Optimizing immunotherapy for head and neck squamous cell carcinoma: recent advances and future directions.

1.
Introduction
Head and neck squamous cell carcinoma (HNSCC), as the sixth most prevalent malignant tumor globally, affects over 830,000 individuals and results in approximately 430,000 deaths annually [1]. Conventional treatment for HNSCC typically involves a combination of surgery, chemotherapy, and radiotherapy, yet more than half of the patients experience relapse, and the 5-year overall survival rate for late-stage patients remains at only 35-45% [2]. A growing body of research has highlighted the critical role of the host immune system in the progression of HNSCC, influencing tumor growth, vascularization, and metastasis. Tumor-infiltrating cytotoxic T cells and natural killer (NK) cells can induce tumor cell death and constrain tumor growth, whereas regulatory T cells (Tregs), myeloid-derived suppressor cells (MDSCs), and tumor-associated macrophages (TAMs) facilitate tumor growth via immune evasion mechanisms in HNSCC [3–5]. In recent years, immune checkpoint blockade (ICB), particularly targeting programmed death-1 (PD-1) and programmed death ligand-1 (PD-L1), has shown substantial promise for patients with recurrent or metastatic (R/M) HNSCC. Clinical trials such as Checkmate-141 and KEYNOTE-040 have demonstrated that PD-1 blockade can enhance overall survival in platinum-refractory R/M HNSCC patients compared to cytotoxic chemotherapy [6,7]. Additionally, KEYNOTE-048 revealed that pembrolizumab monotherapy could improve overall survival in HNSCC patients with PD-L1 positive tumors and offers a superior safety profile compared to cetuximab combined with chemotherapy [8]. However, it is well recognized that only a small fraction of patients are suitable candidates for single immune checkpoint inhibitor therapy. The objective response rates for monotherapy remain low, reported at 13.3% for nivolumab and 14.4%-16.9% for pembrolizumab. This indicates that a significant proportion of HNSCC patients do not respond to single-agent immune checkpoint blockade therapy. Moreover, the clinical response to immune checkpoint inhibitors often requires extended observation periods, which can delay effective treatment for patients who do not benefit from the therapy and might even experience tumor progression or hyperprogression. Additionally, immune-related adverse events (irAEs) cannot be ignored, with 60% of HNSCC patients experiencing immunotherapy-related side effects, and over 17% suffering from severe adverse events. Conversely, multi-pronged approaches that incorporate multiple immune checkpoint inhibitors or combine immunotherapy with other conventional strategies have shown enhanced synergistic effects. Nowadays, most ongoing clinical trials are focusing on combined immunotherapy. Although the knowledge of combined immunotherapy is still evolving, many studies have proved that combined immunotherapy has improved the anti-tumor efficacy.
Currently, scientific efforts are underway to explore the optimal protocols of combination treatment schemes to enhance the clinical outcomes of HNSCC. Here, we comprehensively review the recent studies on combined immunotherapy strategies, including combination of multiple immunotherapies, and combination of immunotherapy with radiotherapy, chemotherapy, and surgery. The underlying mechanisms of the synergistic effects caused by the combined immunotherapy are elucidated respectively. Focused on the distinct immune status of individual patients, many potential immune biomarkers are concluded to indicate the most suitable treatment for HNSCC patients. Additionally, the treatment efficacy, potential adverse reactions, and practical considerations for the implementation of these combined therapies are also summarized in this review. Furthermore, some critical details about dose, fraction, and sequences of the combined pattern are summed up for reference purposes. It is evident that the future of immunotherapy for HNSCC lies in personalized medicine. Given the heterogeneous nature of HNSCC, characterized by a complex immune environment and diverse molecular pathology, a universal solution remains elusive. We hope the review can offer valuable insights to both clinicians and HNSCC patients. Tailoring optimally combined therapies according to the distinct immune status of HNSCC patients will enable more effective adjustments to treatment regimens, especially for those who have not responded to previous treatments. Precision and personalized combined immunotherapy, based on the immune status of individual patients, promises to bring renewed hope to those affected by HNSCC.

2.
Combination immune checkpoint inhibition
Immune checkpoints, often described as the physiological ‘brakes of the immune system’, maintain immune homeostasis by restraining the activity of immune cells, thereby preventing unnecessary damage to normal tissues. However, this immune checkpoint-mediated immunosuppression can also facilitate immune escape, thereby contributing to successive tumor progression. Therefore, immune checkpoint inhibitors (ICIs) represented by anti-PD-1/anti-CTLA-4 have heralded the dawn of the immunotherapy era, bringing revolutionary advancements to cancer treatment. However, monotherapy has gradually revealed its significant limitations, which are specifically manifested in a higher likelihood of developing drug resistance, constraints associated with targeting a single pathway, and a slower onset of action. Consequently, medical researchers have shifted their focus towards combination therapies involving various immunotherapeutic agents. Multiple clinical trials have demonstrated that combinations of multiple immune checkpoint inhibitors might achieve better treatment efficacy than monotherapy for HNSCC patients (Figure 1).
2.1.
Anti-PD-1 combined with anti-CTLA-4
The combination therapy of anti-PD-1 with anti-CTLA-4 is among the most extensively studied approaches. Immunotherapies targeting both CTLA-4 and PD-1 have demonstrated unprecedented efficacy across several cancer types. In a study by Montler et al., T cells derived from peripheral blood lymphocytes (PBL) and tumor tissues of 29 surgically treated patients were analyzed for expression of OX40, PD-1, and CTLA-4. The analysis revealed a substantial proportion of PD-1, CTLA-4, and OX40 triple-positive cells within the regulatory T-cell population in HNSCC [9]. Furthermore, a case report described a 46-year-old patient with refractory HNSCC who was successfully treated with nivolumab (3 mg/kg every 2 weeks) in combination with ipilimumab (1 mg/kg every 6 weeks). Computed tomography examinations revealed a significant reduction in tumor volume at 8 weeks, with nearly complete remission observed 4 months after treatment initiation. Serum PD-L1 decreased due to the response to the therapy, followed by an increase at the time of progression [10]. The overall survival (OS) of HNSCC patients treated with both anti-PD-1 and CTLA-4 antibodies was found to be 7.6 months, compared to 6.0 months with anti-PD-1 monotherapy and 5.5 months with anti-CTLA-4 monotherapy. These findings suggest potential therapeutic advantages of targeting PD-1 and CTLA-4 in combination for HNSCC patients. According to the report by Haddad and his colleagues, the CheckMate 651 trial demonstrated that anti-PD-1 plus anti-CTLA-4 had a better safety profile compared to the EXTREME regimen in recurrent/metastatic SCCHN. However, no significant survival benefit was observed, indicating that novel treatment strategies remain an unmet need for patients with R/M SCCHN [11].

