ROS and Cu release	[255]
Cu2O@SiO2-Ce6 (PDT)	Not specified	Overcoming PDT resistance	Acidic TME releases Cu for inducing cuproptosis	[413]
CJS-Cu NPs (PDT)	Not specified	Inhibited metastasis	H2O2 and light trigger Cu(II) release; ∙O2⁻ reduces it to Cu(I) to trigger cuproptosis	[414]
PV-Cu2O/photosensitizer (PDT)	BC	Inhibited metastasis	Type-I photosensitizer depletes GSH and suppresses Cu efflux	[415]
Cu and Aloe emodin (PDT)	NSCLC	Enhanced PDT efficacy	FA-modified NPs release Cu and AE in LC tissues to induce cuproptosis	[416]
Cu and Ferrihydrite (PDT)	Melanoma	Multi-mode cell death	MHC-I enriched membranes induce cuproptosis, ferroptosis, and DC/T cell maturation	[417]
Cu(II) and ZnPc and DOX and 1-MT	Primary & Distant tumors	Apoptosis-Cuproptosis-Immunotherapy	Cu(II) catalyzes O2 for PDT and triggers protein aggregation; 1-MT reverses immunosuppression	[418]

Limitations of combined therapy strategies

When copper-targeted therapeutic strategies are integrated with cytotoxic chemotherapy, radiotherapy, immune checkpoint blockade, metabolic interventions, photodynamic therapy or sonodynamic therapy, these combinatorial regimens bring substantial hurdles to clinical translation, an issue that should be fully acknowledged. In terms of dose-limiting toxicities, combination therapies frequently produce overlapping organ toxicities of the individual agents. Copper ionophores can worsen myelosuppression and hepatotoxicity induced by cytotoxic chemotherapy, while copper-encapsulated nanoplatforms can exacerbate radiation-induced oxidative stress in normal tissues. Notably, combining cuproptosis-inducing agents with ICIs further increases the risk of grade 3–4 immune-related adverse events, which are commonly observed when ICIs are combined with other systemic therapies [419]. Combinatorial treatment also complicates tumor resistance mechanisms, allowing tumors to concurrently activate diverse immune evasion and drug resistance pathways against different partnered therapies. As reported, multiple myeloma cells acquire resistance to proteasome inhibitors through the suppression of MUC20-dependent cuproptosis, suggesting that the inactivation of cuproptosis signaling is a pivotal contributor to acquired multi-modal therapeutic resistance [147]. Furthermore, metabolic reprogramming driving cuproptosis resistance largely overlaps with chemo- and radioresistance-related mechanisms, which narrows the synergistic therapeutic window [317]. Currently, biomarker-based patient stratification for such combination regimens remains largely theoretical. Retrospective analyses in HCC, glioma, and other malignancies have suggested that cuproptosis-associated gene signatures are closely correlated with clinical prognosis and treatment responses to immune checkpoint blockade [319, 420]. However, no multiparametric panels integrating copper-related markers with therapy-specific predictors such as PD-L1 expression have undergone prospective clinical validation. Further large-scale prospective clinical studies are therefore urgently needed to confirm their clinical applicability and predictive efficacy.

Controversies, unresolved questions and further perspectives

Since the term cuproptosis was coined in 2022, research into the tumor-biological functions of copper has expanded rapidly. The ensuing exploration of novel therapeutic strategies has also opened new avenues for cancer treatment. Nevertheless, important controversies and unresolved questions remain regarding the mechanistic basis of cuproptosis and its clinical translation, which we discuss in this section.

