Targeting the NKG2D axis in head and neck squamous cell carcinoma: from molecular insights to clinical applications Authors: Parsa Alijanizadeh, Mahshid Shahmoradi, Amirhossein Kamroo, Tara Shahmoradi, Fatemeh Tajafrooz, Pegah khodaee, Niloufar Yazdanpanah, Kiarash Saleki, Nima Rezaei Published: 2026 ABSTRACT AbstractHead and neck squamous cell carcinoma (HNSCC) is an aggressive cancer that evades immune surveillance by impairing the NKG2D receptor on natural killer (NK) cells and cytotoxic T lymphocytes. NKG2D recognizes stress-induced ligands (e.g., MICA, MICB, ULBPs) on tumor cells, but HNSCC tumors counteract this by shedding soluble ligands and secreting immunosuppressive cytokines like TGF-β1, reducing NKG2D expression and function. This review explores therapeutic strategies targeting the NKG2D axis to enhance HNSCC treatment. Cytokine therapies, such as IL-2 or IL-15, boost NKG2D-dependent NK cell activation. Monoclonal antibodies targeting MICA/B or ULBPs prevent ligand shedding and induce antibody-dependent cellular cytotoxicity (ADCC). Engineered cell therapies, including NKG2D-based chimeric antigen receptor (CAR)-T and CAR-NK cells, show potent preclinical efficacy against HNSCC. Combining these approaches with EGFR-targeted antibodies (e.g., cetuximab) or checkpoint inhibitors enhances NKG2D-mediated immunity. Despite promising results, challenges persist, including heterogeneous ligand expression, persistent shedding, and the immunosuppressive tumor microenvironment, which limit clinical efficacy. Future directions include multimodal strategies like bispecific NK-cell engagers and integration with chemoradiotherapy to overcome these barriers. By restoring NKG2D-driven immunity, these therapies hold potential to improve outcomes for HNSCC patients, though further research is needed to optimize their clinical translation.Keywords: NKG2D, Head and neck squamous cell carcinoma (HNSCC), Natural killer cells, Immune evasion, Immunotherapy Introduction Head and neck squamous cell carcinomas (HNSCCs), originating from the mucosal epithelial surface of the digestive and upper respiratory systems, are among the top 10 most prevalent types of cancer [1], causing 90% of head and neck malignancies [2]. Prognosis for HNSCC patients remains disappointingly poor; only about half of the patients have 5-year survival rates due to the highly invasive nature of HNSCCs, and even early-stage tumors tend to have lymphogenic metastasis [3]. The immune cells, especially cytotoxic T lymphocytes (CTLs) [4], helper T lymphocytes [4], antigen-presenting cells (APCs) [5], and natural killer (NK) cells [6] are essentially involved in anti-cancer immunity and exert anti-tumor effects. Meanwhile, HNSCCs are able to bypass the protective mechanisms by numerous mechanisms to disrupt the defensive protocol and transform the immunity toward a pro-tumorigenic status [7]. HNSCCs embed a heavily immunosuppressive tumor micro-environment (TME) that hampers the anti-tumor immunity either by secreting immunoinhibitory cytokines [8, 9] or by recruiting immunoinhibitory cells such as T regulatory lymphocytes (Tregs) and myeloid-derived suppressor cells (MDSCs) [10]. Besides, HNSCCs are capable of expressing checkpoint inhibitors such as programmed death-ligand 1 (PD-L1) [11] while cytotoxic T-lymphocyte-associated protein 4 (CTLA-4) contributes to immune suppression [12] to inhibit T lymphocyte mechanisms, along with the capability of modulating the cell-surface antigens, including reducing the levels of MHC-II [13], or elevating CD47 levels [14]. To address the existing problems with immunoinhibition by HNSCCs, immunotherapeutic approaches can be utilized, such as cytokine therapies and checkpoint inhibitors [15, 16]. Checkpoints inhibitors’ function is based on the targeted neutralization of immunosuppressive axes such as PD-1/PD-L1 and CTLA-4/CD80 (B7-1) or CTLA-4/CD86 (B7-2) [15, 16]. Cytokine therapies are alternative methods that use the benefit of certain immunostimulating cytokines to reinvigorate the anti-cancer immunity [15, 16]. Natural killer group 2D (NKG2D) receptors, involved in both innate and adaptive immune systems, are as of interest for immunotherapy in recent years [17]. NKG2D has the role of recognition of ligands, including MICA/B, and ULBP1/2/3/4, which are expressed in stressful situations, such as infections and malignant changes [18, 19]. NKG2D operates as a vital receptor in numerous immune cells, namely CD8+ T cells, γδ T lymphocytes, + and specifically NK cells [20]. NKG2D receptors of NK cells are activated when discerning the corresponding ligands, making the way for the induction of cytotoxic effects and the discharge of positive immunomodulator cytokines [19]. Hence, NKG2D-based immunotherapies aim at the recovery of suppressed NK cells’ immune activities to reestablish the tumoricidal effects [17]. The basis of NKG2D-based immunotherapy is either the elimination of shed NKG2D ligands (NKG2DLs) of plasma, contributing to poorer outcomes of HNSCCs [17], or the application of drugs for the inhibition of the HNSCCs proteasome system to increase tumor-surface levels of NKG2D for further recognition by NK cells [21]. In addition, combinatory therapies of NKG2D ligand immunotherapy with adjuvant application of chemoradiation therapy (CRT) or tumor-targeted antibodies are being explored for anti-HNSCC treatment [22, 23]. In this review article, the current biological function of NKG2D, the role of NKG2D in anti-cancer immunity and its suppression, and current clinical and subclinical perspectives of NKG2DL for HNSCC immunotherapy are reviewed. Biology and molecular structure of NKG2D NKG2D is one of the key players activating receptors against tumors and infections, which can regulate adaptive and innate immune responses [24]. NKG2D is a homo-dimeric receptor made of two type II transmembrane proteins disulfide-linked to each other. Its transmembrane domains contain positively charged amino acid residues, and its intracellular tails are very short and lacks intrinsic signaling motifs. This protein is categorized as a C-type lectin-like receptor, and unlike other members of the NKG2 family, it appears to have