2.2.
Anti-PD-1 combined with anti-LAG-3
LAG-3, typically expressed on the lymphocyte membrane, plays a crucial role in regulating T cell activation and proliferation [12]. A pilot clinical study has further revealed that elevated soluble LAG-3 levels are correlated with unfavorable prognosis in HNSCC [13]. Analysis of HNSCC tissue specimens demonstrates upregulated LAG-3 expression on tumor-infiltrating lymphocytes, with its overexpression significantly associated with higher pathological grade, increased tumor size, and greater lymph node involvement [14]. In several pre-clinical murine models of cancer, anti-LAG-3 is synergized with anti-PD-1 to improve CD8 T cell responses and CD4 T cell function [15]. Consequently, LAG-3 is considered a promising immune checkpoint for synergistic combination with PD-1 blockade in HNSCC therapies. Interestingly, Forster et al. reported that Efti, a soluble LAG-3 protein, in combination with pembrolizumab, demonstrates encouraging antitumor activity and pharmacodynamic effects in patients with second-line HNSCC [16]. This finding appears to contradict the aforementioned observations and warrants further investigation. Although the combination immunotherapy of anti-PD-1 and anti-LAG-3 antibodies has shown therapeutic efficacy in HNSCC, not all patients respond to it. Wang and colleagues investigated the resistance mechanism to anti-LAG-3 plus anti-PD-1 therapy in a murine model of HNSCC. They discovered that the resistance is mediated by an interaction between SOX9+ tumor cells and Frp1+ neutrophils, which leads to impaired anti-tumor immunity [17].

2.3.
Anti-PD-1 combined with anti-GITR
GITR, expressed on T cells, is a type 1 transmembrane protein of TNFR superfamily [18]. Interaction of GITR with its ligand (GITRL) promotes T cell proliferation and prevents activation-induced cell death [19]. A clinical phase I/IIa trial has assessed the efficacy of BMS-986156 (anti-GITR) and or in combination with nivolumab in solid tumors. The combination of anti-GITR with anti-PD-1 showed a similar safety profile and therapeutic effect to anti-PD-1 monotherapy [20]. Of note, GITR expression in the tumor microenvironment is heterogeneous across tumor types, with HNSCC showing relatively lower expression levels compared to other malignancies, suggesting limited suitability of anti-GITR monotherapy in this context [21]. Nonetheless, multiple preclinical studies have demonstrated synergistic antitumor activity with combined anti-GITR and anti-PD-1 treatment in murine models [22]. Therefore, whether combining anti-GITR and anti-PD-1 could yield synergistic benefits in HNSCC remains an open question.

2.4.
Anti-PD-1 combined with anti-TIM-3
TIM-3, a negative regulator of Th1 immunity, is predominantly expressed on CD8+PD-1+ T cells of tumor patients. In a phase II clinical trial involving HNSCC patients receiving cetuximab, Jie et al. assessed immune checkpoint receptor expression on CD8+ tumor-infiltrating lymphocytes (TILs). The study revealed that elevated frequencies of PD-1+TIM-3+ CD8+ TILs were significantly associated with unfavorable clinical outcomes. These results indicate that dual PD-1/TIM-3 blockade could represent a promising strategy to improve prognosis in HNSCC [23]. Additionally, TIM-3+PD-1+ tumor-infiltrating lymphocytes (TILs) were found to exhibit the most severe exhausted phenotype, characterized by an inability to proliferate and secrete IL-2, TNF, and IFN-γ in mice bearing solid tumors [24]. Consequently, the combination of anti-TIM-3 and anti-PD-1 therapies can be more effective than either treatment alone [25]. Despite this, most ongoing clinical trials investigating Anti-PD-L1 in combination with Anti-TIM-3 are focused on solid tumors broadly. Efficacy and toxicity data for a dedicated HNSCC cohort have not yet been released, making it impossible to evaluate the true benefit-risk profile of this regimen.

2.5.
Anti-PD-1 combined with anti-TIGIT
TIGIT, a member of the Nectin family, is predominantly expressed on activated T cells and NK cells [26,27]. The number of TIGIT+ T cells and CD155- (the high-affinity ligand of TIGIT) tumor cells is increased compared to healthy individuals. Interestingly, TIGIT was not only highly expressed on TILs but also PBMCs of HNSCC, unlike PD-1, CTLA-4, LAG-3, and TIM-3 [28]. Blocking TIGIT/CD155 signaling has been shown to enhance anti-tumor immunity in murine models [29]. CD155 was upregulated by anti-PD-L1 treatment alone and anti-TIGIT treatment also increased the expression level of PD-L1 on myeloid-derived suppressor cells in an HNSCC murine model [30]. Based on these mechanistic insights, the dual blockade of TIGIT and PD-1 has shown promising clinical efficacy in trials involving other cancer types. Several clinical trials are currently underway to evaluate this combination strategy in HNSCC, and data from these studies are expected to validate its therapeutic potential in the near future.

