In OSCC, DNA hypomethylation drives tumorigenesis by activating oncogenes and pro-metastatic factors, dysregulating key signaling pathways, and remodeling the TME (17). The following sections detail the roles of specific hypomethylated genes in these processes.

Dysregulated DNA hypomethylation plays a crucial role in reshaping the immunosuppressive TME (68). A key mechanism involves tumor-derived factors that actively reprogram the epigenome of immune cells. For instance, calnexin, an endoplasmic reticulum chaperone upregulated in OSCC, is expressed on the tumor cell membrane. It impairs antitumor immunity by interacting with T cells and inducing DNA hypomethylation of the programmed cell death protein 1 (PD-1) promoter CpG island. This epigenetic alteration leads to increased PD-1 expression on CD4⁺ and CD8⁺ T cells, driving their functional exhaustion and suppressing cytokine production. Consequently, calnexin-expressing tumors exhibit reduced T-cell infiltration and poorer patient survival, illustrating how a tumor cell surface protein can exploit DNA demethylation to enforce an immunosuppressive TME (69).

A primary oncogenic effect of DNA hypomethylation is the transcriptional reactivation of genes that drive tumor aggression and metastasis. For instance, the promoter of Wnt1 inducible signaling pathway protein 1 (WISP1) is significantly hypomethylated in OSCCs with lymph node metastasis, leading to its high expression which promotes cancer spread and correlates with poorer survival (70). Similarly, hypomethylation at the homeobox protein CDX-1 (CDX1) motif reactivates the transcription factor gene homeobox C9 (HOXC9) in OSCC. HOXC9 upregulation promotes tumor invasion and metastasis by driving the expression of MMP13 through the ITGA6/PI3K/Akt signaling axis and is associated with advanced disease stages (71). However, the proposed HOXC9-driven PI3K-Akt/MMP13 axis represents a potential pathological association requiring further direct validation, and the regulatory interplay between HOXC9 and upstream factors remains to be fully elucidated.

The activation of pivotal oncogenic signaling pathways is another major consequence of gene-specific hypomethylation. For instance, in higher-grade OSCC samples, the dickkopf Wnt signaling pathway inhibitor 2 (DKK2) and DKK4 genes, which are inhibitors of the Wnt signaling pathway, were found to be hypomethylated. This hypomethylation is postulated to lead to their increased expression, potentially facilitating tumor cell invasion and progression through the modulation of the Wnt pathway, a key driver in oral carcinogenesis (72). Furthermore, in OSCC associated with oral lichen planus, promoter hypomethylation drives the overexpression of SRY-box transcription factor 11 (Sox11), which in turn activates the PI3K/AKT signaling pathway and enhances glycolysis to promote tumor growth (73). Integrated bioinformatic analyses corroborate that numerous hypomethylated and upregulated genes in OSCC are key components of the PI3K/AKT and epithelial-mesenchymal transition (EMT) pathways (74). Therefore, identifying additional oncogenic pathways activated by this mechanism remains a crucial goal for future research.

Beyond the regulation of individual genes, widespread hypomethylation exerts a profound influence on the TME. Global DNA hypomethylation, measurable through long interspersed element 1 (LINE-1) repetitive elements, is a common feature of OSCC, and low LINE-1 methylation levels in pre-malignant lesions predict a higher risk of progression to cancer (75). This hypomethylated state extends to immune-related genes. Genome-wide studies have identified unique sets of hypomethylated promoters enriched for immune response genes, suggesting that DNA hypomethylation can facilitate lymphocyte infiltration and modulate anti-tumor immunity (76). This is supported by findings that promoters of various immune genes are significantly more hypomethylated in OSCC tissues compared to normal mucosa, irrespective of HPV status (77).

