Among the hallmarks of cancer, the acquisition of resistance to cell death plays an important role in cancer initiation and progression to high-grade malignant states, which are frequently unresponsive to conventional anti-cancer therapies [1]. In recent years, there has been a significant improvement in the understanding of different mechanisms of cell death beyond the traditionally observed apoptosis and necrosis ^[2][3][2,3]. Alternative cell death pathways, collectively called programmed cell death (PCD), have been identified in a variety of pathological processes, including oral cancer [4]. These pathways not only contribute to the understanding of cancer pathophysiology [5] but also reveal an intricate balance between molecules involved in cell survival and death, with significant implications, for example, as prognostic biomarkers and in the development of new therapeutic strategies that are beginning to be explored [6].

While malignant cells develop strategies to escape or limit conventional cell death pathways, understanding of their ability to escape death by other mechanisms is much more limited [5]. Since 2018, the Nomenclature Committee on Cellular Death has classified PCD into 12 subtypes of death that differ in molecular mechanisms but may share small similarities in their morphological characteristics, ranging from a necrotic profile, that is, unprogrammed and with a disordered appearance, to an apoptotic profile with an organized profile ^[2][7][2,7]. However, an update proposed by Yan, Elbadawi and Efferth [3] divided cell deaths into three large groups based on activation of specific signaling pathways: (1) Non-programmed cell death (NPCD) or necrosis, (2) apoptotic programmed cell death (APCD) and (3) non-apoptotic programmed cell death (NAPCD), which differs from the apoptotic form because it does not maintain the integrity of the cell membrane and is independent of caspases. Similar to apoptosis, NAPCD has a highly regulated molecular machinery that can be targeted or modulated by molecular strategies [2].

The most studied emerging NAPCD types in oncology are ferroptosis, pyroptosis and necroptosis [8]. Ferroptosis involves the accumulation of lipid peroxides due to disrupted cellular antioxidant defenses, leading to oxidative stress-induced cell death [9]. Cancer cells can escape ferroptosis by enhancing antioxidant defenses and modifying lipid metabolism, enabling them to survive and proliferate despite conditions that typically trigger ferroptotic cell death [10]. Meanwhile, during pyroptosis and necroptosis, the intracellular content is expelled from the cell through membrane pores formed by proteins such as those from the Gasdermin family (GSDM) in pyroptosis and mixed lineage kinase domain-like (MLKL) in necroptosis, resulting in recruitment of inflammatory cells ^[11][12][11,12]. The pro-inflammatory environment induced by the extravasation of intracellular contents may promote the transformation and progression of tumor cells ^[13][14][13,14]. However, the role of NAPCD in cancer progression is only partially understood, and the literature is conflicting.

2. Ferroptosis

Ferroptosis is a distinguished type of NAPCD characterized by iron-dependent lipid peroxide accumulation, particularly of polyunsaturated fatty acids ^[15][16][97,98]. The research landscape of this type of cell death and its implications in diseases is relatively recent ^[17][18][99,100] but promising for increased knowledge on mechanisms and therapeutic strategies in cancer ^[17][19][99,101]. Ferroptosis can occur through two major pathways: the extrinsic pathway or transporter-dependent, and the intrinsic or enzyme-regulated pathway ^[17][20][99,102] (Figure 1). Despite being distinct pathways, it is important to highlight that one can influence the other, as both rely on iron metabolism and glutathione (GSH)-dependent antioxidant mechanisms ^[21][103]. The extrinsic mechanism depends on the balance of iron and amino acid transport across the cell membrane ^[20][102]. A higher intracellular iron level increases the production of reactive oxygen species (ROS) through the Fenton reaction ^[22][104]. Moreover, inhibition of the Xc-system reduces cystine uptake, which is essential for synthesizing GSH ^[20][102]. The depletion of GSH reduces the cell’s antioxidant defenses and favors lipid peroxidation. In turn, the intrinsic pathway is mainly induced by inhibiting glutathione peroxidase 4 (GPX4), an enzyme that plays a pivotal role in reducing lipid hydroperoxides to non-toxic lipid alcohols using GSH as a cofactor ^[23][105]. The inhibition of GPX4 activity occurs in an unchecked accumulation of lipid hydroperoxides, leading to cellular damage and eventual ferroptosis cell death ^[20][23][102,105]. Tumor cells are more susceptible to ferroptosis