2.6.
Anti-PD-1 combined with anti-VISTA
VISTA, an immunological checkpoint protein containing an Ig domain, is predominantly expressed on hematopoietic cells. With sequence homology to PD-1 and PD-L1, VISTA is considered one of the most promising targets for immunotherapy [31]. However, VISTA has been shown to play a non-redundant role in regulating T cell activation independent of the PD-1/PD-L1 pathway. Kato et al. reported that the protein level of VISTA increased following combined therapy with anti-PD-1 and anti-CTLA-4 [32]. And combination therapy using anti-VISTA and anti-PD-L1 demonstrated optimal anti-tumor efficacy in murine models [33]. Of note, VISTA was found to overexpress in HNSCC specimens and correlate with lymph node status [34]. These findings suggest that optimal tumor-clearing therapeutic efficacy may be achieved through the combination of VISTA blockade. Nevertheless, clinical evidence specifically dedicated to evaluating the anti-PD-1 and anti-VISTA combination in HNSCC is currently lacking, as this area of research is still in its infancy.
The effectiveness of the combined immune checkpoint inhibitors typically depended on multiple factors, including the co-expression level of immune checkpoints, immune response, and immunogenicity. CD155 (the high-affinity ligand of TIGIT) and VISTA are overexpressed following treatment with anti-PD-1 and anti-CTLA-4. The phenotype above hints that a stronger synergistic effect would emerge if TIGIT or VISTA inhibitors are added. Regardless, we must consider that even with the simultaneous blockade of these immune checkpoints, the compensatory upregulation of other checkpoints could create a ‘whack-a-mole’ effect, leading to only transient efficacy of the combination therapy. Given the specificity of CTLA-4, which primarily regulates T cell activation, anti-PD-1 combined with anti-CTLA-4 is recommended for HNSCC patients whose immune response is insufficiently activated. Moreover, anti-PD-1 should be combined with some immune checkpoints such as TIGIT, TIM-3, and LAG-3 which triggered milder phenotypes when intolerable immune-related events occurred. Therefore, many factors mentioned above should be considered to improve the response rate of combined immune checkpoint inhibitors. Notably, multiple parallel studies have been conducted on novel treatment approaches for patients with R/M HNSCC, including therapeutic vaccines, bispecific antibodies/fusion proteins, multi-targeted kinase inhibitors, and antibody-drug conjugates (ADCs) in recent years. These novel agents leverage concurrent targeting and modulation of the tumor microenvironment to harness anti-tumor immunity and inhibit pro-tumor signaling pathways, yielding promising outcomes. Taking bispecific antibodies as example, Volrustomig also known as MEDI5752, is a bispecific antibody targeting both programmed death-1 (PD-1) and cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4). This dual-targeting bispecific antibody has the potential to enhance therapeutic efficacy while reducing the risk of toxicity typically associated with CTLA-4 inhibitors. Recently, a phase III, randomized, open-label, multicenter global study of Volrustomig as sequential therapy for participants with unresected locally advanced HNSCC has been initiated (NCT06129864). Although these trials are still underway, they represent the pioneering breakthrough in the therapy of R/M HNSCC. And further novel insights and significant findings are anticipated to emerge from these clinical trials [35].

