Cold and hot tumors: from molecular mechanisms to targeted therapy

Tumor-specific antigens (TSAs) play a critical role in initiating T cell-mediated anti-tumor immune responses.^26,27 TSAs primarily arise from non-synonymous mutations in tumor cells, collectively known as the tumor mutation burden (TMB).²⁸ Beyond DNA coding sequence mutations, the generation of new antigens can result from gene fusion events, mutations in non-coding regions, alternative splicing, and deficient mismatch repair (dMMR), leading to microsatellite instability (MSI).^29,30 Individuals with high TMB or dMMR/MSI tumors typically exhibit enhanced responses to immunotherapy with programmed death-1 (PD-1)/programmed cell death ligand 1 (PD-L1) inhibitors or cytotoxic T lymphocyte-associated protein 4 (CTLA-4) checkpoint blockade.^31–35 Tumor-derived neoantigens are displayed on the tumor cell surface by human leukocyte antigen (HLA) molecules for interaction with T cells. Altered expression of the major histocompatibility complex class I (MHC-I) molecules on tumor cells can occur via genetic, epigenetic, and transcriptional mechanisms.^36–40 Loss of heterozygosity in HLA-I genes contributes to immune evasion against tumor neoantigens in approximately 17% of pan-cancer analyses.³⁶ Epigenetically, the polycomb repressive complex 2 (PRC2) methylates lysine 27 on histone H3 (H3K27) to repress the transcription of MHC-I antigens, aiding in immune evasion.^41,42 Various genes, including B2m, H2-K1, Tap2, Nlrc5, Tapbp as positive regulators, and Ezh2, Med13, Tada3, Traf3 as negative regulators, control MHC-I transcription at the level of gene regulation.³⁸ Experimental evidence has shown that Traf3-deficient melanoma cells exhibit increased MHC-I expression, enhancing T cell-mediated killing.³⁸ Autophagy-mediated degradation of MHC-I on tumor cells can be targeted for therapeutic intervention to restore MHC-I presentation and increase cytotoxic T lymphocyte (CTL) presence.³⁹ Combined use of autophagy inhibitors with PD-1 or CTLA-4 immune checkpoint blockade enhances CTL presence in pancreatic ductal adenocarcinoma (PDAC) and non-small cell lung cancer (NSCLC) murine models, suggesting a potential treatment strategy for HLA-I-deficient tumors.^39,43 Human leukocyte antigen-G (HLA-G), abundantly expressed on various cancer cells, plays a crucial role in immune evasion and immunosuppressive cell proliferation. The release of HLA-G from cell membranes can modulate the tumor microenvironment, highlighting its potential as a next-generation immune checkpoint for cancer therapy.⁴⁴ This discovery illuminates the strategies employed by tumors to evade immune surveillance and underscores the value of T cell-focused immunotherapies for HLA-I-deficient tumors.

Dendritic cells (DCs) play a crucial role in initiating immune responses against tumors by capturing and processing tumor-specific antigens, facilitating cross-presentation, and activating naïve T cells. The recruitment of DCs is mediated through C-C chemokine ligand 5 (CCL5) and X-C motif chemokine receptor 1 (XCL1), with their activation triggered by various danger signals, such as cytosolic DNA, RNA, ATP, calreticulin (CRT), and high mobility group box 1 (HMGB1).^45,46 For instance, DCs recognize intracellular tumor-derived DNA through the cyclic GMP-AMP synthase (cGAS) receptor, which activates the stimulator of interferon genes (STING) pathway, leading to the production of type I interferon (IFN-I) and enhancing DCs maturation.⁴⁷ However, the maturation of DCs within tumors is often limited and hindered by various mechanisms, including T-cell immunoglobulin and mucin domain 3 (TIM3)-mediated inhibition of extracellular DNA uptake, disruption of cGAS-STING pathways, and the inhibitory effects of sialic acid-binding immunoglobulin-type lectins (SIGLEC) through intracellular immune receptor tyrosine phosphorylation sites.⁴⁸ Tumor cells can also avoid DC phagocytosis by upregulating the “don’t eat me” signal CD47 or expressing proteins like stanniocalcin 1 (STC1) and glycosylated B7-H4 to mask calreticulin, which is essential for the exposure of the “eat me” signal.^49,50 Additionally, the secretion of gelsolin (sGSN) by tumor cells can impede the recognition and cross-presentation of tumor antigens by conventional type 1 dendritic cells (cDC1s) through the DNGR-1 receptor, hindering the stimulation of anti-tumor CD8⁺ T cells.⁵¹ Studies suggest that DCs can process novel antigens via the N⁶-methyladenosine (m⁶A)-binding protein YTHDF1.⁵² Inhibition of YTHDF1 in DCs has shown potential therapeutic benefits in enhancing tumor-specific CD8⁺ T cells, as evidenced in melanoma-bearing Ythdf1^−/− mice and colorectal cancer patients with low YTHDF1 expression, highlighting a promising approach for immunotherapy.⁵²

In a murine model of pancreatic ductal adenocarcinoma (PDAC), conventional type 1 dendritic cells (cDC1s) within the tumor microenvironment (TME) exhibit reduced numbers, increased apoptosis, and impaired maturation. These attributes result in compromised antigen presentation, consequently influencing the activation of CD8⁺ T cells.⁵³ Notably, the incomplete maturation of DCs leads to a deficiency in the generation of costimulatory molecules, accompanied by an upregulation of coinhibitory receptors aiming to suppress T cell activation.⁵⁴ Moving to lung cancer, mature DCs display a significant presence of PD-L1, which negatively impacts co-stimulation by hindering the interaction between CD80 on DCs and CD28 on T cells.⁵⁴ In a murine model of hepatocellular carcinoma (HCC) characterized by high immunogenicity and driven by MYC, but with a deficiency in Trp53, the overexpression of β-catenin within tumor cells obstructs the recruitment of DCs, leading to immune evasion and resistance to ICB therapy targeting the PD-1 antibody.⁵⁵

Dendritic cells (DCs) found in tumors expressing chemokine receptor 7 (CCR7) demonstrate lower immunogenicity compared to DCs in non-malignant tissues.⁵⁶ This reduced immunogenicity is marked by decreased interleukin 12 (IL-12) production, heightened levels of PD-L1 expression at both mRNA and protein levels, and upregulated CMTM6 expression—a critical regulatory factor responsible for stabilizing PD-L1.⁵⁷ Studies suggest that the efficacy of PD-L1 ICB therapy is influenced by mature dendritic cells expressing PD-L1 and CD8⁺ T cells expressing PD-1, primarily located in tumor-draining lymph nodes (TDLNs).⁵⁸ Evaluation in murine models of mesothelioma and colorectal cancer demonstrates that local administration of a small dose of PD-L1 monoclonal antibody effectively disrupts the PD-L1/PD-1 pathway in regional TDLNs, resulting in significant tumor regression and enhanced survival rates.⁵⁸ These findings indicate potential functional deficiencies in DCs concerning recruitment, activation, maturation, antigen presentation, and immune initiation, potentially elucidating the phenomenon of high TMB but low T cell inflammation in specific cancer patients. Further investigations are required to unravel the precise underlying mechanisms.

Following initial activation, chemokines and their corresponding receptors in draining lymph nodes recruit T cells. The process of T cell infiltration and persistence in the tumor microenvironment is complex and tightly regulated due to barriers posed by blood vessels and tumor stroma.^59,60 Precise control is necessary to coordinate the behavior of immune cells as they travel to the tumor site.⁶¹ Chemokines such as C-X-C motif chemokine ligands 9 (CXCL9) and C-X-C motif chemokine ligands 10 (CXCL10) recruit CD8⁺ T cells, natural killer (NK) cells, and helper T cell 1 (Th1) cells by binding to the C-X-C motif chemokine receptor 3 (CXCR3).⁶⁰ Notably, the secretion of CCL5 by NK cells is crucial for recruiting cDC1s, leading to the production of CXCL9 and CXCL10, which promote T cell migration.⁴⁵ Tumor-derived CCL5 enhances the migration of CD8⁺ T cells, while DC-produced CXCL9 facilitates the infiltration of tumor-infiltrating lymphocytes (TILs).⁶² TAMs recruit naive CD4⁺ T cells through CCL18, driving their differentiation into Treg cells.⁶³ However, cancerous tumors often exploit chemokines to attract immunosuppressive T cells.⁶⁴ For instance, colorectal cancer cells upregulate the secretion of chemokines like CCL17, CCL22, and CXCL12 to attract regulatory T cells, as well as CXCL1, CXCL2, and CXCL3 to recruit MDSCs.⁶⁴ Some chemokines, like CCL5, simultaneously recruit both anti-tumoral and immunosuppressive T cell subsets.⁶⁵ The complexity of cell migration responses to chemokine networks is influenced by various factors, such as cell diversity, tumor cell types and quantities, and interactions among different chemokines within the tumor microenvironment. It is important to note that while chemokines are pivotal in the tumor microenvironment, they do not solely dictate T cell recruitment. Research has demonstrated that the protein GTPase activator regulator of G protein 1 (RGS1) in tumor-specific circulating T cells acts as a suppressor of chemokine G protein-coupled receptors (GPCRs) signaling, leading to reduced T cell motility and decreased levels of infiltration by CTL and Th1 cells in mouse models, breast cancer, and lung cancer patients,⁶⁶ underscoring the significance of regulating chemokine receptor function.

Tumor blood vessels display characteristics such as incomplete vascular development and increased leakiness, accompanied by stromal elements including fibroblasts and extracellular matrix (ECM). These factors collectively create barriers that impede the infiltration of T cells into the tumor.^67,68 Studies have shown that CD8⁺ T cells are excluded from metastatic urothelial carcinoma (mUC) patients with abundant fibroblasts and collagen.⁶⁹ Additionally, inadequate responses to ICB have been observed in patients with mUC and colorectal cancer (CRC), possibly due to the activation of transforming growth factor β (TGF-β).^69,70 In a mouse model simulating triple-negative breast cancer (TNBC), researchers have identified the involvement of discoidin domain receptor family member 1 (DDR1), a collagen receptor, in promoting the rearrangement of collagen fibers within the ECM. This restructuring of the tumor microenvironment hinders the migration of CD4⁺ and CD8⁺ T cells towards the central tumor zone.⁶⁸ Studies have shown that inhibiting TGF-β with neutralizing antibodies or genetic deletion of DDR1 can enhance T cell infiltration in murine models of mUC and TNBC, effectively overcoming immune exclusion within the central tumor area.⁶⁸ Understanding these mechanisms sheds light on immune exclusion, providing avenues for identifying new immune therapeutic targets to enhance immune cell infiltration. Moreover, studies in a mouse model of PDAC have demonstrated the crucial role of CXCL1 released by tumor cells in diminishing the CD8⁺ T cell population. Abolishing CXCL1 in tumor cells could improve the migration of CD8⁺ T lymphocytes, thereby boosting the effectiveness of ICB and effectively managing tumor progression.⁷¹ Various intrinsic factors in tumors, such as PTEN loss combined with PI3K pathway activation in tumors, can obstruct T cell infiltration.^72,73 Overall, the components within the tumor microenvironment (TME) regulate the movement and infiltration of anti-tumor T cells through intricate processes, leading to immune evasion and the development of “cold” tumors. These pivotal molecules present promising prospects for enhancing the efficacy of ICB therapies.

CD8⁺ T cells play a crucial role in mediating the immune response within “hot” tumors. The effectiveness of immune checkpoint blockade (ICB) therapy stems from the rejuvenation of CD8⁺ T cells, which detect tumor antigens presented by MHC-I molecules on tumor cells through a peptide-specific mechanism.⁷⁶ Interferon (IFN) can enhance antigen presentation and MHC expression.⁷⁷ Upon recognizing antigens, CD8⁺ T cells secrete perforins, granzymes, and IFN-γ, leading to the death of tumor cells.⁷⁸ The significant infiltration of CD8⁺ T lymphocytes serves as an indicator of an immune response within “hot” tumors. Nonetheless, the definition of “hot” tumors remains vague, often merely characterized by the proximity of T cells to tumor cells.^79–81 Recent advanced studies utilizing single-cell sequencing and spatial transcriptomics have unveiled substantial diversity in the composition and features of infiltrating CD8⁺ T cells in the tumor microenvironment (TME).^82,83

In various types of cancer, NK cells, known for their cytotoxic abilities against tumor cells, have been associated with improved overall survival rates. The significance of NK cells is underscored by the limited T cell responses imposed by HLA-I in most tumors that lack HLA-I expression.^94,120 Studies have shown that the presence of NK cells expressing both PD-1 and PD-L1 plays a crucial role in determining the efficacy of PD-1/PD-L1 blockade therapy in mouse tumor models.⁷⁴ The efficacy of PD-L1 blockade in suppressing tumor growth and impeding tumor progression substantially diminishes with a reduction in NK cell levels.⁷⁴ Moreover, cancer patients exhibit shared inhibitory receptors, such as TIGIT,¹²¹ NKG2A,¹²² KLRB1,¹²³ and IL18BP,¹²⁴ on both NK cells and T cells, suggesting the potential for a dual targeting approach in the context of ICB therapy. Flow cytometric analysis in patients with renal cell carcinoma (RCC) has revealed two distinct subgroups: one group characterized by increased CD3 expression and the other group demonstrating elevated levels of NK cells.¹²⁵

The NK-high subgroup showed a significant association with reduced intratumoral T cell infiltration in patients.¹²⁵ These findings suggest that tumors characterized by high NK cell infiltration may potentially exhibit a favorable response to immune checkpoint inhibition, regardless of T cell infiltration levels and HLA-I expression. In addition to their direct cytotoxic effects, NK cells also modulate adaptive immune responses by producing IFN-γ, which notably promotes DCs maturation.¹²⁶ In melanoma, NK cell-derived CCL5 and XCL1 play a crucial role in recruiting cDC1s. Dendritic cells play a key role in presenting tumor-specific antigens effectively and activating CD8⁺ T cells, underscoring the significance of NK cells in enhancing T cell-driven immune responses against tumors. This underscores the substantial importance of NK cells in boosting T cell-driven anti-tumor immune responses in the context of melanoma.⁴⁵ NK cells and CD8⁺ T cells are vital effector cells in the anti-tumor immune response, serving as crucial targets for immune checkpoint molecules. They contribute to two distinct forms of anti-tumor immunity: adaptive immunity mediated by T cells targeting HLA-I-expressing tumors and innate immunity mediated by NK cells targeting HLA-I-deficient tumors. Furthermore, targeting NK cells has demonstrated efficacy against tumors lacking HLA-I expression. Therefore, exploring the potential of NK cell therapy and inhibiting NK cell immune checkpoints show promise and warrant further investigation.

The cyclic GMP-AMP synthase (cGAS) recognizes double-stranded DNA in a sequence-independent manner, leading to the activation and production of cyclic GMP-AMP (cGAMP).¹³¹ Subsequently, cGAMP binds to and activates the Stimulator of Interferon Genes (STING). This activation initiates the transcription of Interferon Regulatory Factor 3 (IRF3) and NF-κB, culminating in the generation of inflammatory cytokines and chemokines.¹³⁰ The cGAS-STING signaling pathway also plays a critical role in enhancing endogenous antigen presentation by upregulating the levels of co-stimulatory molecules.^132,133 Prolonged STING pathway activation is associated with the modulation of gene transcription in the immune system.¹³⁴ Apart from inducing transcriptional responses, STING activation can stimulate processes like autophagy and cell death, aiding in the clearance of pathogens or their derivatives during infections.^135,136

Various mechanisms have evolved to prevent the unintended activation of the internal immune system by limiting self-DNA recognition and termination of downstream signal transduction. These mechanisms include cellular and extracellular clearance processes for self-DNA, such as three prime repair exonuclease 1 (TREX1), lysosomal DNase II, and adenosine deaminase 2 (ADA2) to prevent cGAS-dependent autoimmunity.^137,138 Interactions between nucleosomes and chromatin structure proteins also play a role in evading intact genomic DNA sensing.^139,140 Transport restrictions by ABCC1 and other transport proteins limit intracellular STING-dependent activation at the level of cGAMP.¹⁴¹ Extracellularly released cGAMP is enzymatically degraded by the membrane-bound extracellular nucleotide phosphodiesterase ENPP1.¹⁴² The intracellular transport of STING regulates its crucial role in activation and regulation, as it moves from the endoplasmic reticulum to the Golgi apparatus for efficient degradation in lysosomes.¹⁴³ Cells utilize various intracellular and extracellular protective mechanisms to maintain a balance and promptly resolve immune responses triggered by the cGAS-STING pathway.