Mechanistic controversies

The seminal work by Tsvetkov et al. and numerous subsequent studies have described cuproptosis as a process that depends strictly on FDX1. In this context, the copper ionophore Elesclomol is thought to selectively deliver copper to mitochondria. Zulkifli and colleagues investigated the mechanism by which Elesclomol traffics copper and found that not all ES–Cu complexes enter mitochondria in an FDX1-dependent manner. Instead, a fraction of Elesclomol appears to deliver copper to cytosolic targets such as CCS/SOD1. Although the precise mechanism remains unclear, this phenomenon results in intracellular copper accumulation even under conditions of FDX1 knockdown, suggesting that copper dyshomeostasis may influence mitochondrial proteotoxic stress through FDX1-independent routes, rather than relying exclusively on FDX1-mediated copper reduction [421]. In 2025, Lewis et al. further demonstrated in AML that inhibition of heme biosynthesis leads to collapse of mitochondrial complex IV function and disruption of the copper chaperone system. These findings indicate that, beyond the canonical FDX1-centered reductive axis, the integrity of the mitochondrial respiratory chain itself may critically determine cellular susceptibility to cuproptosis [54]. Collectively, these observations prompt reconsideration of the relative importance of FDX1-dependent versus FDX1-independent mechanisms of cuproptosis across distinct tumor contexts. However, a unified model that systematically delineates the dominant pathways and their interplay across diverse metabolic states and copper-homeostasis perturbations is still lacking. These advances also raise a fundamental question about the definition of cuproptosis: should it be defined by upstream triggers, such as copper overload or metabolic dysregulation, or by downstream molecular hallmarks, such as oligomerization and aggregation of lipoylated proteins? The convergence of multiple metabolic perturbations onto a shared cuproptotic endpoint suggests that cuproptosis may represent a final common pathway of mitochondrial proteotoxic stress, rather than a discrete cell-death modality driven by a single initiating event.

Key bottlenecks in the clinical translation of cuproptosis-based cancer therapy

In addition, this field remains constrained by a series of technical and methodological limitations. These constraints compromise the reproducibility of experimental findings, and substantially restrict clinical translation.

To date, investigations into the core molecular mechanisms and regulatory networks governing cuproptosis have been derived predominantly from in vitro models. These in vitro systems typically feature controlled copper treatment concentrations and homogeneous metabolic states, while lacking the regulatory complexity of the microenvironment. However, an emerging paradigm in oncology posits that cancer represents a systemic disease rather than an isolated population of cells, with its initiation and progression profoundly influenced by systemic metabolic states, inter-organ signaling networks, and dynamic immune regulation [422]. Consequently, mechanistic insights into cuproptosis derived only from in vitro models may not fully recapitulate its biological behavior and therapeutic responses within the native TME or at the organismal level. Future investigations may leverage patient-derived organoids to more faithfully simulate copper homeostasis and metabolic heterogeneity within the TME, thereby enhancing the generalizability and clinical relevance of research findings [423, 424]. Beyond developing novel experimental models, technical innovations represent another promising avenue to enhance the feasibility of in vivo investigations. A representative example is laser ablation multi-collector inductively coupled plasma mass spectrometry (LA-MC-ICP-MS) coupled with stable ⁶⁵Cu isotope tracing. In rabbit VX2 hepatic tumor models, this technique discriminates tumor tissue from adjacent normal liver parenchyma with per-mil-level precision based on ⁶⁵Cu isotopic signatures, providing a methodological foundation for clinical translation in HCC detection [425]. Moreover, in contrast to apoptosis, which can be readily visualized through Annexin V–PET or caspase reporter molecules, and ferroptosis, for which lipid peroxidation probes provide convenient readouts, cuproptosis currently lacks comparable in vivo imaging tools. We therefore advocate for the development of in vivo visualization platforms for cuproptosis by integrating copper isotope tracing, spatial metallomics, and molecular biomarker probes.

