Macrophages can be polarized into classically activated macrophages (M1) and alternatively activated macrophages (M2) ( Figure 2 ) (13). M1 TAMs polarization is driven by lipopolysaccharides (LPS), interferon-gamma (IFN-γ), tumor necrotic factor alpha (TNF-α), interleukin (IL)-12, and IL-18 ( Figure 2 ). M1 TAMs exhibit pro-inflammatory and anti-tumor responses through secreted cytokines, recruitment of cluster of differentiation (CD) 8+ T cells and natural killer (NK) cells, and increased antigen presentation ( Figure 2 ). IL-4, IL-10, IL-13, and transforming growth factor beta (TGF-β) promote the polarization of M2 TAMs that displays an anti-inflammatory response and promotes tumor growth, metastasis, and angiogenesis ( Figure 2 ) (14, 15).

OSCC cells contributes to the genesis of M2 TAMs through the p38/MKP-1 signaling axis ( Figure 2 ) (16). Ono et al. found that heat shock protein 90 (HSP-90)-enriched metastatic oral cancer-derived extracellular vesicles (MEV) can also tune TAMs to M2 phenotype ( Figure 2 ) (17). Plasminogen activator inhibitor-1 (PAI-1) and IL-8 derived from OSCC cells have been proved to conduce to the production of CD206+ TAMs (18). Zhou et al. reported that CCR7 in OSCC cells facilitate M2 polarization of macrophages by increasing production of CCL19 and CCL21 ( Figure 2 ) (19). Moreover, stromal cells (SCs), verrucous OSCC−associated SCs, and ED−type OSCC-associated SCs are conducive to the differentiation towards M2 phenotype (20). OSCC cells-derived HMGB1and macrophage endogenous HMGB1 facilitate the polarization of M1 macrophages via TLR4/NF-κB signaling (21). Chen et al. demonstrated that OSCC cells-derived THBS1 in exosome promotes M1 macrophages polarization (22). These studies indicate that OSCC cells-derived proteins, enzymes, cytokines and chemokines in TME can induce M2 or M1 polarization of macrophages. However, there are few studies on macrophage polarization induced by other cells besides OSCC cells, and the mechanism has not been fully elucidated. In addition, the studies involved in the regulation of TAMs by HMGB1 secreted during cell death is still controversial.

M2c and M2d TAMs can elevate glycolysis and initiate angiogenesis with secreted chemokines, fibronectins, and growth factors (35). CSCs, endothelial cells (ECs), and TAMs constitute interrelation through the exchange of cytokine mediators, such as vascular endothelial growth factor (VEGF), IL-6, IL-8, and CSF-1, leading to microvessel density improvement and progression of OSCC (36). ALDH3A1 overexpression in OSCC cells suppresses CRI regulated by TAMs, resulting in inhibition of angiogenesis (37). TAMs play an important role in angiogenesis by producing pro-angiogenic factors. CSCs, ECs, and TAMs construct a sophisticated cellular network, utilizing mediators to regulate angiogenesis in OSCC. Targeting TAMs is an effective way to disrupt the formation of tumor vascular networks and thus inhibit the progression of OSCC. However, the detailed signaling pathways related to the production of pro-angiogenic factors in TAMs and targeted drug interventions need to be further investigated.

M2 TAMs promote the immune escape of OSCC through upregulating the checkpoint programmed death-ligand 1 (PD-L1) expression ( Table 1 ) (31). OSCC cells-derived GM-CSF modulates the level of PD-L1 expression in TAMs via JAK2/STAT3 signaling cascade to promote immune escape ( Table 1 ) (32). Endogenous HMGB1 in TAMs promotes immunosuppression and immune escape via the IL-6/STAT3/PD-L1 pathway ( Table 1 ) (21). M2 TAMs induced by cancer stem cell-derived UCA1 targeting LAMC2 promotes immunosuppression by suppressing CD4⁺ T-cell proliferation and IFN-γ production ( Figure 2 , Table 1 ) (33). A high level of Perilipin 2 (PLIN2) derived from TAMs induces less CD8+ T cells but more TAMs and Foxp3+ Tregs, inducing immunosuppression in OSCC (38). Foxp3+ Tregs contribute to the evasion of immune surveillance through the secretion of cytokines by M2 TAMs in TME (39).

