The Evasion Mechanisms of Cancer Immunity and Drug Intervention in the Tumor Microenvironment

1The immune system maintains its defense against infected antigens and self-antigens through an appropriate balance between activation and inhibition of immune response. At its core, this process involves receptor-ligand binding between T cells activated by antigens and various cells within the immune system, and the immune response is known to play an important role in carcinogenesis and anti-tumor immunity.

2The immune response is the intraorganism defense mechanism that responds to threats to self. This defensive reaction takes place in three stages: recognition, processing, and reaction. Non-self elements are recognized as foreign substances by cells or active molecules involved in the immune response. Cells or active molecules involved in innate immunity recognize molecular patterns of non-self antigens. Innate immunity includes cells such as natural killer (NK) cells, macrophages, dendritic cells (DC), and neutrophils, as well as active molecules such as complement system in serum. The lymphocytes and antibodies involved in adaptive immunity have the ability to recognize the structure of non-self antigens or amino acid sequences in detail with high accuracy. In order to induce an effective immune response that can fight infectious diseases or cancer, cells or active molecules in the innate immune system and the adaptive immune system must interact in a coordinated manner. When cells from both systems respond properly at the early stage of cancer, cancer cells can be removed and the tissue returns to its normal state.

3However, cancer cells grow and spontaneously proliferate faster than the immune system can handle through strategies that deceive the immune system and prevent attacks from immune cells. To do this, cancer cells begin to build a microenvironment starting at the time of cancer occurrence, and in most cases, when a cancer mass is found, a tumor microenvironment (TME) is present, making it difficult for the immune system to efficiently eliminate cancer cells.

4Cancer immunology first began in 1957 when Burnet proposed the concept of tumor immunosurveillance. Tumor immunosurveillance is when tumor cells are first recognized by T cells at the occurrence and growth stage, and then are killed by immune cells and secreted interferon-gamma (IFN-γ). In addition, tumor cells actively induce T-cell tolerance, resulting in an immunoediting process that facilitates tumor re-distribution and growth. Accordingly, in cancer immunology, efforts have been made to develop immunotherapy against tumors by targeting the interactions between cancer cells and immune cells. Many recent studies have revealed that various cancers diminish normal immunity from having effective anti-cancer activity, as exemplified by findings of decreased absolute counts of lymphocytes, an increasd regulatory T cells and tumor-associated macrophages, down-regulation of antigen expression in tumor-associated cells, and apoptosis of cytotoxic T cells. In addition, direct inhibition of the immune mechanism occurs due to vascular endothelial growth factor, prostaglandin E2, transforming growth factor-β (TGF- β), and interleukin (IL)-10 secreted from cancer cells, further decreasing the function of anti-cancer immunity. The immune response is also indirectly involved in tumor development through a reduction of the Th1 response, which induces anti-cancer effects, cytotoxic T cell differentiation, and an increase in the antagonistic Th2 response.

5Interestingly, direct and indirect studies have reported that chemotherapy and radiation therapy, which are direct treatments for cancer and its progression, also regulate the immune system’s response against cancer cells. When external radiation therapy is performed for malignant melanoma, lymphoma, and renal cancer, improvement also occurs outside the treated area, and a recent report showed that these effects of cancer treatment might be caused by the activation of cancer-specific antibodies and T cells. It has also been reported that the immune response is suppressed in high-dose chemotherapy, while in low-dose chemotherapy, the anti-cancer immune response is stimulated or induced.

6Cancer immunotherapy is research on the effects of existing cytotoxic chemotherapy and radiation therapy on the host immune system based on an understanding of the relationship between cancer and the immune system. Although cancer immunotherapy has been attempted since 1990 in preclinical and early clinical studies with immunostimulating cytokine therapy, cancer vaccine therapy, and adoptive T cell therapy, it has not been introduced into practical use in clinical settings due to the local delivery of drug, systemic toxicity, and a lack of remarkable treatment results. Nevertheless, there has recently been a rapid increase in interest in immunotherapy, as exemplified by research on the involvement of immune mechanisms in the treatment of cetuximab, diverse research results on immune evasion mechanisms, and early clinical studies of monoclonal antibodies to block immune checkpoints that showed satisfactory results not only for malignant melanoma, but also for head and neck cancer and non-small cell lung cancer, which is a non-immune solid cancer.

7Tumor tissue, besides tumor cell also includes a variety of other cell types such as T cells, NK cells, macrophages, fibroblasts, dendritic cells, and adipocytes. The tumor microenvironment is characterized by cancer cells that exhibit diverse metabolic variations due to their heterogeneity, in accordance with their proliferation and metastasis. Therefore, improvement of tumor immunotherapy is tightly linked with improvement of our understanding of the TME.