The selective eradication of tumor cells while sparing copper-dependent normal organs remains an unresolved challenge with two interrelated dimensions. On one hand, NPs are readily sequestered by the reticuloendothelial system, leading to off-target effects and unintended accumulation in non-target organs, particularly the liver, spleen, and lungs, thereby posing potential toxicity concerns. On the other hand, copper serves as an essential cofactor for cuproenzymes such as cytochrome c oxidase, dopamine β-hydroxylase, and lysyl oxidase, which are indispensable for the proper function of the central nervous system, liver, and heart [342, 426]. Consequently, systemic and non-selective modulation of copper inherently operates within a narrow therapeutic window. From the perspective of copper depletion, a rapid decline in systemic copper levels compromises cuproenzyme activity, which is precisely why clinical management of Wilson’s disease requires meticulous dose titration to prevent iatrogenic copper deficiency and the ensuing neurological sequelae [320]. From the perspective of copper induction, emerging evidence indicates that, beyond tumor cells, murine neurons can also undergo cuproptosis under conditions of cerebral copper accumulation, ultimately leading to cognitive impairment [427]. Collectively, these observations underscore the imperative to further enhance the tumor selectivity of copper-based therapies. Specifically, copper-loaded nanocarriers can be functionalized with tumor-targeting ligands or engineered as TME-responsive platforms that release their payloads in response to acidic pH, elevated GSH, or hypoxic cues, thereby achieving precise spatial delivery [335]. In parallel, increasing efforts have focused on using the spatiotemporal control offered by external physical stimuli to selectively activate cuproptosis. For instance, near-infrared light, ultrasound, and magnetic fields can be non-invasively focused onto tumor lesions. Harnessing these modalities to trigger localized copper release and sensitize tumor cells to cuproptosis confines cytotoxicity to the tumor compartment and substantially broadens the therapeutic safety window [428].

The strict dependence of cuproptosis on the TCA cycle and OXPHOS provides an intrinsic mechanistic basis for resistance in glycolysis-prone tumor cells. Beyond hypoxia-driven resistance mediated through the HIF-1α/PDK1/3/DLAT axis [136], the complement C5a/C5aR pathway has recently been shown to confer cuproptosis resistance by upregulating ATP7B via Wnt/β-catenin signaling and thereby enhancing copper efflux, a phenotype that can be reversed in murine models by combining a C5aR antagonist with CuS-mediated photothermal therapy [429]. Collectively, these findings indicate that cuproptosis resistance differs mechanistically from conventional multidrug resistance (MDR) or apoptosis evasion, but is rather rooted in tumor metabolic heterogeneity and microenvironmental signaling networks. Several promising strategies for overcoming this bottleneck have been investigated in preclinical models. The first is metabolic pre-sensitization, in which glycolytic inhibitors such as 4-OI and 2-DG enforce a metabolic shift toward OXPHOS, thereby augmenting cuproptosis sensitivity [154, 393]. In addition, single-cell metabolic atlases enable phenotypic prediction of intratumoral heterogeneity and support the construction of precision-stratification models that identify metabolic subpopulations likely to give rise to residual resistant clones [390]. Such stratification could ultimately inform the selection of tailored combinatorial regimens within a precision-medicine framework. Finally, the rational combination of cuproptosis induction with ferroptosis or apoptosis inducers may close off the escape routes exploited by resistant tumor populations [430].

Conclusions

In conclusion, this review systematically delineates recent advances in copper homeostasis and its pivotal role in TME remodeling. Mechanistically, copper can not only drive tumor progression through copper-dependent proliferation but also elicit tumor-suppressive effects via cuproptosis, with broad implications including immune regulation. This review emphasizes anticancer strategies targeting copper metabolism, with a focus on nanomaterial-based cuproptosis induction combined with multimodal therapies—including chemotherapy, radiotherapy, immunotherapy, metabolic therapy, sonodynamic therapy, and phototherapy—highlighting recent breakthroughs in overcoming therapeutic resistance and enhancing treatment efficacy. Given the central position of cuproptosis in tumor initiation, progression, and therapeutic response, future investigations should comprehensively elucidate its regulatory networks and develop novel copper-targeting strategies to accelerate the clinical translation of cuproptosis-based therapeutics.

Several critical questions warrant prioritized investigation within the next five years. Mechanistically, the in vivo copper concentration thresholds governing the functional switch between copper-driven proliferation and cuproptosis-mediated cell death remain to be quantitatively defined, and validated predictive biomarkers, including FDX1/LIPT1/DLAT axis integrity, serum Cu/Zn ratio, and emerging Cu-isotope signatures, are urgently needed to enable precision therapy. Technologically, the rapid maturation of spatial metallomic imaging and CRISPR-based metabolic mapping is expected to resolve copper distribution and cuproptosis susceptibility at single-cell and tumor-niche resolution. Clinically, the most promising near-term strategies center on copper-metabolism–guided patient stratification and rational combinations of copper modulators with other cancer therapy strategies, especially immunotherapy. Coordinated advances along these mechanistic, technological, and clinical axes will be essential to translate cuproptosis from a biological phenomenon into a clinically actionable therapeutic paradigm.