M2 TAMs promote the immunosuppression of OSCC by inducing the expression of PD-L1 and directly inhibiting CD4+ and CD8+ T cells. Inhibition of M2 macrophages polarization can be used as a long-term therapeutic strategy to control or reverse immunosuppression and immune escape. However, the correlation between the density of specific subtypes of TAMs and immune escape remains controversial. The surface receptors and signaling pathways of TAMs that mediate M2 TAMs polarization as well as TAMs-secreted mediators associated with immunosuppression are the focus of further research. Meanwhile, it is necessary to verify whether HMGB1 and PLIN2 can be used as potential therapeutic directions for OSCC.

Currently, immunotherapies targeting TAMs mainly concentrate on three aspects: hindering macrophages from polarizing, removing M2 TAMs from the TME, and repolarizing M2 TAMs towards M1 TAMs (39, 40). Myeloid-based Dusp1-deficiency can reduce the accumulation and polarization of TAMs, which may be a novel target to reprogram TAMs and inhibit the progression of OSCC (41). Targeted inhibition of eukaryotic initiation factor 5A2 (eIF5A2) impedes tumor growth and polarization towards M2 TAMs, which is correlated with the function of modulating cell proliferation, CSC properties, and EMT in oral cancer ( Figure 2 ) (42). PY314, an antibody against TREM2, can exhaust TREM2+ TAMs to decrease exhausted CD8+ T cells and the method has already been applied in clinic (43). Regulator of G protein signaling 12 (RGS12) suppresses OSCC by driving M1 TAMs polarization via MYCBP2/KIF2A signaling pathway ( Figure 2 ) (44). Dietary polyphenols can promote polarization towards M1 TAMs and curcumin can reprogram M2 TAMs to M1 TAMs to inhibit OSCC ( Figure 2 ) (45, 46). Receptor for activated C kinase 1 (RACK1) fosters the polarization towards M2 TAMs improving the M2/M1 TAMs ratio by means of regulating NF-κB, which can be used as a potential therapeutic target ( Figure 2 ) (47).

A significant breakthrough has been made in cancer immunotherapy through advancements in nanomedicine. Nanomedicines targeting TAMs plays a very important role in anti-tumor immunotherapy, mainly focusing on inhibiting the recruitment of TAMs, killing TAMs, reprogramming the TAMs phenotype, and promoting the immune function of TAMs (48, 49). Mannose-modified hyaluronic acid nanocapsules targeting TAMs enable precise drug delivery and reprogramming of M2 TAMs (50).

Given the high infiltration of TAMs in OSCC and their intimate involvement in OSCC progression, targeting TAMs has emerged as a promising area of research for therapeutic intervention. However, the diverse subtypes of TAMs and complexity of TME have significantly limited the advance of immunotherapy specifically targeting TAMs in OSCC. Further elucidation of the regulatory mechanism for TAMs polarization and mediator secretion, and the development of TAMs targeting drugs using nanomedicine has become the focus of further research.

CAFs are activated fibroblasts with mesenchymal cell lineage as well as heterogeneity of phenotype and function in TME (51, 52). CAFs secrete a variety of active factors that are involved in tumor biologic behaviors such as formation and maintenance of tumor mass, generation and maintenance of cancer cell stemness, immune regulation, angiogenesis, metabolic response, ECM remodeling, homeostasis, and treatment resistance (53).

TGF-β activates fibroblasts into CAFs, and lncRNA PVT1 expression further stimulates CAFs proliferation by increasing infiltration of TGF-β (54). Exosomes released from HNSCC cells can convert normal fibroblasts into CAFs (55). Numerous evidences showed that OSCC cells-derived extracellular vesicles (EVs) can activate fibroblasts into CAFs, although EVs vary in content and may include RNAs, proteins and TGFβ (56). CAFs can activate OSCC cells and suppress immune cells, which is related to their secretion of RNAs, proteins, TGF-β and their own regulation of glucose metabolism ( Figure 3 ).

CAFs-derived exosomal immune-related genes (hsa-miR-139-5p, ACTR2 and EIF6) induce immunosuppression and promote OSCC proliferation by regulating the expression of PIGR, CD81, UACA, and PTTG1IP in cancer cells (72). CAF-derived ACLP suppresses CD8+ T cell migration and facilitates OSCC progression (73). Zhao et al. reported that the low number of CD68+ CAFs, a novel subtype of CAFs, exhibited a negative association with the proportion of Tregs (74).