8Cancer cells change through mutations or abnormal regulation, while immune cells change through specific mechanisms related to external invasion. In addition, cancer cells require sufficient nutrient uptake to maintain their metabolism and regulate their growth, whereas immune cells are relatively flexible compared to cancer cells because they can maintain their function through a balance of various metabolic systems. Thus, studies aiming to inhibit cancer metabolism by regulating the signaling pathway of immune cells through interactions between immune cells and cancer cells have recently been highlighted. Cancer cells affect the metabolism of T cells through the activation and proliferation of immune cells, depletion of glucose and amino acids (e.g., leucine, serin, tryptophan, and glutamine), conditions of high acidity and high lactic acid levels, and up-regulation of immune checkpoints.

9Cancer cells affect the metabolism of T cells in various ways, and through the activation and proliferation of immune cells, the depletion of glucose and amino acids, high acidity and lactate, and upregulated immune checkpoint, consequently inhibiting glycolysis, thereby affecting the metabolism of T cells. To summarize the metabolic mechanism between cancer cells and immune cells, M1 macrophages, effector T cells, and normal cancer cells regulate proliferation and anabolism using a high proportion of glycolysis and glutaminolysis to synthesize proteins, nucleic acids, and amino acids, while M2 macrophages, regulatory T cells, and quiescent cancer cells regulate catabolism using the oxidation of fatty acids to synthesize ATP. Among them, tumor-associated macrophages, corresponding to M2 macrophages, are the most abundant immune cells in the TME (accounting for 50%), and mainly affect cancer progression and resistance by supplying or supporting nutrients to malignant cancer cells. In particular, because tumor-associated macrophages are responsive to pharmacological agents that improve bacterial and oxidative function, mechanisms of regulating their metabolism are highly promising targets for novel cancer treatment. Therefore, recent studies have considered the mechanisms through which tumor-associated macrophages are involved in the overall metabolism of the TME for this goal.

10In recent years, substantial advances have been made in various immune-based treatments, including the administration of specific cytokines, antibodies to immune checkpoints, and cell therapies such as CAR-T and CAR-NK, as a new paradigm of cancer treatment. These advances can be mainly summarized as attempts to modulate immune cell activity through the regulation of T cells using adoptive cell delivery and monoclonal antibodies. To date, among these immune-based treatments, immune checkpoint inhibitors are most widely used as an effective immunotherapy for a variety of solid tumors and malignant hematologic tumors. Immune checkpoint inhibitors are defined as drugs that upregulate the immune response using monoclonal antibodies targeting CTLA4 and PD-1, which are located on the cellular membrane of T cells and cancer cells. Although there have been many studies on the mechanism and utilization of immunotherapy with immune checkpoint inhibitors, immunotherapy targeting only these single antigens has limitations due to the high immunosuppression of the TME and the low immunogenicity of cancer cells. A recent study showed that a combination of anti-PD1 and anti-CTLA4 to treat metastatic melanoma and non-small cell lung carcinoma (NSCLC) resulted in a better response than was achieved with conventional single antibody immunotherapy, as shown by increased patient life expectancy and the inhibition of new metastasis. In addition, combination of immune checkpoint inhibitors with other chemotherapy agents, has recently attracted attention as a new strategy for anti-cancer treatment. Nevertheless, tumor-infiltrating lymphocytes (TILs) need to overcome not only the immune checkpoints, but also a wide range of metabolic checkpoints within the TME that weaken their functions. Thus, it is important to improve our comprehensive understanding of the mechanisms that reduce anti-cancer immunity within the metabolically hostile TME.

11In fact, cancer cells can regulate immune cells by upregulating nutrient uptake and metabolite production, and as a result, constructing an immunosuppressive TME that promotes cancer cell growth and immune evasion. However, some recent studies have suggested that combining the immune checkpoint inhibitors and metabolic regulating agents can decrease cancer cell metabolism more effectively than either approach used in isolation. Caution is needed to ensure that treatments of this type only target specific tumor sites to prevent side effects such as systemic toxicity. Although cancer metabolism has been studied long enough to be considered the basis of cancer, active research is still undergoing and the mechanisms of cancer metabolism remain to be fully elucidated. Therefore, various aspects of cancer metabolism and its metabolites are noteworthy for future cancer research.