3.
Combination of radiotherapy with immunotherapy
Radiotherapy (RT) is one of the mainstay treatment methods for most patients with HNSCC [36]. It is assumed that more potent anti-tumor response can be induced in HNSCC patients when immunotherapy combined with radiotherapy. Notably, in the combination of radiotherapy with immunotherapy, radiotherapy appears to exert its effects through multiple mechanisms. First, irradiation induced the vaccine-induced CD8 T cells to the threshold level to trigger the potent anti-tumor response. Second, tumor vascular permeability increased with the augmented pericyte coverage and intercellular cell adhesion molecule-1 (ICAM-1) expression in the combination therapy. And the vascular-induced changes were found to occur in a dose-dependent manner after a single dose of irradiation. Third, immunologic cell death such as dendritic cells (DCs) induced by radiotherapy also contributed to the synergy between radiotherapy and immunotherapy [37]. Here, some clinical and preclinical evidence is provided to discuss how to maximize therapeutic potential for HNSCC through combining radiotherapy and immunotherapy.
3.1.
Preclinical and clinical evidence about the combination immunotherapy with radiotherapy
Radiotherapy induces an increase in antigen presentation, cytotoxic T cell infiltration, and PD-L1 expression on tumors in response to T cell-derived IFN-γ, acting as a mechanism of adaptive resistance. This phenomenon provides a compelling rationale for the concurrent administration of PD-1 or PD-L1 antibodies with radiotherapy to enhance anti-tumor immunity and establish immunological memory. Leidner and colleagues demonstrated the priority of combining stereotactic body radiation therapy (SBRT) with anti-PD-1 to PD-1 blockade alone in a phase Ib clinical trial. In their study, the combination of neoadjuvant SBRT with nivolumab resulted pathological downstaging in 90% of patients with previously untreated locally advanced HNSCC. Among the entire cohort, the major pathological response (MPR) and pathological complete response (PCR) rate was 86% and 67%, respectively. Sari and colleagues also verified that concurrent SBRT and immunotherapy could achieve higher survival and local control rates in patients with R/M HNSCC. They reported that six months’ overall survival rates and progression-free survival rates were 93% and 86%, respectively and the local control rate in the site of SBRT was 96% [38]. These data revealed that the combination of radiotherapy with immunotherapy is both safe and superior to immune checkpoint inhibitor therapy alone. However, McBride et al. and his colleagues conducted a single-center, randomized phase II trial to investigate whether radiotherapy can synergize with anti-PD-1 therapy in patients with metastatic or recurrent HNSCC and improve survival through abscopal effects. A total of 62 patients were enrolled in the study, with 30 patients randomly assigned to the nivolumab group and the remaining 32 patients allocated to the nivolumab plus SBRT group. Each patient had at least two metastatic lesions: one that could safely receive radiotherapy and another that was measurable by RECIST version 1.1. The results showed no significant differences between the two groups in terms of ORR (34.5% [95% CI, 19.9% to 52.7%] v 29.0% [95% CI, 16.1% to 46.6%]; p = 0.86), overall survival (p = 0.75), progression-free survival (p = 0.79), or response duration (p = 0.26). Therefore, the study concluded that in patients with metastatic HNSCC, nivolumab plus SBRT did not significantly improve survival compared to nivolumab alone, and no notable abscopal effects were observed [39]. Furthermore, to investigate whether the addition of anti-PD-L1 to chemoradiotherapy could improve survival in patients with unresected locally advanced HNSCC, Lee et al. conducted a randomized, double-blind, placebo-controlled phase III clinical trial. In this study, 697 patients enrolled in the study were from 196 hospitals and cancer treatment centers across 22 countries. A total of 350 patients were randomly assigned to the avelumab group, while 347 patients were assigned to the placebo group (only treated with chemoradiotherapy). However, the median progression-free survival follow-up time was 14.6 months (IQR 8.5–19.6) in the avelumab group and 14.8 months (11.6–18.8 months) in the placebo group. These findings indicate that incorporating avelumab into chemoradiotherapy does not extend survival in patients with locally advanced HNSCC [40]. Tao et al. conducted the first evaluation of the potential synergistic effects of combining pembrolizumab with radiotherapy versus the standard of care (SOC) of cetuximab combined with radiotherapy in patients with locally advanced HNSCC. The study included a total of 133 patients, with 66 patients randomized to the cetuximab combined with radiotherapy group and the remaining 67 patients allocated to the pembrolizumab combined with radiotherapy group. The results showed no significant differences between the two groups in progression-free survival (HR = 0.85, 95% CI 0.55–1.32, p = 0.47) and overall survival (HR = 0.83, 95% CI 0.49–1.40, p = 0.49). This demonstrates that pembrolizumab combined with radiotherapy does not improve survival compared to cetuximab combined with radiotherapy in patients with locally advanced HNSCC. However, it is noteworthy that the incidence of adverse events including mucositis, radiation dermatitis, and rash was significantly lower in the pembrolizumab combined with radiotherapy group (74%) compared to the cetuximab combined with radiotherapy group (92%) [41].
The inconsistent results reported in different clinical trials may be due to the dual-edged nature of radiotherapy’s impact on immunotherapy mechanisms. To achieve truly precise and effective radio-immunotherapy, it is essential to comprehensively consider the multifaceted effects of radiotherapy on cancer cells, immune cells, and various stromal cells within the microenvironment. Moreover, Kobayashi and colleagues have confirmed the feasibility of combining NKT cell therapy with radiation therapy in HNSCC. They observed that the number of NKT cells remained unchanged after radiotherapy, and the proliferative responses of these NKT cells were enhanced in comparison to those collected before radiation therapy. This finding indicates that NKT cells are relatively resistant to radiation and may be suitable for use as adjuvant immunotherapy in conjunction with radiotherapy [42].
Furthermore, the progress in combining proton therapy with immunotherapy deserves attention. Compared to conventional photon radiotherapy, proton therapy more effectively preserves immune cells and allows for dose escalation without increasing additional toxicity. This theoretical advantage suggests that, when combined with immunotherapy, it may achieve a stronger abscopal effect with lower synergistic toxicity, thereby creating a synergistic treatment effect [43]. This promising prospect is supported by preliminary clinical evidence: a Phase II study by Kim et al. enrolled 31 patients with R/M HNSCC, who received intravenous Durvalumab (1500 mg) and Tremelimumab (75 mg, once every four weeks, for four doses), followed by proton therapy (total dose of 25 Gy in 5 fractions) to one lesion after the first treatment cycle. Among the 23 patients who completed proton therapy, the objective response rate (ORR) was 30.4%, indicating that the combination regimen was well-tolerated and demonstrated encouraging antitumor activity [44]. Future efforts should focus on larger Phase III clinical trials to validate its survival benefits and further optimize the timing and dosage of the combination therapy, with the ultimate goal of establishing this as a standard treatment paradigm.

3.2.
Precondition tumor environment to improve the synergetic efficacy
For clinical translation, preconditioning the tumor environment with therapies designed to counteract the immunosuppressive effects of radiotherapy could enhance the overall synergistic efficacy. Numerous studies suggest that modulating the tumor immune microenvironment is a promising strategy to improve treatment outcomes in HNSCC and other solid tumors.
Although immunoradiotherapy had been proved to be superior to immune checkpoint alone, the response to combined radiotherapy and immune checkpoint blockade was found to be only transient, and adaptive resistance developed quickly. Treg accumulation in tumor and the circulation was reported to correlate with recurrence, metastasis, and poorer survival in HNSCC. The treatment with dual checkpoint blockade (anti-TIM3 or anti-CTLA-4 with PD-L1 blockade) and radiotherapy indeed reduced Tregs to 15%, but more than 80% of these Tregs were Ki-67+ ,indicating the extensive proliferation of local Tregs [45]. Remarkably, targeted depletion of Tregs would restore an anti-tumor response of radiotherapy and dual checkpoint blockade combined in the preclinical orthotopic models. Nevertheless, their data continued to show that Treg depletion alone could not induce an anti-tumor immune response in a so-called cold tumor, likely because the absence of Tregs was insufficient to expose tumor neoantigens and attract CD8+ T cells [46]. Targeted Treg inhibitor may contribute to overcoming resistance to the combined treatment of radiotherapy and immunotherapy in HNSCC. And the potent synergistic effect of radiotherapy and Treg depletion combined should also function through distinct aspects of anti-tumor immunity.
Inducible nitric oxide synthase (iNOS) is one of the tumor-mediated inflammatory molecules that is highly upregulated in numerous solid tumors including HNSCC and promotes tumor growth by inducing and recruiting MDSCs. Inhibition of iNOS has been shown to induce both immune-dependent and independent anti-tumor effects. Cyclophosphamide (CTX) serves as an ideal complement to iNOS inhibition due to its capacity to deplete Tregs. Consequently, the CTX/LNIL combination is commonly employed to positively recondition the tumor immune microenvironment. In Hanoteau and colleagues’ study, CTX/LNIL was demonstrated to enhance the susceptibility of treatment-refractory tumors to chemoradiotherapy (CRT) by using a preclinical model of human HPV-associated HNSCC. This combined treatment effectively remodels the tumor myeloid immune environment, characterized by an increase in anti-tumor immune cell types and a decrease in immunosuppressive granulocytic MDSCs [47]. Furthermore, they observed that radiotherapy, whether used alone or in combination, was insufficient to eradicate established tumors or to reverse the unfavorable balance of effector to suppressor cells within the tumor immune microenvironment.