due to altered metabolism and increased iron uptake ^[16][98]. Although ferroptosis is expected to be associated with tumor suppression ^[17][99], evidence demonstrates that ferroptosis regulatory pathways may promote tumor growth or progression ^[24][106]. The p53 gene can promote ferroptosis by repression of solute carrier family 7 member 11 (SLC7A11), a component of the cystine/glutamate antiporter, reducing GSH and increasing cellular oxidative stress and lipid peroxidation ^[25][107]. In addition, tumor cells display increased iron uptake, which can increase the Fenton reaction, producing reactive oxygen species and lipid peroxidation ^[26][108]. In other circumstances, the oxidative stress, a prominent feature of ferroptosis, can activate the nuclear erythroid factor 2-related factor 2 (NRF2), a well-known transcription factor activated in response to oxidative stress ^[27][109]. Although NRF2 may be cytoprotective, chronic NRF2 activation in cancer cells supports proliferation, metabolic reprogramming and resistance to therapy ^[28][29][110,111]. The hypoxia inducible factor 1 subunit alpha (HIF-1α) has also been implicated in increasing iron storage and transport proteins, influencing cellular propensity for ferroptosis ^[30][112]. While HIF-1α is primarily known for its role in the cellular response to hypoxia, oxidative stress can also stabilize HIF-1α, resulting in the transcription of genes involved in angiogenesis, glucose metabolism, and cell survival, and supporting tumor growth, angiogenesis, and metastasis ^[31][113]. The tumor microenvironment is also influenced by ferroptosis once tumor cells release damage-associated molecular patterns (DAMPs) that recruit and activate immune cells such as dendritic cells and macrophages, initiating an antitumor immune response ^[32][114]. Increased lipid peroxidation resulting from ferroptosis may also increase T cell-mediated cytotoxicity against tumor cells ^[33][115]. Targeting ferroptosis in OSCC may offer a new path in tumor treatment ^[34][116], especially those resistant to traditional therapy ^[17][99], and holds the potential to be integrated into combination therapies for enhanced efficacy ^[35][117]. Cancer cells, due to their high metabolic activity and dependency on iron and lipid metabolism ^[36][37][118,119], are vulnerable targets for ferroptosis induction, which could inhibit their growth and proliferation. This is further enhanced by targeting CSCs, which are often responsible for tumor initiation and recurrence ^[38][39][120,121]. Iron metabolism plays a crucial role in CSC maintenance and survival ^[40][41][122,123]. These cells exhibit an “iron addiction” by showing a higher concentration of iron compared to the non-stem cell population in the tumor ^[26][108]. By hijacking iron metabolism in these cells, ferroptosis becomes a potential tumor suppressor agent to control a significantly powerful population of cells within cancer. Several strategies have been described to induce ferroptosis in this manner, such as manipulation of tumor-suppressor p53 ^[42][43][124,125], cysteine deprivation ^[44][126], and inhibition of ferritinophagy, a selective autophagic process capable of repressing accumulation of iron and lipid ROS ^[45][127]. In a similar manner, many studies use the collectively termed “ferroptosis inducers” in OSCC as a potential therapeutic approach, essentially by interfering with intracellular iron levels or ROS accumulation ^[46][50]. This has been evaluated both in vitro and in vivo by tampering with the signaling cascades that trigger ferroptosis or silencing of ferroptosis inhibitors, which can be achieved through direct genetic modifications or chemical compounds. Some of the compounds used to achieve such effects are erastin ^[47][48][49][21,42,57], carnosic acid ^[50][34], piperlongumine ^[51][58] and Disulfiram ^[52][63]. Some studies used strategies to induce ferroptosis in OSCC by silencing target genes. One study silenced the circular RNA FNDC3B (circFNDC3B) with interference RNA, which inhibited GPX4 and SLC7A11 expression (negative regulators of ferroptosis), inducing intracellular ROS and iron accumulation ^[53][29]. Another study silenced eukaryotic translation initiation factor 3 subunit B (EIF3B), commonly associated with unfavorable head and neck squamous cell carcinoma (HNSCC) prognosis ^[54][44]. EIF3B knockdown resulted in decreased invasion and migration in OSCC, as well as induction of cell death ^[54][44]. The knockdowns of enhancer of zeste 2 polycomb repressive complex 2 subunit (EZH2) and SLC7A11, both highly expressed in OSCC, resulted in ferroptosis induction ^[55][61]. Other target genes whose knockdown promoted ferroptosis were heat shock protein family A (Hsp70) member 5 (HSPA5) ^[56][54], adipocyte enhancer-binding protein 1 (AEBP1) ^[57][48], glutaredoxin 5 (GLRX5) ^[58][23], and miR-7-5p ^[59][27]. AEBP1 silencing was especially effective after sulfasalazine treatment, which significantly increased levels of ROS and free intracellular iron ^[57][48]. Another study targeting GPX4 with circular RNA showed increased levels of ROS and intracellular iron, meanwhile repressing tumor growth in OSCC cells ^[53][29]. Inhibition of GPX4 in Erlotinib-tolerant persisted cancer cells (erPCC) was also effective in increasing sensitivity to ferroptosis ^[60][30]. Many studies evaluated the potential of compounds or nanoparticles in controlling tumor progression by inducing ferroptosis. The recent study by Wu et al. ^[61][59] reported the effect of A-GSP (aqueous-soluble sporoderm-removed G. lucidum spore power) in tumor suppression by activating ferroptosis, which was confirmed by the assessment of GSH, malondialdehyde (MDA) and ROS levels, as well as ferroptosis-marker expression and mitochondrial morphological alterations—key in confirming the occurrence of this pathway ^[62][128]. Additionally, the authors evaluated this compound in vivo in a tumorigenesis assay, and their results showed that there was a decrease in tumor growth among the treated groups while maintaining low toxicity to the treated animals. These results strongly suggest the effectiveness and specificity of this compound in suppressing the tumor through ferroptosis induction. Another study achieved similar ferroptosis induction using nanoparticles termed ginseng-based carbon dots ^[63][56]. Another interesting study was performed by Huang et al. ^[64][19] using on-oxidized zero-valent iron (ZVI) nanoparticles. The authors treated several oral cancer cell lines with ZVI nanoparticles, and cell death was observed, together with ROS accumulation and mitochondrial damage. However, some cell lines managed to acquire resistance to treatment, and further examination revealed that ferroptosis-related genes were associated with this resistance. Glutathione reductase replenishes cellular GSH stock and circumvents ferroptotic cell death ^[65][129]. Therefore, by targeting these cells with ferroptosis-inducers, the authors were able to overcome treatment resistance without affecting the viability of other non-tumoral cells in vitro ^[64][19]. These results provide further evidence for the role of ferroptosis as a target while minimizing toxic side effects. Ferroptosis induction is especially powerful in treatment-resistant cell lines, as was shown in a study that overcame Cetuximab resistance in oral cancer by inducing ferroptosis ^[66][52]. It was noticed that the utilization of Cetuximab in resistant cells was not enough to suppress tumor growth or reduce viability. However, when Cetuximab was associated with RSL3, a ferroptosis-inducer agent, there was mitochondrial damage and increased cellular sensitivity to ferroptosis. RSL3 acts by depletion of GTX4, which is responsible for regulating a GTX4 protein depletor, which is, in turn, responsible for reducing the amount of intracellular lipid peroxide ^[67][130]. However, the study does not suggest through which mechanism this chemical synergy occurs, despite this effect being described in other studies ^[68][69][131,132]. Ferroptosis can also be induced by hyperbaric oxygen and X-ray radiation in a synergic effect ^[70][38]. It has been evaluated as a mechanism to overcome radiotherapy resistance by oral cancer cells, as has been done in other cancers ^[71][133]. Another interesting approach is the development of carrier particles to increase intracellular iron and trigger cell death ^[72][62]. OSCC-specific nanoprobes carrying iron can be internalized by endocytosis, much like “Trojan horses”. The acidic lysosomal conditions lead to the release of iron from the nanoprobes, inducing ferroptosis ^[72][62]. Non-thermal plasma treatment also showed an effect in oral squamous cell carcinoma cells by promoting lipid peroxidation and, ultimately, cell death by ferroptosis ^[73][20]. Moreover, studies show that regulation of ferroptosis inhibitor proteins is a promising therapeutic approach, as is the case of caveolin 1 (CAV1) downregulation ^[74][40]. CAV1 is known for its ferroptosis-inhibiting capabilities ^[74][40]. By knocking down CAV1, authors increased ROS and intracellular iron levels while reducing tumor cell growth in vitro. One of the main obstacles to inducing ferroptosis is assuring target specificity in order to avoid non-tumoral cell toxicity and side effects. In most studies, this has been achieved with a certain degree of success, such as the use of A-GSP, which was able to promote ferroptosis in oral cancer cells by inducing Fe²⁺ influx and GSH depletion, as well as lipid peroxide and ROS accumulation, while not producing significant toxic side effects ^[61][59].