12In order for an anti-cancer immune response to effectively kill cancer cells, a series of steps must be repeated and amplified. This series of steps is called the cancer-immune cycle: 1) Neoantigens produced during tumor formation are released from dead cancer cells, and dendritic cells (DCs) acquire them and move to the nearest draining lymph node (DLN). 2) DCs present the acquired antigen to T cells through major histocompatibility class I (MHC-I) and MHC-II molecules. 3) Effector T cells are activated by recognizing this antigen. 4) Antigen-recognizing cancer-specific T cells in the DLN express the cell adhesion molecules and chemokine receptors necessary for migration and infiltration into the tumor, and then leave the DLN and move through the blood toward the tumor tissue. 5) These T cells then infiltrate into the tumor and 6) recognize and bind the MHC-I–antigen complex presented by cancer cells through the T cell receptor (TCR). 7) Through these processes, cancer cells are finally killed. CD8+ T cells kill cancer cells through granule-exocytosis mechanisms mediated by perforin and granzyme and interactions between apoptotic ligands-receptors such as FasL-Fas. When cancer cells die and additional neoantigens are released, the immune reaction continues the cycle again from the first step, and it is amplified compared to the previous reaction; for this reason, this mechanism is termed the cancer-immune cycle (Figure 1). In cancer patients, at least one of these steps is defective and the cancer-immune cycle does not work properly.

13Successful antigen recognition and activation (steps 1–3) of T cells, which are the early stages of the cancer-immune cycle, and infiltration (steps 4–5) of T cells into the tumor are essential conditions for “hot” tumors. This type of immune infiltration is usually characterized by reactivity to immune checkpoint inhibitors. In contrast, in “cold” tumors, characterized by immune non-infiltration, there is a defect in the initial stage of the cancer-immune cycle, and most cold tumors do not respond to immune checkpoint inhibitors. Therefore, antigen recognition and migration of T cells to the tumor can be assumed to be the main factors influencing reactivity to immune checkpoint inhibitors.(1) Decrease in tumor immunogenicity (evasion of step 1): During tumor formation, immunosurveillance removes cancer clones that express strong immunogenic neoantigens. At this point, the tumor evades anti-cancer immune responses by eliminating immunogenic antigens or maintaining cancer clones without cancer antigens so that they are not recognized by T cells. In other words, cancer clones that evade immunosurveillance have fewer immunogenic antigens.(2) Inhibition of dendritic cell maturation (evasion of step 2): Damage-associated molecular patterns, such as ATP and high-mobility group box 1 (HMGB1), released from dead cancer cells can induce DC maturation. Cancer inhibits DC maturation through tumor derived-factors such as IL-10, macrophage colony-stimulating factor (MCSF), vascular endothelial growth factor (VEGF), prostaglandin, TGF-β, and indoleamine 2,3-dioxygenase (IDO). In addition, immunosuppressive cells in the TME, such as Treg and myeloid-derived suppressor cells (MDSCs) express inhibitory factors that suppress DC maturation, reducing the expression of MHC and co-stimulatory factors in DCs, resulting in reduced production of inflammatory cytokines, such as IL-12, and ultimately inhibiting the proliferation of T cells and IFN-γ.(3) Suppression of T cell activity (evasion of step 3): For full activation of T cells, both antigen recognition and co-stimulatory signals are required. Co-stimulatory interactions between DC and T cells include B7.1/B7.2:CD28, 4-1BBL:4-1BB, OX40L:OX40, CD70:CD27, and GITRL:GITR. These co-stimulatory interactions promote proliferation, differentiation, survival, cytotoxic function, memory formation, and cytokine generation of T cells. Tumors inhibit the activity of T cells by reducing the expression of co-stimulatory factors and MHC, limiting the co-stimulation required for T cells. When the TCR is activated without co-stimulation, excessive activity of calcium/nuclear factor of activated T-cell (NFAT) signals induces the expression of negative modulating factors and T cells become unresponsive (T cell anergy).(4) Inhibition of T cell migration and infiltration (evasion of steps 4-5): T cells express chemokine receptors such as CXCR3 on the cell surface in response to IFN-γ during the activation process. As a major evasion mechanism at this stage, cancer cells reduce the expression of CXCR3 ligands such as CXCL9, CXCL10, and CXCL11, and/or carry out posttranslational modification or decomposition of CXCR3 ligands, thereby inhibiting the migration of CD8+ T cells to the tumor. These fragments of cleaved CXCR3 ligands may also act as antagonists of the receptor. As another mechanism for inhibiting the migration of T cells, tumors transform nearby blood vessels. Tumors produce neoplastic factors such as VEGF, reducing the expression of adherent factors in endothelial cells (ECs), which are important for the migration of T cells. In addition, IL-10 and prostaglandin E2, which are immunosuppressive factors, are produced and promote Fas ligand expression along with VEGF, thereby inducing apoptosis of CD8+ T cells infiltrating the tumor. Moreover, the endothelin-B receptor expression of tumor ECs is increased to inhibit the migration of T cells. However, even if CD8+ T cells move toward tumor tissue, they may not be able to infiltrate the center of the tumor. This is because immunosuppressive immune cells and cancer-associated fibroblast (CAF) around the tumor produce extracellular matrix (ECM) proteins to physically suppress T cells or produce chemokines such as CXCL12, inhibiting the migration of T cells to tumors. In fact, analyses of human lung cancer tissue have confirmed that fibroblasts or collagen accumulates in the tumor substrate to prevent interactions between T cells and tumor cells.