Abbreviations

4-OI

4-Octyl itaconate

ADRB1

Adrenoceptor beta 1

ALDH1

Aldehyde dehydrogenase 1

AML

Acute myeloid leukemia

ATOX1

Antioxidant 1 copper chaperone

BACH1

BTB and CNC homology 1

BC

Breast cancer

CCND1

Cyclin D1

CCS

Copper chaperone for superoxide dismutase

CDK16

Cyclin-dependent kinase 16

CD8

Cluster of differentiation 8

circFRMD4A

Circular RNA FRMD4A

COX11

Cytochrome c oxidase assembly protein 11

COX17

Cytochrome c oxidase copper chaperone 17

CRC

Colorectal cancer

CSCs

Cancer stem cells

CTR1

Copper transporter 1 (SLC31A1)

CuET

Copper diethyldithiocarbamate

CuO

Copper oxide

DSF

Disulfiram

EMT

Epithelial–mesenchymal transition

ESCC

Esophageal squamous cell carcinoma

EVs

Extracellular vesicles

GAPDH

Glyceraldehyde-3-phosphate dehydrogenase

GBM

Glioblastoma

GCPM

Gastric cancer peritoneal metastasis

GSCs

Glioblastoma stem cells

GSH

Glutathione

H3K27ac

Histone H3 lysine 27 acetylation

HCC

Hepatocellular carcinoma

HK2

Hexokinase 2

ID1

Inhibitor of DNA binding 1

JAK1

Janus kinase 1

LC

Lung cancer

LIAS

Lipoic acid synthetase

LOX

Lysyl oxidase

MT1E

Metallothionein 1E

MT1X

Metallothionein 1X

NCF1

Neutrophil cytosolic factor 1

NF-κB

Nuclear factor kappa B

NPs

Nanoparticles

NSCLC

Non–small cell lung cancer

OXPHOS

Oxidative phosphorylation

PC NPs

Polydopamine-copper nanoparticles

PD-1

Programmed cell death protein 1

PD-L1

Programmed death-ligand 1

PDAC

Pancreatic ductal adenocarcinoma

PI3K

Phosphoinositide 3-kinase

PKM2

Pyruvate kinase M2

PTBP3

Polypyrimidine tract binding protein 3

REGO

Regorafenib (in HFn-Cu-REGO nanoplatform)

ROS

Reactive oxygen species

RT

Radiotherapy

SLC31A1

Solute carrier family 31 member 1

SOD3

Superoxide dismutase 3

STING

Stimulator of interferon genes

TCA

Tricarboxylic acid cycle

TCF4

Transcription factor 4

TCF12

Transcription factor 12

TICs

Tumor-initiating cells

TME

Tumor microenvironment

TM

Tetrathiomolybdate

TMZ

Temozolomide

TRIM21

Tripartite motif containing 21

XIAP

X-linked inhibitor of apoptosis protein

YAP

Yes-associated protein

Author contributions

Guo-Qing Li: Writing–original draft, Conceptualization, Visualization Wen-Long Wang: Writing–review & editing, Conceptualization, Funding acquisition. All authors reviewed the manuscript.

Funding

This study was supported by the National Natural Science Foundation of China (No. 82403230), the China Postdoctoral Science Foundation (No. 2024M763714), the Postdoctoral Fellowship Program of CPSF (GZC20233156), the Science and Technology Program Foundation of Changsha City (kq2403019), the Special Funding for the Construction of Innovative Province in Hunan (2022SK2041), the Natural Science Foundation of Hunan Province of China (2024JJ6664 and 2024JJ9133), and the Beijing Medical Awards Foundation (YXJL-2024–1352-0612).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Declaration of Generative AI Use

Generative AI and AI-assisted technologies were NOT used in the preparation of this work.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.