CAFs can indirectly or directly inhibit immune cells to promote the progression of OSCC. The indirect inhibition effect of CAFs on immune cells is achieved by inducing immunosuppressive microenvironment. Genome sequencing and mRNA-miRNA interaction network analysis provide evidence for the correlation between CAFs-derived exosomal immune-related genes and immunosuppression. However, their detailed regulatory mechanism in OSCC has not been fully clarified due to the lack of comprehensive molecular biological analysis. The next step is to further elucidate the specific mechanism of CAFs inhibiting immune cells through in vitro and in vivo experiments, while focusing on the regulation and inhibition of CAFs on T lymphocytes.

Inflammatory cells within TME secrete chemokines like C-X-C motif chemokine (CXCL)12 and CCL17 to recruit Tregs (10, 75, 76). Alternatively, in non-inflammatory conditions, mutations in genes such as epidermal growth factor receptor (EGFR) can modulate signaling pathways, leading tumor cells to produce chemokines like CCL22 that enhance Tregs infiltration (77). Furthermore, the level of Tregs in the TME is also intricately linked to tumor-draining lymph nodes (TDLNs) (78).

Forkhead box P3 (FOXP3) serves as a significant identifier for the presence and function of regulatory T cells (Tregs), which are essential modulators for the immune system (79). Chatzopoulos et al. found that Foxp3+ T cells took part in the immunosuppressive microenvironment in TSCC (80). Hayashi reported that Foxp3+ T cell expression might play a role in nodal metastasis and exhibited a significant correlation with the weak immune response, onset of OSCC, and enhancement of tumor malignancy (81). However, Koike et al. reported that Foxp3+ T cells might not suppress immune response but might perform site-specific anti-tumor response in OSCC (82). The Foxp3+ T cell density at the invasive front exhibits a positive correlation with overall survival (OS), disease-specific survival (DFS), recurrence-free survival, and metastasis-free survival, which means Foxp3+ T cells may be a novel prognostic factor for OSCC (82). IL-23R- Tregs display a more immunosuppressive profile compared to IL-23R+ Tregs, which is regulated by IL-23 secreted by other cells in TME (83).

The expansion of CD4+ cytotoxic T lymphocytes with specific gene expression profiles may inhibit the tumor immunity in OSCC (84). Increased PD-1 levels and glycolysis in CD4+ T cells are positively associated with lymph node metastasis of OSCC ( Figure 4 ) (85). Statistical analysis revealed that the level of CD8+ T cells in TSCC exhibited a crucial correlation with the tumor inflammation signature and cytotoxicity signature (80). CD8+ T cells with the overexpression of PD-1 or TOX in OSCC have low cytotoxicity and weak ability to proliferate, which impairs the function of immunosuppression ( Figure 4 ) (79). Exhausted T cells are characterized by upregulated immunosuppressive checkpoints including PD-1, TIM-3, and CTLA-4 ( Figure 4 ) (86). Exhausted CD4+ T cells which appear earlier and more in number than exhausted CD8+ T cells foster the development of oral cancer (86).

A high ratio of CD45RO-expressing tumor-infiltrating lymphocytes (TILs) was associated with a disease-free/overall survival improvement in OSCC patients, suggesting that CD45RO-expressing TILs are useful biomarkers for OSCC (87). Caspase activation in TILs is associated with lymph node metastasis in OSCC (88). malignant cells and T cells overexpress T lymphocyte proliferation regulator-related genes (RANs) to maintain the immunosuppressive microenvironment by impeding the functionality of plasmacytoid dendritic cells (pDCs) and CD8+ T cells but enhancing the activity of Th2 cells and Th cells ( Figure 4 ) (89).

In conclusion, Tregs exhibit immunosuppressive properties, yet there are also reports indicating their anti-tumor effects. CTLs possess the capability to eliminate tumor cells, thereby inhibiting OSCC and being linked to a positive prognosis. The role of Tregs in OSCC is intricate, and their impact on the prognosis of OSCC patients remains contentious. Given this complexity, there is a pressing need for more comprehensive and in-depth research into Tregs.

TNF-α and IL-8 are prominently overexpressed in OSCC and OPMDs, significantly contributing to the advancement of malignant transformation ( Table 2 ) (105). Gupta et al. reported that both TNF-α and TNF-β took an active part in the progression of precancerous lesions and OSCC in the North India population, but the influence of TNF-α was stronger ( Table 2 ) (106). OSCC patients have abundant rare in situ cytokine-secreting T cell populations such as CD4+ Th cells secreting TNF-α and CD8+ T cells secreting IFN-γ ( Table 2 ) (107). The antigen-reactive memory T cells promote tumor immune control through the secreted type I cytokines, especially TNF-α ( Table 2 ) (107).