14(1) Decrease in tumor immunogenicity (evasion of step 1): During tumor formation, immunosurveillance removes cancer clones that express strong immunogenic neoantigens. At this point, the tumor evades anti-cancer immune responses by eliminating immunogenic antigens or maintaining cancer clones without cancer antigens so that they are not recognized by T cells. In other words, cancer clones that evade immunosurveillance have fewer immunogenic antigens.

15(2) Inhibition of dendritic cell maturation (evasion of step 2): Damage-associated molecular patterns, such as ATP and high-mobility group box 1 (HMGB1), released from dead cancer cells can induce DC maturation. Cancer inhibits DC maturation through tumor derived-factors such as IL-10, macrophage colony-stimulating factor (MCSF), vascular endothelial growth factor (VEGF), prostaglandin, TGF-β, and indoleamine 2,3-dioxygenase (IDO). In addition, immunosuppressive cells in the TME, such as Treg and myeloid-derived suppressor cells (MDSCs) express inhibitory factors that suppress DC maturation, reducing the expression of MHC and co-stimulatory factors in DCs, resulting in reduced production of inflammatory cytokines, such as IL-12, and ultimately inhibiting the proliferation of T cells and IFN-γ.

16(3) Suppression of T cell activity (evasion of step 3): For full activation of T cells, both antigen recognition and co-stimulatory signals are required. Co-stimulatory interactions between DC and T cells include B7.1/B7.2:CD28, 4-1BBL:4-1BB, OX40L:OX40, CD70:CD27, and GITRL:GITR. These co-stimulatory interactions promote proliferation, differentiation, survival, cytotoxic function, memory formation, and cytokine generation of T cells. Tumors inhibit the activity of T cells by reducing the expression of co-stimulatory factors and MHC, limiting the co-stimulation required for T cells. When the TCR is activated without co-stimulation, excessive activity of calcium/nuclear factor of activated T-cell (NFAT) signals induces the expression of negative modulating factors and T cells become unresponsive (T cell anergy).

17(4) Inhibition of T cell migration and infiltration (evasion of steps 4-5): T cells express chemokine receptors such as CXCR3 on the cell surface in response to IFN-γ during the activation process. As a major evasion mechanism at this stage, cancer cells reduce the expression of CXCR3 ligands such as CXCL9, CXCL10, and CXCL11, and/or carry out posttranslational modification or decomposition of CXCR3 ligands, thereby inhibiting the migration of CD8+ T cells to the tumor. These fragments of cleaved CXCR3 ligands may also act as antagonists of the receptor. As another mechanism for inhibiting the migration of T cells, tumors transform nearby blood vessels. Tumors produce neoplastic factors such as VEGF, reducing the expression of adherent factors in endothelial cells (ECs), which are important for the migration of T cells. In addition, IL-10 and prostaglandin E2, which are immunosuppressive factors, are produced and promote Fas ligand expression along with VEGF, thereby inducing apoptosis of CD8+ T cells infiltrating the tumor. Moreover, the endothelin-B receptor expression of tumor ECs is increased to inhibit the migration of T cells. However, even if CD8+ T cells move toward tumor tissue, they may not be able to infiltrate the center of the tumor. This is because immunosuppressive immune cells and cancer-associated fibroblast (CAF) around the tumor produce extracellular matrix (ECM) proteins to physically suppress T cells or produce chemokines such as CXCL12, inhibiting the migration of T cells to tumors. In fact, analyses of human lung cancer tissue have confirmed that fibroblasts or collagen accumulates in the tumor substrate to prevent interactions between T cells and tumor cells.