The report highlights the activation of cyclic GMP-AMP synthase (cGAS) during malignant transformation and treatment processes by DNA originating from various sources. Defects in DNA damage recognition, signaling, or repair, including the DNA damage response (DDR), are considered hallmarks of cancer.¹⁴⁴ Numerous studies emphasize the relationship between intrinsic DDR defects and the immuno-stimulatory properties of tumor cells activated through the cGAS-stimulator of interferon genes (STING) pathway.¹⁴⁴ The interaction of DNA damage with cellular cGAS primarily relies on two key mechanisms: the excessive production of abnormal DNA fragments in the nucleus and the formation of micronuclei (MN) or age-related chromatin fragments in the cytoplasm are plausible events.^131,145,146 Extensive DNA damage caused by intrinsic carcinogenic processes or exposure to mutagens such as radiation or chemotherapy leads to the generation of atypical double-stranded DNA segments involved in DNA repair.^131,147 Cancer cells dependent on the cGAS-STING pathway display prominent characteristics of type I interferon in these segments. In this context, the distinct structure of extracellular DNA generated in tumor cells can elicit a strong type I interferon response, as it relies on the cGAS-STING pathway.¹⁴⁸ Considering the sustained presence of cGAS in the nucleus, the induction of nuclear DNA may also partly promote cGAS activation, although cytoplasmic sources are predominantly acknowledged.¹⁴⁹

Unrepaired genomic damage or chromosome segregation defects may lead to nuclear and cytoplasmic abnormalities, giving rise to so-called micronuclei.¹⁴⁷ Additionally, unresolved DNA breaks and chromosome fragments during mitotic arrest provide additional chromosome substrates for cGAS activation and can be observed in pre-cancerous senescent cells or tumor cells.¹⁴⁷ Micronuclei or chromosomal fragmentation expose chromatin to the cytoplasm, causing DNA damage and chromosome breakage, providing highly immune-stimulatory double-stranded DNA segments for cGAS binding and activation.¹⁵⁰ Increased DNA damage is attributed to the loss of nuclear membrane integrity, cleavage of chromosomal bridging DNA mediated by three prime repair exonuclease 1 (TREX1), and DNA processing mediated by apurinic/apyrimidinic endonuclease 1 (APE1) in the absence of excision repair cross-complementation group 1 (ERCC1) or breast cancer type 1 and 2 (BRCA1/2).^137,141 Studies suggest that DNA recognition within micronuclei or chromosomes explains the phenomenon of autonomous cGAS activation in various cancers.^146,151

DNA damage induced by radiation exposure can lead to micronuclei formation and trigger cGAS-mediated innate immune response in tumor cells in various environments,¹⁵² including those with ERCC1 or BRCA1/2 deficiencies. PARP inhibitors exploit the cGAS-STING signaling pathway to boost the immunogenicity of tumor cells.¹⁵³ Further investigation into micronuclei sheds light on their immunogenic potential. Upon nuclear membrane degradation, regulatory factors are activated, capable of either inhibiting or stimulating cGAS activity. Specifically, the nuclear exonuclease TREX1 degrades DNA within micronuclei, thereby suppressing type I interferon responses.¹⁵⁴ Conversely, abnormal ESCRT-III mechanisms increase damaged micronuclei levels, stimulating the expression of pro-inflammatory genes linked to heightened micronuclear membrane permeability.¹⁵⁵ Additionally, the inherent properties of chromatin can influence the recruitment and activation of cGAS, as different types of induced micronuclei exhibit notable variances in attracting cGAS.¹⁵⁵

Besides aberrant DNA sources, other potential DNA origins also play a role in triggering the cGAS-STING pathway in cancer. Disruption of transcription networks by endogenous retrotransposon elements within the “viral response” process is observed to promote cancer progression.^156,157 This process involves nucleic acids from activated retrotransposon elements participating in type I interferon responses through STING and mitochondrial antiviral signaling protein (MAVS) in tumor cells.¹⁵⁸ Retrotransposon elements may compromise genome integrity, indirectly enhancing cGAS-STING activation, suggesting a potential strategy for enhancing cancer immunogenicity via epigenetic drugs.¹⁵⁸

In conclusion, various forms of “foreign DNA” are commonly implicated as promoters of cancer development and intrinsic immune triggers within cells. These instances underscore the significance of abnormal DNA in cancer immunity; however, further research is warranted to deepen our understanding of how tumor-related processes influence the immunostimulatory properties of endogenous DNA. For example, downregulation of TREX1 expression in tumor cells may elevate DNA accumulation, ribosomal collisions, and translational stress, thereby augmenting cGAS-dependent DNA recognition.¹³⁸

The DNA accumulated in apoptotic tumor cells and non-neoplastic cells within the tumor microenvironment (TME) is believed to act as the initiator for the cGAS pathway (Fig. 2).¹⁵⁹ This has led to proposals for enhancing the activation of the cGAS-STING pathway in the immune response against tumors through radiotherapy, targeted therapy, and ICB therapy. While there is an understanding of the relationship between DNA release from dying cells and the triggering of cGAS-STING signaling in phagocytic cells, the precise regulatory mechanisms remain unclear.¹⁶⁰ The transfer of engulfed tumor cell fragments by macrophages to DCs rather than to other macrophages may contribute to the immunostimulatory effects. The efficiency of cell corpse processing in DCs may be relatively low, potentially leading to DNA leakage into the cytoplasm and subsequent immune stimulation.¹⁶¹ The interaction between cGAS and other innate signaling pathways, particularly the recognition of DNA by Toll-like receptor 9 (TLR9), plays a crucial role in the immune response to apoptotic cells.¹⁶² Tissue-resident macrophages specialized in clearing apoptotic cells exhibit limited responsiveness to TLR9 stimulation in vivo.¹⁶³ This limited responsiveness may extend to tumor-infiltrating macrophages, explaining the heightened effectiveness of the cGAS-STING pathway compared to the TLR9 cascade upon exposure to extracellular DNA released by apoptotic tumor cells within the TME.^163,164

The intercellular transfer of diffusible cyclic dinucleotide cGAMP facilitates a variety of cGAMP-dependent effects within the tumor microenvironment (TME) (Fig. 2). Tumor cells have the capability to transmit cGAMP directly to astrocytes or dendritic cells via gap junctions, thereby influencing either inhibitory or promotive effects on tumor growth.¹⁶⁵ Moreover, continuous activation of cGAS within tumor cells leads to the constant release of cGAMP into the extracellular space, consequently promoting STING transcription activation in adjacent immune cells.¹⁶⁶ It is important to note that the ability of tumor cells to secrete cGAMP is significantly heightened by ionizing radiation (and other cancer therapies), which significantly contributes to the rationale behind the STING-dependent immune response against the tumor.¹⁶⁶ The movement of cGAMP into and out of cells is facilitated by specific transmembrane channels that vary between cell types. Notably, members of the solute carrier group such as SLC19A1 and SLC46A2 play a crucial role in facilitating the entry of cGAMP into diverse cell populations like monocytes and macrophages. Conversely, chloride channel complexes are pivotal in controlling the uptake of cGAMP in vascular cells and bone marrow-derived macrophages.^167–169 Regulating the influx and efflux of cGAMP provides an additional mechanism to fine-tune the innate STING responses within cancer cells, subsequently influencing immune responses in the TME. For example, the expulsion of cGAMP from malignant cell cytoplasm may attenuate internal STING signaling.¹⁴¹ Dendritic cells (DCs) and macrophages possess the ability to internalize external DNA released from apoptotic tumor cells, leading to direct interaction with cGAS. This interaction enhances the expression of co-stimulatory markers (such as CD80 and CD86) and MHC molecules, thereby bolstering their capacity to activate cytotoxic CD8⁺ T cells.^161,170

In the realm of cancer, the cGAS-STING pathway plays a pivotal role in facilitating a variety of functions that may sometimes be conflicting. These functions have a significant impact on tumor immunogenicity, influencing both the responses of tumor cells and the communication between tumor cells and nearby cells in the TME (Fig. 2). The cGAS-STING pathway is crucial for enhancing therapeutic antitumor immune responses, particularly through the triggering of a robust IFN response. This response is crucial for activating immune cells, particularly DCs, present in the tumor microenvironment.¹⁷¹ Notably, this pathway’s activation has been observed in DCs, T cells, and NK cells, indicating a broad effect.¹⁷² Apart from IFN, cytokine signals regulated by NF-κB are also vital for modulating tumor growth as they work synergistically with IFN responses to boost the activity of NK cells in controlling tumor progression.¹⁷³ Furthermore, the expression of cGAS and STING is not limited to immune cells but also extends to stromal cells, where STING triggers a cytokine response that enhances the inflammatory environment, leading to tumor necrosis and additional anti-tumor effects.¹⁷⁴

The cGAS-STING pathway is critical in various aspects of cancer progression, with a key role in regulating the senescence-associated secretory phenotype (SASP).¹⁷⁵ This signaling pathway functions to suppress the proliferation of damaged cells and improve the elimination of precancerous cells by immune cells through cytokine signaling.¹⁷⁶ Cells that manage to evade senescence face additional hurdles in transitioning to cancer, including replicative crisis characterized by telomere shortening, chromosomal abnormalities, and significant cell mortality. The enzyme cGAS, serving as a detector of telomeric DNA damage, is believed to promote cell death linked to replicative crisis. This process relies on STING’s enhancement of autophagy rather than its involvement in cytokine signaling.¹⁷⁷ The full cGAS-STING signaling pathway inhibits tumor development by stimulating the production of NK cell ligands (e.g., RAE1) on the cell surface as tumor cells advance.¹⁷⁸ Recent studies have demonstrated that the cGAS-STING signaling pathway autonomously triggers a direct anti-proliferative effect or cellular apoptosis, particularly notable in T-cell leukemia.^179,180 These responses boost the production of IFN and chemokines crucial for tumor cells, fostering an immune stimulating milieu within the tumor microenvironment.¹⁸¹

The cGAS-STING signaling pathway has been implicated in promoting tumor proliferation and metastasis. Initial studies showed that mice lacking STING displayed resistance to carcinogen-induced skin cancer.¹⁸² The continued engulfment of dying cells by phagocytic cells is believed to trigger inflammatory cytokines, thus contributing to the pro-tumor effect of STING. This observation is consistent with the widely accepted concept that untreated chronic inflammation aids in cancer development.¹⁸³ The cGAS-STING pathway not only initiates cancer but also stimulates pro-tumor actions post-tumor formation. In cases of increased chromosomal instability in tumors, persistent activation of the cGAS-STING pathway promotes invasion and metastasis by transitioning from type I interferon and classical NF-κB signaling to noncanonical NF-κB pathways.¹⁵⁰ Studies have shown that non-traditional NF-κB signaling pathways or prolonged type I interferon stimuli can dampen the anti-tumor immune responses triggered by radiation, achieved through dendritic cell suppression and myeloid-derived suppressor cell enhancement.¹⁸⁴ The immunosuppressive effects are counteracted by cGAMP, inhibiting M2 macrophage and MDSC recruitment. Conversely, continuous STING pathway activation impairs dendritic cell function and facilitates myeloid-derived suppressor cell mobilization, creating an immunosuppressive tumor environment. Moreover, within stromal and endothelial cells, STING-mediated responses aid in enhancing the anti-tumor effect by boosting the inflammatory setting, attracting immune cells, and guiding tumor necrosis. Thus, ongoing STING pathway stimulation in tumors may advance cancer progression by altering the immune-suppressive tumor environment.

In acute scenarios, lymphocyte depletion caused by STING-induced cell death responses can impede tumor clearance. Studies on tumor transplants suggest that tumor cells effectively exploit the cGAS-STING pathway to eliminate T cells, possibly through cGAMP release.^185,186 Similarly, endothelial cells exhibit a tendency to undergo apoptosis when exposed to cGAS-STING triggers.¹⁸⁷ The documented immune response modulation by various cell types may diminish the therapeutic efficacy of STING activation against tumors. The overall impact of STING on tumor progression and regression is heavily influenced by the surrounding microenvironment, the timing, and the intensity of STING activation. The response of target cells to STING stimulation plays a critical and distinct role in shaping treatment outcomes. Consequently, the overall impact of STING on tumor progression and regression is shaped by these factors. Acute and moderate STING involvement is beneficial for inhibiting tumor effects, while prolonged or excessive STING activation results in immune suppression and adverse outcomes. Furthermore, targeting STING in dendritic cells while avoiding T cell exposure is crucial to enhance persistent tumor-specific T cell responses. This knowledge is currently being applied to enhance the effectiveness of STING agonist treatments in cancer immunotherapy. Initially, the use of small molecule inhibitors to block cGAS or STING was primarily aimed at mitigating harmful inflammatory aspects linked to autoimmune disorders.¹⁸⁸ Recent findings suggest that targeting cGAS or STING inhibition could play a pivotal role in preventing inflammation-induced tumor progression or serve as a therapeutic strategy against metastasis in cancers with high chromosomal instability.¹⁸⁹ Lastly, cGAS is also believed to influence tumor progression and suppression independently of STING. It has been proposed that cGAS accelerates tumorigenesis and increases DNA damage by interfering with homologous recombination DNA repair.¹⁹⁰ Conversely, the absence of cGAS, regardless of STING, compromises the integrity of the intestinal barrier and impedes colon cancer progression, underscoring cGAS’s unique role in guarding against inflammation-induced tumorigenesis.¹⁹¹

Research on the anti-tumor properties of STING agonists obtained from cGAMP has highlighted the crucial role of IFN in their effectiveness against tumors.¹⁹² It has been found that the production of type I interferon by dendritic cells is essential for eliciting endogenous T cell responses. Additionally, endothelial cells have been observed to contribute to local type I interferon responses.¹⁹³ In addition to promoting activated immune phenotypes, cGAMP has shown the ability to counteract immune-suppressive phenotypes, such as polarizing M2 macrophages.¹⁶¹ Therefore, targeted manipulation of innate immune cells, particularly dendritic cells, to avoid non-specific targeting within the tumor microenvironment, requires further investigation. Current research suggests encapsulating cGAMP into extracellular vesicles derived from viruses or through genetic engineering as a strategy to selectively target antigen-presenting cells to prevent immune cell dysfunction. Moreover, nanocarriers containing STING agonists have demonstrated improved efficacy in preclinical cancer models. These advancements hold promise for enhancing drug bioavailability, optimizing pharmacokinetics, and enabling systemic delivery.¹⁹⁴ While challenges persist in utilizing STING agonists as adjunct therapeutic approaches, significant progress has been achieved in this realm. Therapeutic mRNA vaccine strategies leveraging the STING pathway to enhance tumor immunogenicity have been explored. This approach involves customization of specific lipid components for mRNA delivery, independent of cGAS activation of STING.¹⁹⁵ Furthermore, the use of mRNA encoding cGAS or STING has been found to activate antigen-presenting cells involved in tumor infiltration, enhancing IFN signaling pathways to enhance CD8⁺ T cell responses.¹⁹⁶ Despite the highest adjuvant activity observed in patients with the STINGV155M mutation associated with early-onset STING-associated vasculopathy of infancy (SAVI), potential T cell cytotoxic side effects should be considered.¹⁹⁶

Various cellular protective mechanisms along the DNA-cGAS-cGAMP-STING pathway have been explored as potential therapeutic targets, in addition to STING agonists.¹³⁸ Upstream, enhancing DNA levels by inhibiting the degradative activity of extracellular nucleases like TREX1 has been shown to improve tumor control by radiotherapy. Altering radiation dosage protocols has been demonstrated to enhance type I interferon production, dendritic cell activation, and endogenous cytotoxic T lymphocyte responses against tumors by preventing TREX1 induction.¹³⁸ Cancer cells have been found to downregulate cGAS or STING expression to effectively evade immune surveillance, inhibiting intracellular pathway activation. This has led to the development of strategies to reactivate natural cGAS-STING signaling in tumor cells, such as introducing mRNA encoding STING or utilizing pharmaceutical agents to antagonize epigenetic suppression mechanisms to enhance cancer cell immunogenic properties.¹⁹⁷ Another strategy to boost STING activation in the tumor microenvironment is by inhibiting ENPP1, an enzyme that degrades extracellular cGAMP.¹⁹⁸ Initial studies have indicated that genetic knockout or pharmacological inhibition of ENPP1 increases tumor immunogenicity. Enhancing local anti-tumor immunity by targeting ENPP1 significantly reduces systemic toxicity resulting from excessive cGAMP degradation. Furthermore, blocking the anti-phagocytic signal mediated by CD47 can activate STING, leading to dendritic cell stimulation and CD8⁺ T cell activation.^199,200 Blocking TIM3 can enhance DNA uptake by dendritic cells and modulate STING-mediated immune signaling.⁴⁸ Similarly, defects in LC3-related autophagic processes are crucial for cellular balance and immune responses and can enhance T cell activation in a STING-dependent manner.²⁰¹ Effective integration of cancer treatment methods that activate cGAS-STING activity is essential for treatment optimization.

The diverse roles of the cGAS-STING pathway as an innate immune signaling pathway in detecting intracellular imbalances and initiating strong anti-tumor immune responses have garnered considerable attention. Moreover, the interplay between cGAS-STING and cancer is intricate, encompassing functions in immune suppression as well as facilitation of metastasis. Researchers have commenced the development of next-generation STING agonists with the goal of creating pharmaceuticals possessing enhanced bioavailability and tailored toxicity profiles. The prospect of translating these research findings into clinical applications is indeed promising.