different exclusivity [25]. The KLRK1 gene, which is located on chromosome 12p13.2, can encode NKG2D at varying levels in healthy human cells; however, this encoding can lead to constitutive expression of NKG2D on all NK cells, γδ T cells, CD8+ T cells, and cytotoxic T cells in the case of tumor risk or infection [20, 26, 27] This receptor’s importance lies in identifying a self-induced stress ligand group that can increase the chance of cancer or infection in cells [26, 28, 29]. The expression of ligands associated with NKG2D increases immune response to stressed cells, thereby triggering NK cells to use cytotoxicity, cytokine release, and killing of tumor cells. However, various mechanisms inhibit the action of the NKG2D-NKG2DL to enable immune escape of tumor cells [30]. NKG2D can also play a role as a co-receptor, which depends on the cell type and context [31]. NKG2D does not contain intracellular signaling motifs; instead, it forms a complex with adaptor proteins to initiate signaling in a stressed cell. NKG2D can recognize various types of ligands with different types of bonding structures. The function of NKG2D is usually mediated through interactions with ligands, and it triggers intracellular signaling cascades [28, 32]. These ligands cannot imply signals equally; upon engagement with these various ligands, NKG2D can send different responding signals. These ligands are mainly categorized into two main human groups: MHC class I-related molecules (MICA- MICB) and ULBP ligands (ULBP1–6) [27]. NKG2D activity begins with the cooperation of an adaptor protein molecule. This adaptor protein is usually a DNA X-activating protein of 10 kDa (DAP10) in humans. This protein associates with NKG2D, creating the ability to accept different types of molecules and create different ligands for NKG2D by shaping a receptor complex. Building a complex formation is essential for NKG2D to start functioning, and the structure of NKG2D is shown in Fig. 1 [28]. NKG2D binding triggers the PI3K pathway, leading to the release of perforin, granzymes, and cytokines that kill target cells and boost immune responses [30, 31]. Thus, NKG2D provides immune surveillance, and cells under genotoxic/oncogenic stress express these ligands [33]. Human NKG2D ligands, their structural features, expression patterns, and stress inducers are summarized in Table 1. Structure of NKG2D and ligands Table 1Human NKG2D ligands, structural features, expression patterns, and stress inducersLigandStructural featuresExpressionStress inducersReferencesMICAMHC I–like (α1–α3)TM & cytoplasmic tailGPI form (MICA*008)Shedding (ADAM/MMP), exosomal releaseLow in normalHigh in tumorsHeat shock (HSF1/HSE)DNA damageViral (HIV, HCV)[34]MICBMHC I–like (α1–α3)TM & cytoplasmic tailADAM/MMP sheddingLow in normalTumor/stress-inducedHeat shockDNA damageViral infection[35]ULBP1α1–α2 onlyGPI-anchoredResistant to proteasesPI-PLC cleavableAbsent/low in normalGlioma, prostate, lymphomaDNA damageViral (HCMV, HIV, HCV)[36]ULBP2α1–α2 onlyGPI & TM isoformsTM cleaved by ADAM/MMPGPI PI-PLC cleavableAbsent/low in normalGlioma, ovarian, melanoma, neuroblastoma, HNSCCDNA damageViral (HCMV, HIV)[37]ULBP3α1–α2 onlyGPI-anchoredExosomal releaseLow in normalGlioma, prostate, lymphoma, HNSCCDNA damageViral (HCMV)[38]ULBP4α1–α2 onlyTM-anchoredAbsent/low in normalProstate, lymphomaStress-related[39]ULBP5 (RAET1G)α1–α2 onlyTM & GPI formsTM cleaved by proteases, GPI PI-PLC cleavableAbsent/low in normalSome tumors (prostate)Not defined[39]ULBP6 (RAET1E)α1–α2 onlyGPI-anchoredProtease resistant, PI-PLC cleavableLow in normalTumors (prostate)Not defined[39]Overview of Human NKG2D Ligands, Highlighting Structural Features, Expression Patterns, and Stress Inducers in Normal and Tumor Contexts Human NKG2D ligands, structural features, expression patterns, and stress inducers MHC I–like (α1–α3) TM & cytoplasmic tail GPI form (MICA*008) Shedding (ADAM/MMP), exosomal release Low in normal High in tumors Heat shock (HSF1/HSE) DNA damage Viral (HIV, HCV) MHC I–like (α1–α3) TM & cytoplasmic tail ADAM/MMP shedding Low in normal Tumor/stress-induced Heat shock DNA damage Viral infection α1–α2 only GPI-anchored Resistant to proteases PI-PLC cleavable Absent/low in normal Glioma, prostate, lymphoma DNA damage Viral (HCMV, HIV, HCV) α1–α2 only GPI & TM isoforms TM cleaved by ADAM/MMP GPI PI-PLC cleavable Absent/low in normal Glioma, ovarian, melanoma, neuroblastoma, HNSCC DNA damage Viral (HCMV, HIV) α1–α2 only GPI-anchored Exosomal release Low in normal Glioma, prostate, lymphoma, HNSCC DNA damage Viral (HCMV) α1–α2 only TM-anchored Absent/low in normal Prostate, lymphoma α1–α2 only TM & GPI forms TM cleaved by proteases, GPI PI-PLC cleavable Absent/low in normal Some tumors (prostate) α1–α2 only GPI-anchored Protease resistant, PI-PLC cleavable Low in normal Tumors (prostate) Overview of Human NKG2D Ligands, Highlighting Structural Features, Expression Patterns, and Stress Inducers in Normal and Tumor Contexts NKG2D-mediated activation and anti-tumor cytotoxicity NKG2D-triggered NK cell activation After the binding of NKG2D to its ligands, an array of outcomes happens in the activated immune cells, such as the degranulation of perforin/granzyme and secretion of IFN-γ, TNF-α, and other pro-inflammatory cytokines [40]. Despite the existing trend in the balance of stimulatory/inhibitory signals, stimulating NKG2D signals can shift the balance toward activation without facing a significant weakening or blockage in the initial signal. For illustration, the expression of normal MHC I over the external surface of the cancer cells is not able to fully block the NKG2D-mediated elimination; the inhibitory signal of the normal MHC I is not sufficient to prevent the NKG2D-mediated elimination [41]. Similar for the inhibitory signal from Lymphocyte antigen 49 (Ly49) in mice, stimulatory NKG2D signal can dominate the balance over inhibitory signaling [33, 42]. This overriding phenomenon becomes vitally important in the pathologically transformed cells, such as infected or cancerous cells. Cancer/infected cells upregulate the ligands of NKG2D, such as MICA/B and ULBPs, in response to cellular stress; thereby, the immune cells