The consistent pattern of DNA hypomethylation in OSCC offers considerable potential for clinical translation into biomarkers (78). Studies on oral brushing samples have identified hypomethylated genes like miR-296 and telomerase reverse transcriptase (TERT) in OSCC and pre-malignant lesions, demonstrating their utility for early and non-invasive detection (79). In mouse models, hypomethylation and overexpression of Fgf3 occur during the early stages of oral carcinogenesis, marking it as a potential early detection biomarker (80). From a prognostic perspective, hypomethylation of the ornithine aminotransferase (OAT) gene promoter is associated with a radio-resistant TME and poorer survival after radiotherapy (81), whereas the hypomethylated state of the tubulin polymerization promoting protein family member 3 (TPPP3) promoter, which favors its tumor-suppressive expression, is an indicator of good prognosis (82). The lncRNA H19 also exhibits promoter hypomethylation and high expression in OSCC, which is associated with a significantly lower 5-year survival rate, underscoring its role in disease progression (83). It promotes disease progression through diverse mechanisms, including acting as a competitive endogenous RNA (ceRNA) to sponge miRNAs such as miR-138, miR-29b, and let-7a. This sponge activity leads to the upregulation of downstream targets like zeste homolog 2 (EZH2), zinc finger E-box binding homeobox 1 (ZEB1), and 6-phosphofructo-2-kinase/fructose-2,6-biphosphatase 3 (PFKFB3), thereby driving EMT, enhancing cell proliferation, invasion, and glycolysis, and contributing to immunosuppressive TME (84).

In conclusion, DNA hypomethylation is a pervasive and driving force in OSCC pathogenesis. It functions not only by activating specific oncogenes and pathways that enhance tumor cell proliferation, invasion, and metastasis, but also by reshaping the global genomic and immune landscape, most notably by reshaping immunosuppressive TME, to ultimately favor tumor survival and progression. The wealth of hypomethylated genes and loci, many with clear clinical correlations, positions DNA hypomethylation as a rich source of mechanistic insights and a promising foundation for developing diagnostic, prognostic, and therapeutic strategies against OSCC.

DNA hypermethylation of tumor suppressor genes is a fundamental epigenetic mechanism driving OSCC pathogenesis (85). A key consequence of this silencing is the blunting of anti-tumor immune responses and the fostering of immunosuppressive TME, which is increasingly recognized as a critical step in OSCC progression (86). Recent studies have identified a range of genes silenced by this mechanism, which can be broadly grouped by their disrupted cellular functions, underscoring the critical role of epigenetic dysregulation in OSCC.

A primary mechanism by which promoter hypermethylation directly establishes immunosuppressive TME in OSCC is the silencing of genes that regulate key immune checkpoint molecules (87). The paired box 1 (PAX1) gene is downregulated due to arecoline-induced hypermethylation. This loss of PAX1 function is a pivotal event that not only enhances cancer stem cell (CSC) -like properties but also actively promotes an immunosuppressive TME by upregulating interferon induced protein with tetratricopeptide repeats 1 (IFIT1) and the immune checkpoint programmed death ligand 1 (PD-L1), thereby facilitating tumor immune evasion. Future validation in an orthotopic animal model is needed to confirm the involvement of the IFIT1/PD-L1 signaling pathway (88). Similarly, promoter hypermethylation-induced downregulation of the tumor suppressor miRNA miR-34b/c drives OSCC aggressiveness and is a marker of adverse clinical outcomes, including shortened survival (89). Given the established role of the miR-34 family in directly targeting and suppressing immune checkpoints such as PD-L1 in other cancers (90), its silencing in OSCC likely contributes directly to immune dysregulation within the TME. This epigenetic event is strongly associated with advanced tumor stage, nodal metastasis, cancer recurrence, and poor survival outcomes particularly in HPV-negative OSCC patients (89). The silencing of PAX1 and miR-34b/c exemplifies how hypermethylation can be leveraged by OSCC to directly suppress anti-tumor immunity.

A significant group of hypermethylated genes disrupts apoptosis and cell proliferation, which can indirectly alter immune surveillance. The pro-apoptotic gene homeobox A5 (HOXA5) is frequently hypermethylated in OSCC tissues leading to its downregulation, contributing to reduced cell death. Reactivation of HOXA5 expression not only induces cell death but also enhances chemosensitivity both in vitro and in vivo (91). Another key gene, PAX9, a differentiation-associated tumor suppressor, is silenced through hypermethylation. Pharmacological inhibition of DNMTs can reactivate PAX9 expression, which in turn triggers apoptosis, inhibits cell growth, and suppresses cancer stemness through an autophagy-dependent pathway (92). The evasion of apoptosis and enhanced stemness mediated by the silencing of these genes contribute to a tumor cell population that is resistant to immune cell-mediated killing and fosters a pro-tumorigenic niche.