Molecular biomarkers are indeed crucial in predicting the response to tumor immune therapy, especially in the realm of personalized cancer immunotherapy.²⁰⁵ Understanding the regulatory mechanisms of epigenetic events sheds light on the significance of specific epigenetic changes as potential biomarkers for immunotherapy.^206,207 Inherent epigenetic modifications within tumor cells are strongly linked to cancer progression, development, and resistance to treatment.^208,209 Additionally, it is important to note that therapy itself can trigger epigenetic changes, such as variations in DNA methylation patterns of CD8⁺ T cells post-immunotherapy.²¹⁰ Epigenetic biomarkers offer numerous advantages, including minimal invasiveness, the possibility of liquid biopsy, and the ability to measure DNA methylation changes in body fluids.²¹¹ Particularly in diseases like lung cancer, epigenetic biomarkers can fulfill various roles as diagnostic indicators, prognostic factors, predictive markers, and tools for treatment monitoring. This drives progress in therapeutic diagnostics and precision medicine. While the potential of using epigenetic changes as biomarkers is substantial, it is crucial to exercise caution and adhere to specific prerequisites. Testing candidate epigenetic biomarkers in well-defined and homogeneous patient cohorts is essential. This necessitates the development of sensitive and precise detection methods to uncover novel epigenetic features that could further enhance our understanding of cancer biology and treatment outcomes.

The development of cancer is primarily driven by genetic mutations and abnormal changes in chromatin structure, disrupting the usual functioning of cells and potentially initiating and fostering tumor growth.²¹² Common mutations found in cancer often involve alterations in the activity of chromatin-modifying enzymes, stemming from abnormal expression or mutations.^213,214 Tumor cells frequently exhibit heightened acetylation of promoters, leading to the overexpression of oncogenes. Many cancer types also display a widespread loss of DNA methylation, with a particular trend towards gaining DNA methylation at CpG island sites.²¹⁵ As tumor cells undergo significant alterations in the epigenome and the chromatin landscape of the tumor microenvironment and immune cells, the regulation of the intensity and efficacy of anti-tumor immune responses come into play. This intricate interplay may have implications for responses to immunotherapy and overall disease outcomes.^216,217 Immune suppression within the tumor microenvironment can promote tumor progression and impact immune cells at an epigenetic level, potentially facilitating tumor evasion.²¹⁸ In tumors characterized as “hot” effective anti-tumor immune responses, including those involving helper type I T cells and linked to the interferon response, are observed in malignancies like melanoma and head and neck squamous cell carcinoma (HNSCC).^219,220 Therefore, understanding how tumors influence immune suppressive tumor microenvironments is pivotal. CD8⁺ T cells commonly exist in a state of exhaustion due to prolonged antigen stimulation,²²¹ featuring upregulation of genes like PTCD1 (encoding PD-1) and activation of the IL-10 signaling pathway.^222,223 Recent clinical evidence underscores that solely applying anti-PD-1 antibodies locally is insufficient to fully reverse CD8⁺ T cell exhaustion.²²⁴ Emphasizing the necessity of bolstering a robust and effective immune response against tumors, it is crucial to recognize that, as tumors progress, CD8⁺ T cells targeting specific tumor antigens may experience dysfunction, as evidenced by the release of IFN-γ and TNF-α.⁹³ Additionally, the accumulation of double-stranded RNA (dsRNA) in tumor cells can trigger IFN responses in the tumor microenvironment, bolstering a potent antitumor immune response.²²⁵ Inhibiting the functions of methyltransferase G9a and DNA methyltransferases (DNMTs) in ovarian cancer cell lines can enhance the expression of endogenous retrovirus (ERV) transcripts, consequently prompting the activation of virus defense genes such as IRF7 and STAT1.²²⁶ Notably, treating colon cancer cell lines with the demethylating agent 5-Azacytidine (5-AZA) boosts the expression of interferon response factors like IRF7 and OASL by increasing dsRNA levels and initiating the MDA5/MAVS/IRF7 signaling cascade.²²⁷ Therefore, leveraging epigenetic therapies to target tumor cells may induce a viral-like response, enhancing antitumor reactions and improving therapeutic outcomes when combined with ICB therapy.

Epigenetic biomarkers linked to immune evasion or response traits are valuable indicators for evaluating the efficacy of immunotherapy.²²⁸ While most studies have focused on exploring at the tissue level, a deeper understanding at the cellular level is still ongoing (Fig. 3). Employing transcriptomic and epigenomic data for systematic analysis proves to be a powerful method in discovering epigenetic markers. Through comprehensive whole-genome bisulfite sequencing (WGBS) on virus-targeting CD8⁺ T cells from a mouse model with persistent infection by LCMV strain 13, the DNA methylation profile regulated by the newly generated DNMT3a for important genes (such as Ifng, Myc, Tcf7, Ccr7, and Tbx21) was revealed, with this methylation pattern associated with fatigue.²²⁹ DNA methylation analysis has unveiled abnormal activation of the TGF-β pathway in fibroblasts associated with tumors, potentially leading to immune suppression in initially “hot” tumors and hindering effective responses to ICB.²³⁰ In a study focusing on the cohesive tissue of tumor cells in the murine ID8 model of ovarian cancer, inhibiting the epigenetic controllers EZH2 and DNMT1 increased expression levels of Th1 chemokine genes Cxcl9 and Cxcl10 within the tumor cells.⁶² Through concurrent application of epigenetic and immunotherapy via inhibition of EZH2 and DNMT1, enhanced infiltration of CD8⁺ T cells and improved efficacy of the anti-PD-L1 antibody in treatment were observed.²³¹ Epigenetic regulation has the potential to counteract immune avoidance due to epigenetic suppression of Th1 chemokine gene expression.²³¹ Inhibiting the LSD1 histone demethylase gene in melanoma cells resulted in increased penetration of CD4⁺ and CD8⁺ T lymphocytes in tumor tissues, enhancing tumor immunogenicity compared to normal controls.²³² In ovarian cancer in humans, heightened expression of CCL5 and CXCL9 led to increased infiltration of CD8⁺ T cells into tumors, prolonging patient survival outcomes and augmenting the effectiveness of anti-PD-1 antibody ICB treatment compared to cases with lower chemokine expression.⁶² Methylation status of Ccl5 DNA in ID8 mouse ovarian cancer tissues led to reduced attraction of tumor-infiltrating lymphocytes and macrophages, presenting a ‘cold’ tumor characteristic with impact on animal survival and anti-PD-1 antibody treatment response.⁶² In a small-cell lung cancer (SCLC) mouse model, tumor cells showed increased resistance to CD8⁺ T cells by inhibiting MHC-I genes through PRC2, while inhibiting EZH2 restored HLA gene expression⁴². Melanoma cells treated with IFN-γ displayed elevated STAT-1 binding at sites such as MHC-II and CD274, dampening the effectiveness of anti-CTLA-4 antibodies.²³³ Epigenetic mechanisms regulate molecules like Adora 2 A and galectin-3 in HNSCC, influencing immune modulation and suggesting potential as biomarkers for patient stratification.²³⁴ Modulation of TNFRSF9 expression in immune cells infiltrating tumors through DNA methylation boosts anti-tumor immune responses, associated with extended progression-free survival (PFS) and positive responses to PD-1 antibody treatment.²³⁵ These findings support the use of epigenetic “biomarkers” and transcriptomic programs to enhance or synergize with anticancer therapies, particularly in immunogenic malignancies like melanoma.In immunogenic malignancies such as melanoma, the DNA hypomethylation of genes associated with the immune synapse, namely HLA, CD40, CD86, and CD80, contributes to the attraction of effector CD4⁺ and CD8⁺ T cells. Additionally, this mechanism regulates immune tolerance within the TME.²³⁶ The upregulation of Ezh2 expression in the tumor has been demonstrated in three distinct melanoma mouse models (B16-F10, RIM3, and NrasQ61KInk4a^−/−) following anti-CTLA-4 antibody therapy.²³⁷ The upregulation leads to the epigenetic silencing of immune-related genes in tumors, such as Cxcl9, in contrast to the control group. Consequently, the efficiency of antigen processing and presentation to immune cells is diminished.²³⁷ Additionally, a subgroup of CD8⁺ T cells with high PD-1 expression and decreased chromatin accessibility in the TCF7 locus was pinpointed using single-cell transcriptomic profiling. These cells are associated with the suboptimal outcomes observed in melanoma patients experiencing fatigue and unresponsiveness to checkpoint inhibitor therapies.²³⁸ This evidence may support the use of epigenetics as “biomarkers” and transcriptional programs to dissect potential drivers or complement treatment strategies for cancer therapy.

In clinical settings, patients undergoing immunotherapy often display related epigenetic modifications. A comparative analysis of histone modification profiles in human gastric tumors showed a decrease in tumor antigenicity due to altered H3 lysine 4 methylation and H3 lysine 27 acetylation, indicating an immune evasion mechanism regulated by epigenetic factors.²³⁹ A study on non-small cell lung cancer patients receiving treatment with EGFR tyrosine kinase inhibitors and nivolumab (an anti-PD-1 antibody) unveiled a correlation between the methylation levels of the PDCD1LG1 promoter in tumor cells and the development of resistance to anti-PD-1 antibody therapy.²⁴⁰ Furthermore, lower methylation levels of CTLA4 in malignant melanoma tissues have been linked to improved treatment outcomes with anti-PD-1 and anti-CTLA-4 antibodies, leading to enhanced overall survival rates.²⁴¹ This underscores the significant role of epigenetics in predicting and influencing immune responses to immunotherapy. Another study identified a relationship between reduced DNA methylation and the presence of CD8⁺ T cells within tumors.²⁴² Experiments with CAR T cells targeted against CD19 (CART19) demonstrated heightened cytotoxicity upon re-stimulation in laboratory tests and resulted in complete recoveries in patients.²⁴³ Additionally, distinct activation levels were observed at 2732 promoter sites associated with resistance to immune checkpoint inhibition by anti-PD-1 antibodies, suggesting specific promoters could potentially serve as predictive biomarkers for immunotherapy.²⁴⁴ Utilizing epigenetic changes that govern tumor-associated immune responses to bolster the effectiveness of immunotherapy in cancer patients holds potential and warrants additional investigation, given the myriad unknown factors, especially in forecasting responses among various tumor types.²⁴⁵ Recognizing the significance of identifying reliable biomarkers linked to favorable outcomes in immunotherapy, investigating epigenetic changes that might indicate treatment efficacy or predict responses could prove beneficial and help refine precise diagnostic and stratification strategies.

Metabolic shifts can induce exhaustion and dysfunction in effector immune cells, thereby influencing their differentiation (Fig. 4a, b). The progression of cancer triggers a conflict between immune cells and tumor cells, leading to decreased oxygen and glucose levels. In a series of cancer models, bone marrow cells exhibit the highest capability in absorbing glucose from the tumor, followed by T cells and cancer cells.²⁵⁰ Insufficient glucose can diminish MHC-I antigen presentation on tumor cells, reducing their responsiveness to IFN-induced cytotoxic actions.²⁵¹ Similarly, glucose scarcity can impair the functionality of T lymphocytes infiltrating the tumor microenvironment.²⁵² Specific transcription factors and oncogenic signaling pathways, including AKT,²⁵³ KRAS,²⁵⁴ and MYC,²⁵⁵ regulate immune checkpoint proteins like CD47 and PD-L1, impacting glycolysis-related gene expression and promoting immune evasion. Metabolites can directly influence the expression of immune inhibitory molecules.²⁵⁶ The function and viability of NK cells heavily rely on glycolysis, as evidenced by studies demonstrating NK cell dysfunction upon glycolysis inhibition and reduced fructose 1,6-bisphosphatase 1 (FBP1) expression due to TGFβ, affecting their survival.²⁵⁷ Moreover, compromised NK cell function arises from lipid peroxidation-induced oxidative stress impeding glucose metabolism.²⁵⁸ These findings underscore the critical role of glucose availability and glycolytic activity in supporting the effector functions of T cells and NK cells within the tumor microenvironment (TME).

In tumor cells, glycolysis converts excess pyruvate and NADH into lactate and NAD⁺ via lactate dehydrogenase A (LDHA) to produce ATP to some extent. In comparison to normal tissues, tumors exhibit an increased glycolytic flux, known as the Warburg effect. However, this increase is insufficient to compensate for the low flux through the TCA cycle in terms of ATP production.²⁵⁹ Consequently, solid tumors typically generate ATP at a slower rate than normal tissues, contrary to the widely held perception of them being highly metabolically active. Accumulation of lactate in highly glycolytic tumors impedes CD8⁺ T cell cytotoxicity, facilitating immune evasion.²⁶⁰ Moreover, tumor microenvironment lactate presence affects genes linked to NAD⁺ salvage pathways, impacting NK cell cytotoxicity efficacy and persistence.²⁶¹ High lactate levels at metastatic sites hinder T cell activation and NK cell function.²⁶² Increased lactate uptake through monocarboxylate transporter 1 (MCT1) upregulates PD-1 in regulatory T cells, reducing effector T cell responsiveness to immune checkpoint blockade.²⁶³ Additionally, lactate disrupts anti-tumor M1 macrophage metabolic activities while reinforcing pro-tumor M2 macrophage inhibitory functions.²⁶⁴ Impaired glycolysis in TAMs induces hypoxia, affecting endothelial function, subsequently leading to metastasis and nutrient deprivation.²⁶⁵ Unlike glucose-deprived activated T cells, T cells in acidic TME face lactate export challenges via MCT1, impeding proliferation. Acidosis stimulates pro-tumor neutrophil activation, elevating microenvironment acidity through proton release.²⁶⁶ The complex metabolic interplay between lactate-producing and lactate-consuming cells in tumors significantly influences resistance to anti-angiogenic treatment and shapes immune responses against tumors.²⁶⁷

Besides glucose, the availability of amino acids can significantly influence the activity and differentiation of immune cells (Fig. 4c). Effector T cells and glutamine-dependent tumor cells within tumors both rely on glutamine, indicating a competition for glutamine metabolism in the tumor microenvironment.²⁶⁸ Glutamine is crucial for fueling mTORC1 activity in T cells, a process essential for their differentiation into effector T cells by providing the required energy.²⁶⁹ Restricting external glutamine can prompt the differentiation of CD4⁺ naive T cells into Treg cells, even in conditions conducive to generating other T cell subsets. Supplementation with the glutamine-derived metabolite α-ketoglutarate can boost Th1 cell differentiation.²⁷⁰ On the flip side, interrupting the glutamine metabolism of tumor cells in a triple-negative breast cancer model can amplify the expansion and cytokine secretion of effector T cells.²⁷¹ Research suggests that increasing arginine supply may help sustain effector T cell functionality, potentially augmenting the effectiveness of checkpoint blockade therapy synergistically.²⁷² Upon activation, T cells exhibit heightened arginine metabolism through the regulation of arginase 2 (ARG2), facilitating the formation of central memory-like T cells characterized by prolonged longevity and robust anti-tumor effectiveness. Additionally, enhanced arginine-sensing mechanisms aid in preserving T cell survival.²⁷² Thus, the absence of external amino acids in the tumor microenvironment could impact the development, function, and maturation of immune cells.

Tumors elevate levels of crucial catabolic enzymes like ARG1 or IDO1, leading to decreased arginine and tryptophan levels in the tumor. This depletion plays a vital role in regulating T cell differentiation and proliferation.²⁷³ Lactate originating from tumors can drive macrophage polarization towards the M2 phenotype, potentially inducing the upregulation of arginase-1 in macrophages, causing arginine deprivation in T cells and NK cells.²⁷⁴ Both M1 and M2 polarized macrophages utilize arginine, albeit through distinct metabolic routes. While M1 macrophages employ inducible nitric oxide synthase (iNOS), M2 macrophages utilize Arg1 to metabolize arginine.²⁷⁵ The anti-tumor efficacy is attributed to nitric oxide (NO) produced by iNOS, while metabolites generated by ARG1 promote tumor cell proliferation and suppress NO synthesis.²⁷⁶ Furthermore, glutamine utilization is implicated in regulating the M2 macrophage polarization process, suggesting that inhibiting glutamine synthetase can shift M2 macrophage polarization towards an M1-like phenotype.²⁷⁷ These findings underscore the critical role of nutrient provision in modulating immune cell proliferation and specialization, thereby influencing the suppressive environment of tumors.