shall be motivated into imposing the anti-tumoral and anti-infection mechanisms, without facing a noise signal from other inhibitory signals [43]. Co-stimulation of CD8⁺ T cells On cytotoxic T lymphocytes (CTLs, CD8⁺ T cells), NKG2D functions as a costimulatory receptor rather than a standalone stimulator. Standalone activation of NKG2D is not able to motivate a CD8⁺ T cell into cellular elimination; however, the activation of T cell receptors (TCRs) costimulated with the NKG2D considerably strengthens the cytotoxic effects of the CTLs [44]. In other words, the exposure of human CD8⁺ T cells with NKG2D ligands (NKG2DLs) is able to boost the immunologic effects via raising the production of IFN-γ andTNF-α, and stronger cytotoxicity [45]. NKG2D-mediated signaling axes NKG2D possesses a short cytoplasmic tail and heavily depends on intracellular adaptors to transduce the extracellular recognition into intracellular signals. In human NK cells and resting mouse NK cells, NKG2D signal transduction is mediated by DAP10 homodimers; in mice, the splice variant (NKG2D-S) associates to DAP12 for further signal transduction [46]. When NKG2D binds NKG2DL, the receptor starts clustering and establishing contact with adaptor proteins. In human NK cells and resting mouse NK cells, DAP10’s cytoplasmic tail becomes phosphorylated at the YNIM motif to recruit the phosphoinositide 3-kinase (PI3K) (p85 subunit) and Grb2–Vav1 complexes [47]. PI3K and Grb2-Vav1 call for other downstream signaling, such as the Rho-family GTPases, and promotes intracellular Ca²⁺ mobilization [48]. Finally, these downstream signaling, modify the intracellular actin filaments to make room for the degranulation of lytic enzymes toward the targeted cell. Within other mouse NK cells, that are dependent on NKG2D-S and DAP12, the NKG2D-s isoform recruits an ITAM-motif-containing DAP12. The ITAM motifs further call for Syk and ZAP70 kinases, that results in the activation of the NK cell [49]. Both of DAP10- and DAP12-based signal converge at the immune synapse of the entry of calcium and targeted degranulation of granzyme and perforin. Besides, both of the signaling pathways result in elevated production of cytokine [50, 51]. The NKG2D signaling pathways are shown in Fig. 2. Dual role of the NKG2D–ligand axis in head and neck squamous cell carcinoma (HNSCC) Preclinical evidence Extensive in vitro and in vivo studies have established NKG2D as an important component of tumor surveillance mechanism.NKG2D-deficient mice show increased tumor growth compared with wild-type mice, indicating impaired tumor surveillance [52]. Besides, impaired NKG2D-mediated immune surveillance mice have severely impaired tumoricidal measures: NKG2D-deficient mice were several-fold more sensitive to spontaneous prostate cancer in the TRAMP model than wild-type [53]. In vitro experiments show that NKG2D blockade substantially reduces NK cell lysis of ligand tumors; in contrast, artificial cross-linking of NKG2D enhances cytotoxicity and cytokine release of the NK cells. To further study the role of NKG2D axis, chimeric antigen receptor (CAR) immune cells and bispecific antibodies have been developed in the animal models. CAR immunotherapy refers to the targeted engineering of the immune cells to reprogram them into active targeting of the antigen of interest. This manipulation enables the immune cells to boost their surveillance capabilities along with an enhance tumoricidal effects [54]. In the context of NKG2D, the NKG2D-based CARs refer to the engineered T or NK cells embracing an NKG2D receptor linked to an intracellular signaling domain, that are reprogrammed to target NKG2DLs and tumor cells expressing the NKG2DLs [55–57]. Furthermore, engineering approaches taking advantage of bispecific antibodies cross-linking NKG2D on effectors to a tumor antigen display increased tumor killing in animal models. For instance, a bispecific antibody cross-linking NKG2D on NK/CTLs to “Human Epidermal Growth Factor Receptor 2” (HER2) on tumor cells induced potent NK-cell-mediated lysis in vitro [40, 58, 59]. These findings point to the potential for therapeutic targeting of NKG2D by inducing its ligands on tumors to augment anti-cancer immunity. NKG2D Downregulation and tumor immune evasion Shedding of NKG2D ligands Tumors can evade NKG2D-mediated recognition and elimination by actively shedding NKG2D ligands. Cancer cells express stress‐inducible NKG2DLs (MICA/B, ULBPs), and the expressed NKG2DLs are capable of initiating the NK-mediated recognition and elimination. However, the tumoral proteases, such as metalloproteases (ADAM10, ADAM17, MMPs) and disulfide isomerases (ERp5), contribute to the release of NKG2DLs from tumor cells [60]. The ligand proteolysis phenomenon, as well as the release of NKG2DL-containing exosomes, depletes the tumor external surface from NKG2D targets. Consequently, shed ligands allow for a meticulous immune evasion [61, 62]. The process is highly regulated: endogenous inhibitors such as tissue inhibitor of metalloproteinases 3 (TIMP3) suppress the activity of cancer protease ADAM17, the results in the reduced release of soluble NKG2D ligands (sNKG2DLs), such as the soluble MICA (sMICA), soluble MICB (sMICB), and soluble ULBP2 (sULBP2) [63]. Clinically, high levels of sNKG2DLs in patient sera correlate with advanced disease stage and poor prognosis [64, 65]. Importantly, the sNKG2DLs, can contribute to the immune exhaustion. Immune exhaustion refers to the functional failure of the immune system, caused by the prolonged exposure to antigen, and is importantly seen within the context of chronic infections and/or cancers [66]. Importantly, immune exhaustion opens the avenue for accelerated tumor development and resistance to treatment [67]. From the point of view of sNKG2DLs, the prolonged exposure of sNKG2DLs to the immune cells, especially the NK cells, shifts them toward a dysfunctional phenotype, and severely impairs their tumoricidal capabilities [68]. Altogether, the presence of sNKG2DLs can impose a paradoxical effect via disarming the anti-tumor arm of the immune system, shifting them toward an exhausted phenotype, and distract the immune cells by establishing