18When the initial 5 stages of the cancer-immune cycle are successfully performed, immune infiltration into the TME is formed. Similar to the initial stage of the cancer-immune cycle, the mechanism of immune evasion of the last stage also takes place in the TME. PD-L1:PD-1 interactions are only one of several causes of immunosuppression, so reactivity to immune checkpoint inhibitors is sometimes absent even when the TME has immune infiltration. In addition, even in patients who initially respond to an immune checkpoint inhibitor, secondary immune evasion can occur when the tumor inhibits immune reaction in response to attacks by T cells.(1) Inhibition of recognition by immune cells (evasion of step 6): Cancer cells remove, reduce, or transform MHC-I on the surface of cancer cells as a mechanism to evade recognition by T cells. Cancer cells directly regulate MHC-I genes or proteins or indirectly inhibit peptide-MHC components. In addition, cancer cells downregulate the expression of antigens, proteasome components, TAP1/TAP2, MHC-I, and β2-microglobulin through mutation, genetic loss, transcription inhibition, or epigenetic inhibition of gene expression. Recent cancer genome studies have confirmed that the reduction of peptide-MHC-I expression on the surface of cancer cells due to somatic mutations in human leukocyte antigen. Even if cancer cells can evade recognition by T cells through reduced MHC-I expression, NK cells cannot be evaded. This is because NK cells can induce an immune response to abnormal cells by recognizing the degree of MHC-I expression on the cell surface. As an alternative to this, cancer cells release ligands to NKG2D, an active receptor of NK cells, to evade lysis by NK cells.(2) Immune checkpoint molecule expression (evasion of step 7): The CD8+ T cells that infiltrate a tumor can simultaneously express several additional co-inhibitory receptors in addition to PD-1, including B and T lymphocyte attenuator (BTLA), lymphocyte activation gene 3 protein (LAG-3), T-cell immunoglobulin domain, mucin domain-3 (TIM3), T-cell immunoglobulin, and immunoreceptor tyrosine-based inhibitory motif domain (TIGIT). Other co-inhibitory receptors are expressed simultaneously, and T cells become exhausted T cells (Tex) that are unresponsive to immune checkpoint inhibitors.(3) Immunosuppressive cells (evasion of step 7): Immunosuppressive cells in the TME are also an important mechanism of immune evasion. The TME induces macrophages to differentiate into M2-type tumor-associated macrophages that promote tumor formation, and tumor-associated macrophages generate IL-10 instead of IL-12 to inhibit the CD8+ T cell response. Tumor-associated macrophages directly inhibit immune checkpoint inhibitor responses by removing anti-PD-1 antibodies from PD-1+ CD8+ T cells in an FcγR-dependent manner. MDSCs are a group of heterogeneous cells that can strongly inhibit the Teff response and induce Treg. MDSCs inhibit the immune response by generating arginase, inducible nitric oxide synthase (iNOS), and TGF-β. In particular, TGF-β inhibits the cytotoxic activity of cytotoxic T cells and NK cells by reducing the expression of cytotoxic factors such as perforin and granzyme. Treg cells are among the immunosuppressive cells of the TME; when their number increases, they are known to inhibit the CD8+ T cell response and promote tumor progression. A high Treg frequency is generally associated with a poor prognosis. For example, strong anti-cancer immune responses have been reported in Treg-deficient mouse models, and these results suggest that Treg cells play an important role in inhibiting anti-cancer immunity. IDO, an immunosuppressive enzyme expressed in myeloid cells and various cancer cells, induces kynurenine, which is a tryptophan metabolite that exerts immunosuppressive actions. This is known to strongly suppress Teff function while promoting the generation and activity of Treg and MDSCs. Another immunosuppressive enzyme, arginase 1, inhibits the function of DCs through cooperation with the IDO mechanism. Other metabolites (e.g., glucose consumption, lactate generation, cholesterol metabolism) and inflammatory mechanisms (e.g., cyclooxygenase-2/prostaglandin E2) are also known to affect cancer cells and immune cells.

19(1) Inhibition of recognition by immune cells (evasion of step 6): Cancer cells remove, reduce, or transform MHC-I on the surface of cancer cells as a mechanism to evade recognition by T cells. Cancer cells directly regulate MHC-I genes or proteins or indirectly inhibit peptide-MHC components. In addition, cancer cells downregulate the expression of antigens, proteasome components, TAP1/TAP2, MHC-I, and β2-microglobulin through mutation, genetic loss, transcription inhibition, or epigenetic inhibition of gene expression. Recent cancer genome studies have confirmed that the reduction of peptide-MHC-I expression on the surface of cancer cells due to somatic mutations in human leukocyte antigen. Even if cancer cells can evade recognition by T cells through reduced MHC-I expression, NK cells cannot be evaded. This is because NK cells can induce an immune response to abnormal cells by recognizing the degree of MHC-I expression on the cell surface. As an alternative to this, cancer cells release ligands to NKG2D, an active receptor of NK cells, to evade lysis by NK cells.

20(2) Immune checkpoint molecule expression (evasion of step 7): The CD8+ T cells that infiltrate a tumor can simultaneously express several additional co-inhibitory receptors in addition to PD-1, including B and T lymphocyte attenuator (BTLA), lymphocyte activation gene 3 protein (LAG-3), T-cell immunoglobulin domain, mucin domain-3 (TIM3), T-cell immunoglobulin, and immunoreceptor tyrosine-based inhibitory motif domain (TIGIT). Other co-inhibitory receptors are expressed simultaneously, and T cells become exhausted T cells (Tex) that are unresponsive to immune checkpoint inhibitors.