Tumor cells disrupt not only the levels of glucose and amino acids in the tumor microenvironment (TME) but also impact the types and quantities of fatty acids (FAs) (Fig. 4d).²⁷⁸ Within the TME, infiltrating immune cells may adjust their metabolism to confront metabolic constraints by increasing lipid droplets and extracellular fatty acids. This metabolic adaptation facilitates the efficient function of CD8⁺ T cells even under low glucose and oxygen levels, as fatty acid breakdown is sustained by peroxisome proliferator-activated receptor (PPAR)-α signaling. This process helps counteract the decrease in CD8⁺ T cells caused by metabolic stress.²⁷⁹ Notably, linoleic acid plays a pivotal role in enhancing the metabolic flexibility of CD8⁺ T cells, protecting them from exhaustion and promoting a transition to a memory cell-like phenotype.²⁸⁰ The utilization of oxidized fats can lead to ROS accumulation associated with lipids, hindering CD8⁺ T cell activity by upregulating the expression of scavenger receptor CD36.¹¹⁷ Studies indicate that Treg cells show a preference for utilizing fatty acid oxidation²⁸¹ or short-chain fatty acids derived from the microbiota to regulate Treg cells differentiation.²⁸² In contrast to CD8⁺ T cells, Treg cells sustain their survival and immunosuppressive function through increased fatty acid usage mediated by CD36.¹¹⁹ Mechanistically, within the tumor milieu, Treg cells selectively upregulate CD36, which, in turn, enhances mitochondrial health via PPARβ signaling, allowing for Treg cells reshaping to better adapt to the lactate-rich TME.¹¹⁹ The immune suppressive function of Treg cells can be hindered by limiting fatty acid transportation using sulfosuccinimidyl oleate (SSO) or impeding fatty acid oxidation with etomoxir.²⁸³ Lipid synthesis is essential for Treg cells to accumulate intracellular lipids, supporting their function by providing glycolytic products for expansion.^284,285 Notably, the resistance of Treg cells to PD-1 therapy is strengthened by their utilization and availability of fatty acids.²⁸⁶

Moreover, lipid-rich environments also impact myeloid cells. In liver metastasis contexts, M2 macrophages contribute to tumor progression by absorbing long-chain fatty acids released by tumor cells via the CD36 receptor.²⁸⁷ Accumulation of lipids in DCs triggers endoplasmic reticulum (ER) stress, activating the XBP1 factor and consequently impairing the presentation of tumor-associated antigens.²⁸⁸ Furthermore, DCs educated by tumor cells with elevated fatty acid synthase (FASN) levels exhibit defects in T cell activation.²⁸⁹ Glycerol, a crucial component for lipid synthesis, is derived from the glycolytic intermediate dihydroxyacetone phosphate (DHAP). Similarly, to DCs, TAMs activate through lipolysis by upregulating CD36 expression.²⁹⁰ In addition to conventional fatty acid uptake, TAMs can recognize β-glucosylceramide via the Ca2±dependent Mincle receptor. This interaction not only fosters a pro-tumor phenotype but also enhances cell survival by triggering the ER stress response.²⁹¹ With the pivotal roles played by ATP citrate lyase (ACLY) in fatty acid metabolism, cholesterol synthesis, protein acetylation, and histone acetylation, there is growing interest in developing anticancer drugs targeting this enzyme.²⁹²

The critical role in metastatic diseases is determined by the heterogeneity of tumor cells and the immune response, impacting the spatiotemporal advantages of tumor cell subclones.²⁹³ The selective pressure from CD8⁺ T cells drives clonal evolution within tumor cells.^293,294 Moreover, the fate of metastatic tumor cells, whether eradicated or forced into a dormant state, depends on the presence of CD8⁺ T cells at the metastatic site.^275,295 Factors present in the tumor environment, such as cytokines and metabolic by-products, play a pivotal role in immune modulation, potentially influencing the interplay between the tumor microenvironment (TME) and immune surveillance. Studies suggest a close relationship between epigenetic regulation, cellular metabolic activity, and immune surveillance in the TME. Specific molecules that permeate cells, like α-KG, can induce changes in DNA methylation patterns, impacting T cell differentiation.²⁹⁶ Additionally, findings have demonstrated that certain metabolic compounds, such as palmitic acid, can prompt H3K4 methylation, thereby influencing transcriptional alterations in cells.²⁹⁷

As a key regulator of metabolism and epigenetics, MYC significantly influences the TME by modulating the expression of CCL9 and IL-23, which in turn alters the immunosuppressive stroma, affecting the regulation of innate and adaptive immunity.²⁹⁸ The presence of IFNγ can induce epigenetic reprogramming in melanoma cells, leading to a rapid upsurge in MYC expression levels.^299,300 Additionally, type I interferons can reshape the characteristics of cancer cells by inducing the demethylation of the epigenetic regulator, KDM1B, thereby influencing the reprogramming of cancer cells during immunogenic chemotherapy.³⁰¹ Consequently, metabolic regulations may drive tumor cells into a dormant state, impacting anti-tumor immune responses, and indicating an alternative potential mechanism of immune modulation driven by metabolic changes through epigenetic modifications.

Research indicates that IFNγ from T cells can trigger malignant metabolic reprogramming, including heightened aerobic glycolysis and increased MYC expression, through the FGF2 signaling pathway. Furthermore, the enzymatic activity of PKM2 is impeded, causing a reduction in NAD⁺ levels, leading to elevated beta-catenin acetylation and fostering cancer progression.³⁰² Additionally, infiltrating Th2 cells with KRAS mutations in PDAC express IL-4 or IL-13 cytokines, activating the IL2rγ–IL4rα and IL2rγ–IL13rα1 receptor complex signaling via the STAT6 pathway, driving MYC-mediated glycolytic reactions.²⁵⁴ Moreover, evidence suggests that the release of IFNγ by T cells or NK cells can alter cellular lipid composition and fatty acid metabolism, influencing sensitivity to ferroptosis by modulating long-chain acyl-coenzyme A synthetase 4 expression levels.¹¹⁸ Throughout cancer progression, tumor cells undergo metabolic adaptations to ensure survival under diverse nutritional conditions, potentially affecting the immune system’s surveillance capabilities. In essence, cytokine release by immune cells can regulate metabolic cascades involved in immune evasion, emphasizing the role of metabolic reprogramming in immune modulation and cancer therapy.

Current clinical trials are increasingly focusing on targeting metabolic activity as a therapeutic approach in cancer treatment, showing promising responses that bolster the effectiveness of cancer immunotherapy. Notably, inhibitors of IDO (NCT02752074↗),³⁰³ CD73 (NCT03794544↗),³⁰⁴ arginase (NCT01266018↗),³⁰⁵ ODC (NCT01059071↗),³⁰⁶ iNOS (NCT02834403↗),³⁰⁷ mutant IDH (NCT02719574↗),³⁰⁸ and (NCT02577406↗),³⁰⁹ have exhibited good patient tolerability and are progressing towards phase II/III trials (Table 1). Enhancing the efficacy of ICB therapy could be achieved by adjusting the metabolic profile of effector T cells, curtailing glycolysis, or reducing LDHA levels in cancer cells, thereby improving the functionality of activated CD8⁺ T cells in the tumor microenvironment.³¹⁰ Studies have illustrated that ICB can suppress immune cell metabolism by inhibiting glycolysis while boosting fatty acid oxidation and lipolysis.³¹¹ The integration of metabolic interventions with ICB presents a novel avenue for enhancing anti-tumor efficacy. In hepatocellular carcinoma patients with urea cycle defects, the combined use of arginine restriction and GCN2 inhibition has shown significant inhibitory effects.³¹² Treatment with dichloroacetate (DCA) can enhance oxidative phosphorylation in p53-positive tumor cells, induce the expression of stress ligands (such as MICA/B), and strengthen the efficacy of CAR T cell or allogeneic NK cell therapy. These findings underscore the value of focusing on cancer metabolism to intensify the effectiveness of immunotherapy and mitigate relapse.³¹³ In cases of refractory patients, highly resistant tumor cells may require the use of metabolic medications or imposition of nutrient restrictions to heighten the vulnerability of tumor cells to cytotoxic lymphocyte attacks.³¹³ However, challenges persist in implementing metabolic-targeted therapies in clinical settings. For instance, co-administration of IDO inhibitors with ICB has not resulted in amplified therapeutic efficacy.³⁰³

Targeting cancer metabolism poses challenges, as certain metabolic interventions may impede crucial processes relied upon by immune surveillance. Upon T cell activation, there is a shift to aerobic glycolysis, prompting increased glucose and glutamine uptake to accelerate proliferation and cytotoxicity. Nonetheless, inhibiting glycolysis may render T cells inactive. Furthermore, complete blockage of glutamine uptake in tumors could potentially foster Treg cells expansion, limiting the effectiveness of T cell therapy. The development of resistance to ICB can induce distinct metabolic conditions in tumors, characterized by diminished CD8⁺ T cell infiltration and reduced IFN-γ gene expression, potentially influencing the efficacy of metabolic treatments. Identifying unique metabolic nodes vital for cancer cells to evade the immune system is crucial. By doing so, we can determine the optimal timing for modifying immune and cancer metabolism to significantly enhance existing immunotherapeutic strategies. Subsequently, further exploration in this evolving field will establish crucial foundations for maximizing the efficiency of metabolic-targeted cancer therapies.

The release of immune stimulatory signals, such as high mobility group box 1, calreticulin, ATP, and oxidized phospholipids, by tumor cells undergoing ferroptosis is a key aspect of immunogenic cell death (Fig. 5).^320–324 These signals play a vital role in promoting dendritic cell maturation and enhancing macrophage-mediated phagocytosis of ferroptotic tumor cells.^321,323 For instance, oxidized phospholipids exposed on the surface of ferroptotic cancer cells can be recognized by Toll-like receptor 2 on macrophages, leading to increased engulfment of these cells.³²⁴ Studies have shown that these interactions can induce polarization of macrophages towards the M1 phenotype and potentially elicit responses akin to vaccines, thereby boosting anti-tumor immunity.^321,323,325 Conversely, the inhibition of ferroptosis in tumor cells may hamper T cell infiltration and activity in the tumor microenvironment, dampening T cell-mediated anti-tumor reactions.^118,326 Additionally, ferroptotic tumor cells could impede dendritic cell maturation and compromise their ability to present antigens, potentially negatively impacting adaptive immune responses.³²⁷ The impairment of dendritic cell maturation and function may be attributed to products of phospholipid peroxidation or other lipid peroxidation byproducts, such as 4-hydroxynonenal. Interestingly, in colorectal cancers lacking GPX4, there is an increase in dendritic cell infiltration, a phenomenon not observed in hepatocellular carcinoma.³²⁸ In conclusion, the debate surrounding the immunogenicity implications of ferroptotic tumor cells is context-specific and remains a topic of discussion in the field of ferroptosis research. Some studies have utilized ferroptotic tumor cells cultured in vitro as vaccination agents.³²¹ However, achieving maximal efficacy in inducing ferroptotic cell death at the vaccination site may be challenging, potentially allowing viable tumor cells to persist and complicating tumor progression.

In the context of anti-tumor immunity, macrophages, neutrophils, and NK cells play pivotal roles in the immune response against iron-induced cell death (Fig. 5).³²⁹ Recent studies have revealed that M1 macrophages exhibit greater resistance to iron-induced cell death compared to M2 macrophages.³³⁰ This heightened resistance is attributed to the elevated expression of inducible nitric oxide synthase in M1 macrophages, leading to increased production of nitric oxide molecules. Nitric oxide radicals can effectively neutralize lipid radicals, thus protecting M1 macrophages from ferroptosis-induced damage.³³¹ Iron nanoparticles have been shown to convert M2 macrophages into an M1 phenotype, enhancing their anti-cancer properties.^332–334 Notably, the tyrosine protein kinase receptor TYRO3 significantly influences ferroptosis within the tumor microenvironment. Inhibiting TYRO3 attenuates iron-mediated cell death in cancer cells, resulting in a shift in the M1/M2 macrophage ratio and creating a tumor-promoting environment.³³⁵ Conversely, inhibiting TYRO3 promotes tumor ferroptosis, alters the M1/M2 macrophage balance, and improves the tumor response to PD-1 therapy.³³⁵ Several studies have elucidated the complex relationship between TAMs and ferroptosis. It has been suggested that ferroptosis can drive the recruitment and polarization of TAMs towards the M2 macrophage phenotype, while inhibiting ferroptosis may impede the development of an immunosuppressive tumor microenvironment.³³⁶ Factors that increase iron-induced cell apoptosis, such as elevated iron intake or deficiency of GPX4, could regulate macrophage migration and orientation through the STING-triggered DNA detection pathway, potentially impacting M2 macrophages and facilitating the progression of PDAC.³³⁷ These findings underscore the intricate and paradoxical effects of iron depletion on macrophages in the context of anti-tumor immunity.

Myeloid-derived suppressor cells (MDSCs) are well-known for their potent ability to induce immunosuppression.³³⁸ The resistance of MDSCs to ferroptosis is predominantly attributed to the inhibitory effect of n-acylsphingosine amidohydrolase 2 (ASAH2) on the signaling pathway involving p53 and heme oxygenase 1 (HO-1).³³⁹ Inhibition of ASAH2 can induce ferroptosis in MDSCs, enhance the infiltration of CD8⁺ T cells into tumors, and boost the inhibitory impact on tumor progression.³³⁹ Enhancing cancer immunotherapy efficacy could be achieved by combining ferroptosis induction and MDSC inhibition.³²⁸ Interestingly, restricted GPX4 loss in hepatocytes alone does not impede hepatocellular tumor formation; GPX4-related ferroptosis triggers immune-mediated tumor suppression in hepatocytes. Within the tumor microenvironment, neutrophils, polymorphonuclear-myeloid-derived suppressor cells (PMN-MDSCs), serve as crucial primary immunosuppressive agents. Genetically or pharmacologically inhibiting ferroptosis can abolish the suppressive function of PMN-MDSCs, resulting in reduced tumor progression. This strategy synergizes with immune checkpoint blockade to effectively impede tumor growth.³⁴⁰ Recent studies have demonstrated that tumor-infiltrating neutrophils with immunosuppressive characteristics, resembling PMN-MDSCs, manifest notable resistance to ferroptosis. This resistance may be associated with elevated levels of aconitate decarboxylase 1 (ACOD1) in neutrophils. Elevated ACOD1 levels upregulate aconitate decarboxylase expression, activating NRF2-mediated protective responses against ferroptosis. Conversely, ACOD1 deficiency suppresses neutrophil migration, augments the antitumor immune response, and enhances the efficacy of immunotherapeutic interventions.³⁴¹ The vital need for a comprehensive understanding of the complex interplay between ferroptosis and immune cells in the tumor microenvironment is underscored by this discrepancy. Additionally, NK cells play a crucial role in the antitumor immune response.³⁴² Dysfunction of NK cells in the tumor microenvironment is frequently linked to oxidative stress from lipid peroxidation. Activation of the NRF2 transcription factor could potentially rejuvenate NK cell function and bolster their effectiveness against tumors.²⁵⁸ Conversely, inhibiting ferroptosis may prolong the survival of NK cells in tumors.³⁴⁰ In essence, these investigations shed light on the susceptibility of NK cells to ferroptosis within the tumor microenvironment.

The collective results illuminate the intricate impact of ferroptosis on diverse innate immune cell subsets within the tumor microenvironment. The study highlights the resilience of M2 macrophages and infiltrating neutrophils to ferroptosis, the susceptibility of immune-stimulating NK cells to this process, and the nuanced responses of MDSCs to ferroptosis. A comprehensive comprehension of the complex interplay between iron metabolism in the tumor microenvironment and the innate immune system is vital for enhancing and refining approaches to cancer immunotherapy.

Within the tumor microenvironment (TME), the adaptive immune system is mainly composed of B cells and T cells, which play a crucial role in controlling tumor progression by reprogramming mechanisms.^343,344 Iron-related products, such as lipid peroxides released by cancer cells, hold the potential to influence the functionality of adaptive immune cells and their ability to recognize tumor-specific antigens. Similar to innate immune cells, the response of adaptive immune cells to iron-induced products varies significantly based on specific conditions (Fig. 6). B cells are pivotal in the immune response against tumors, as they produce anti-tumor antibodies and modulate T cell responses.³⁴⁵ Innate-like B cells, including B1 cells and marginal zone B cells, exhibit dynamic lipid metabolism. These cells’ vulnerability to ferroptosis triggered by GPX4 depletion stems from GPX4’s critical role in maintaining antibody response and overall cellular function.³⁴⁶ Further research is needed to fully grasp the significance of iron-induced cell death within the realm of B cell-driven anti-tumor immunity. This discovery points towards a promising avenue for exploring novel treatment strategies that target GPX4 in specific malignant B cell malignancies.

Conversely, T cells are less susceptible to ferroptosis, supporting the notion that selectively inducing ferroptosis in tumor cells does not significantly impede T cell-mediated anti-tumor immunity. In murine models, the depletion of SLC7A11 impedes tumor progression while leaving in vivo T cell expansion and anti-tumor immunity unaffected, thereby enhancing the efficacy of immune checkpoint blockade in preclinical studies.³⁴⁷ Inducing ferroptosis, activated by cystine deficiency or arachidonic acid introduction, can also impede tumor progression without affecting the expansion and functionality of T cells in the TME. On the contrary, blocking iron-induced cell death mediated by ACSL4 leads to a decrease in T cell-mediated immune responses against tumors.³⁴⁸ Additionally, stimulating ferroptosis through various nanoparticle-based methods enhances T cell infiltration in the tumor microenvironment.^349,350 These findings underscore the potential enhancement of T cell-mediated immune responses towards tumors by simultaneously activating multiple pathways that induce ferroptosis.