a ligand-receptor connection in the wrong place far from the tumor [69]. Chronic engagement or soluble ligands impairing NKG2D function The soluble ligands have further deleterious effects when being chronically engaged with NKG2D: desensitization of NK and CD8⁺ T cells. Soluble MICA/B or ULBPs extensively bind NKG2D on lymphocytes and trigger receptor internalization and downregulation [35]. Zingoni et al. showed that tumor-derived soluble ligands reduce cell-surface expression of NKG2D; hence, their NKG2D-mediated cytotoxic responses become impaired. Functionally, this means reduced cellular elimination and weakened cytokine production. In addition, Chronically NKG2D‐engaged NK cells may exhibit impaired CD3ζ signaling and show reduced other activating receptors, and CD8⁺ T cells with persistent NKG2D stimulation become hypo-responsive and secrete less IFN‐γ [70]. For example, in tumor models sMICB was shown to deplete peripheral NK cells from NKG2D, that ultimately resulted in an accelerated cancer progression; whereas, tumor‐restricted NKG2D ligands preserved NKG2D‐dependent immunity [71]. Similarly, human tumor-infiltrating lymphocytes exposed to abundant NKG2DLs exhibit downregulated NKG2D expression and impaired cytotoxicity [70]. In conclusion, tumors evade NKG2D‐driven immunity by shedding their ligands: This not only removes stress ligands from the tumor cell surface, but also over-saturates the receptor of immune cells with useless, distracting soluble ligands that persistently engage and downmodulate NKG2D on NK and CD8⁺ T cells. The final result would be an impaired anti‐tumor response of NK and T effectors to reduced NKG2D expression and cytotoxic functions [62]. Exosomes and their role in ligand presentation Tumor-derived exosomes first trigger a transient boost in NK activation but then functional impairment in some cancers, like HNSCC [72, 73]. A similar suppressive effect has also been observed in a cancer study. This research investigated the expression level of NK cell-activating receptors, including NKG2D, NKp30, and NKp46. In the initial results, the aforementioned NK cell receptors were decreased, on the other hand ,the expressions of activating receptors NKG2D, NKp30, and NKp44 decreased, and the cytotoxic activity of NK cells was consistent with the changes in NK cell-activating receptors [74]. Exosomes have also shown immunosuppressive effects on NK cells. The plasma of patients who have cancer divided in two groups: melanocyte-derived exosomes (MTEX) and normal cell-derived exosomes (non-MTEX) through immunocapture. On primary NK cells both total exosomes and MTEX can reduce the expression of NKG2D. In sharma et al. study, it was revealed that by adding anti-TGF-β monoclonal antibody, the function of NK cells was restored, suggesting that TGF-β might be present in tumor-derived exosomes, and the expression level of NKG2D receptor is inhibited by exosomes [75]. Beyond suppressing NKG2D through cytokines, exosomes can also damage NK function by presenting ligands such as MICA/MICB and ULBPs. Exosomes can carry MICA/MICB, and are contribute to cancer progression and immune evasion. Tumors have developed mechanisms by shedding these ligands through exosomes to evade immune surveillance, which can disrupt function of immune cell. NKG2D receptors are downmodulated from the NK cell surface when tumor-derived exosomes and NK cells were co-cultured, and this causes immunosuppression of NK cells by reducing their ability to recognize and eliminate cancer cells [62, 76]. When NK cells are exposed to exosomes carrying the MICA*008 protein (found in HeLa cells), their ability to kill cancer cells is reduced. Thus, exosome release may be an alternative pathway to metalloproteinase-mediated shedding of MICA/MICB molecules, contributing to immune evasion. MICA*008 is released as a full-length, membrane-anchored molecule. This ligand was released in exosomes instead of proteolysis [62]. Tumor cells enhance their ability to eliminate NKG2D ligands by inclusion into exosomes. These exosomes may deliver a mixture of signals, like multiple NKG2D ligands, through general or specific adhesion molecules [77]. These findings suggest exosomes help tumors escape immune attack by suppressing NK cell activity through NKG2D-related pathways. HNSCC-mediated disruption of NKG2D in the TME As implied earlier, NK cells play an important role in anti-tumoral immunity. Adaptive NK cells are involved in increasing IFN-γ secretion and antibody-dependent cellular cytotoxicity through NKG2D [78]. HNSCCs can also disrupt this signaling axis in the TME: Tumor-originated cytokines modulate NKG2D expression. The most important NKG2D-disrupting element in the context of the TME is TGF-β. TGF-β a well-known immunosuppressive factor, and is upregulated in HNSCC-infected tissues compared with normal tissues, that ultimately helps the immune escape procedure through NKG2D downregulating [79]. TGF-β downregulates NKG2D on NK cells and CD8+ cells. Accompanied by the sNKG2DLs disabling effect, TGF-β and sNKG2DLs synergically suppress the NKG2D pathway in HNSCC patients [80, 81]. For clarification, TGF-β also enhances the inhibitory effect of sNKG2DLs on NKG2D expression [79], and sNKG2DLs-mediated impairment of the immune system was enhanced by TGF-β levels [82]. In parallel with the TGF-β, TNF-α and IL-1β can also impair the NKG2D signaling pathway. Although being pro-inflammatory cytokines, these cytokines downregulate the expression of NKG2D when released from the tumor-associated macrophages. The downregulation of NKG2D weakens the surveillance capabilities of the NK cells and compromise their tumoricidal mechanisms [83]. Other cytokines, such as adenosine or IL-2 depletion, have not been directly studied for their anti-NKG2D measures within the TME; however, recent reaserch hypothesize that the inhibitory effects of adenosine over the immune system [84], and the NK cells’ infiltration/activation may be affected by the IL-2 depletion [85], could be possibly involved in disrupting the NKG2D axis. Further study of the TME-involved cytokines could illuminate their inhibitory effects over the NKG2D axis. Polymorphisms studies of