3.3.
Appropriate radiotherapeutic regimen in combination immunotherapy with radiotherapy
Combining radiotherapy with immunotherapy has been demonstrated to suppress tumor growth and enhance survival in HNSCC. However, determining the optimal radiation dose and fractionation to induce potent anti-tumor immunity requires further investigation. Traditionally, doses of 50–70 Gy delivered in multiple fractions of 1.8-2 Gy per fraction are preferred in the curative setting of solid tumors. The response of macrophages to radiotherapy is dose-dependent, and tumor-associated macrophages are associated with poor prognosis in relapsed HNSCCs. However, there is currently no commended radiation dose in the combination of radiotherapy and immunotherapy for optimal efficacy. Many preclinical evidence suggests that hypofractionated radiotherapy can enhance anti-tumor immunity when combined with immune checkpoint blockade. Conversely, low-dose daily fractionated radiotherapy used for HNSCC may result in lymphopenia and immunosuppression. Morisada and colleagues have revealed that 8 Gy in two fractions combined with anti-PD-1 therapy, rather than 2 Gy in ten fractions could reverse adaptive immune resistance and achieve CD8 T cell-dependent control of both primary and abscopal tumors [48]. Consistent with these data, Oweida and colleagues observed an increase in effector CD4 and CD8+ T cells with a single dose of 10 Gy, but not with fractionated 10 Gy, in orthotopic mouse models of HNSCC [46]. The head-to-head comparison between hypofractionated high dose and daily fractionated low dose has significant implications for the design of clinical trials combining radiotherapy and immunotherapy. In addition to radiation dose and fractionation, the scheduling of treatments is also critical to the synergistic effect of combination immunoradiotherapy. Dovedi et al. discovered that PD-L1 blockade, when administered concurrently rather than sequentially with fractionated radiotherapy, generates effective anti-tumor immunity and long-term tumor control [49]. Therefore, it is equally important to consider the radiation dose, fractionation, and schedule of radiotherapy when combined with immunotherapy to achieve better therapeutic efficacy.
The abscopal effect induced by RT, although rare and elusive, has garnered wide interest with the advent of immune checkpoint blockade. Similar with anti-tumor immunity, the abscopal effect also requires priming of tumor-specific T cells and a cytotoxic T cell-permissive tumor environment. Unfortunately, immune suppression within the tumor environment ensues after immunotherapy with the concurrent radiotherapy due to the negative feedback on T cell effector function including Treg activity mentioned above. This phenomenon may explain the lack of improved response and the absence of an abscopal effect observed with the addition of SBRT to nivolumab in patients with metastatic HNSCC in a randomized phase II trial [50]. Moreover, most clinical trials assessing the efficacy of combined radiotherapy and immunotherapy employ single-site irradiation or distant approaches. However, substantial preclinical and clinical evidence suggests that these methods yield suboptimal results. In response, Brooks et al. advocate for the exploration of comprehensive radiotherapy targeting multiple or all lesions to enhance the likelihood of achieving meaningful clinical outcomes, thereby avoiding the potentially premature abandonment of combined radiotherapy and immunotherapy due to failure with single-site approaches [51]. Clinically, neck dissection and elective nodal irradiation (ENI) are commonly employed strategies to reduce regional and distant metastasis. However, this approach may lead to a decrease in the number of antigen-specific cytotoxic immune cells within the tumor-draining lymph nodes. However, the essence of immunotherapy relies on the immune system. Therefore, some researchers believe that immunotherapy might be more effective if comprehensive ENI are avoided. To explore whether ENI compromises the efficacy of radiotherapy combined with immunotherapy, Marciscano and colleagues employed a preclinical model to compare stereotactic radiotherapy (Tumor RT) with or without ENI to investigate the immunologic differences between radiotherapy techniques that either spare or irradiate the draining lymph nodes (DLN). Their results demonstrated that the addition of elective nodal irradiation could attenuate chemokine expression, restrain immune infiltration, and adversely affected survival when combined with ICB, especially with anti-CTLA-4 therapy [52]. Nevertheless, whether nodal irradiation should be incorporated into the procedure of immunoradiotherapy remains understudied. In addition, Darragh et al. utilized a preclinical model to assess whether mice subjected to SBRT targeting only the primary tumor exhibit superior tumor control compared to those receiving ENI or neck dissection. Their experimental results revealed that, in the context of radiotherapy combined with immunotherapy, ENI leads to reduced efficacy by diminishing the systemic CD4 and CD8 effector T cells capable of controlling the tumor. Notably, they also found that mice eradicating both local and distant tumors following SBRT frequently experienced local recurrence only, but concurrent sentinel lymph node dissection or irradiation was sufficient to mitigate regional spread while preserving both local and systemic immune responses. Therefore, the authors advocate for the clinical reduction of ENI and comprehensive neck dissection [52].