T cells, particularly those lacking Gpx4 due to T cell-specific Gpx4 deletion in mice, exhibit heightened sensitivity to iron-induced cell death triggered by GPX4 inhibition. This increased sensitivity is attributed to the rapid accumulation of intracellular lipid peroxides, ultimately leading to cell death through ferroptosis.³⁵¹ CD8⁺ T cells, in particular, show heightened sensitivity to iron-induced cell death caused by GPX4 inhibitors.³⁵² Furthermore, the increased expression of CD36, a crucial player in enhancing the absorption of fatty acids in CD8⁺ T cells infiltrating tumors, may lead to lipid peroxidation and iron-mediated cell death in these T cells, thereby impairing their ability to combat tumors. Conversely, eliminating the CD36 gene or preventing iron-mediated cell death in CD8⁺ T cells has the potential to restore their antitumor function.^116,117 The role of GPX4 in inducing ferroptosis also extends to CD4⁺ T cells.³⁵³ Regulatory T cells, a subset of immunosuppressive T cells, compromise the immune surveillance of tumors. Treg cells display inherent resistance to ferroptosis, possibly linked to their elevated GPX4 expression. Consequently, selectively deleting Gpx4 in Treg cells could enhance anti-tumor immune responses by instigating ferroptosis.³⁵⁴ Follicular helper T (TFH) cells, a specific subset of CD4⁺ T cells known for promoting anti-tumor responses, display susceptibility to iron-induced apoptosis, akin to Treg cells.³⁵⁵ Further investigation is necessary to comprehend the interaction between ferroptosis and TFH cells in the tumor microenvironment. Ferroptosis, induced by SLC7A11 depletion or GPX4 abnormalities, exerts unique effects on T cell function and the immune response to tumors. T cells seem to rely more on GPX4 than on SLC7A11, possibly due to the relatively low expression level and non-essential role of SLC7A11.^356,357 Repressing SLC7A11 and/or stimulating iron-mediated cell death by depleting cystine or providing additional arachidonic acid seems to have no detrimental effects on the immune response against tumors by CD8⁺ T cells. Conversely, the iron-dependent cell death triggered by inhibiting GPX4 has varying effects on different T cell subpopulations: it enhances antitumor immune responses in the immunosuppressive Treg cells but dampens antitumor immune responses in CD8⁺ T cells and TFH cells.

Recent research has revealed a significant discovery that inducing ferroptosis in tumors or enhancing their susceptibility to this process can significantly enhance the effectiveness of immunotherapy. Conversely, tumors that exhibit resistance to ferroptosis are also likely to be resistant to immunotherapy. A noticeable enhancement in the anti-tumor immune response, leading to a more efficient tumor inhibition, is observed when anti-PD-L1 therapy is administered concurrently with either cyst(e)inase or arachidonic acid treatment, as opposed to when each treatment is used individually.¹¹⁸ Similarly, the decreased expression of SLC7A11 increases tumors’ vulnerability to anti-PD-L1 treatment either alone or in combination with radiotherapy.³⁵⁸ Furthermore, periodic dietary restriction of methionine upregulates the glutathione-specific enzyme CHAC1, depleting GSH and exacerbating ferroptosis induced by cystine limitation, thereby rendering tumors more responsive to anti-PD-1 therapy.³⁵⁹ Surprisingly, the combined application of intermittent fasting targeting methionine, the SLC7A11 inhibitor IKE, and PD-1 therapy significantly inhibits tumor progression and prolongs the survival of animal subjects.³⁵⁹

A recent study has identified a beneficial relationship associated with the protein kinase C beta (PKCβII)-ACSL4 signaling pathway. In this process, PKCβII phosphorylates and activates ACSL4 to detect and enhance lipid peroxidation, ultimately contributing to the promotion of ferroptosis. Disruption of this signaling pathway impedes ferroptosis and the T cell-mediated immune response against tumors, leading to resistance to anti-PD-1 treatment by deleting PKCbII or ACSL4, or modifying phosphorylation sites of ACSL4.³⁶⁰ Preclinical investigations have shown that sensitizing tumors to immunotherapy by targeting GPX4 with FIN is feasible. Additionally, the synergistic effect of GPX4 inhibitors in combination with anti-PD-1/PD-L1 therapy has shown increased anti-tumor immune response and tumor inhibition.³⁶¹ Exploring the potential integration of the newly developed effective GPX4 inhibitor JKE-1674 with anti-PD-1/PD-L1 treatment, which enhances in vivo drug metabolism, presents an enticing opportunity. Targeting ferroptosis in immunotherapy also involves other therapeutic approaches capable of inducing cancer cell ferroptosis. By suppressing phosphoglycerate mutase 1 (PGAM1), the downregulation of lipocalin-2 (LCN2) promotes ferroptosis in hepatocellular carcinoma cells and facilitates the recruitment of CD8⁺ T cells. Consequently, the combined inhibition of PGAM1 with anti-PD-1 therapy demonstrates promising synergistic effects in inhibiting tumor progression.³⁶²

In preclinical research, the bifunctional compound BEBT-908, targeting both PI3K and histone deacetylase, has been shown to induce immunogenic ferroptosis in cancerous cells. This mechanism involves enhancing p53 acetylation and inhibiting NRF2 activity, leading to the creation of an inflammatory microenvironment that promotes an anti-tumor immune response and enhances the efficacy of anti-PD-1 therapy.³⁶³ Furthermore, the effectiveness of immunotherapy has been demonstrated by certain nanomaterials possessing ferroptosis-inducing properties.^363,364 For instance, in a preclinical model of hepatocellular carcinoma (HCC) with ascites, the use of an injectable hydrogel drug delivery system incorporating sulfasalazine has successfully induced significant immunogenic ferroptosis. Combining this strategy with anti-PD-1 treatment has synergistically inhibited the dissemination of malignant ascites in the peritoneal region.³⁶⁵ These findings highlight the potential of integrating immunotherapy with ferroptosis inducers in cancer treatment to enhance therapeutic outcomes.

When tumor cells undergo ferroptosis, they release signaling molecules such as arachidonic acid metabolites and HMGB1, which play a crucial role in enhancing the immune response against tumors by facilitating the recruitment of lymphocytes and activation of antigen-presenting cells. It is important to recognize that infiltrating immune cells within tumors can also release immunosuppressive molecules.³⁶⁶ Therefore, the overall outcomes depend significantly on the tumor-associated microenvironment. Notably, “cold” tumors are favorable for triggering ferroptosis, a process that can leverage the advantages of “immunogenic cell death” in cancer cells while mitigating immune suppression effects. Overcoming the limitations and toxicities associated with each approach is essential to ensure the efficacy of iron-targeted therapy as an effective treatment strategy. Anticipated advancements include the development of novel tools for detecting biomarkers related to iron metabolism in the tumor microenvironment, which will be instrumental in refining combined approaches to modulate iron levels in the TME and improve therapeutic outcomes for individuals with advanced cancer.

Tumors arise from acquired genetic mutations in normal cells, driving uncontrolled cell proliferation.¹⁸³ The progression of tumors is often influenced by interactions with immune cells, fibroblasts, endothelial cells, and neurons within the tissue, which cancer cells exploit to their advantage. Furthermore, immune cells recruited to the site, exhibiting suppressive traits, aid in tumor evasion, reflecting behaviors similar to those observed in wound healing or tissue remodeling resolution stages.³⁷¹ Various cell types within the tumor microenvironment play crucial roles in regulating cancer advancement, including TAMs, TANs, MDSCs, and Treg cells. While tumors comprise tumor cells themselves, their growth can be impeded by NK cells, which possess potent cytotoxic capabilities against distressed cells.³⁷² Additionally, CD4⁺ Th1 cells and CD8⁺ T cells have the potential to eliminate cancer cells by recognizing tumor-specific or newly formed antigens.³⁷³ The initiation of anti-cancer T cell responses heavily relies on APCs, primarily DCs, that capture and process tumor-specific antigens to present to T cells.³⁷⁴ Throughout the process of tumor development, chemokines are indispensable for regulating the stimulation, mobilization, properties, and functions of immune cells within the tumor microenvironment (TME). Therefore, the substantial impact of chemokines and their receptors on the effectiveness of both tumor-promoting and tumor-inhibiting immune responses is clearly evident.

In “cold” tumors, myeloid cell infiltration may be present, yet CD8⁺ Teff cell infiltration is often lacking, indicating a reduced level of anti-tumor immunity. In this context, the activation of β-catenin in tumors inhibits the production of CCL4 by upregulating the transcription repressor ATF3, thus restricting the migration of conventional cDC1 towards the tumor, suggesting that tumors avoid immune surveillance by disrupting chemokine functionality.^45,375 PGE2 diminishes the activity of NK cells, hindering their capability to produce inflammatory chemokines. Consequently, the abundance of cDC1 is limited, fostering tumor progression.³⁷⁶ Moreover, the decrease in cDC1 density within tumors over time implies that advanced tumors may possess the capacity to impede cDC1 migration.³⁷⁴ Whether a specific activation threshold is necessary to initiate CCR7-dependent migration of cDC1 towards tumors in sterile tissues remains unclear. Prior studies have highlighted the crucial involvement of atypical chemokine receptor 4 (ACKR4) in regulating DC ingress into lymphatic vessels and their localization in T cell regions within draining lymph nodes. This regulation is achieved through the modulation of the biological activity of CCL19 and CCL21, alongside the maintenance of functional chemotactic gradients.³⁷⁷ The impact of tumors on ACKR4 function may hinder DC infiltration into tumor-draining lymph nodes, though this effect remains uncertain. Certain tumors exhibit a limited presence of CD8⁺ T cells but attract substantial numbers of immunosuppressive cells like TAMs, MDSCs, and Treg cells.³⁷⁸ These tumors stimulate tumor growth by upregulating the expression of inflammatory mediators, creating a tumor microenvironment conducive to tumor survival. Notably, an elevated expression of NF-κB-dependent inflammatory genes is observed in the absence of the tumor suppressor p53.³⁷⁹ Studies suggest that NF-κB-driven chemokines may promote tumor progression, recurrence, and resistance to therapy. For instance, NF-κB activity induces the expression of CCL5, a chemokine that attracts macrophages expressing CCR5 and enhances collagen production, thereby promoting the proliferation of breast cancer cells.³⁸⁰ Activation of NF-κB in TAMs leads to the secretion of CCL22, which attracts CCR4⁺ Treg cells.³⁸¹ Additionally, TAMs induce the release of CCL20 through NF-κB activation, recruiting CCR6⁺ Treg cells.³⁸²

“Hot” tumors are characterized by the abundant presence of CD4⁺ and CD8⁺ T cells surrounding tumor cells,³⁸³ emphasizing the critical need to effectively activate these anti-tumor T cells and produce a high concentration of T cell-attracting chemokines in the peritumoral lymph nodes.³⁸⁴ The positioning of T cells within tumors closely correlates with the expression of chemokines such as CCL2, CCL4, CCL5, CXCL9, and CXCL10. Stimulation of TNFSF14 initiates signaling via the lymphotoxin-β receptor, further enhancing an inflammatory microenvironment enriched with T cells. Notably, CD8⁺ effector T cells can enhance recruitment by releasing chemokines that attract similar cells already present in the tumor microenvironment.³⁸⁵ Tumor cell death can activate TLR3, prompting the production of type I interferons and subsequently CXCL10. Compared to “cold” tumors, “hot” tumors exhibit a higher mutational burden, leading to increased production of TAAs that enhance the immune system’s ability to recognize and eliminate tumor cells.³⁸⁵ In contrast to non-inflammatory tumors, inflammatory tumors demonstrate elevated expression of immunosuppressive genes like IDO, Foxp3, and PD-L1, activated as a response to positive anti-tumor immune reactions. Consequently, over time, T cells often face exhaustion, influenced by the chemokines present in the tumor microenvironment, the full extent of which remains incompletely understood.

Throughout various stages of tumor growth, the activation or inhibition of specific chemokine networks significantly impacts the immune milieu within the tumor microenvironment (TME), consequently influencing tumor biology and responses to therapy. However, the mere presence of chemokines cannot solely determine immune composition. The establishment of the tumor immune landscape is shaped by numerous factors, including genetic mutations within tumors, epigenetic changes, the host’s genetic makeup, the tissue origin of the tumor, microbial profile, and the intricate dynamics during cancer immune remodeling processes. Despite the complex interplay, evaluating cytokine and chemokine levels within tumors at distinct time points, along with the array of immune and non-immune cells present, is vital in specific contexts. This comprehensive evaluation assists in tailoring personalized therapeutic approaches for individual patients.

The initial recognition of the involvement of chemokines in tumor formation stemmed from investigations using viral and bacterial infection models, highlighting potential anti-tumor effects throughout tumor progression (Fig. 6). For instance, cDC1s mature through the processing of TAAs and response to inflammatory signals. This maturation process triggers the upregulation of CCR7 expression, vital for the migration of cDC1s from tumor sites to TDLNs. Deficiency in CCR7 within the cDC1 subset hinders anti-tumor T cell response, fostering tumor growth.³⁸⁶ Additionally, tumor-derived ligands binding to the liver X receptor (LXR) have been found to hinder the migration of DCs to TDLNs.³⁸⁷ Recent findings indicate that the CCR7-dependent migration of cDC2 from the TME to TDLNs is crucial for triggering the anti-tumor activity of CD4⁺ T cells. However, the recruitment of cDC2 is often obstructed by the suppressive activity of Treg cells.³⁸⁸ Tumor development induces changes in chemokine expression not only within the TME but also in surrounding tissues and lymphoid organs, which ultimately compromises immune cell activation. The TDLNs play a pivotal role in activating anti-tumor T cells, acting as a central hub for initiating this critical process. To effectively activate CD8⁺ T cells and CD4⁺ T cells targeting TAAs, a series of chemokine-mediated interactions with both mobile and stationary DCs in various lymph node regions is essential.³⁸⁹ Notably, factors originating from tumors have the ability to suppress the production of the stromal cell-secreted chemokine CCL21 in lymph nodes responsible for tumor drainage. The reduction in CCL21 expression, crucial for attracting T cells and DCs to lymph nodes, is apparent in TDLNs. This reduction in CCL21 levels is correlated with alterations in the distribution of immune cell populations, potentially leading to impaired T cell activation.³⁹⁰ Tumor-induced factors disrupt the normal process of hematopoiesis in the bone marrow, leading to the generation of granulocytic MDSCs.³⁹¹ The regulation of immune cell development and maturation in the bone marrow is significantly influenced by the CXCR4-CXCL12 chemokine axis.³⁷⁵ Solid tumors boost the expression of CXCL12 in stromal cells, resulting in the accumulation of myeloid cells expressing CXCR4, ultimately causing disruptions in hematopoiesis.³⁹² Further research is required for a comprehensive understanding of how tumor-derived elements impact the expression of chemokines in lymphatic organs like the spleen and thymus, and their role in the generation and activation of immune cells.

It is intriguing to observe that the same chemokine pathways often attract immune cells that both support tumor growth and inhibit tumor progression (Fig. 6). For example, chemokines like CXCR3 and its associated molecules CXCL9 and CXCL10, along with CCR5 and its ligand CCL5, can promote the recruitment of Treg cells and CD8⁺ T cells. As a result, the composition of immune cells may be shaped by the changing levels of chemokines in the TME and alterations in the expression of chemokine receptors on immune cells. There is variation among Treg cells in tumors, with different Treg subsets dominating different stages of tumor development.³⁹³ In the context of infection, the ability of Treg cells to suppress T cell responses depends on their expression of specific transcription factors and chemokine receptors that correspond to various T cell subsets.³⁹⁴ Treg cells expressing CXCR3 gather in the TME before Treg cells expressing CCR4 and CCR8.³⁹⁵ The emergence of CXCR3⁺ Th1 and CD8⁺ T cells has raised questions about the recruitment of CXCR3⁺ Treg cells. Additionally, more clarity is needed regarding the functional roles of different Treg cells subgroups in tumor progression. NK cells possess diverse functional characteristics, as seen in the unique expression patterns of CXCR3 and CX3CR1 in CD27^low and CD27^hi subsets. The expression of chemokines in the TME and the presence of corresponding chemokine receptors on NK cells may impact the recruitment efficiency of various NK cell subgroups into tumors and thus potentially influence the overall clinical outcome. The composition of immune cell subpopulations in the TME is significantly influenced by the presence of diverse immune cell subsets, adding complexity to TME interactions. For instance, the migration of conventional cDC1s to tumors relies on chemokines released by NK cells, such as XCL1 and CCL5.⁴⁵ Notably, tumor-infiltrating cDC1s play a crucial role in producing CXCL9 and CXCL10, key chemokines that attract CD8⁺ effector T cells to the TME.^396,397 Unlike other chemokine axes with overlapping functions, the XCL1-XCR1 axis uniquely recruits cDC1, indicating a potential therapeutic target to enhance cDC1 and T cell mobilization, and facilitate their interaction, thereby boosting T cell expansion and effector function acquisition.³⁹⁸ Moreover, an increased presence of MDSCs is associated with decreased CXCL11 expression levels and reduced infiltration of CD8⁺ T cells.³⁹⁹ Chemokine-mediated cellular feedback mechanisms may amplify subtle distinctions in early anti-tumor and pro-tumor responses as tumors progress.