In summary, several studies have shown promising results in using the existent ferroptosis-related cellular machinery as potential therapeutic strategies. Both knockdown strategies targeting key proteins involved in promoting/regulating this type of cell death and application of compounds with activity to induce ferroptosis of the OSCC cells were used. Additionally, the characterization of several gene signatures of ferroptosis-associated genes for both prognostication and response to treatment is relevant, but further verification in large-scale clinical studies is required. Thereafter, the ongoing exploration of ferroptosis in oral cancer not only deepens our understanding of cancer mechanisms but also holds the potential to translate this knowledge to improve patient outcomes.

3. Pyroptosis

Pyroptosis, an inflammatory type of caspase-mediated cell death, can modulate the immunogenic potential of specific cancers ^[75][82]. The role of pyroptosis in OSCC is an area of ongoing research, and the mechanisms and implications are not fully understood, but there is evidence to suggest that pyroptosis may play a role in both development and progression of OSCC ^{[75][76][77][78][79][80][81][82]}[76,77,78,79,80,81,82,83]. Pyroptosis derives its name from the combination of “pyro” and “ptosis”. “Pyro” signifies fire, highlighting its inflammatory properties, while “ptosis” refers to falling, which aligns with other forms of programmed cell death ^[83][141]. There are notable similarities between pyroptosis and apoptosis, including features like DNA damage and chromatin condensation ^[84][142]. Interestingly, pyroptotic cells exhibit swelling and numerous bubble-like protrusions on the cellular membrane before rupture, a phenomenon reminiscent of membrane blebbing observed in apoptosis. Pyroptosis was officially defined as Gasdermin-mediated programmed cell death in 2015 ^[83][141]. The Gasdermin superfamily in humans includes Gasdermin A/B/C/D (GSDMA/B/C/D), Gasdermin E (GSDME, also known as DFNA5), and DFNB59 (Pejvakin, PJVK). Inflammasomes are responsible for initiating pyroptosis via two distinct pathways (Figure 2). The canonical inflammasome pathway is reliant on the activation of caspase-1, and the noncanonical inflammasome pathway involves the activation of caspase-4, caspase-5, or caspase-11 ^[85][143]. Furthermore, certain studies have demonstrated that proapoptotic caspase-3 activation can also initiate pyroptosis by cleaving GSDME ^[86][87][144,145]. Inflammation is a critical component of tumor progression ^[88][146], and inflammation intensified by chemotherapy can lead to therapy failure and metastasis ^[89][147]. In the Nod-like receptor family, pyrin domain containing 3 (NLRP3) inflammasome is one of the critical components of the innate immune system and plays an important role in cancer ^[90][91][148,149]. Many factors can activate NLRP3 inflammasomes, including potassium efflux, intracellular calcium, endoplasmic reticulum (ER) stress and ROS ^[91][149]. Chronic inflammation has the potential to impact every phase of the carcinogenic process, increasing the risk of tumorigenesis with prolonged exposure to an inflammatory milieu ^[92][150]. Pyroptosis, as a form of lytic cell death, amplifies the release of mature interleukin-1 (IL-1) and interleukin-18 (IL-18), potentially influencing the development of cancer ^[93][151]. Furthermore, pyroptosis serves as the mechanism for inflammatory cell death in cancer cells, thereby restraining the proliferation and migration of these cancer cells ^{[78][79][94][95][96][97][98]}[28,33,68,69,70,78,79]. Consequently, pyroptosis assumes a dual role, both promoting and inhibiting tumorigenesis ^[99][152]. Previous findings revealed that 5-fluorouracil (5-FU) treatment increased NLRP3 expression in OSCC, which mediated drug resistance. It was also proven that NLRP3 could promote tumor growth and metastasis in OSCC ^{[97][100][101]}[64,69,153]. Activation of pyroptosis has also been directly associated with increased chemoresistance to cisplatin and 5-FU treatment ^[100][64], and inhibition of pyroptosis has been associated with increased sensitivity of neoplastic cells to cisplatin treatment ^[102][66]. Methods of pyroptosis inhibition are gaining the attention of the scientific community ^{[97][103][104][105]}[65,69,74,154]. Yang et al. ^[97][69] highlighted that extracellular vesicles derived from bitter melon led