Another critical pathway affected by hypermethylation involves genes that control cell invasion, metastasis, and stemness. Methylation-mediated silencing of miR-124–3 represents one mechanism. The suppression of this miRNA leads to the overexpression of its target, the oncogene leucine rich repeat containing 1 (LRRC1), which subsequently drives OSCC cell proliferation and migration (93). The homeobox gene HOXA3 also exhibits epigenetic regulation, with its expression showing an inverse correlation with promoter methylation levels. Hypermethylation of its 3’ untranslated region is notably associated with poor overall survival in advanced-stage OSCC patients (94). Moreover, HOXA3 expression may also be subject to post-transcriptional regulation by ncRNAs and RNA-binding proteins, although this complex regulatory network requires definitive experimental confirmation. Furthermore, the tumor suppressor gene transglutaminase 3 (TGM-3) shows significant promoter hypermethylation in OSCC. This epigenetic alteration is quantitatively associated with advanced tumor stage and higher histological grade, highlighting its role in cancer progression (95). The concerted hypermethylation of this group of genes thus equips OSCC cells with enhanced invasive, metastatic, and stem-like capabilities, all of which are hallmarks of an aggressive TME that supports tumor dissemination and therapy resistance.

Beyond the silencing of protein-coding genes and miRNAs, hypermethylation also affects genes that regulate the epigenetic machinery itself. The methylation status of DNMT3A and tet methylcytosine dioxygenase 2 (TET2), a key demethylation-initiating enzyme, is aberrant in OSCC. Functionally, knockdown of DNMT3A and overexpression of TET2 can inhibit the proliferation and migration of OSCC cells, indicating their crucial roles in maintaining epigenetic balance (96). On a different level, the protein methyltransferase SET domain containing 6 (SETD6) is upregulated in OSCC. Silencing SETD6 inhibits OSCC tumorigenesis by reducing the promoter methylation of its substrate proteins P21 (RAC1) activated kinase 4 (PAK4) and RELA proto-oncogene, NF-κB subunit gene (RelA), thereby counteracting their pro-tumorigenic activities (97). This category reveals a self-reinforcing loop in OSCC, where the epigenetic regulators themselves become targets of dysregulation, further amplifying the global epigenetic disruption that underpins a malignant and immunosuppressive TME.

In summary, the landscape of hypermethylated genes in OSCC is diverse, encompassing key regulators of apoptosis, invasion, stemness, drug resistance, and the epigenetic apparatus itself, which significantly reshapes the TME of OSCC (Table 1). The collective evidence from these studies strongly supports the potential of these hypermethylated genes as valuable biomarkers for diagnosis and prognosis, as well as promising targets for epigenetic therapy in OSCC.

Histone methylation, a dynamic process regulated by methyltransferases and demethylases, plays a pivotal role in OSCC pathogenesis (102). Beyond driving cell-autonomous malignant behaviors, emerging evidence underscores its profound impact on shaping the TME, particularly in fostering immunosuppression and therapeutic resistance (103). The dysregulation of this epigenetic mechanism directly influences cellular responses to TME stressors and activates pathways that collectively mold a pro-tumorigenic niche (104).

A primary mechanism through which histone methylation influences OSCC involves direct modulation of the hypoxic and acidotic TME, a key driver of immunosuppression (105). This TME in OSCC, primarily regulated by hypoxia-inducible factor 1-alpha (HIF-1α), impacts the activity of histone-modifying enzymes, thereby reshaping the histone modification landscape. A foundational study utilizing LC-MS-based proteomics directly demonstrated that hypoxic, acidotic, and combined stress conditions in the CAL27 OSCC cell line induce distinct, position-dependent alterations in histone methylation and acetylation marks, such as histone H3 trimethylation at Lys36 (H3K36me3), histone H2A lysine 9 acetylation (H2AK9Ac), and H4K16Ac (105). Building upon this, a subsequent mass spectrometry-based proteomic study revealed that the combination of the HDAC inhibitor vorinostat and the thioredoxin-1 (Trx-1) inhibitor PX-12 further alters a spectrum of histone methylation and acetylation marks under hypoxic conditions (106). These findings position histone methylation as a crucial epigenetic interface between the hypoxic and acidotic TME and cancer cell adaptability, suggesting that targeting these modifications may reverse TME-mediated therapy resistance.