The increasing recognition of the pivotal role of chemokines in modulating the properties and efficacy of immune cells within the TME is gaining momentum. An initial investigation reveals that the binary classification of cells as either pro-tumor or anti-tumor is inadequate to fully comprehend the complexity of cellular dynamics in the TME. For example, the phenotype of CD8⁺ T cells targeting tumors evolves throughout tumor progression. Distinct signatures of CD8⁺ T cells, such as TCF-1⁺ and CX3CR1⁺, are correlated with effective anti-tumor immune responses.^98,99,238 Similarly, specific features of Treg cells, like CCR8⁺ Treg cells, may be linked to tumor advancement^400,401 or response to PD-1 immunotherapy (IFNγ⁺ Treg cells).⁴⁰² The heterogeneous and malleable characteristics of TAMs and TANs result in distinct pro-tumor (M1 TAMs and N2 TANs) and anti-tumor (M2 TAMs and N1 TANs) phenotypes. This diversity potentially encompasses a spectrum of intermediate cellular states.^403–405

The maintenance of diverse cellular states within the TME and the potential for therapeutic manipulation of these states remain uncertain. The TME exhibits variations in oxygen levels, nutrient availability, and acidity, known factors that impact the characteristics and functions of immune cells. Therefore, the control of immune cell distribution in the TME through chemokines could have a profound influence on immune cell properties. Substantial alterations in chemokine receptor expression are commonly linked to the maturation and specialization of immune cells. For instance, as TCF-1⁺CD8⁺ T cells transition into effector CD8⁺ T cells and subsequently into exhausted T cells, there is a reduction in specific chemokine receptors such as CXCR5 and CXCR3, alongside an elevation in others like CXCR6 and CX3CR1.^225,406 Moreover, in human tumors, TCF-1⁺CD8⁺ T cells are situated in a microenvironment rich in antigen-presenting cells.⁴⁰⁷ The upregulation of CXCR3 on these cells is crucial for their migration into the TME facilitated by the CXCL9 and CXCL10 ligands produced by cDC1s, promoting their revitalization and expansion in the TME.^408,409 Exposure to IL-12 allows these cells to acquire effector functions.⁴¹⁰ Additionally, the CXCR3 pathway and its ligands are postulated to play a role in directing the positioning of active immune cells in proximity to cancerous cells.⁴¹¹

The characteristics exhibited by TAMs and TANs are likely influenced by specific cues within the tumor microenvironment (TME) (Fig. 6). In the TME context, the interaction between CXCR4 and CXCL12, for example, facilitates the differentiation of recently arrived TAMs into perivascular TAMs.⁴¹² Resident TAMs play a crucial role in enhancing vascular permeability and assisting in the intravasation of tumor cells in the TME. Moreover, the inhibitory functions of TAMs in the TME can be modulated by the CCL5-CCR5 pathway. Disruption of CCR5 could induce a shift in TAM characteristics towards a pro-tumor profile, offering significant therapeutic benefits for individuals with colorectal cancer.⁴¹³ Key chemokine signaling pathways, such as CCR2-CCL2 and CCR5-CCL5, play vital roles in mediating communication between tumor cells and TAMs, thus promoting tumor stemness, metastasis, and resistance to therapies.^414,415 The immunosuppressive role of Treg cells in the TME is intricately linked to antigen recognition, which could trigger the recruitment of CCR8 and/or CCR4-dependent Treg cells towards DCs within the TME.⁴¹⁶ The maintenance of a suppressive phenotype in Treg cells is supported by CCR8 expression. This is attributed to the enhancing effect of the CCL1-CCR8 signal on Foxp3 expression, a relationship associated with Treg cells phenotype stability.^402,417 Additionally, CCR8⁺ Treg cells display elevated levels of suppressive markers, including CD25, CTLA4, CD39, TIGIT, PD1, ICOS, OX40, and Helios, compared to CCR8^– Treg cells.⁴¹⁸ Conversely, CCR4 plays a crucial role in facilitating communication between Treg cells and DCs within lymph nodes, suggesting a potentially similar role within the TME.⁴¹⁹ In the TME setting, activation of CCR6 signaling can promote the local expansion of Treg cells.⁴²⁰ Exploring the impact of chemokines on shaping cellular niches in the TME offers a promising avenue for future research. Although addressing chemokines and their receptors as therapeutic targets presents challenges, understanding their complexities may aid in identifying key cellular components involved in fostering or impeding tumor growth and immune responses. The burgeoning field of TME research is increasingly captivating, focusing on the intricate interplay between immune cells and chemokines, offering valuable therapeutic insights for modulating immune responses in cancer.

The essence of immunotherapy lies in triggering the patient’s immune system to combat tumor cells by rectifying effector cell dysfunctions and reducing the inhibitory immune cell population. Research indicates that the immune landscape of tumors can forecast the responsiveness to immunotherapy - favorable tumor environments (“hot” tumors) typically exhibit better responses to treatment, whereas cold tumors with minimal immune infiltration often present suboptimal responses. To bolster the efficacy of immunotherapy, there is a growing emphasis on integrating multiple cancer treatment modalities to overcome tumor drug resistance and enhance the reactivity of tumors less sensitive to traditional immunotherapy. A critical aspect of this approach involves regulating chemokine expression during tumor treatment, a factor crucial for both treatment effectiveness and patient tolerance.^421,422

Immune checkpoint blockade (ICB) is a therapeutic strategy that enhances anti-tumor responses by freeing T cells from the constraints imposed by checkpoint molecules on their surface, thereby restoring T cell functionality. The success of checkpoint blockade therapy is closely linked to the intrinsic immune responses of patients.⁷⁹ Patients with “hot” tumors, particularly those adept at attracting a substantial T cell influx, tend to derive greater benefits from these interventions.^62,423 The effectiveness of cancer treatment hinges not only on the infiltration of T cells into the tumor microenvironment. Recent studies have highlighted the correlation between the efficacy of checkpoint inhibition therapy and the presence of specific T cell subpopulations expressing the TCF-1 transcription factor within tumor sites.^95,424 Moreover, the chemokine receptor CXCR3 and its ligand CXCL9 play a pivotal role in enhancing the effectiveness of anti-PD-1 therapy in tumor studies.⁴²⁵ Inhibition of PD-1 leads to an upsurge in CXCL9 expression on conventional dendritic cells (cDC1), aiding in the targeted activation of CXCR3 expressing CD8⁺ T cells. This tactic helps prevent T cell exhaustion and initiates anti-tumor responses that effectively eliminate tumor cells. CXCL9 expression not only underpins clinical responses in the context of anti-PD-1 therapy but also significantly influences the success of TIM-3 blockade therapy in preclinical experiments. By stimulating CXCL9 production driven by cDC1, anti-TIM-3 therapy activates robust CD8⁺ T cell responses within the tumor, heightening sensitivity to paclitaxel chemotherapy.⁴²⁶ Therefore, the limited efficacy in highly inflamed tumor patients could be attributed to the reduced presence of TCF-1⁺ cells and/or impediments hampering their interaction with cDC1 in the tumor microenvironment.

Therapies utilizing CAR-T cells and ex vivo expanded autologous T cell infusions have demonstrated potential therapeutic benefits. However, a significant challenge confronting these treatments for solid tumors is the abnormal vascular structure and the suppressive tumor microenvironment, greatly impeding the infiltration of transplanted T cells into the tumor mass. Consequently, for patients with “cold” tumors and those undergoing TME phenotypic alterations, cell transfer therapy alone may not suffice for effective treatment. Research endeavors are underway to bolster the tumor penetration and functionality of T cells by investigating various chemokine systems in murine tumor models, including targeting CCR4,⁴²⁷ CXCR2,⁴²⁸ and CX3CR1.⁴²⁹ Despite these endeavors, there is an urgent need to translate preclinical findings into clinical practice to ascertain whether modulating chemokine systems can indeed enhance the efficacy of immune cell transfer therapy in cancer treatment.

Initially intended for tumor cell destruction, radiotherapy and chemotherapy are now acknowledged for their additional capacity to incite immune responses. By boosting the presentation of tumor-specific antigens on tumor cells, these treatments stimulate immune responses from anti-tumor T cells. Ongoing studies are exploring the synergistic effects of combining ICB with radiotherapy or chemotherapy. Hence, persistent efforts are being made to devise innovative radiotherapy and chemotherapy strategies to effectively augment T cell responses. Nevertheless, the collaborative interplay between these two treatment modalities may be hindered by CCR7⁺ cDC1 cells.^430,431 Therefore, the efficacy of treatment protocols may be restricted if obstacles in recruiting cDC1 cells impede their accumulation in the tumor microenvironment, unless the treatment itself can enhance this process. Animal model research has revealed that anthracycline drugs can elevate the levels of CCL2 in the tumor microenvironment, thereby attracting antigen-presenting cells and facilitating robust immune responses against the tumor.⁴³²

Radiotherapy and chemotherapy have been demonstrated to elevate the release of chemokines in the tumor microenvironment. For example, radiotherapy application in a murine breast cancer model increases the expression of CXCL16, a key element facilitating the recruitment of CXCR6⁺CD8⁺ effector T cells to the tumor microenvironment.⁴³³ In murine melanoma models, radiotherapy induces the release of both type I and type II interferons, leading to heightened levels of CXCL9 or CXCL10, thereby promoting the infiltration of effector T cells expressing CXCR3.⁴³⁴ Chemotherapy induces the secretion of chemokines like CCL5, CXCL9, and CXCL10 in the tumor microenvironment, attracting CD4⁺ and CD8⁺ T cells to the tumor site. The expression of these chemokines correlates with the influx of CD4⁺ and CD8⁺ T cells, influencing tumor management and patient prognosis.⁴³⁵ Furthermore, in chemotherapy treatment regimens involving anthracyclines and conventional therapies, tumor cells exhibit increased levels of CXCL10 expression, a crucial factor in enhancing anti-tumor T cell responses.³⁸⁵

Tumor cells inherently resist treatment due to immune restrictions at therapy initiation. Additionally, tumors evolve continuously throughout treatment, incorporating various factors related to the local tumor microenvironment and the host. Treatment-induced alterations in chemokine expression may potentially contribute to resistance or recurrence in tumors. Elevated expression of chemokines such as CCL2 and CCL5 post-radiotherapy attracts immunosuppressive cell subgroups, including monocytes expressing CCR2 and CCR5, MDSCs, and Treg cells expressing CCR2, thereby assisting in cancer progression.^436–439 Furthermore, therapy-triggered recruitment of CCR5-dependent macrophages contributes to tumor recurrence.⁴⁴⁰ Resistance to anti-BRAF inhibitor therapy in melanoma patients is linked to the CCL2-CCR2 pathway, where elevated CCL2 levels are associated with poor treatment response.⁴⁴¹ Studies in a mouse melanoma model suggest a potential connection between anti-BRAF inhibitor resistance and CCR2⁺MDSC infiltration into the tumor microenvironment.⁴⁴² Recent research identifies a novel chemotherapy resistance mechanism involving CCL2, where chemotherapy drugs prompt tumor cells to release extracellular vesicles, inducing CCL2 expression in vascular endothelial cells, attracting CCR2⁺Ly6C⁺ monocytes, and promoting tumor growth.⁴⁴³

Moreover, the CCR6-CCL20 axis may contribute to treatment resistance by recruiting CCR6⁺ Treg cells.⁴⁴⁴ Elevated levels of CCL20⁺ and CCR6⁺ Treg cells have been closely correlated with chemotherapy resistance in patients with colorectal cancer, and triple-negative breast cancer.^445,446 In a mouse model of bladder cancer, resistance to PD-L1 immunotherapy is mainly linked to Gr1⁺ neutrophil infiltration into the tumor microenvironment, dependent on the CCL20 for immune cell attraction.⁴⁴⁷ This underscores the frequent occurrence of chemotherapy resistance in T cell therapy, the detailed mechanisms of which require further exploration. Considering the dual role of chemokines in modulating immune responses in cancer, their complex functions in treatment outcomes and resistance, and their impact on pro-cancer immune responses and immune suppression, chemokines play a critical role in delivering immune responses to the tumor microenvironment. Moreover, specific chemokine expression may influence treatment responses, potentially serving as biomarkers for predicting and monitoring treatment outcomes. Inducing or redirecting chemokine receptor expression in anti-tumor T cells represents a novel emerging therapeutic approach.

Tumors are complex structures composed of a variety of cells, including tumor cells and various stromal components. The interactions between tumor cells and stromal elements play a crucial role in shaping the behavior of solid tumors. Tumor cells possess the ability to alter the surrounding stroma, creating a supportive microenvironment for their own growth. Interestingly, tumor cells can undergo transdifferentiation, transforming into cells resembling stromal cells through various signaling pathways.⁴⁵³ This transformation enhances tumor angiogenesis and drives cancer progression.⁴⁵⁴ The tumor stroma significantly impacts tumor formation, cancer progression, and resistance to therapy, thereby influencing various characteristics of cancer. Key stromal elements comprise the extracellular matrix, vascular system, activated cancer-associated fibroblasts, mesenchymal cells, and other cell components, which influence anti-tumor immune responses and ultimately determine the trajectory of tumor progression (Fig. 7).

When considering targeted therapy based on the tumor stroma, the focus shifts towards non-malignant cellular components, rather than solely on tumor cells as in conventional approaches. The advancements in precision medicine have accelerated the clinical applications of molecular targeted therapy. The occurrence and metastasis of cancer result in the development of a complex stromal environment within the host tissues, directly driving cancer progression. Therefore, the tumor stroma is seen as a crucial target for advancing therapeutic approaches, providing the opportunity to improve current treatment options and customize treatments to individual needs. Consequently, the tumor stroma is deemed a critical focal point for designing successful therapeutic strategies to enhance current treatment modalities and achieve personalized treatment goals. Strategies that target the tumor stroma involve directly influencing the cells or non-cellular elements within the stroma, as well as implementing novel approaches to remodel or normalize the tumor stroma to hinder or reverse tumor progression. By summarizing the latest research advancements on stromal components, this article highlights their potential therapeutic value and advocates for the translation of laboratory research into clinical applications.

Tertiary lymphoid structures (TLS) are formed in response to persistent inflammatory stimuli, such as infections, transplant rejections, autoimmune diseases, and malignant tumor tissues.⁵⁷⁸ These structures structurally resemble secondary lymphoid organs (SLO) and comprise B cell follicles enclosing follicular dendritic cells (FDCs). Within the T cell zone, high endothelial venules (HEVs) aggregate together, accompanied by fibroblastic reticular cells (FRCs) present within TLS. In contrast to secondary lymphoid organs (SLOs), TLS lack capsules or afferent/efferent lymphatic vessels.⁵⁷⁹ The predominant T cell types in TLS include CD4⁺ follicular helper T cells (TFH), CD8⁺ T cells, CD4⁺ Th1 cells, and CD4⁺ Treg cells.⁵⁷³ Notably, stromal cell-like T cells are widely distributed throughout TLS, suggesting a significant role in their differentiation, support, and protection.⁵⁸⁰ Dendritic cells (DCs) are situated within the T cell zone, and their abundance correlates with the degree of Th and CD8⁺ T cell infiltration, thereby promoting B cell expansion and antibody production.⁵⁸¹ The stromal cells in TLS consist of fibroblasts, FDCs, and FRCs. FDCs facilitate communication between antigen-presenting cells and B cells in the primary lymphoid follicles, while FRCs, located in the T cell zone, provide structural support for TLS.⁵⁸²

Cancer is commonly perceived as a continuous and dynamic inflammatory process involving tissue damage, cytokine production, chemotaxis of immune cells, cell proliferation in response to growth factors, and stimulation of tissue repair.⁵⁸³ Tertiary lymphoid structures (TLS) within tumors are recognized as a vital immunogenic trait especially evident in highly immunogenic tumors.⁵⁸⁴ The immune response within TLS is governed by tumor genetic characteristics, encompassing neoantigens, tumor mutational burden (TMB), expression of immune checkpoints, and immune cell infiltration.^585,586 Cellular therapies commonly address neoantigens, which encompass both tumor-associated and tumor-specific antigens.⁵⁸⁷ Additionally, immune checkpoint expression levels within TLS can trigger responses that facilitate tumor progression, such as activation of indoleamine 2,3-dioxygenase 1 (IDO1),⁵⁸⁸ upregulation of PD-L1,⁵⁸⁹ recruitment of MDSCs,⁴³⁸ and induction of T-cell exhaustion.⁵⁹⁰

The maturation and localization of TLS are influenced by various factors, shaping their functionality within tumors.⁵⁹¹ TLS functionality is significantly impacted by cellular diversity at different developmental stages. Primary follicular TLS structures, rich in B cells, exhibit a potential target for antibody therapies. In contrast, secondary follicular TLS structures comprise mature T and B cells, rendering them more responsive to T-cell therapies and immune checkpoint blockade.⁵⁹² Mature TLS, in opposition to immature TLS, demonstrate heightened immune activity and prognostic significance.⁵⁹³ The importance of understanding the influence of tumor immune characteristics on TLS maturation is highlighted by the observed disparities between mature and immature TLS across various malignancies. Predominantly located in the peritumoral area, mature TLS display T-cell activation signals, plasma cell expansion, and CXCL13 expression.⁵⁹⁴ In melanoma, IKZF1 has been recognized as a key player in the formation of immature TLS, significantly impeding TLS maturation without affecting melanoma cell proliferation.⁵⁹⁵ Notably, CD8⁺ T cells exhibit distinct functionalities based on their origin from either the inflammatory environment or the tumor itself. Tumor-infiltrating lymphocytes originating from inflammation tend to create a conducive milieu for tumor cell proliferation, while those arising from the tumor site typically support antitumor immunity.⁵⁹⁶ Hence, further exploration of the spatial distribution of immune cells within TLS is warranted. The establishment and efficacy of TLS within tumors and adjacent tissues are substantially influenced by the tumor’s genetic attributes and the nature of the inflammatory response. The composition, localization, and maturation of TLS have been correlated with their immune response.