NKG2D-related elements in HNSCC Studies indicate that genetic polymorphism in NKG2D ligands are associated with HNSCC risk. Particular polymorphism of MICA/B genes are associated with NKG2D soluble ligands and reduce NK cell-dependent response in the oral squamous cell carcinoma microenvironment [86]. Ryouji Tani et al. showed that the MICA allele, A5.1, was significantly more frequent in OSCC patients. This was correlated with the age of patients in a way that younger patients were more at risk of OSCC. Having A5.1/A5.1, people are more prone to early OSCC incidence than other genotypes, but their prognosis is better [87]. Ben Chaaben et al. also showed that MICA-129 Val/Val polymorphism is associated with a higher risk of nasopharyngeal cancer. Val allele in position 129 of the MICA protein decreases its binding strength to NKG2D. In the meantime, it eases cell surface translocation and the shedding process, which increases sMICA plasma levels are associated with anti-tumoral immunity. Additionally, sMICA level rises with tumor stage [88]. sMICA in HNSCC patients before and two weeks after chemo-radiotherapy was significantly higher than that of healthy individuals. Patients whose post-treatment sMICA remained elevated show earlier disease progression than those with reduced levels. Meanwhile, the overall survival rate was insignificant [86, 89]. MICB expression is usually induced in cells under stress or tumoral conditions, which results in innate and adaptive immunity. Its upregulation might reduce oral cancer risk due to an anti-tumoral immune response via the MICB-NKG2D axis [90]. ULBP-1 is an NKG2D activating ligand. It contributes to decreased sensitivity to NK cell tumoricidal effects due to its downregulation in esophageal squamous cell carcinoma (ESCC) stem-like cells at protein and mRNA levels. Overexpression of ULBP-1 also induced more NK cell cytotoxicity [91]. Plasma levels of ULBP1/2/3 were also raised in HNSCCs derived from initial tumor shedding after proteolytic cleavage [92]. Targeting the NKG2D axis and therapeutic implications in HNSCC Targeting approaches for the NKG2D axis The NKG2D–NKG2DL axis has widely gained attention for its possible application in HNSCC immunotherapy. As mentioned earlier, NKG2D is an (co-)activating receptor on T/NK cells, thus various therapies are being studied in hope for the restoration of this vital axis in HNSCC [93]. Different therapeutic strategies focus on resurrecting the NKG2D tumoricidal measures, and the blockage of immune evasion from the NKG2D axis [94, 95]. Adoptive cell therapies are also in progress. NKG2D-based CARs are redirected NK/T lymphocytes to eliminate NKG2DL-positive tumors, which have demonstrated potent tumoricidal measures in preclinical models (e.g. CYAD-01 CAR-T). However, the previous challenges associated with CAR immunotherapy, such as hampered solid tumor infiltration and immune evasion, remain with NKG2D-based CARs [96]. Monoclonal antibodies against MICA/MICB offer another strategy. Antibodies targeting the α3 shedding domain can prevent ligand release and mediate ADCC. Likewise, selective ADAM/MMP inhibitors can reduce MICA/MICB cleavage, preserving surface ligands and sensitizing tumors to NK attack [97, 98]. The third option is cytokine modulation; cytokine modulation augments the NKG2D axis. For instance, IL-15 modulation is a successful cytokine modulation, which can reinvigorate the NK cells via Jak3/DAP10 pathways and boost NKG2D-dependent elimination. Altogether, IL-15 super-agonists are being developed for safer NK activation [98, 99]. These approaches may be combined with standard therapies. Overall, this multi-modal strategy illustrates promising avenues to restore NKG2D-mediated immunity in HNSCC. Clinical studies and applications Experimental approaches include adoptive NK/NKT cell transfers, cytokine or small-molecule immunomodulators, and engineered cell therapies that directly exploit the NKG2D–ligand interaction. Early-stage clinical trials in HNSCC patients demonstrate that such strategies can increase NKG2D expression on immune cells, increase tumor infiltration by NK cells, and enhance cytotoxic markers like perforin [100]. HNSCC is frequently defined by immune evasion mechanisms that impair NK cell function, particularly by interfering with the NKG2D pathway [64]. In a recent study by Klöß et al. in 55 patients with HNSCC, there were considerable immunosuppressive changes, including impaired NK cell cytotoxicity and immune cell subset variations compared with controls. Elevated plasma concentration of sMICA and transforming growth factor β1 (TGF-β1), both interfering with the function of NKG2D. Both were associated with disease relapse and progression. These soluble ligands downregulated the NKG2D expression on NK cells as well as their cytotoxic activity against tumor cells. Functional assays determined that patient plasma containing high levels of sMICA and TGF-β1 suppressed NKG2D-dependent NK cell activity and increased caspase-3 activation with apoptosis in NK cells. Interestingly, in vitro neutralization of such molecules restored NK function, and this implicates their immunosuppressive activity [82]. Expanding on these findings HNSCC tumors evade immune surveillance by exfoliating sNKG2DL, which inhibits NK cell function and invasion of the tumor. Another study by Weil et al. showed high levels of plasma sNKG2DLs correlated with the inhibition of NK cells and the progression of disease. Depletion of sNKG2DLs restored NK cell activity in vitro and after tumor resection. A proof-of-concept study in rhesus monkeys confirmed that adsorption apheresis can reduce circulating sNKG2DLs, and hence, suggests it as an efficient preconditioning protocol to enhance NK cell-based immunotherapy in HNSCC [101]. In cases of relapsed HNSCC, elevated plasma sMICA and TGF-β1 levels block NK cell activity by NKG2D downregulation and reduced cytokine secretion. However, functional CD16 expression is retained by NK cells, and restoration of cytotoxicity via antibody-dependent cellular cytotoxicity (ADCC) can be achieved. Cetuximab, a monoclonal anti-EGFR antibody, could rearm sMICA-inhibited NK cells, enhancing IFN-γ and