4.
Combination of chemotherapy and immunotherapy
4.1.
Clinical trials of immunochemotherapy
For the most prevalent epithelial solid malignancies at metastatic stages, chemotherapy has traditionally been regarded as a complementary therapy to radical surgery, offering only modest prolongation of survival at best. Before the advent of immunotherapy, second-line treatment options for recurrent HNSCC included cetuximab, taxanes, and methotrexate, with response rates varying from 4% to 14% and a median progression-free survival (PFS) of only 2–3 months [53]. Currently, the most striking therapeutic effects have been observed with the combination of checkpoint inhibitors and chemotherapy. In the phase III CheckMate 141 trial, nivolumab significantly improved median overall survival versus chemotherapy (7.5 vs. 5.1 months) in platinum-refractory R/M HNSCC. This survival benefit was accompanied by a reduced incidence of grade 3–4 toxicities and maintained quality of life, compared to a decline in the chemotherapy group. Thus, treatment with nivolumab after chemotherapy leads to longer overall survival compared to chemotherapy alone in patients with recurrent HNSCC [6].
The activity of pembrolizumab was subsequently confirmed in the phase II KEYNOTE-040 trial, which randomized 495 patients to receive either pembrolizumab or standard-of-care chemotherapy. Cohen and colleagues found that median overall survival in the intention-to-treat population was 8.4 months with pembrolizumab, which was evidently longer than 6.9 months with standard of care chemotherapy for HNSCC patients. In fact, there was a confounding effect of subsequent ICIs in the standard of care arm, potentially decreasing the observed magnitude of benefit in overall survival in the pembrolizumab arm. Furthermore, fewer patients treated with pembrolizumab experienced grade 3 or worse treatment-related adverse events compared to those receiving standard-of-care chemotherapy. Importantly, all four complete responses and 30 of 32 partial responses in the pembrolizumab group occurred in patients with a PD-L1 combined positive score of 1 or higher, suggesting that PD-L1 expression could serve as an enrichment strategy in future trials of PD-1 blockade. The clinically meaningful prolongation of overall survival and favorable safety profile of pembrolizumab in recurrent and metastatic HNSCC patients support its further evaluation as part of combination therapy in earlier stages of the disease [7]. In the pivotal KEYNOTE-048 trial, pembrolizumab plus chemotherapy demonstrated superior progression-free survival and a higher objective response rate (36% vs 16%) compared to pembrolizumab monotherapy. Despite the early advantage of chemotherapy, the identical 2-year progression-free survival rate between arms confirmed that chemotherapy does not compromise the long-term durability of pembrolizumab. The combination exhibited a manageable safety profile, with grade 3–5 adverse events (71%) comparable to the EXTREME regimen (69%), supporting its establishment as a standard first-line treatment for R/M HNSCC[8].
Many important factors should be considered in treatment selection for patients with R/M HNSCC including prior exposure to systemic therapy, tumor burden and location, symptom burden, PD-L1 expression, and toxicity. Pembrolizumab plus chemotherapy should be preferred in cases where a rapid response is critical, such as for symptomatic diseases, or patients with bulky locoregional disease at risk of airway or bleeding complications. Conversely, for patients with only lung metastases and high PD-L1 expression, pembrolizumab monotherapy is highly attractive due to its similar efficacy and reduced toxicity.

4.2.
Underlying mechanisms of synergetic effect of immunochemotherapy
The potential mechanisms underlying the synergy between immunotherapy and chemotherapy include immunogenic tumor cell death, selective depletion of myeloid immunosuppressive cells, and lymphopenia, which reduces regulatory T cell populations and allows for the proliferation of effector T cells. Certain chemotherapeutic drugs, such as anthracycline, are identified to induce immunogenic cell death [54]. Chemotherapy drugs like carboplatin and paclitaxel were considered to reduce abnormally high numbers of immunosuppressive myeloid cells that could impair T cells in diverse ways in both mice and patients, which made the cancer vaccines more effective [55]. Furthermore, cyclophosphamide and doxorubicin have been reported to favor M1 macrophage differentiation, potentially mediating anti-tumor functions. However, chemotherapeutic agents can also downregulate growth factors (GM-CSF, G-CSF, and M-CSF), which may promote the differentiation of protumoral M2 macrophages [56]. Lymphotoxic chemotherapy, typically involving agents such as cyclophosphamide and fludarabine, might lead to the differentiation of cytotoxic T cells that target tumors, a phenomenon better observed in combination with T-cell therapy approaches. Additionally, certain chemotherapies like cyclophosphamide selectively target and kill specific FOXP3+ T cell subpopulations.
Notably, dosing and sequence are critical to immunomodulatory effects of cyclophosphamide. Medium-dose intermittent chemotherapy has been suggested as optimal. However, high-dose chemotherapy beyond a poorly understood threshold typically results in overall immune functional depression, probably due to the elimination of anticancer effector and the demise of stem-like T cells. Importantly, lymphopenia and neutropenia caused by chemotherapy may interfere with mechanism of checkpoint inhibitors by inhibiting the clonal expansion of effector lymphocytes. A report on locally advanced esophageal squamous cell carcinoma demonstrated that simply altering the sequence of chemotherapy and immunotherapy could dramatically improve treatment outcomes. Traditionally, chemotherapy and immunotherapy drugs are administered concomitantly. Immunotherapy can relieve T cell inhibition and restore T cell activity, but activated T cells are more vulnerable to being killed by chemotherapy drugs, which can significantly impair the synergistic effect. Conversely, administering antibodies after chemotherapy drugs metabolized in the body can minimize chemotherapy mediated damage to T cells. Subsequently, T cells activated by antibodies could play a prominent role in killing tumor cells. Based on these findings, whether the recommended sequence of immunotherapy combined with chemotherapy could be applied in HNSCC patients awaits more similar clinical trials (Figure 2).