TGF-α has been linked to facilitating cell proliferation, and metastasis, suppressing immune responses, and promoting angiogenesis ( Table 2 ) (108). TNF-α promotes TGF-β-induced endothelial-to-mesenchymal transition (EndMT) via TGF-β signal augmentation in OSCC ( Table 2 ) (109). Furthermore, Maldonado et al. demonstrated that suppressing the secretion of TGF-β in TME would interfere with the polarization of M2 TAMs, inhibiting invasion and chemotaxis of OSCC cells ( Table 2 ) (110).

Lopez-Labady et al. have reported that the level of IL-1β and IL-8 in OSCC was comparatively less than that found in normal mucosal tissue, indicating that the chemotactic activity and anti-tumor function are decreased in OSCC (111). IL-6 can contribute to tumor growth by up-regulating anti-apoptotic genes and promoting cell proliferation via a STAT-dependent pathway ( Table 2 ) (112). IL-6 can facilitate the migration ability of OSCC cells undergoing EMT through the interaction of IL-6-ATATT3 axis and Rac1 axis ( Table 2 ) (113). IL-6 and IL-8 may alter TME by recruiting inflammatory cells and reducing TIMP-1 production, and may serve as potential prognostic biomarkers for OSCC ( Table 2 ) (114). OSCC-derived EVs can promote tumor proliferation and invasion by altering the levels of inflammatory cytokines in TME, and activating the IL-17/TRAF6/AP-1 signaling pathway ( Table 2 ) (115).

In OSCC, overexpressed VEGF participates in angiogenesis through the VEGF/VEGFR pathway, which is related to the occurrence of lymph node and distant metastasis and poor prognosis of OSCC ( Table 2 ) (116).

Cytokines, including IL-8, TNF-α, TNF-β, CXCL9, CXCL10, CCL5, CCL17, CCL22, and IL-10, exhibit close correlations with the progression of OPMDs to OSCC by fostering inflammation, angiogenesis, and immune modulation. Similarly, TGF, IL, TNF, and VEGF play pivotal roles in OSCC development by stimulating the proliferation and migration of OSCC cells, modulating the immune response and angiogenesis, and influencing bone resorption. As a result, these cytokines hold promise as potential novel prognostic biomarkers for clinical diagnosis and treatment of OSCC. However, the precise role of various cytokines in oral squamous cell carcinoma (OSCC) remains incompletely understood. Therefore, future research is necessary to delve deeper into the specific mechanisms and effects of these cytokines in OSCC, with the aim of identifying potential targets for immunotherapy.

Hypoxia biomarkers exhibit an association with early dysplastic epithelial changes, which indicates that hypoxia has a potential effect on oral carcinogenesis (117). Hypoxia exhibits a significant association with immune cell infiltration and gene mutations in OSCC (118). Hypoxic TME induces changes to histone acetylation and methylation in OSCC (119). Under hypoxia, the upregulated Tumor protein D52 (TPD52) expression at the center of OSCC is beneficial to cell viability and proliferation by inhibiting autophagic signal transduction (120). Overexpression of special AT-rich sequence-binding protein 2 (SATB2) in SCC9 cells induced by hypoxia increases the stemness of SCC9 cells and accelerates autophagy by affecting the level of Beclin-1 and the conversion of LC3-I (121). The level of regulated in development and DNA damage responses 1 (REDD1) that induces hypoxia considerably increases in OSCC leading to rising microvessel density and lymphatic metastasis (122). Suprabasin elevates the invasion, migration, and angiogenic ability of OSCC in hypoxia (123). Carbonic anhydrase 9 (CA9) is frequently expressed in TSCC due to hypoxia, which can regulate oncogenes and signaling pathways such as apoptosis, hypoxia, PI3K/AKR/mTOR signaling (124). The expression of Phosphoglycerate kinase 1 (PGK1) is also increased because of hypoxia, which engages the AKT signaling pathway to foster glycolysis, facilitate cell growth, and contribute to metastasis by promoting EMT (125). Vacuolar ATPase (V-ATPase), an enzyme that influences cellular glycolysis, primarily regulates potential of hydrogen (pH) within TME, and autophagy triggered by V-ATPase may contribute to the development of chemotherapy resistance in OSCC (126). Additionally, the inhibitory effect of odecylmethylaminoethyl methacrylate (DMAEM) on OSCC is enhanced by the low pH in hypoxia conditions (127).