TNFα secretion and tumor infiltration in vitro. These results supports the union of NK cell-based immunotherapy and monoclonal antibodies to overcome NKG2D pathway inhibition in HNSCC [102, 103]. The use of NK cells is a promising approach in cancer immunotherapy; however, combining irradiation and cisplatin as a chemoradiotherapy (CRT) regimen is being investigated to augment the effects of NK cell immunotherapy in HNSCC. These results indicated an increase NK cells and expression of activation markers such as NKG2D during the expansion process, which is positively correlated with the NK cell infiltration and overall survival in patients with HNSCC. On the other hand, NK cells have a potent anti-tumor effect in combination with CRT against HNSCC [104]. Further supporting this approach, combined with local IL-2 gene therapy, cisplatin enhances antitumor efficacy against HNSCC and reduces the dose-limiting toxicity of cisplatin. Cisplatin alone suppresses NKG2D expression on lymphocytes, but by IL-2 gene transfer, this suppression is prevented, and NK cell and CTL cytotoxicity is maintained. This combination interacts with tumor apoptosis and growth inhibition, highlighting NKG2D’s activation of antitumor immunity and the rationale for further developing combined chemo-immunotherapy strategies [105]. Immune regulation through Regulatory T cells (Tregs) is another factor influencing NKG2D-related therapy. Tregs suppress NK cell ADCC and NKG2D expression via TGF-β1. In patients receiving cetuximab, peripheral Tregs increase over time, potentially limiting NK cell function. Ipilimumab, a monoclonal anti-CTLA-4 antibody, can enhances NK cell ADCC by depleting and targeting Tregs and restoring NK cell function in co-culture systems with HNSCC cells [106]. In Johnson et al. phase I trial of ipilimumab and nivolumab definitive radioimmunotherapy in locally advanced HNSCC, 3-year PFS was 74% and OS 96%, outperforming standard chemoradiotherapy. However, patients experienced high levels of treatment-related toxicities. These findings show promise for the ability of ipilimumab to augment NK cell–mediated immunity and patient outcomes, but also raise toxic concerns [107]. An interesting approach to directly reverse immunosuppression is the application of recombinant vaccinia virus-expressing IL-2 vaccine (rvv-IL-2). Rvv-IL-2 restored NK-mediated tumoricidal immunity through the upregulation of NKG2D and CD16; also, rvv-IL-2 decreased TGF-β production, which is an important immunoinhibitory cytokine. These results validate the promise of IL-2–based immunotherapy for recovering NKG2D-mediated cytotoxicity caused by the TGF-β–enriched HNSCC TME [108]. With increasing evidence, the incorporation of NKG2D-based treatments into standard or multimodal therapies could lead to better patient outcomes. Subsequent trials need to address how best to optimize these approaches by biomarker-driven trial design and rational combination therapy. A number of clinical and subclinical studies have demonstrated the promising outcomes of NKG2D therapies for HNSCC management (Table 2). Table 2Clinical and preclinical outcomes of NKG2D-based therapies in HNSCCTargetType of studyCellular/Molecular findingsClinical notesReferenceNKG2D receptor – ULBP1Subclinical; Cellular: Human cell line HeLa, FaDu, PCI-15 A, DM14, CAL27, PCI-15B, SK-MES-1, and Tu167Expression of MICA/MICB, ULBP2/3 in HNSCCNo expression of ULBP1/4 in HNSCC• Cisplatin or bleomycin did not change the expression of NKG2D• Bortezomib increased the expression of ULBP1 [21]HSPA1A and HMOX1 - MICA/BSubclinical; Cellular: Human cell line Het-1 A, Cal 27, and FaDuSubclinical; Animal: wild-type C57BL/6 mice-• Sulforaphane was able to induce oxidative stress proteins, including HSPA1A and HMOX1• Sulforaphane upregulated MICA/B via NRF2• Sulforaphane enhanced NK cytotoxic effects against HNSCC [109]Glycolysis and Krebs enzymes, NK-stimulating receptors, such as NKG2DSubclinical; Cellular: murine squamous cell carcinoma cells MOC2, P029, and LY2Subclinical; Animal: Balb/c and C57Bl/6 miceUpregulation of PD-L1 in CD4+, CD8+, and regulatory T cells• Radiation and PD1-IL2v imposes anti-HNSCC measures• Radiation and PD1-IL2v enhances NK surveillance and cytotoxicity• PD1-IL2v upregulates glycolysis enzymes including lactate dehydrogenase A and glyceraldehyde 3-phosphate dehydrogenase, and Krebs cycle like succinate dehydrogenase• PD1-IL2v upregulates NK-stimulating receptors such as NKG2D [110]CXCL14 – NKG2D receptorSubclinical; Cellular: Human HNSCC samplesDownregulation of CXCL14 in HNSCCPoor expression of CXCL14 in NK cells• IL-2 increased the expression of CXCL14 in NK cells• Upregulation of CXCL14 resulted in NKG2D elevation and enhanced cytotoxic measures of NK cells• Upregulation of CXCL14 resulted in an increase of TNF-α and IFN-γ [111]MICA/B – ULBP1/2/3/4/5Database study; cellular: TCGA normal and HNSCC human samplesUpregulation of MICA/B and ULBP1/2/3/4 in HNSCCDownregulation of ULBP5 in HNSCC- [112]NKG2D receptorSubclinical; Cellular: Human NK cell line UM-SCC-1G/−11AG/−47, UPCI: SCC-90NKG2D expression in haNK• haNK cells effectively eliminated HNSCC• Avelumab and cetuximab enhanced the ADCC-dependent cytotoxic effects of NK cells [113]NKG2D receptorSubclinical; Cellular: PBMCs and HNSCC-• Metformin upregulated NKG2D, increased the number of NK cells, a shift in CD8+ T cells• Decrease of STAT3 was hypothesized as a potential moderator in NK cell strengthening [114]NKG2D ligands and NKG2D receptorSubclinical; Cellular: SSC15 and FaDu human HNSCC, PBMCs and feeder cellsSubclinical; Animal: 6-to-8-week femail mice, NOD/SCID IL-2Rγnull (NSG) female miceUpregulation of LFA-1, CD69, and NKG2D in eNK• Significant increase in NKG2D ligands and LFA-1 ligands in FaDu cell line after chemoradioation• Minimum changes in NKG2D ligands and LFA-1 ligands in LFA-1 cell line after chemoradioation• Increased migration of eNK after chemoradioation• Enhanced cytotoxic measures of eNK after chemoradioation via increase in NKG2D [115]MICA/B, ULBP1/2, NKG2DSubclinical; Cellular: PCI-13, SNU-899, and SNU-1041 HNSCCUpregulation of NKG2D and LFA-1 in eNK• Cisplatin and radiation therapy shortened the longevity