Beyond the direct hypoxic response, histone methylation fuels OSCC progression by activating specific oncogenic axes that enhance tumor aggressiveness and indirectly reshape the TME. A compelling signaling cascade involves the nuclear receptor interacting protein 1 (NRIP1)/nuclear receptor binding SET domain protein 2 (NSD2)/DiGeorge critical region 8 (DGCR8) axis. NRIP1, an aberrantly expressed transcription factor, activates the transcription of the methyltransferase NSD2, which in turn increases DGCR8 transcription by modulating histone methylation near its promoter. This axis significantly augments OSCC cell proliferation, migration, invasion, and in vivo metastatic potential (107). However, a key limitation of this work is that the critical downstream effectors of DGCR8, which are most likely specific miRNAs or signaling pathways, need to be further elucidated. Similarly, lysine demethylase 3A (KDM3A) facilitates OSCC proliferation and invasion by removing the repressive H3K9me2 mark from the promoter of doublecortin like kinase 1 (DCLK1), a CSC marker, thereby upregulating its expression (108). However, this established link is likely to represent a part of a broader regulatory network. A critical next step is to investigate the full spectrum of transcription factors targeted by KDM3A and to unravel the downstream signaling mechanisms through which DCLK1 executes its oncogenic functions. By promoting this stem-like phenotype, the KDM3A-H3K9me2-DCLK1 axis indirectly sustains the immunosuppressive TME, a consequence of the well-documented immunomodulatory and therapy-resistant properties of CSCs.

Specific histone methylation marks have clinical and prognostic significance in OSCC, indicating their role in the TME. Immunohistochemical analyses have consistently linked aberrant histone methylation marks to aggressive disease and poor survival. The upregulation of the heterochromatin mark H3K9me3 in OSCC tissues is associated with advanced disease features like depth of invasion (109). Future studies should dissect the functional relationship between these epigenetic marks and key oncogenic pathways, particularly the poorly explored link with PI3K/AKT signaling in OSCC. Furthermore, high levels of the repressive mark H3K27me3, either alone or in combination with the active mark H3K27ac, are powerful predictors of shorter survival in OSCC patients (110). The prognostic value of these marks underscores their potential as biomarkers and therapeutic targets.

In summary, histone methylation drives OSCC progression through dual mechanisms: directly by modulating cellular responses to the TME, such as hypoxia, and indirectly by activating oncogenic pathways that remodel the tumor stroma. Key marks like H3K9me3 and H3K27me3 are linked to poor prognosis and likely foster an immunosuppressive TME. Targeting these epigenetic regulators presents a promising strategy to simultaneously curb tumor growth and counteract immunosuppression in OSCC.

Histone acetylation serves as a critical epigenetic mechanism in OSCC pathogenesis, with its dysregulation extending beyond cell-autonomous effects to actively shape the TME and immune response (111). Evidence increasingly shows that targeting acetylation pathways can directly reverse immunosuppressive networks and overcome therapy resistance.

Epigenetic modulation of histone acetylation can directly reshape the immune landscape of the TME (112). Pharmacological inhibition of HDAC6 by tubastatin A (TSA) presents a compelling case. It suppresses the secretion of the pro-tumorigenic cytokine interleukin 1β (IL-1β) and concurrently reprograms the immune landscape. This is evidenced by a reduction in myeloid-derived suppressor cells (MDSCs) and M2-type tumor-associated macrophages (TAM) alongside an increase in M1-type TAM, effectively alleviating immunosuppression (113). Furthermore, the oncogenic lncRNA deleted in lymphocytic leukemia 1 (DLEU1) influences immune signaling by modulating the active enhancer mark H3K27ac. DLEU1 knockdown reduces H3K27ac levels and suppresses interferon-stimulated genes, linking this specific acetylation mark to the regulation of JAK/STAT signaling and TME-associated immune responses (114). However, the mechanism behind the concomitant downregulation of genes unrelated to interferon signaling remains an open question, hinting at a broader epigenetic function for DLEU1 beyond the JAK/STAT pathway.

The regulatory network of histone acetylation involves complex interactions with lncRNAs, which can act as critical modulators of specific histone marks (115). As mentioned previously, the lncRNA DLEU1 contributes to OSCC by maintaining H3K27ac levels to activate oncogenic transcriptional programs (114). Conversely, a separate study in head and neck squamous cell carcinoma identified the interferon-alpha-induced lncRNA lncMX1-215, which negatively regulates immunosuppression. This lncRNA exerts its function by directly binding to the histone acetyltransferase general control non-depressible 5 (GCN5), a known writer of the H3K27ac mark. This interaction interrupts the binding of GCN5 to H3K27ac sites on the promoters of immunosuppressive molecules such as PD-L1 and galectin-9, thereby inhibiting their transcription (116). These findings demonstrate that lncRNAs can directly interface with the histone acetylation machinery, either by facilitating or by obstructing the deposition of specific acetyl marks to control gene expression programs in OSCC.