21(3) Immunosuppressive cells (evasion of step 7): Immunosuppressive cells in the TME are also an important mechanism of immune evasion. The TME induces macrophages to differentiate into M2-type tumor-associated macrophages that promote tumor formation, and tumor-associated macrophages generate IL-10 instead of IL-12 to inhibit the CD8+ T cell response. Tumor-associated macrophages directly inhibit immune checkpoint inhibitor responses by removing anti-PD-1 antibodies from PD-1+ CD8+ T cells in an FcγR-dependent manner. MDSCs are a group of heterogeneous cells that can strongly inhibit the Teff response and induce Treg. MDSCs inhibit the immune response by generating arginase, inducible nitric oxide synthase (iNOS), and TGF-β. In particular, TGF-β inhibits the cytotoxic activity of cytotoxic T cells and NK cells by reducing the expression of cytotoxic factors such as perforin and granzyme. Treg cells are among the immunosuppressive cells of the TME; when their number increases, they are known to inhibit the CD8+ T cell response and promote tumor progression. A high Treg frequency is generally associated with a poor prognosis. For example, strong anti-cancer immune responses have been reported in Treg-deficient mouse models, and these results suggest that Treg cells play an important role in inhibiting anti-cancer immunity. IDO, an immunosuppressive enzyme expressed in myeloid cells and various cancer cells, induces kynurenine, which is a tryptophan metabolite that exerts immunosuppressive actions. This is known to strongly suppress Teff function while promoting the generation and activity of Treg and MDSCs. Another immunosuppressive enzyme, arginase 1, inhibits the function of DCs through cooperation with the IDO mechanism. Other metabolites (e.g., glucose consumption, lactate generation, cholesterol metabolism) and inflammatory mechanisms (e.g., cyclooxygenase-2/prostaglandin E2) are also known to affect cancer cells and immune cells.

22Cancer transforms its own metabolic processes to maintain rapid growth and survival. Normal cells generate energy through oxidative phosphorylation (OXPHOS) in mitochondria, while cancer cells generate ATP through aerobic glycolysis even in an environment with sufficient oxygen, which is known as the Warburg effect. The Warburg effect promotes the growth and survival of cancer cells, facilitates adaptation to changes in oxygen concentration in tissues, and uses intermediate products of glycolysis to maintain cell proliferation. In addition, lactate produced through glycolysis maintains TME oxidized, contributing to cancer infiltration and immune evasion. However, recent studies have shown that OXPHOS and mitochondrial metabolic processes are also important for cancer metabolism. Some tumors increase OXPHOS by oxidation of glucose, protein, amino acids (e.g., glutamine or tryptophan), and fatty acids, and some tumors use waste products such as ammonia and lactate as energy sources. These changes in cancer metabolism are controlled by carcinogenic mutations (e.g., MYC, phosphatase and tensin homologue; PTEN, AKT serine/threonine kinase, and phosphoinositide 3-kinase; PI3K). In addition, carcinogenic mutations are controlled by changes in surrounding factors or specific metabolic enzymes (e.g., isocitrate dehydrogenase 1; IDH1, succinate dehydrogenase; SDH, and IDO). Changes in metabolite concentration activate specific metabolic pathways, affect cell function through metabolite-mediated signaling pathways, and change the epigenetics of cancer cells. For this reason, even within a single tumor tissue, metabolic changes vary from cancer cell to cancer cell, and the use of nutrients in the TME is also quite different. These complex and diverse tumor metabolic changes construct a TME that threatens the survival of immune cells, and in response, immune cells activate specific metabolism to control their own survival and cancer cell growth. Therefore, further research should aim to elucidate the diversity of tumor metabolism in order to develop effective anti-cancer immunotherapy.

23The concentration of metabolites such as glucose, lactate, and glutamine in the TME affects the function and activity of tumor-infiltrating immune cells. Cancer cells deplete glucose in the TME, inhibiting the anticancer immunity of T cells, NK cells, macrophages, and DCs that use glucose for anti-cancer activity. Mechanistically, glycolysis is necessary to regulate IFNγ production by T cells, and the restricted use of glucose in the TME limits Ca2+ signaling, glycolysis, and cytokine production of TILs. In fact, resistance to T cell therapy appears in cancer patients with overexpressed glycolysis enzymes. If the available glucose concentration is increased by inhibiting glycolysis in the tumor, Teff cells are activated, and the production of immunosuppressive cytokines by cancer cells is inhibited, thereby improving anticancer immunity. Like glucose, lactate produced by the glycolysis of cancer cells or immune cells also inhibits immune cell function. Lactate inhibits the function of Teff cells, but not Treg cells, ultimately inducing an immunosuppressive environment. Glutamine in the TME also allows Treg cells to accumulate, rather than the Th1 response, if its concentration is limited. Reducing glucose and glutamine in the TME limits UDP-GlcNAc synthesis by T cells and promotes differentiation into Treg rather than Th17. Since cytotoxic cells such as CD8+ T cells and NK cells are sensitive to amino acid restrictions, their function is suppressed when glutamine, serine, and glycine are depleted or when branched-chain amino acids, especially leucine and isoleucine, are restricted. Therefore, restrictions on the use of nutrients in immune cells are immunologically associated with cancers that do not undergo immune infiltration (cold tumors) or malignant tumors.