to a significant decrease in the expression of NLRP3, reducing OSCC resistance to 5-FU treatment. The study developed by Yue et al. ^[103][65] evaluated the effect of anthocyanin on OSCC. Anthocyanin reduced the viability of OSCC cells and inhibited migration and invasion capacity, concomitantly increasing pyroptosis. Simultaneously, activation of pyroptosis was associated with increased expression of NLRP3, caspase-1 and interleukin-1β (IL-1β). After administration of caspase-1 inhibitors, anthocyanin-activated pyroptosis was suppressed, and cell viability, migration and invasion rates increased concomitantly. In vitro studies in monoculture do not show the real dimension of the role of pyroptosis in OSCC. Therefore, conflicting results can be seen depending on the methodology used in different studies. The poor prognosis of pyroptosis is associated with its ability to activate inflammation; however, the development of in vitro and in vivo studies capable of more broadly evaluating the tumor microenvironment and all mechanisms triggered by the activation of pyroptosis should be encouraged. The inflammation associated with pyroptosis can lead to the recruitment of immune cells and other factors that support tumor growth and metastasis ^[105][154]. The pro-inflammatory environment can also contribute to resistance to therapy and promote angiogenesis, which is the formation of new blood vessels that supply nutrients to the tumor. The study developed by Xin et al. ^[104][74] used The Cancer Genome Atlas (TCGA) dataset to investigate the predictive value of pyroptosis-related lncRNAs in the prognosis of OSCC. The authors identified eight pyroptosis-related lncRNAs associated with overall survival in patients with OSCC by multivariate regression analysis. Taken together, the analysis indicates that the number of studies exploring pyroptosis in OSCC is still restricted, limiting our knowledge of the mechanisms of how molecules related to this type of cell death affect OSCC cells. However, the studies highlighted the potential of pyroptosis in the control of OSCC development and progression and described drugs and molecules related to both blockage and induction of it. Moreover, two studies demonstrated that pyroptosis gene clusters are correlated with clinical characteristics, infiltration of immune cells, susceptibility to chemotherapy and immunotherapy, and prognosis of patients with OSCC ^[104][106][74,75]. It will be important to validate these risk models and to explore potential drugs that could induce cell death through pyroptosis and concomitantly inhibit the pro-tumor inflammatory response in OSCCs.

4. Necroptosis

Necroptosis is considered a programmed form of necrosis mediated by receptor-interacting protein kinase 1 (RIPK1) and receptor-interacting protein kinase 3 (RIPK3) ^[107][108][155,156]. The stimulus is initiated by tumor necrosis factor (TNF) binding to its receptors ^[109][110][157,158]. The interaction of ligand and receptor leads to the formation of a signaling complex, which may include adaptor proteins such as Fas-associated protein with death domain (FADD) and TNF receptor-associated death domain (TRADD). RIPK1 is recruited to the signaling complex and is activated by phosphorylation. Depending on cellular conditions, RIPK1 can associate with caspases, promoting apoptosis, or with RIPK3, initiating the necroptosis pathway. If the necroptosis pathway is activated, RIPK1 recruits RIPK3 and the protein MLKL, forming the necrosome complex. RIPK3 phosphorylates MLKL, activating it. Activated MLKL oligomerizes and translocates to the plasma membrane, where it causes damage, leading to membrane rupture and necroptosis (Figure 3) ^[111][159]. The formation of an MLKL oligomer opens a pore in the membrane, allowing the entrance of ROS and DAMPs ^[112][160]. In this manner, death by necroptosis, despite being activated by specific signals, leads to a similar end as necrosis and ferroptosis, triggering the entrance of ROS in the cell and leading to membrane damage ^[112][113][160,161]. To allow for necroptosis to occur, the apoptotic pathway must be impaired or damaged ^{[107][114][115]}[155,162,163], which can be quite common in cancer since the inhibition of healthy apoptosis as a control allows for the unchecked reproduction of damaged cells ^[116][164]. This phenomenon holds significant implications for oral cancer, where dysregulation of cell death