The clinical significance of these mechanisms is underscored by specific acetyl marks serving as robust prognostic biomarkers. Hyperacetylation of H3K18 and H3K9 is linked to advanced invasion and high T stage (109), while elevated H3K27ac levels are associated with shorter patient survival (110). These marks epitomize the aggressive, TME-shaped tumor phenotype. Notably, the expression and therapeutic response may exhibit sex-related differences, as female mice showed distinct dynamics of H3K9ac and H3K14ac during carcinogenesis (117), pointing to personalized therapeutic considerations.

Evidence indicates histone methylation and acetylation interact closely in OSCC (110). The lysine methyltransferase 2D (KMT2D), which catalyzes H3K4me1, also promotes H3K27ac enrichment at enhancers of key oncogenes like KLF transcription factor 7 (KLF7), activating their transcription (118). This shows one histone modifying enzyme can directly influence the deposition of another mark. This dynamic crosstalk represents a key epigenetic feature of OSCC, suggesting combination therapies targeting this network may be more effective than single agents.

In summary, histone acetylation is crucially implicated in molding the immunosuppressive TME of OSCC. It operates by directly reprogramming immune cells, mediating adaptation to TME stressors, and fueling the chemo-resistant CSC niche. The potent anti-tumor and immune-modulating effects of HDAC inhibitors, evidenced by both preclinical and clinical investigations (119), solidifying the targeting of acetylation pathways as a promising strategy to disrupt the pro-tumorigenic TME and improve OSCC treatment outcomes.

lncRNAs are a class of transcripts longer than 200 nucleotides with limited or no protein-coding potential (125). They have emerged as critical regulators of gene expression at epigenetic, transcriptional, and post-transcriptional levels (126). In cancer, lncRNAs play pivotal roles in diverse biological processes, including cell proliferation, apoptosis, metastasis, and immune responses, functioning either as oncogenes or tumor suppressors (127, 128). In the context of OSCC, numerous lncRNAs have been identified to be dysregulated and contribute significantly to tumor initiation and progression (Figure 6).

miRNAs function as pivotal regulators in the pathogenesis of OSCC by modulating key oncogenic and tumor-suppressive pathways (Figure 7) (157). Their expression is frequently dysregulated in OSCC, influencing critical cellular processes such as proliferation, apoptosis, invasion, and angiogenesis (158).

circRNAs are a novel class of endogenous ncRNAs that have emerged as pivotal regulators in OSCC (173). They frequently function as molecular sponges for miRNAs, thereby derepressing miRNA targets and influencing various aspects of tumor biology (174). Functioning as key miRNA sponges, specific circRNAs are central regulators in OSCC. Their dysregulation promotes hallmark malignant phenotypes, including proliferation, metabolic reprogramming, metastasis, and therapy resistance, by orchestrating oncogenic signaling and remodeling the TME, notably through angiogenesis (Figure 8) (175).

DNA methylation inhibitors represent a promising class of epi-drugs for OSCC. These agents primarily function by inhibiting DNMTs, thereby reversing the aberrant hypermethylation of tumor suppressor genes and restoring their anti-tumor functions (67). The exploration of these inhibitors spans from well-characterized drugs like 5-aza (5-azacytidine) to novel applications of existing compounds such as statins.

The foundational mechanism of DNA methylation inhibitors involves the reactivation of silenced tumor suppressor genes. Treatment with DNMT inhibitors like 5-aza leads to the demethylation and subsequent upregulation of critical tumor suppressive miRNAs. For instance, 5-aza treatment reactivated miR-6741-3p, which exerts its tumor suppressive function by directly targeting and downregulating the oncogene SRSF3 (199). Beyond nucleoside analogs, the local anesthetic procaine also acts as a DNMT inhibitor. It reactivated the differentiation gene PAX9 by suppressing DNMT activity, which in turn inhibited cell growth, triggered apoptosis, and reduced cancer stemness through an autophagy-dependent pathway both in vitro and in vivo (92). A key future direction is to delineate the upstream regulatory mechanism, specifically how nuclear factor E2-related factor-2 (NRF2) contributes to sustaining PAX9 expression and the ensuing antitumor response, which will provide a more complete understanding of procaine’s action.