5.
Combination of surgery and immunotherapy
5.1.
Neoadjuvant immunotherapy: promising combined setting
Currently, surgery remains the mainstay of treatment for HNSCC, primarily through the excision of the primary tumor and metastatic lymph nodes. However, the combination of surgery and immunotherapy is not yet well understood. Immunotherapy could be administered before (neoadjuvant) or after (adjuvant) surgery. The most common approach is to add immunotherapy following surgical resection, akin to postoperative chemotherapy or radiotherapy. Some clinical trials have showed that immunotherapy may be better tolerated than chemotherapy and recommended for patients with inability to tolerate chemotherapy post-surgery. Nevertheless, prognosis remains poor when recurrence and metastasis occur, even with the addition of post-surgical immunotherapy [6]. Recent research suggests that neoadjuvant immunotherapy may offer more substantial benefits. First, neoadjuvant immunotherapy could preserve the critical structures such as blood vessels and lymphatic vessels prior to surgery. Intact vascular channels ensure more reliable drug delivery, thereby achieving adequate drug concentrations necessary for tumor cell eradication. Similarly, intact lymphatics serve as conduits for newly activated T cells from lymph nodes to infiltrate the tumor [57]. Second, neoadjuvant immunotherapy is advantageous for downgrading the clinical stage before surgery. It is widely accepted that tumors at an advanced clinical stage generally result in poor prognoses. Therefore, the reduction in clinical stage achieved through neoadjuvant immunotherapy could indicate a more favorable clinical outcome. Additionally, the excision of large tumors in the maxillofacial region can severely impact the aesthetics and function of HNSCC patients. Neoadjuvant immunotherapy can significantly improve the quality of life and alleviate the secondary pain associated with surgery in these patients. Third, neoadjuvant immunotherapy can target micrometastases, such as circulating tumor cells (CTCs), before surgery. CTCs, as a critical factor for metastasis, can be detected even in early-stage patients. Neoadjuvant immunotherapy can effectively eliminate these CTCs, thereby preventing metastasis before surgery. Lastly, neoadjuvant immunotherapy exhibits a longer ‘tailing effect.’ Given that adjuvant therapies such as radiotherapy and chemotherapy typically commence one month (4–6 weeks) post-surgery, the tailing effect of neoadjuvant immunotherapy can bridge the therapeutic gap between surgery and subsequent adjuvant treatments. There is evidence that this tailing effect can persist for several years post-surgery in some responding patients, preventing recurrence (Figure 3). Moreover, neoadjuvant immunotherapy offers numerous opportunities for uncovering potential mechanisms and therapeutic principles. Exposing the tumor to immunotherapy drugs creates favorable conditions for identifying therapeutically effective biomarkers. Concurrently, advanced imaging technologies can be utilized to assess disease status in response to immunotherapy. Biopsy specimens obtained before immunotherapy, along with surgically excised specimens post-immunotherapy, can provide valuable insights into the key events during immunotherapy, including mechanisms of tumor remission, progression, and even hyperprogression. Therefore, further exploration of neoadjuvant immunotherapy in combination with other therapeutic modalities is warranted.

5.2.
Clinical and preclinical evidence about neoadjuvant immunotherapy
In many progressing tumors, immunoediting can lead to the loss of strong antigens and the growth of tumor cells only expressing subdominant antigens. However, neoadjuvant immunotherapy has the potential to reverse this trend. In a multicenter phase II trial, Uppaluri et al. evaluated the safety and feasibility of neoadjuvant pembrolizumab for patients with resectable locally advanced HNSCC. Their study revealed a one-year relapse rate of 16.7% among patients with high-risk pathology, significantly lower than the historical rate of 35%. They also observed that pathologic tumor response (PTR), the percentage of the overall tumor bed, in response to neoadjuvant pembrolizumab may serve as a predictive biomarker for a lower disease relapse rate. And PTR correlated with baseline tumor PD-L1 expression, immune infiltration, and IFN-γ pathway activity, but not with tumor mutational burden (TMB). Although several biomarkers were found to correlate with PTR, it was still unable to distinguish which biomarker was best to predict PTR [58]. Knochelmann and colleagues reported neoadjuvant nivolumab achieved a meaningful ORR of 33% in a phase-II trial of OSCC. They observed significant infiltration of CD4 and CD8+ T cells in both the best-responding and worst-progressing patients, indicating that the presence of immune cells was not the limiting factor for response. However, PD-L1 staining with tumor cells showed less overlap in the surgical tumor sample of the responding patients, while coinciding in progressing patients. Their study also addressed whether surgical margins should be planned based on the pre- or post-therapy tumor size, concluding that planning resection around the margins of the untreated tumor was the most prudent strategy [59]. L Ferris et al. also confirmed that neoadjuvant nivolumab could induce pathologic regressions in both HPV+ (23.5%) and HPV- (5.9%) tumors. And they suggested that continued postoperative therapy was warranted to seek a combinatorial neoadjuvant treatment regimen for enhancing the over the survival of HNSCC patients [60]. In addition, Friedman and colleagues provided mechanistic evidence supporting the superiority of neoadjuvant immunotherapy. Using oral cavity carcinoma models lacking immunodominant antigens, they demonstrated that pre-surgical immunotherapy could induce T cell responses against tumor cells with subdominant antigens in HNSCC [61]. Based on current evidence, neoadjuvant immunotherapy represents a promising preoperative strategy for resectable head and neck squamous cell carcinoma by inducing pathological tumor response, modulating the tumor immune microenvironment, and activating T-cell responses against subdominant antigens.