of HNSCC• Cisplatin and radiation therapy increased the levels of MICA/B and ULBP1/2, with increased cytotoxic measures• Cisplatin and radiation therapy increased eNK and HNSCC cell-to-cell interaction mediated via NKG2D and partially via LFA-1 [104]NKG2DSubclinical; Animal: orthotopic murine-• Cisplatin with IL2 gene therapy reduced tumor proliferation, and increased rates of cellular apoptosis• Cisplatin with IL2 gene therapy enhanced the activation of cytotoxic T lymphocytes• Cisplatin alone did not significantly increased number of NK cells, however, cisplatin with IL2 gene therapy enhanced the activation of NK cells• mNKG2D activation by IL2 gene therapy and cisplatin [116] Clinical and preclinical outcomes of NKG2D-based therapies in HNSCC Subclinical; Cellular: Human cell line HeLa, FaDu, PCI-15 A, DM14, CAL27, PCI-15B, SK- MES-1, and Tu167 Expression of MICA/MICB, ULBP2/3 in HNSCC No expression of ULBP1/4 in HNSCC • Cisplatin or bleomycin did not change the expression of NKG2D • Bortezomib increased the expression of ULBP1 Subclinical; Cellular: Human cell line Het-1 A, Cal 27, and FaDu Subclinical; Animal: wild-type C57BL/6 mice • Sulforaphane was able to induce oxidative stress proteins, including HSPA1A and HMOX1 • Sulforaphane upregulated MICA/B via NRF2 • Sulforaphane enhanced NK cytotoxic effects against HNSCC Subclinical; Cellular: murine squamous cell carcinoma cells MOC2, P029, and LY2 Subclinical; Animal: Balb/c and C57Bl/6 mice • Radiation and PD1-IL2v imposes anti-HNSCC measures • Radiation and PD1-IL2v enhances NK surveillance and cytotoxicity • PD1-IL2v upregulates glycolysis enzymes including lactate dehydrogenase A and glyceraldehyde 3-phosphate dehydrogenase, and Krebs cycle like succinate dehydrogenase • PD1-IL2v upregulates NK-stimulating receptors such as NKG2D Downregulation of CXCL14 in HNSCC Poor expression of CXCL14 in NK cells • IL-2 increased the expression of CXCL14 in NK cells • Upregulation of CXCL14 resulted in NKG2D elevation and enhanced cytotoxic measures of NK cells • Upregulation of CXCL14 resulted in an increase of TNF-α and IFN-γ Upregulation of MICA/B and ULBP1/2/3/4 in HNSCC Downregulation of ULBP5 in HNSCC • haNK cells effectively eliminated HNSCC • Avelumab and cetuximab enhanced the ADCC-dependent cytotoxic effects of NK cells • Metformin upregulated NKG2D, increased the number of NK cells, a shift in CD8+ T cells • Decrease of STAT3 was hypothesized as a potential moderator in NK cell strengthening Subclinical; Cellular: SSC15 and FaDu human HNSCC, PBMCs and feeder cells Subclinical; Animal: 6-to-8-week femail mice, NOD/SCID IL-2Rγnull (NSG) female mice • Significant increase in NKG2D ligands and LFA-1 ligands in FaDu cell line after chemoradioation • Minimum changes in NKG2D ligands and LFA-1 ligands in LFA-1 cell line after chemoradioation • Increased migration of eNK after chemoradioation • Enhanced cytotoxic measures of eNK after chemoradioation via increase in NKG2D • Cisplatin and radiation therapy shortened the longevity of HNSCC • Cisplatin and radiation therapy increased the levels of MICA/B and ULBP1/2, with increased cytotoxic measures • Cisplatin and radiation therapy increased eNK and HNSCC cell-to-cell interaction mediated via NKG2D and partially via LFA-1 • Cisplatin with IL2 gene therapy reduced tumor proliferation, and increased rates of cellular apoptosis • Cisplatin with IL2 gene therapy enhanced the activation of cytotoxic T lymphocytes • Cisplatin alone did not significantly increased number of NK cells, however, cisplatin with IL2 gene therapy enhanced the activation of NK cells • mNKG2D activation by IL2 gene therapy and cisplatin Challenges and limitations HNSCC is characterized by considerable inter- and intra-tumoral heterogeneity [117]. This diversity expands to NKG2DLs expression. Indeed, MICA/B and ULBP family members are differentially regulated by stress signals and tissue context [118]. Such existing heterogeneity, suggests that only a subset of malignant clones may display sufficient one-of-a-kind ligand for effective NKG2D engagement of the NK/T lymphocytes. In practice, therapies that rely on uniform NKG2D–ligand interactions may miss a massive number of tumor cells in the considerably heterogeneous tumor environment [119, 120]. Another important point that should be importantly taken care of is the clear discrimination between acute and chronic exposure of NKG2D to NKG2DLs. As previously stated, acute ligand engagement typically activates cytotoxicity, but chronic or excessive NKG2D stimulation can induce tolerance or even promote protumor inflammation. Hence, it should be considered for the immunotherapies that the prolonged NKG2D exposure to ligands can paradoxically enhance the immune evasion capabilities [45]. Another challenge is the clear lack of predictive biomarkers for NKG2D-based immunotherapies. Conventional biomarkers have not been studied in the context of the NKG2D axis, and the levels of sNKG2DLs have shown inconsistency in some studies that tried to link their levels with clinical outcome [121, 122]. The existing complexity to identify a single biomarker of prediction mostly stems from the intra-tumoral heterogeneity and the intricate immunology in this context. Without a validated biomarker specific to the NKG2D axis, it becomes highly challenging to select patients or monitor responses in clinical trials. Last but not least, CAR-based and antibody therapies targeting NKG2D or its ligands are likely to face the known limitations of these immunotherapies in solid tumors and their TMEs. Similar to other solid tumors, HNSCC and its TME profoundly suppresses NKG2D and blunts effector function [123]. These challenges limit the effectiveness of current NKG2D-directed CAR or antibody strategies in HNSCC. Future perspectives NK-based treatment of HNSCC via NKG2D methods can be integrated with other conventional treatments of cancer. Radiation therapy followed by cisplatin is capable of imposing anti-HNSCC measures and enhancing the anti-tumor effects of NK cells; an integrated approach which is dependent on MICA/B, ULBP1/2/3, and ICAM-1/2/3 [104]. Moreover, NKG2D-based therapies can be synergized with other