24Tumor-infiltrated immunosuppressive cells (Treg cells, tolerogenic DCs, and MDSCs) and cancer vascular endothelial cells also reduce nutrients in the TME, potentially contributing to an immunosuppressive environment. Treg cells act competitively against glucose, inducing replicative senescence of CD4+ and CD8+ T cells. The activity of TLR8 hinders the corresponding action of Treg cells, thereby improving anticancer immunity. In addition, Treg cells in the TME convert ATP into adenosine, inhibiting the activity of immune cells in tumors.

25Taken together, the immune state in the TME is determined by the competitive action against nutrients, the accumulation of metabolites that inhibit immune response, and cellular signaling that changes metabolic processes. These changes in the TME ultimately induce the recruitment of tumor-associated macrophages, which secrete various cytokines that induce angiogenesis, metastasis of cancer cells, and immunosuppression. Understanding this process will be important for the success of anti-cancer immunotherapy.

26The TME is composed of various cells around cancer cells as mentioned above. These various cells in TME play a complicated role in inducing epithelial-mesenchymal transition (EMT) in tumor progression and metastasis. EMT mechanistically was studied in some cancers such as breast, thyroid, and colon cancer, and related signaling pathways were found PI3K/AKT/PKB, MAPK/ERK, WNT/β-catenin, and NF-κB pathways.

27In tumor microenvironments, the cells supporting cancer cells, such as CAF, CD4+ T cells, Treg, MDSC, and TAM, inhibit the epithelial state of tumor, promote and activate the mesenchymal state. In this case, these cells also inhibit anti-cancer immune cells such as CD8+ T cells, NK cells, and activated M1 macrophages. Cancer cells that come up to the mesenchymal state can up-regulate the expression and activation of various immune cells that affect tumor progression. Treg, M2 macrophages, and MDSC activated by cancer cells directly inhibit the anti-cancer function of T cells and NK cells to support tumor progression. These EMT-mediated communications between cancer cells and immune cells have been the object of intensive investigation in recent years, considering the potential efficacy of immunotherapy for various cancer, suggesting that EMT increases evasion of cancer cell removal processes by immune cells. In fact, melanoma cells expressing SNAIL have been reported to secrete TGF-β and thrombospondin 1 to induce the activity of Treg and reduce the ability of dendritic cells to present antigens. These melanoma cells expressing SNAIL showed resistance to immunotherapy, and inhibition of SNAIL promoted recovery of anti-cancer immune response and sensitivity to immunotherapy. EMT-mediated immune evasion effects have also been observed in other cancer. While a large number of CD8+ T cells infiltrated into the epithelial state of breast cancer, there were many Treg cells and TAM in the mesenchymal state of breast cancer. In addition, only epithelial state tumors were inhibited by anti-CTLA4 ICI treatment, while mesenchymal state tumors were resistant to the same treatment. Treg and TAM continued to infiltrate even in mixed tumors with only 10% of mesenchymal state tumors, and anti-CTLA4 ICI showed no effect. The resistance of mesenchymal state cancer cells to attack cytotoxic T cells was also demonstrated in human breast cancer cell lines. It has been reported that MCF-7 cells activated the response of T cells when co-cultured conditions with T cells, however, T cell function decreases when MCF-7 cells express SNAIL. Even though these studies clearly demonstrate the relationship between EMT and immunosurveillance, the mechanism for immune evasion of mesenchymal state tumors has not been understood clearly.