Unprecedented responses to ICB therapy have been observed across various types of cancers.⁵⁷⁹ However, not all patients demonstrate favorable responses, emphasizing the pressing need to identify precise biomarkers for response prediction. The tumor mutation burden is recognized as a potential predictive indicator linked to treatment response, although current biomarkers lack the ability to consistently forecast responses to immunotherapy.⁵⁸² Notably, the presence of pre-existing anti-cancer T cells and the level of PD-L1 expression within the tumor microenvironment are pivotal factors in predicting the efficacy of anti-PD-1 therapy for patients with melanoma and non-small cell lung cancer (NSCLC).^581,597 In the context of ICB therapy, PD-L1 plays a critical role in revitalizing CD8⁺ T cells by inhibiting their activity, thereby transforming them into potent cytotoxic cells that target tumor cells. This mechanism involves the expression of PD-L1 on both tumor cells and myeloid cells.

Tertiary lymphoid structures (TLS) serve as crucial sites for generating memory effector cells responsible for managing tumor recurrence, presenting a unique opportunity to inform forthcoming clinical investigations in the rapidly evolving field of immuno-oncology. A study conducted in patients with non-small cell lung cancer demonstrated that those who underwent neoadjuvant therapy with anti-PD-1 agents showed an increased formation of TLS within tumors after surgical intervention. Additionally, the identification of PD-1⁺CD8⁺ T cells within TLS prior to treatment initiation had the potential to predict the efficacy of anti-PD-1 therapy.⁵⁹⁸ Furthermore, the induction of TLS in regressing lesions is associated with the response of patients with cervical intraepithelial neoplasia to human papillomavirus (HPV) vaccines, the primary causal agent of cervical cancer.⁵⁹⁹ TLS not only serve as markers of therapeutic immune responses in cancer but also contribute to enhancing anti-tumor immune reactions. The induction of TLS in tumors is believed to aid in recruiting lymphocytes and controlling tumor progression, potentially broadening the application of adjuvant immune therapy in both immunologically “cold” and “hot” tumors.

While TLS are observable in diverse malignancies, further research is warranted to clearly characterize, distinguish, and quantify TLS, as well as to conduct large-scale, multicenter studies. Moreover, the integration of single-cell sequencing and spatial transcriptomics is essential for exploring the relationship between cancer driver genes and TLS formation. Improving the understanding of TLS functionality and its roles across different tumor types amidst the complexity of cancer can hasten the development of enhanced therapeutic strategies. This, in turn, can boost tumor responsiveness to immunotherapy, leading to improved treatment outcomes and enhanced survival rates.

While each tumor carries a distinct microbial composition, the majority in cancer typically consist of the phyla Firmicutes and Proteobacteria.⁶⁰⁴ Additionally, non-gastrointestinal tumors, like those in breast, lung, and ovarian cancers from the Actinobacteria phylum, often exhibit a prevalence of Micrococcaceae and Corynebacterium.⁶⁰⁵ Helicobacter pylori, a widely recognized cancer-associated bacterium, acts as a significant risk factor for gastric cancer and can also colonize colorectal tumors, thereby facilitating liver metastasis.^606,607 The intricate ecosystem within tumors comprises a diverse microbial community, and obtaining a comprehensive understanding of its composition is vital for unveiling the bacterial contribution to cancer pathogenesis and tumor progression. In addition to bacteria, fungi play a significant role in cancer susceptibility. In pancreatic ductal adenocarcinoma (PDAC), the fungal community shows significant enrichment in Malassezia species.⁶⁰⁸ Moreover, fungi have been detected in colon, lung, prostate, stomach, and skin cancers.⁶⁰⁹ Chlamydial infections are widespread across various cancer types, with a notable increase observed in small-cell lung cancer tumor tissues compared to healthy controls.⁶¹⁰ The pathological impact of Chlamydia in the tumor microenvironment remains a contentious topic despite its potential for malignant transformation and tumorigenesis. Viral infections are also intricately linked to cancer development. High-risk HPV types, including HPV16 and HPV18, exert a substantial pathogenic influence on cervical cancer.⁶¹¹

The interaction between the microbiota and the tumor microenvironment is complex and diverse, encompassing direct cell-microbe interactions and metabolite-mediated crosstalk. Pathogen-associated molecular patterns (PAMPs) are conserved elements found in bacteria, viruses, and fungi that are recognized by the pattern recognition receptors (PRRs) of the host innate immune system.⁶¹² Microbial compounds containing PAMPs can be classified into categories such as carbohydrates, nucleic acids, lipids, and proteins. Lipopolysaccharides (LPS), extensively studied PAMP-carrying molecules present in various Gram-negative bacteria, are notable for their potent inflammatory properties, often referred to as endotoxins.⁶¹² PRRs, which include Toll-like receptors (TLRs), C-type lectin receptors (CLRs), RIG-I-like receptors (RLRs), and NOD-like receptors (NLRs), are conserved receptors located on cell membranes or within cells.⁶¹³ PRRs recognize PAMPs, initiating intracellular signaling cascades in immune cells, leading to the production of inflammatory mediators that drive systemic immune responses and adaptive immunity. The crucial interaction between PAMPs and PRRs plays a key role in stimulating either anti-tumor or pro-carcinogenic innate immune responses within the tumor microenvironment.⁶¹⁴ Integration of cancer vaccines featuring PAMPs can reshape the immunosuppressive tumor microenvironment, enhancing anti-tumor immune responses and the establishment of immune memory.⁶¹⁵ Overall, microbial-derived PAMPs play a significant role in modulating the tumor microenvironment and immune responses.

Studies have revealed that small molecule metabolites play crucial regulatory roles, capable of influencing cellular metabolism and shaping the immune system, leading to either positive or negative effects in the tumor microenvironment. Short-chain fatty acids (SCFAs), a subset of fatty acids generated through bacterial fermentation, are involved in gene expression modulation. This function is achieved by inhibiting histone deacetylases (HDACs) and activating G protein-coupled receptors (GPCRs). Such regulation impacts essential processes including metabolism, inflammatory responses, and tumor development.⁶¹⁶ Bile acids are steroidal compounds primarily found in bile and are classified into primary bile acids like cholic acid and chenodeoxycholic acid, and secondary bile acids such as deoxycholic acid and lithocholic acid. Primary bile acids are synthesized by the liver, while secondary bile acids are formed from the modification of primary bile acids by bacteria in the colon.⁶¹⁷ Bile acids have a critical role not only in digestion but also in cancer regulation, with secondary bile acids recognized as significant promotors of tumor growth. Polyamines, low-molecular-weight metabolites containing multiple amino groups, are produced by gut bacteria and include compounds like putrescine, spermidine, and cadaverine.⁶¹⁸ Due to their antioxidant properties, polyamines can shield cells from oxidative stress and modulate immune responses by influencing macrophage function and polarization.⁶¹⁸ Methylglyoxal (MGO) is an active carbonyl compound generated endogenously during glycolysis, and reactive carbonyls like MGO cause DNA damage, known as DNA glycation, leading to mutations, DNA breaks, and cell toxicity. Elevated MGO levels can compromise cell viability, while lower levels benefit tumor cell proliferation.⁶¹⁹ Concerning secondary metabolites, colibactin, a mutagenic compound produced by pathogenic Escherichia coli, can promote colorectal cancer development by inducing DNA double-strand breaks.^620,621 Peptidyl aldehydes, produced by bacteria such as Escherichia coli, Streptomyces, and Bacillus subtilis, enhance carcinogenicity by inhibiting protease activity.⁶²² Thiopeptides, known for their potent antibacterial properties, are metabolized by intestinal clostridia, urogenital lactobacilli, and cutibacterium acnes on the skin. These compounds may also display anticancer effects through proteasome targeting.^623,624 Overall, microbial metabolites, with their direct and indirect impacts on cancer development, underline the significance of exploring tumor microbiota metabolism. This approach is crucial for advancing our understanding of the tumor microenvironment and developing innovative cancer treatment strategies.

The presence of a tumor microbiome represents an emerging frontier in the field of tumor immunity and may exert a more potent influence compared to the microbiome in the body or intestines due to its strong local effects.⁶²⁵ For example, the tumor microbiome can influence cytokine production, leading to enhanced immune responses. Pattern recognition receptors (PRRs) are crucial in initiating the innate immune response by recognizing conserved microbial components known as pathogen-associated molecular patterns (PAMPs), including lipopolysaccharides and lipoproteins. Activation of the innate immune system serves as the initial recognition step, followed by the activation of the adaptive immune system, vital for the body’s defense against tumors. CD8⁺ T cells, essential components of the adaptive immune system, could potentially be affected by the tumor microbiome, influencing their presence.⁶²⁶ Research indicates that the presence of bacteria in tumors can boost the movement and infiltration of cytotoxic CD8⁺ T cells, potentially leading to improved outcomes for individuals with melanoma. This process is regulated by chemokines such as CCL5, CXCL9, and CXCL10.⁶²⁷ Microbiota-induced IFN can regulate tumor-targeting anti-tumor monocytes, enhance anti-cancer immunity, and boost the efficacy of immune checkpoint blockade, while the absence of microbiota can skew the tumor microenvironment toward pro-tumor macrophages.⁶²⁸

On the other hand, the tumor microbiome may induce T cell dysfunction and immune suppression through cytokine production, upregulation of immune checkpoint inhibitors, and recruitment of immune-suppressive cells.⁶²⁵ In prostate cancer, for instance, Lactobacillus johnsonii has been shown to increase the recruitment of regulatory T cells and enhance the functions of immune-inhibitory elements like PD-L1, CCL17, and CCL18. Consequently, an immune-suppressive tumor environment may be created.⁶²⁹ Furthermore, lipopolysaccharides have the ability to induce IL-6 production by activating the NF-kB signaling pathway, subsequently initiating the JAK-STAT3 signaling cascade. This activation leads to the mobilization of MDSCs and an upsurge in PD-1 expression, ultimately facilitating immune suppression.^630,631 Additionally, microbial metabolites can also contribute to immune suppression. Tryptophan metabolites generated by Lactobacillus can activate the aryl hydrocarbon receptor (AhR) in tumor-associated macrophages. AhR activation is pivotal in the rapid progression and increased fatality rate of pancreatic carcinoma. Conversely, reducing dietary tryptophan intake can lower AhR activity in macrophages, impeding tumor growth.⁶³² In summary, the tumor microbiome and its products can either enhance or suppress immunity, leading to varying effects on immune modulation. These variances stem from the specific microbial composition and distribution across different tumor types. The microbiome plays a critical role in influencing responses to immune therapies, given its diverse impact on tumor immunity. Current research suggests that the gut microbiome may affect the efficacy of anti-PD-1/L1 and anti-CTLA-4 therapies.⁶³³ Besides the gut microbiome, the characteristics of the tumor microbiome can significantly influence responses to immune therapies. CP1, a patient-derived prostate microbiota with local immune-stimulatory properties, can reprogram “cold” tumor microenvironments and enhance the therapeutic outcomes of anti-PD-1 immune therapy.⁶³⁴

Short-chain fatty acids (SCFAs) produced through microbial fermentation in the gut are a group of organic acids with diverse therapeutic benefits in fighting various cancers. However, their clinical use is limited due to rapid renal clearance and adverse reactions at high doses.⁶³⁵ An innovative SCFAs prodrug, based on amphiphilic copolymers, was created by conjugating SCFAs (such as butyric acid or propionic acid) to hydrophobic polymer segments via ester bonds. This formulation spontaneously forms nanocarriers suitable for oral administration, effectively inhibiting melanoma growth and metastasis.⁶³⁶ Lipopolysaccharide (LPS), a product of gram-negative bacteria, was targeted by developing a fusion protein incorporated into lipid-protein-DNA (LPD) nanoparticle systems. Selectively blocking LPS signaling led to a significant improvement in anti-PD-L1 therapy.⁶³⁷ Recent studies have shown that D-lactic acid, a gut microbiota metabolite, can convert immunosuppressive M2 macrophages to an M1 phenotype by regulating PI3K/Akt and NF-kB pathways. This study used this approach to design biologically-mimetic PLGA nanoparticles containing D-lactic acid to transform the immunosuppressive microenvironment in hepatocellular carcinoma tumors, offering a combined immunotherapy approach for this condition.⁶³⁸ In addition to microbial metabolites, the use of engineered bacterial flagellin proteins to modulate the tumor microenvironment is a promising and innovative strategy for cancer treatment.⁶³⁹ By engineering an attenuated strain of Salmonella Typhi to overexpress and secrete exogenous pathogenic Vibrio cholerae flagellin protein B (FlaB), an effective immunotherapeutic adjuvant was created that activates the TLR5 signaling pathway to initiate innate immune responses.⁶⁴⁰ Furthermore, specialized immunotherapeutic adjuvant clostridia, with peptidoglycan-remodeling abilities, secretes homologous NlpC/p60 peptidoglycan hydrolase SagA to produce immunomodulatory muropeptides, thus enhancing PD-1 checkpoint inhibitor cancer immunotherapy.⁶⁴¹

Bacterium-based artificial expression and release of therapeutic agents are now widely produced using gene engineering techniques, thanks to the progress in synthetic biology. This is because bacteria have a natural predilection for colonizing hypoxic and necrotic tumor microenvironments.⁶⁴² The technique involves introducing tailored plasmids into bacterial cells to regulate the synthesis of specific proteins. Through the application of this approach, bacteria have been genetically engineered to produce nanobody drugs (such as CD47, PD-L1, and CTLA-4) for the treatment of tumors by locally expressing them. This facilitates targeted delivery of enhanced therapeutic doses while minimizing systemic adverse effects.⁶⁴³ Furthermore, a genetically modified probiotic was precisely engineered to reduce the immunological adverse reactions associated with concurrent treatment using anti-PD-L1 and anti-CTLA-4 agents. This probiotic, created by introducing plasmids into Escherichia coli Nissle (EcN), aims to deliver checkpoint blockade nanobodies directly to tumor sites.⁶⁴³ The probiotic employs a cleavage and release mechanism that relies on the LUXL and 4X174E genes to simultaneously express and release nanobodies resistant to PD-L1 and CTLA-4. Harnessing the crucial role of L-arginine in anti-tumor T cell responses and its synergistic effect with PD-L1 blockade, an engineered EcN strain was developed by deletion of the arginine repressor (ArgR), proficient in colonizing tumors and converting ammonia to L-arginine.⁶⁴⁴ The metabolism of the tumor microenvironment can be effectively regulated by these engineered bacteria, leading to improved outcomes of immunotherapy. In addition to bacteria, viruses and other microorganisms can also be harnessed for the production and delivery of therapeutic agents within the tumor microenvironment, serving as a viable approach for cancer treatment. For instance, a research project detailed the creation of a modified oncolytic vaccinia virus that can express a PD-L1 blocker and granulocyte-macrophage colony-stimulating factor (GM-CSF) in a combined manner. The synergistic action of this combination leads to increased viral replication, inhibition of PD-L1, and stimulation of GM-CSF, all working together to boost the immune response of tumor neoantigen-specific T cells. Consequently, this strategy enables effective local viral administration and eradication of distant tumors.⁶⁴⁵

In summary, microbial communities and their metabolic byproducts play a pivotal role in shaping the tumor microenvironment. The host immune system also influences the microbiota by modulating signals related to microorganisms or metabolic functions, thereby surveilling tumors. Furthermore, cancer can alter the composition of gut microbial communities, consequently impacting the response to immunotherapy, particularly ICB. Precision medicine of the microbiome has emerged as a promising therapeutic domain in recent years. Nonetheless, there are still numerous mysteries and questions to be explored in the future. Conducting high-quality, large-scale studies can provide robust evidence for microbial communities as adjuvants in cancer immunotherapy, potential prognostic markers, or therapeutic candidates.

The effectiveness of immune regulation strategies is closely intertwined with anti-tumor immune responses, involving tumor-related or circulating immune components.^23,32,79 Clinical efficacy of anti-PD-1 monoclonal antibody therapy in human melanoma is linked to immune activation in circulating CD8⁺ T cells that have been depleted.³¹ This discovery implies that the presence of tumor-specific T cells in the bloodstream may serve as a prognostic indicator for treatment effectiveness. Additionally, the application of ICB can induce alterations in the TCR repertoire, fostering the proliferation of specific T cell clones.⁶⁴⁶ Following treatment, examination of TCR profiles in melanoma patients responsive to anti-PD-1 therapy reveals a significant presence of the TCR Vß subgroup associated with the MART1 antigen, a factor previously unidentified before treatment initiation.⁶⁴⁷ This emergence may result from inadequate sensitivity to low-expressing clones or the emergence of new immune activations, subsequently leading to the development of T cells targeting mutated neoepitopes. Preclinical studies have demonstrated that without adequately primed and dedicated antigen-specific T cells, PD-1 blockade is ineffective in stimulating anti-tumor T cell responses.⁶⁴⁸ Cumulatively, these research findings strongly support the existence of an inherent and/or peripheral anti-tumor immune response, enabling subsequent checkpoint blockade therapies to exhibit significant clinical efficacy.