cancer immunotherapies, such as CAR NK. A novel design of a preserved domain of MICA/B in induced pluripotent stem cell (iPSC)-derived NK cells successfully counteracted HNSCC immune escape [124]; the adjuvant application of other anti-HNSCC treatments, such as anti-PD-1 treatments, monoclonal antibodies targeting MICA/B cleavage, nanodrugs or nano-delivery of treatments, exosome-transshipment methods, oncolytic virotherapies, and cancer vaccines, could be suggested. To maximize the efficacy of synergistic approaches, the utilization of in silico analysis, simulators of immune system responses, and machine learning procedure are suggested. NKG2D expression and its ligands can be used as a diagnostic/prognostic biomarker in HNSCC [125]. Another important approach in NKG2D-based immunotherapies could be proteasome inhibitors. With positive results in other experiments [126], taking benefit from other proteasome inhibitors such as MG132, is recommended for future studies. In this context, the application of proteomics, structural dynamic analysis, and artificial intelligence-based tools are suggested to improve the efficacy and scientific accuracy. Lowered NKG2D expression in CD8+ T cells could indicate immune escape measure in HNSCC [127]. In parallel, levels of NKG2DLs are associated with the extent of tumor proliferation [17]. Another study indicated that increased levels of sNKG2DL are positively correlated with poorer prognosis of HNSCC [128], yet the lowered expression of NKG2D ligands is a biomarker of poorer outcome in HNSCC [129]. It could be hypothesized that NKG2D ligands to be used in the detection of HNSCC and in the prediction of its clinical outcomes. Given the considerable role of exosomes in the immunomodulation of HNSCC, the application of liquid biopsies for non-invasive diagnosis and prognostic predictions is recommended. The role of gene therapies and TME remodeling might be underexplored in the context of HNSCC management. Given the promising results of IL2 transfer into tumor cells [116], it could be suggested the engineering, transplantation, or manipulation of other genes regarding MICA/MICB and ULBP1/2/3 for tumor cells. Other gene delivery approaches can be used in terms of NK cells for bypassing the shedding of NKG2D ligands, including viral-vectors, cationic lipids, polymers, ionic complexes, and Sono/Electroporation. In addition, the epigenetic studies to determine the precise sensitizer agents of HNSCC is recommended for future studies. Conclusion In conclusion, NKG2D receptors play a vital role in the detection and elimination of pathologically transformed cells, such as neoplasms and malignancies. Numerous ligands are responsible for activating the NKG2D receptors and induce the cytotoxic effects of NK cells. One mechanism of HNSCC immune evasion is mediated via the manipulation of NKG2D ligands and a reduction in NK cell activation and hindered tumoricidal by other immune elements like T lymphocytes. By the application of NKG2D immunotherapies, immune evasion methods of shedding ligands, suppressed NK cells, and attenuated immune surveillance can be neutralized [130]. However, the utilization of NKG2D-based NK cell-mediated immunotherapies some the challenges for a complete clinical translation. NKG2D polymorphisms may reduce its expression, and without clear markers or consistent effects on tumors, its therapeutic use is complex. Most importantly, NKG2D activation can deteriorate the immune measures of NK cells [131]. The application of NKG2D receptors/ligands as therapeutic agents is at some points followed by hyperinflammation, excessive recruitment of NK cells, and eventually return to the immunoinhibitory status. Being initially presented as a positive technique immunotherapy, clinical application of NKG2D is still in early phases and demands large assays, with attention to prognostic markers, CAR technology, and gene therapy. Schematic showing the molecular pathways and cell–cell interactions that shape NKG2D-driven tumor surveillance and immune escape in HNSCC. In the upper panel, tumor cells increase surface expression of NKG2D ligands (MICA, MICB, and ULBP family members) when exposed to stress signals such as DNA damage from environmental or lifestyle carcinogens, including tobacco smoke, alcohol, HPV or EBV infection, and ultraviolet or ionizing radiation. Engagement of NKG2D on NK cells through the DAP10 adaptor triggers downstream cascades involving VAV1, PI3K, JNK, MEK, and STAT5, ultimately leading to polarized degranulation of perforin and granzymes and the release of pro-inflammatory cytokines. Acknowledgements This study was facilitated by the USERN Research Tour held in the USERN & Companions House and Museum. This study was supported by the USERN Foundation, Houston, TX (Grant No. 2025.0820). This grant was awarded to the Network of Immunity in Infection, Malignancy and Autoimmunity (NIIMA), Universal Scientific Education and Research Network (USERN), Stockholm, Sweden. Illustration created with BioRender.com. Figure 2 Created in BioRender. Saleki, K. (2026) https://BioRender.com/w8sz4fc Figure 1 Created in BioRender. Saleki, K. (2026) https://BioRender.com/7i0lr47. Author contributions P. A., M. S., A. K., T. S., F. T., P. K., N. Y., K. S., and N. R. conceptualized the study and drafted the manuscript. K.S. and T. S. designed figures. K. S. and N. R. supervised the study, critically appraised the manuscript, and prepared the final draft. Funding This study was supported by the USERN Foundation, Houston, TX (Grant No. 2025.0820). Data availability No datasets were generated or analysed during the current study. Declarations Ethics approval and consent to participate Not applicable. Consent for publication All authors provided consent to publish the manuscript. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Tara Shahmoradi, Fatemeh Tajafrooz and Pegah khodaee share co-third authorship and contributed equally to the present work. Parsa Alijanizadeh and Mahshid Shahmoradi share co-first authorship and contributed equally to the present work. References Associated Data Data Availability Statement No datasets were generated or analysed during the current study.