The positive correlation among the expression of CD44, HIF1α, and Snail in advancing grades of OSCC and their metastatic lymph suggests that there is a dynamic interaction between CSCs, hypoxia, and EMT, which facilitates the formation and maintenance of oral CSCs (128). Bharti et al. found the overexpression of HIF-1α in T1-staged OSCC (129). HIF-1α exhibits a crucial correlation with the deterioration of precancerous lesions and the poor prognosis of OSCC (130). Swartz et al. demonstrated the level of HIF-1α expression in OSCC cells is notably higher than that in laryngeal squamous cell carcinoma (LSCC) and oropharyngeal squamous cell carcinoma (OPSCC) (131). They found that HIF-1α had a positive correlation with better OS and DFS which is different from the finding in LSCC and OPSCC (131). There exists an important statistical correlation between the overexpression of HIF-1α and enlargement in tumor size, which means HIF-1α has a potential diagnostic role and maybe a novel auxiliary diagnostic indicator (132).

Hypoxia improves the sensitivity of HSC-2 to lactoferrin (LF) which is an iron-binding glycoprotein causing HIF-1α, HIF-2α, and p53 protein expression upregulated (133). Expression of HIF-1α can modulate the level of Par3 and TJs expression in OSCC, and the crosstalk between partitioning-defective protein 3 (Par3) and tight junctions (TJs) is disrupted during hypoxia, facilitating the metastasis of OSCC by upregulating MMP activity (134). The PER1/HIF-1α negative feedback loop facilitates ferroptosis and suppresses OSCC progression (135). MicroRNA-18a performs a detrimental function in the migration and invasive capabilities of OSCC cells by downregulating the level of HIF-1α (136). HIF-1α serves as a crucial hypoxia indicator that impacts the prognosis of OSCC by regulating TME and the function of OSCC cells. However, the influence of HIF-1α on prognosis remains controversial, necessitating an investigation into the pathway through which HIF-1α operates within the TME, ultimately targeting HIF-1α as a pivotal therapeutic focus.

Lan et al. reported that hypoxia-relative genes contributed to the increase of CD68+/CD163+ TAMs and Albumin Human Serum-coated perfluorocarbon-carrying oxygen was an effective way to alleviate hypoxia (137). Hypoxia-related genes, aldolase A (ALDOA), prolyl 4-hydroxylase subunit alpha-1 (P4HA1), PGK1, and VEGFA, exhibit a significant correlation with OS in OSCC, which can predict the prognosis of patients (138). Lu and Liu found a positive correlation between phosphofructokinase, platelet (PFKP) and serpin peptidase inhibitor, clade E, member 1 (SERPINE1), which could also lead to poor prognosis (139). Furthermore, Han et al. revealed that the mutation of nuclear receptor binding SET domain protein 1 (NSD1) also indicated a high hypoxia potential index in OSCC (118). Hypoxia leads to elevated levels of miR-1825 in OSCC cells, promoting angiogenesis via the exosomal miR-1825/TSC2/mTOR signaling pathway (140). Piwi-interacting RNA-39980 inhibits the proliferation and migration of TSCC by suppressing the oncogene farnesyl-diphosphate farnesyltransferase 1 (FDFT1), disrupting ROS production, fostering a hypoxic TME, elevating levels of damaged DNA, and ultimately triggering cell death (141). In hypoxic conditions, hsa_circRNA_101036 boosts endoplasmic reticulum (ER) stress via the miR-21-3p/transmembrane and tetratricopeptide repeat containing 1 (TMTC1) pathway, ultimately causing cell apoptosis and generating ROS that can be removed by peroxiredoxin 5 (142, 143).

In hypoxic TME, hypoxia-relative genes facilitate cell death by modulating ROS levels, resulting in the inhibition the proliferation and migration. Meanwhile, Hypoxia promoting angiogenesis via the increased exosomal miR-1825. Conversely, the upregulation of numerous hypoxia-relative genes leads to an increase in microvascular density or enhances the serum’s oxygen-carrying capacity, thereby providing adequate nourishment and oxygen to OSCC.