Effector T cells play a crucial role in eliciting anti-tumor defenses. The presence of T cells in primary colorectal cancer is significantly associated with reduced metastasis, decreased invasion, and prolonged overall survival. The success of PD-1 inhibition therapy in advanced melanoma hinges on the presence of CD8⁺ T cells localized at the tumor periphery. The proliferation of CD8⁺ T cells is closely associated with the reduction in tumor volume observed in patients displaying a favorable response post-treatment.^649,650 Furthermore, the efficacy of PD-1 and PD-L1 inhibitors in the management of various cancers like melanoma, lung cancer, and MSI-positive colorectal cancer is closely related to the presence of a substantial mutational load in the TME.^651–653 By disrupting tolerance, ICB can unleash pre-existing immune responses against tumors. ICB are ineffective in rejecting tumors when pre-existing responses, such as “cold” tumors, are absent.^654,655 It is crucial to acknowledge that the quality of pre-existing immune responses also significantly influences the response to ICB.^656,657 An example of this phenomenon is the potential emergence of adaptive resistance due to increased expression of alternative immune checkpoint molecules like CTLA-4 and TIM3 following PD-1 treatment.^657,658 In a real-world scenario, sole inhibition of PD-1 is insufficient for the functional restoration of CD8⁺ T cells co-expressing PD-1 and TIM3, as observed in renal cell carcinoma patients.⁶⁵⁹ As a monotherapy, the response rate to ICB ranges from approximately 10% to 35%. Many late-stage solid tumors demonstrate limited efficacy in response to ICB due to inadequate immune infiltration from the primary tumor.^654,660 Extensive assessment of various immunotherapeutic agents and their combinations with standard treatments in research and clinical trials aims to augment the clinical benefits of ICB.^661,662 These combination approaches offer varying response potentials based on the tumor immune status (including “hot,” “altered,” or “cold” tumors).

In “hot” tumors, immune cells are highly active, and the microenvironment is densely infiltrated by a plethora of T cells. Immune checkpoint inhibitors (ICB) play a crucial role in reactivating the immune response of T cells against the tumor, leading to the destruction of cancer cells and imparting an immunotherapeutic effect.^10,663 Consequently, tumors characterized by an inflammatory phenotype, known as “hot” tumors, exhibit heightened sensitivity to ICB.

The response to immune checkpoint blockade (ICB) in advanced solid tumors is linked to the presence of densely infiltrating T lymphocytes within the tumor microenvironment. This infiltration is characterized by elevated expression levels of CD4 and CD8 markers on histological slides. These tumors are commonly referred to as “hot” due to their inflammation-like features of immune cell infiltration.²¹ On the other hand, non-inflamed tumors are typically unresponsive to ICB. Additionally, “immune excluded” tumors prevent the entry of anti-tumor T lymphocytes into the tumor microenvironment, confining them to the outer regions of the tumor.^10,663 To restore T cell generation in “immune excluded” tumors, various combination therapy strategies can be implemented, utilizing the key mechanisms outlined above:

“Cold” tumors, characterized by low immune scores, present a significant challenge for eradication and are often associated with poor prognoses. One treatment strategy focuses on addressing the limited pre-existing immune response by enhancing targeted therapies that promote T-cell responses. This may involve the use of vaccines, autologous T-cell transplantation (ACT), or techniques that transform tumors into vaccines. Additionally, this approach may entail blocking co-inhibitory signals by reducing ICB and providing co-stimulatory signals such as anti-OX40 or anti-GITR in combination.^10,21,749 However, a concern with this strategy is the simultaneous increase in adverse effects, a common issue in many combination therapies, requiring thorough evaluation. The concurrent activation of multiple pro-tumor mechanisms ultimately contributes to the development of “cold” tumors, necessitating a combination of strategies to achieve clinical benefits.

Current cancer immunotherapy strategies primarily focus on targeting and boosting anti-tumor adaptive immune responses. However, recent data indicate that in cancer patients, activating the tolerant innate immune system could potentially offset tumor-induced immunosuppression and modify the tumor microenvironment. This suggests that stimulating innate immune responses as a new approach to immunotherapy may result in improved treatment outcomes. Nanoparticles play a crucial role in enhancing cancer immunotherapy by targeting various innate immune pathways. For instance, a nanoparticle complex named BO-112 consisting of Poly I:C and polyethyleneimine (PEI) induces type I interferon and CD8⁺ T cell infiltration into the tumor microenvironment, demonstrating potent anti-tumor activity.⁸⁵¹ Studies have shown that local therapy with nanoparticles containing TLR7/8 agonists strongly activates macrophages, promoting proliferation of specific CD8⁺ T cells.⁸⁵² Utilizing pH-responsive poly(lactic-co-glycolic acid) (PLGA) nanoparticles loaded with TLR7/8 agonists has enabled precise intratumoral drug delivery. This strategy has successfully boosted the activation of antigen-specific CD8⁺ T cells and NK cells, leading to enhanced anti-tumor effects in a murine melanoma model.⁸⁵³ Nanoparticles carrying TLR7/8 agonists, such as R848, show potential in enhancing antibody-dependent immunotherapy mediated by NK cells, potentially reshaping the tumor microenvironment and enhancing the effectiveness of cancer immunotherapy.⁸⁵⁴ Nanoparticles co-loaded with two Toll-like receptor (TLR) agonists demonstrate synergistic effects in promoting adaptive immune responses, functioning as a co-delivery system. For instance, polymer nanoparticles carrying both CpG-ODN and Poly(I:C) induce a more robust Th1 immune response when compared to nanoparticles containing only one adjuvant.⁸⁵⁵ Overall, nanoparticles present a promising approach to precisely and efficiently target innate immune pathways, such as NLR signaling, and enhance cancer immunotherapy by delivering antigens, adjuvants, and immune-modulating factors.

NLR functionality, exemplified by NOD1, can facilitate tumor cell apoptosis and contribute to tumor progression in head and neck squamous cell carcinoma.⁸⁵⁶ Targeting NLRs, especially NOD2, in cancer immunotherapy shows potential, and incorporating NOD2 agonists into nanoparticle carriers enhances their immunostimulatory and anti-cancer activities.⁸⁵⁷ Furthermore, the activation of RLRs, such as RIG-I and MDA5, plays a crucial role in antiviral defense mechanisms and may be significant in cancer treatment by promoting immunogenic cell death in tumor cells.⁸⁵⁸ Polymer nanoparticles delivering synthetic RIG-I agonists have shown potential anti-tumor effects, including promoting cell death and modulating the tumor microenvironment.⁸⁵⁹ Designed pH-responsive polymer nanoparticles have enhanced the delivery of RIG-I agonists, demonstrating increased immunogenic cell death, interferon production, and T cell infiltration in the tumor microenvironment.⁸⁶⁰ Research targeting the delivery of dual-function RIG-I agonists to enhance cancer immunotherapy strategies has been conducted.⁸⁶¹ These strategies aim to improve treatment outcomes against tumors by leveraging immune responses, whether used alone or in combination with checkpoint inhibitors. Promising approaches for pancreatic cancer immunotherapy involve encapsulating dual-function dsRNA in lipid-calcium nanoparticles to induce pro-inflammatory Th1 responses and increase CD8⁺ T cell levels.⁸⁶² The addition of Riboxxim to PLGA microparticles has been shown to enhance tumor-specific CD8⁺ T cell responses by activating TLR3 and RIG-I signaling pathways in both mouse and human dendritic cells.⁸⁶³ Another key class of RLRs, MDA5, can activate IRF3 and IFN-β production, leading to MHC-I upregulation and immune responses within the tumor microenvironment. Targeting MDA5 activation by including TLR3/ML8 agonists in nanoparticles has strengthened anti-tumor immunity, prolonged survival in a pancreatic cancer mouse model by increasing CD8⁺ T cell infiltration and immune cell activation.⁸⁶⁴ Additionally, the activation of the cGAS/STING pathway is crucial for cancer immunity and has been investigated as a promising therapeutic target. Various DNA sensors, including cGAS, can detect intracellular DNA, resulting in downstream signaling and interferon production.⁸⁶² Targeting cGAS-STING with nano-delivery systems, such as CDN-Mn2⁺ particles and mesoporous silica nanoparticles, effectively activate anti-tumor immune responses and reshape the TME in mouse models.⁸⁶⁵ The delivery of a biodegradable mesoporous silica nanoparticle carrying STING agonists has significantly enhanced anti-tumor efficacy and promoted immune cell infiltration, thereby inhibiting tumor growth.⁸⁶⁶ Functionalized nanoparticles containing CDG have shown enhanced immune cell infiltration and improved anti-tumor effects in mouse models of breast cancer and melanoma.⁸⁶⁷ Utilizing nanoparticle carriers to target the cGAS/STING pathways offers a comprehensive strategy for modulating both innate and adaptive immune responses. These techniques showcase the capacity of nanotechnology to bolster cancer immunotherapy by directing crucial immune pathways and augmenting immune reactions against tumors. The amalgamation of STING-LNPs with anti-PD-1 treatment has exhibited mutually reinforcing anti-tumor properties in a melanoma experimental setting. A pioneering methodology encompasses encapsulating a STING agonist within the core of iron oxide nanoparticles, which are enveloped by inert gold nanostars.

Moreover, nanotechnology has been explored for its capability to reprogram the TME with the aim of targeting innate immune cells and enhancing cancer immunotherapy. Utilizing nanoparticles to modulate key innate immune cells, such as macrophages, has demonstrated promising potential in enhancing the body’s immune response against tumors. Researchers have investigated methods to directly manipulate innate immune cells such as macrophages, MDSCs, DCs, NK cells, and neutrophils as strategies to improve cancer immunotherapy (Fig. 8). Progress has been made in targeting innate immune cells with nanocarriers, combined with STING activation and immune receptor modulation, aiming to overcome therapy resistance and improve patient outcomes.⁸⁶⁸

The significance of TAMs in cancer immunotherapy underscores their plasticity and the presence of two conflicting subtypes: M1 TAMs activated by LPS and IFN-γ, and M2 TAMs activated by IL-4 and IL-13. The predominance of immunosuppressive M2 TAMs in the tumor microenvironment has been established. Strategies have been developed to regulate TAM functions through nanoparticles, such as reshaping M2 TAMs into M1 anti-tumor macrophages and blocking the CD47-SIRPα pathway to enhance cancer immunotherapy. Promising results have been demonstrated with targeted elimination of TAMs using strategies like dendritic polymer nanoparticles, bisphosphonate calcium nanoparticles, and silicon-coated gold nanoparticles.⁸⁶⁹ Furthermore, targeting signal-regulating monocytes derived macrophages infiltrating tumors with siRNA-loaded lipid or cationic polymer nanoparticles effectively promotes anti-tumor immune responses, such as through the CSF-1/CSF-1R and CCL2/CCR2 pathways. Various nanoparticle-based approaches have been discussed, including glucose-based liposomes, PBAE nanoparticles delivering M1 polarization factor mRNA, and nanoparticles redirecting M2 TAMs towards an anti-tumor M1 phenotype to enhance immunotherapy efficacy.⁸⁷⁰ Delivery of IL-12 through nanoparticles promotes immune modulation in the tumor microenvironment, while vesicles derived from M1 macrophages or bone marrow even help reprogram TAMs into anti-cancer phenotypes.⁸⁷¹ Additionally, nanoparticles targeting the CD47-SIRPα axis like SNPACALR&aCD47 show enhanced effects in macrophage-mediated cancer immunotherapy by promoting phagocytosis of cancer cells and blocking immune escape signals.⁸⁷² These diverse nanoparticle-based strategies aim to reshape TAMs and enhance cancer immunotherapy by immune modulation and promoting anti-tumor immune responses.

Targeting of immature MDSCs, which promote tumor growth, angiogenesis, and immune suppression, has been achieved through various nanoparticle-based approaches to eliminate or inhibit their functions, thereby enhancing the efficacy of cancer immunotherapy.⁸⁷³ Strategies include using synthetic nanoantibodies (SNAbs) to target MDSCs in a triple-negative breast cancer mouse model, PD-1 antibody-conjugated PLGA nanoparticles to reduce circulating and pulmonary MDSCs, and LPH nanoparticles carrying siRNA targeting HMGA1 to reduce MDSCs and increase colon cancer sensitivity to PD-L1 checkpoint inhibition.⁸⁷⁴ Additionally, zinc-doped iron oxide nanoparticles have been employed to convert immunosuppressive MDSCs into pro-inflammatory states, enhancing anti-tumor efficacy when used in conjunction with radiation therapy. Moreover, nanoparticles designed to switch MDSCs into anti-tumor modes increase the production of anti-tumor cytokines, activate T cell activity, and extend mouse lifespan.⁸⁷⁵ Targeting MDSCs with HDL-like nanoparticles in a melanoma metastasis mouse model leads to increased CD8⁺ T cell numbers, reduced regulatory T cell numbers, slowed tumor growth, reduced tumor burden, and extended survival.⁸⁷⁶ Combination therapy delivering anti-cancer drugs via nanoparticles and modulating MDSCs shows promise in improving outcomes of cancer immunotherapy. Nanoparticle-based approaches provide a precise and efficient way to modulate MDSCs in the tumor microenvironment, enhancing anti-tumor immune responses and improving the effectiveness of cancer treatment.

Dendritic cells, known as specialized antigen-presenting cells, possess the capacity to initiate initial T lymphocyte reactions to particular antigens, serving as vital mediators bridging innate and adaptive immune responses. When dendritic cells migrate to the lymph nodes, they process antigens that they have engulfed in order to present them on MHC class I or II molecules to T cells. This process entails the upregulation of MHC I/II molecules and the activation of co-stimulatory molecules. Concurrently, dendritic cells develop the capacity to generate pro-inflammatory cytokines and chemokines that are beneficial for stimulating T cells.⁸⁷⁷ Therefore, the mature state of dendritic cells is crucial for inducing T cell responses. Several strategies have been developed to activate and mobilize dendritic cells, co-delivering tumor antigens and adjuvants to dendritic cells to activate cytotoxic T lymphocytes (CTLs) and generate dendritic cell-based vaccines. Strategies to modulate dendritic cell function using nanoparticles have attracted significant attention. Certain nanoparticles, such as gold nanorods and carbon black nanoparticles, are recognized as adjuvants that can promote dendritic cell maturation and boost humoral and cellular immune responses through synergistic effects.^878–880

Numerous approaches utilizing nanoparticles have been devised to enhance the immune reactions of NK cells towards cancer. For instance, in an in situ VX2 liver cancer model, monitoring through magnetic resonance imaging (MRI) revealed that locally delivered immunomodulatory nanocomplex microspheres (IMM-MS) via hepatic artery intervention selectively targeted liver tumors, significantly increasing NK cell infiltration at the site of liver tumors.⁸⁸¹ Trivalent nanoparticles, referred to as a-EGFR/a-CD16/a-4-1BB NPs, have shown promising capabilities in recruiting and triggering NK cells towards attacking tumor cells with elevated levels of EGFR. The stimulation of NK cells facilitates anti-cancer immune responses. Notably, these nanoparticles are capable of transporting the cytotoxic chemotherapy agent paclitaxel, thus improving treatment efficacy.⁸⁸² Systemic delivery of cationic liposomes carrying TUSC2 plasmid DNA in a Kras mutant mouse lung cancer model results in immune effect NK cell infiltration, providing survival benefits.⁸⁸³ Furthermore, chitosan particles carrying NKG2D and IL-21 genes have been shown to effectively activate NK and T cells, leading to inhibition of tumor growth and prolonged mice survival.⁸⁸⁴ Exosomes derived from dendritic cells also exhibit potent activation effects on NK cells.⁸⁸⁵ Research has demonstrated that Dex can stimulate the expansion and activation of NK cells in secondary lymphoid organs of mice by activating the interleukin 15 receptor alpha subunit (IL-15Ra) and NKG2D-dependent pathways. Significantly, Dex immunization shows potential in boosting NK cell count and efficacy in advanced-stage melanoma patients.⁸⁸⁶

The focus of the discussion is the role of neutrophils in cancer, highlighting different subtypes such as N1/N2 neutrophils, TANs, and PMN-MDSCs.³⁹¹ The primary research focus lies in investigating neutrophils in animal models or studying the functions of neutrophils in peripheral blood in humans, while information regarding TANs in cancer patients is scarce.⁸⁸⁷ Neutrophils primarily engulf nanoparticles through phagocytosis, thereby enhancing prospects for improving cancer immunotherapy by delivering nanoparticles via neutrophils. Various approaches have been explored, with promising results achieved in inhibiting tumor recurrence with drug-carrying neutrophils, particularly in glioblastoma.⁸⁸⁸ Overall, precise targeting of the tumor immune microenvironment with nanoparticles holds promise in enhancing the effectiveness of cancer immunotherapy.