Research has demonstrated that combining DNA methylation inhibitors with other therapeutic modalities can yield synergistic anti-cancer effects. The combination of DNMT inhibitors such as 5-aza-2’-deoxycytidine (5-aza-dC) or N-phthalyl-L-tryptophan (RG108) with the HDAC inhibitor TSA significantly decreased OSCC cell viability. This enhanced cytotoxicity was attributed to cell cycle arrest in the S and G2/M phases and an increase in DNA double-strand breaks (200). However, these findings from 2D culture models require validation in more physiologically relevant systems, such as 3D spheroids and in vivo models. Another study confirmed that the combination of the DNMT inhibitors zebularine and the HDAC inhibitors Valproic acid (Vpa) exhibited significant cytotoxic effects and upregulated tumor suppressor genes including P16 and P21 in OSCC cells. This combination also effectively reduced tumor volume in a xenograft mouse model of well-differentiated OSCC (198). Furthermore, the combination of 5-aza with spirulina-based photodynamic therapy (PDT) showed a synergistic effect, markedly enhancing cell death, apoptosis, and DNA damage while suppressing cell migration and invasion in OSCC cells (201). To advance these promising combination strategies, future efforts should focus on delineating the underlying signaling pathways and rigorously evaluating their efficacy and safety in advanced animal models and clinical trials. Additionally, exploring novel combinations, such as with immunotherapy, or developing more efficient photosensitizers could further unlock their full therapeutic potential.

Emerging evidence indicates that the antitumor effects of DNA methylation inhibition extend beyond cancer cells to modulate the TME. Inhibition of DNMT1 was found to improve the TME in immunocompetent mouse OSCC models. This improvement was characterized by a decrease in MDSCs and an increase in tumor-infiltrating T cells, which collectively contributed to delayed tumor growth (202). This suggests that DNMT1 inhibitors can potentiate antitumor immunity, offering a promising strategy for OSCC treatment.

Some therapeutic agents exert their anti-cancer effects in part through indirect inhibition of DNMTs. Statins, including cerivastatin and simvastatin, were shown to inhibit OSCC cell proliferation by suppressing DNMT1 expression. This suppression led to the upregulation of p21 and subsequent G0/G1 cell cycle arrest, highlighting a novel epigenetic mechanism underlying the anti-proliferative effect of statins (202). Additionally, the tumor suppressor inhibitor of growth family member 4 (ING4) was found to play a repressive role in OSCC by promoting the degradation of NF-κB p65, which subsequently reduced DNMT1 expression. This reduction in DNMT1 led to decreased methylation of the aldehyde dehydrogenase 1 family member A2 (ALDH1A2) gene, thereby inhibiting OSCC progression (203). These studies reveal an indirect epigenetic pathway through which certain agents inhibit OSCC by targeting DNMT activity.

Recent research aims to refine the application of DNMT1 targeting to enhance efficacy and minimize toxicity. A key study demonstrated that DNMT1 knockdown or inhibition in OSCC remodeled a specific global DNA hypomethylation pattern. This altered methylation landscape collaboratively hampered the activation of the PI3K/AKT and CDK2-Rb pathways while inactivating glycogen synthase kinase 3 beta (GSK3β). This targeted approach achieved superior anti-cancer efficacy in vivo compared to a PI3K inhibitor alone and circumvented the adverse effects such as blood glucose dysregulation associated with direct kinase inhibition (17).

In conclusion, DNA methylation inhibitors demonstrate considerable therapeutic potential against OSCC through multifaceted mechanisms. They directly reactivate silenced tumor suppressor genes and miRNAs, synergize with other epigenetic therapies or conventional treatments, and remarkably remodel the immunosuppressive TME. The ongoing exploration of DNA methylation inhibitors, both as monotherapies and as integral components of combination regimens, heralds a promising and evolving frontier in the precision epigenetic therapy of OSCC.

HDAC inhibitors constitute a significant class of epi-drugs under investigation for OSCC (204). These agents function by increasing histone acetylation, which alters chromatin structure and gene expression, leading to various anti-tumor effects including cell cycle arrest, apoptosis induction, inhibition of metastasis, and modulation of the TME (29). Research spans from pan-HDAC inhibitors to isoform-selective agents, exploring their potential both as monotherapies and in combination regimens.