mechanisms can tip the balance in favor of tumor progression. In this context, necroptosis, which is characterized by a regulated inflammatory response, becomes a last-resort mechanism to eliminate aberrant cells ^[117][118][165,166]. However, cancer cells can hijack this mechanism to promote their survival and evade the body’s natural defenses, using pro-tumoral inflammation to their advantage ^[117][165]. Additionally, necroptosis can generate an immunosuppressive tumor microenvironment, which may further contribute to cancer cell survival and progression ^[119][167]. It comes as no surprise that the expression level of key mediators of necroptosis is elevated in cancer ^[117][120][165,168], indicating that necroptosis may play a role in promoting oncogenesis and cancer metastasis ^[121][169]. However, it is possible to interfere in this process by targeting necroptosis as an ally in halting cancer progression and survival, as has been evaluated in oral cancer studies, by targeting focal adhesion molecules ^[122][123][170,171] or even caspase-8 itself, which is responsible for deciding the pathway outcome of apoptosis versus necrosis ^[122][124][170,172]. In the context of OSCC, studies have explored the induction of necroptosis using different agents, such as Obatoclax ^[125][84]. This agent targets members of the BCL-2 family, specifically the myeloid cell leukemia sequence 1 (MCL-1). The study suggests that Obatoclax induces cell death in OSCC cells through autophagy-dependent necroptosis ^[126][173], with mitochondrial stress and dysfunction as detectable upstream events. Additionally, capsaicin was found to inhibit cell proliferation and induce endoplasmic reticulum stress and autophagy in oral cancer ^[127][87]. This mechanism negatively regulates ribophorin II, impairing P-glycoprotein functions and sensitizing cells to anticancer therapy ^[128][174]. The association of capsaicin with anticancer agents promotes necroptosis rather than apoptosis, showcasing a unique pathway for inhibiting OSCC cell viability. Similarly, chelerythrine chloride (CS) demonstrated necroptosis induction in OSCC, impairing cell proliferation and inducing morphological alterations in a dose-dependent manner, such as membrane rupture, and dose-dependent cell death ^[127][87]. Moreover, the development of targeted delivery systems such as PLGA-Dtx (poly-lactic-co-glycolic acid nanoparticles containing docetaxel) has shown enhanced efficacy in inhibiting cancer cell proliferation. This strategy induced both apoptosis and necroptosis in oral cancer cells ^[129][89]. These results altogether suggest a potential role for these compounds in triggering necroptosis in OSCC cells. Studies on necroptosis-related genes may reveal potential targets, either enhancers or inhibitors of this NAPCD. HNSCC studies have associated CASP8 mutations with radioresistance and poor survival outcomes ^[130][86], as is the case for OSCC ^[131][175]. In this manner, knockdown of CASP8 enhances the radiosensitizing effects of certain compounds through the induction of necroptosis. Additionally, as seen in other cell death mechanisms, the investigation of necroptosis-related genes that may be relevant in prognosis prediction has also been explored. Bioinformatic analyses and in vitro experiments identified six genes (hypoxanthine phosphoribosyltransferase 1-HPRT1, PGAM family member 5, mitochondrial serine/threonine protein phosphatase-PGAM5, BH3 interacting domain death agonist-BID, survival of motor neuron 1, telomeric-SMN1, FADD, and KIAA1191) contributing to OSCC development, metastasis, and immune modulation. These genes may play a role in the regulation of cell death pathways, including necroptosis ^[132][90]. Moreover, a study in HNSCC emphasized the prevalence of necroptosis and its association with poor overall survival and progression-free survival. Approximately half of the necrosis in HNSCC was attributed to necroptosis, indicating its significance as an independent risk factor for adverse clinical outcomes ^[133][85]. In summary, necroptosis induction is a promising alternative in the treatment of oral cancers, and the expression of necroptosis-related genes depicts prognostic potential for predicting OSCC outcomes. However, one of the main challenges in inducing necroptosis as a cancer treatment strategy remains in establishing specificity in a manner by which little toxicity is archived ^[124][172].