28Nevertheless, considering the communication system within the tumor microenvironment so far revealed, EMT induction seems to be mainly controlled by multiple cytokines and chemokines exchanged between tumors and various cells in the tumor microenvironment. EMT occurring cancer cells secrete TGF-β, which stimulates resident immune cells in tumor microenvironments to promote the secretion of multiple cytokines and chemokines to evade immunity. For example, melanoma cells expressing SNAIL secrete chemokine CCL2, which leads to the secretion of LCN2 (lipocalin 2). CCL2 and LCN2 reduce the MHC-I expression of T cells and induce the expression of cancer cell immunosuppressive molecules such as PD-L1 to induce exhaustion of cytotoxic T cells. The expression of PD-L1 in cancer cells induced by EMT consequently evaded attacks on CD8+ T cells, increasing cancer metastasis, which was shown to inhibit immune response even in EMT occurred NSCLC patients through increased expression of various immune checkpoint proteins and increased Treg. Prostate cancer cells produce CXCL1, its receptor CXCR1 downstream signaling produces LCN2 in neutrophils. The CXCL1-LCN2 axis activates Src signaling and leads to EMT and contributes to tumor progression. Furthermore, in pancreatic cancer CXCL8 causes cell invasion and promotes metastases. Similarly in gastric cancer, neutrophils like cells expressed CXCL8 and induced EMT through CXCR1/CXCR2 receptors. Another chemokine related to tumor progression and metastasis, CXCL16 has been reported that could promote brain metastases in breast cancer. In addition, it has been reported that EMT causes liver metastasis in colorectal cancer patients. High expression of CXCL16 was associated with M2 macrophage- and angiogenesis-related genes which were poor prognostic factors including a higher TNM staging and the BRAFV600E mutation. Several CC chemokines CCL5 and CCL18 also promote EMT, cell migration and invasion in co-culture experiments involving TAMs and different cancer cells. Furthermore, CCL20 of monocyte-derived macrophages induces EMT and induces tumor metastases in hepatoma cells. In addition, fibroblasts induced CCL17/CCL22 plays a critical role in the malignant progression of prostate, breast and hepatocellular cancer.

29It rarely happens that cancer cells completely transition to a mesenchymal state during human carcinogenesis. Although many studies are underway on the role of tumor microenvironments in inducing EMT, the mechanisms for occurring partial EMT are still unclear. Understanding the signaling pathway for maintaining this condition might be important to develop more effective anti-cancer strategies for most cancers.

30Recently, research on immunotherapy that treats cancer by controlling the immune system is being actively conducted. With the development of immune checkpoint inhibitors at the forefront, the results of various clinical approaches using this immunotherapy have been reported. In particular, research continues to further increase the survival period of cancer patients and maintain the continuity of treatment through combination therapy of various anti-cancer drugs with immune checkpoint inhibitors. However, as mentioned earlier, TME upregulate the expression and secretion of immunosuppressive signal proteins such as IL-10, IDO, and TGF-β. Through this, normal immune cells for removing cancer cells are transformed to immunosuppressive cells such as Treg and MDSC, resulting in down-regulate the anti-cancer effects of drugs.

31A new strategy for immunotherapy is to change the cancer immunity based on the TME and treat it. Various studies earlier, it has shown that the TME has an effect on the anticancer effect of immune checkpoint inhibitors. From this point of view, more recently, various therapeutic approaches have been tried to increase the anti-cancer effect through the combination treatment of TME inhibitors and immune checkpoint inhibitors that suppress mediates in the TME affecting the proliferation of cancer (Table 1).

32Tumor-associated macrophages are immune cells that regulate various factors in the tumor microenvironment and play an important role in connecting cancer with various immune cells in the tumor microenvironment. Recently, these tumor-associated macrophages have also been recognized as biomarkers for the treatment of cancer, and anti-cancer drugs have been developed targeting them (Table 2). This new therapeutic approach is to achieve the effect of reprogramming immune cells to attack cancer cells by combination therapy with immune checkpoint inhibitors and a next-generation immuno-oncology drugs developed based on tumor-associated macrophages. The therapeutic strategy and development of drug targeting tumor-associated macrophages to date is limited to inhibiting the recruitment, polarization, and immune suppression of them. Thus, in order to develop a novel therapeutic strategy and development of drug that encompasses the whole tumor microenvironment, more in-depth research is essentially needed on detailed mechanisms in which macrophages regulate immunity.

33Cancer can occur in otherwise healthy people because many tumors have specialized survival mechanisms to evade the immune response of the host. Furthermore, because tumor cells originally derive from the normal cells of the host, they still have many similar properties to normal cells, and their immunogenicity is lower than that of pathogenic microorganisms. Therefore, many naturally occurring tumors, not those caused by carcinogenic virus infections, only cause very weak immune responses. In tumors that proliferate or migrate quickly, the rate of proliferation and diffusion exceeds the rate of cancer cell removal by the immune system, so it cannot be eradicated by the immune response. Cancer immunologists are racing to increase the effectiveness of cancer treatment by reversing or neutralizing the immune evasion mechanism of these cancer cells.

34Macrophages were originally known to directly kill non-self cancer cells, but recent studies have shown that tumor-associated macrophages promote the proliferation of cancer cells and help cancer cells that cause metastasis to pass through blood vessel walls easily. As the cancer mass grows, the immune response in the TME is severely inclined toward the suppression of the anti-cancer response by tumor-associated immune cells, especially tumor-associated macrophages, impeding the effectiveness of other treatments such as immunotherapy or chemotherapy. Explaining the role of macrophages in tumor biology is important because regulating their activity can open up new opportunities for therapeutic interventions. Similarly, it is also important to understand how macrophages in the TME are affected by current treatments aimed at cancer cells.