HDAC inhibitors demonstrate potent anti-proliferative and pro-apoptotic effects in OSCC models. Apicidin was shown to inhibit cell growth in murine OSCC cells both in vitro and in vivo, an effect linked to its selective suppression of HDAC8 expression and the subsequent induction of apoptosis and autophagy (205). Similarly, the natural compound trichodermin (TCD) suppressed OSCC cell proliferation, induced G2/M phase arrest, and triggered caspase-dependent apoptosis, also involving mitochondrial dysfunction and downregulation of HDAC-2 and its downstream signaling (206). Sodium phenylbutyrate (SPB) also inhibited OSCC cell viability and promoted apoptosis, accompanied by downregulation of the anti-apoptotic protein B-cell lymphoma 2 (BCL-2) and increased caspase-3 cleavage (207). These findings underscore that HDAC inhibitors counter OSCC through a convergence of mechanisms centered on arresting growth and inducing cell death.

Targeting specific HDAC isoforms, particularly HDAC6, has emerged as a promising therapeutic strategy. Pharmacological inhibition of HDAC6 by TSA was reported to overcome cisplatin resistance in OSCC by targeting CSCs. This effect was mediated through increased oxidative stress, DNA damage, and reduced expression of the antioxidant protein peroxiredoxin 2 (PRDX2) (208). Another study confirmed that HDAC6 inhibition with TSA attenuated tumor progression in a mouse model, which was associated with suppressed lysosomal exocytosis of the pro-tumorigenic cytokine IL-1β via tubulin acetylation (113). Furthermore, HDAC6 blockade with TSA was found to attenuate necroptotic signaling by suppressing the receptor interacting serine/threonine kinase 1 (RIPK1)/RIPK3/mixed lineage kinase domain like pseudokinase (MLKL) pathway, thereby reducing cell death and reprogramming the immunosuppressive TME to enhance anti-tumor immunity (209). These studies establish HDAC6 inhibition as a versatile therapeutic strategy that counteracts OSCC through multiple parallel mechanisms.

The combination of HDAC inhibitors with other therapeutic agents often yields superior anti-tumor efficacy (194). The co-administration of the Class I HDAC inhibitor 4SC-202 and the mTORC1/C2 inhibitor INK128 exhibited synergistic effects in suppressing OSCC cell growth, sphere formation, and tumor initiation. This combination repressed the expression of the stemness factor SOX2 through distinct mechanisms, namely miRNA-mediated mRNA degradation and inhibition of cap-dependent translation (29). The rationale for such drug conjugates was supported by the development of Roxyl-ZR, a single molecule combining HDAC and cyclin-dependent kinase (CDK) inhibitory motifs, which showed enhanced efficacy in suppressing OSCC proliferation and metastasis by inducing cell cycle arrest and apoptosis and inhibiting the JAK1/STAT3 pathway (210). Another synergistic interaction was observed between the HDAC inhibitor vorinostat and the Trx-1 inhibitor PX-12 under hypoxic conditions, which was more effective than either agent alone and was linked to specific alterations in histone acetylation and methylation marks (106, 211). Together, these findings demonstrate that HDAC inhibitor-based combinations exert synergistic anti-tumor effects in OSCC by engaging multiple targets and pathways.

HDAC inhibitors also show potential in targeting CSCs and overcoming therapy resistance (212). Research indicated that the HDAC inhibitor vorinostat and the NF-κB inhibitor emetine could both disrupt the CSC subpopulation in OSCC when used individually. Their combined administration did not yield further enhancement, suggesting that each agent independently targets CSCs through distinct pathways, namely increasing histone acetylation and inhibiting NF-κB signaling, respectively (213). Vpa was found to inhibit OSCC cell viability, induce G1 cell cycle arrest, and promote apoptosis, effects that were associated with the regulation of SUMOylation in both in vitro and in vivo models (214). The combination of the DNMT inhibitor zebularine and the HDAC inhibitor vpa also demonstrated significant cytotoxic effects and upregulated tumor suppressor genes in OSCC cells, effectively reducing tumor volume in vivo (198). However, as the xenograft model used cannot fully recapitulate the human oral cancer TME, future studies employing more representative models are warranted to strengthen the clinical relevance of these promising findings.

In conclusion, HDAC inhibitors demonstrate multi-faceted anti-tumor efficacy in OSCC, ranging from direct cytotoxicity and metastasis inhibition to overcoming chemoresistance and modulating immunosuppressive TME. The emergence of isoform-selective inhibitors, particularly targeting HDAC6, and rational combination strategies represent a promising direction for future therapeutic development.