5.3.
Further exploration of neoadjuvant immunotherapy combined with other therapeutics
Certainly, clinicians have not limited their efforts to combining immunotherapy with surgery. In numerous preclinical models, the therapeutic efficacy of PD-1 inhibitors combined with CTLA-4 inhibitors has been shown to surpass that of either agent alone. Based on this, Ferrarotto et al. compared the efficacy of PD-1 inhibitors with PD-1 inhibitors combined with CTLA-4 inhibitors in HNSCC patients. Although they did not observe a higher density of CD8+ TILs in the combination group, 43% of patients exhibited a response, and 29% achieved a major pathologic response. The observed activity and safety in the trial suggested that neoadjuvant anti-PD-1 and anti-CTLA-4 could be used for tumor downgrading and a larger cohort should be required for further investigation [62]. Subsequently, Schoenfeld and colleagues demonstrated the feasibility of neoadjuvant nivolumab or nivolumab plus ipilimumab in untreated HNSCC (NCT02919683). They identified a significant correlation between CD4 T cells and pathological response in the neoadjuvant nivolumab plus ipilimumab group, providing a potential marker for identifying patients most likely to benefit from this combined therapy. Additionally, they observed a high rate of increased fluorodeoxyglucose uptake within cervical lymph nodes that were pathologically negative, suggesting that post-immunotherapy scans should not dictate adjustments in the planned surgical approach [63] Recently, Luoma et al. conducted a systematic kinetic analysis of single-cell RNA sequencing data on tumor-infiltrating and circulating immune cells from the aforementioned clinical trial. Their findings identified tissue-resident memory T cells (Trm) as the primary T cell population responsive to neoadjuvant immunotherapy. Pre-existing Trm cells were responsible for the early intra-tumoral response to immunotherapy, with subsequent activation of new T cells in tumor-draining lymph nodes, which then migrated into the tumor via blood vessels. They also reported that PD-1+ KLRG1- CD8 T cells exhibited strong association with response in both pre-and on treatment blood samples, suggesting that this marker could be effective for identifying patients who will benefit from neoadjuvant immunotherapy [64]. In conclusion, their discovery elucidated how neoadjuvant immunotherapy can enhance local and systemic tumor immunity, combat micrometastases, and prevent recurrence. Vos and colleagues demonstrated a MPR in 35% of patients receiving neoadjuvant nivolumab and ipilimumab, and in 17% of those receiving neoadjuvant nivolumab alone, in a non-randomized phase Ib/IIa trial (IMCISION). Notably, none of the patients with an MPR in either group experienced tumor relapse at a median of two years post-surgery. They further confirmed that an AID/APOBEC-associated mutational profile could serve as a potential biomarker to identify patients more likely to achieve MPR in the trial [65]. Additionally, their colleagues reported that 18 F-FDG-PET-based primary tumor volumetric metabolic response assessment might serve as an early biomarker to identify HNSCC patients with pathological responses to the regimens. A decrease in metabolic tumor volume (MTV) or total lesion glycolysis (TLG) would indicate a lower likelihood of relapse [66].
Multiple clinical trials combining neoadjuvant immunotherapy and chemotherapy or radiotherapy for the treatment of HNSCC are currently underway. Leidner et al. reported a phase Ib trial investigating the combination of neoadjuvant stereotactic body radiotherapy (SBRT) and nivolumab prior to surgery in patients with locoregionally advanced HNSCC. The safety profile of this combination was demonstrated, with a relatively modest rate of grade 3 toxicity compared to historical cohorts treated with conventional chemoradiation. This regimen resulted in a high rate of MPR and clinical and pathological downstaging in 90% of patients. Zhang et al. reported the first application of neoadjuvant immunotherapy (PD-1 inhibitor camrelizumab) combined with chemotherapy in patients with resectable stage III-IVB HNSCC. The study observed a favorable safety profile and promising outcomes, including an ORR of 96.7%, a clinical and pathological downstaging rate of 100%, and an MPR of 74.1%. Furthermore, the level of IL-6 in patients’ plasma was found to be associated with pathological response, suggesting its potential as a biomarker. However, it is important to note that this trial was a single-arm study with only 30 patients and a short follow-up duration of 16.1 months, indicating the need for larger-scale trials to further explore this combination therapy. Recent findings from a phase III clinical trial reported by Uppaluri and colleagues have demonstrated the efficacy of neoadjuvant pembrolizumab in locally advanced HNSCC patients. In this study, patients were randomly assigned 1:1 to receive either pembrolizumab plus standard care or standard care alone. The standard care regimen consisted of surgery followed by adjuvant radiotherapy with or without chemotherapy. The experimental arm received additional treatment with two cycles of neoadjuvant pembrolizumab followed by 15 cycles of adjuvant pembrolizumab (200 mg every 3 weeks). The results showed significantly improved 3-year event-free survival (EFS) with pembrolizumab versus standard care in both PD-L1 CPS ≥10 (59.8% vs 45.9%) and CPS ≥1 (57.6% vs 46.4%) populations. These findings establish that adding neoadjuvant and adjuvant pembrolizumab to standard care significantly enhances event-free survival in patients with locally advanced HNSCC [67].
For the past few decades, immunotherapy has predominantly been administered postoperatively for the treatment of metastatic and recurrent tumors. This approach has provided a critical lifeline for some HNSCC patients, saving lives in otherwise dire circumstances. However, taking the immunotherapy as the last bullet had obviously not made the best use of its hidden potential. Shrinking the tumor size effectively for some certain tumor populations and triggering a more durable T cell response due to existing tumor antigen make neoadjuvant immunotherapy more promising. Currently, administering immunotherapy prior to surgery can shrink the tumor, thereby minimizing surgical trauma and providing an opportunity for surgical excision in advanced HNSCC patients. In future, neoadjuvant immunotherapy should be considered a conventional therapeutic regimen for patients with HNSCC at any stage. What is in accompanying need of improvement is the scope of surgery in the neoadjuvant immunotherapy. While preoperative immunotherapy can reduce tumor volume, it remains debatable whether the extent of primary tumor resection should be determined based on tumor size before or after immunotherapy. This question is crucial for optimizing surgical outcomes and maximizing the benefits of neoadjuvant immunotherapy. In addition, we must acknowledge that the administration of neoadjuvant immunotherapy can also carry the potential risk of converting resectable HNSCC into an unresectable state. This risk arises from several key factors. Firstly, the efficacy of monotherapy immunotherapy is limited, with a considerable proportion of patients not responding to the treatment. If the malignancy is highly aggressive and progresses rapidly, the window for surgical resection may close, rendering the HNSCC unresectable. Secondly, the onset of action for immunotherapy is relatively slow, typically taking 1.5 to 2 months to assess its effectiveness, and sometimes even longer in certain patients. This delay may pose a risk of tumor advancing to an unresectable state during the waiting period. Additionally, some patients may experience hyperprogression, where the tumor grows more rapidly following immunotherapy, further increasing the risk of transitioning from a resectable to an unresectable state. Therefore, it is imperative to closely monitor tumor dynamics in patients to prevent HNSCC from progressing from a resectable to an unresectable state. This vigilance will help ensure timely surgical intervention and improve patient outcomes. Notably, when the clinical or imaging evidence exhibits lymph node metastasis, the conventional neck dissection is performed to excise both positive and negative lymph nodes for HNSCC patients. However, negative lymph nodes in which T cell response is extensively triggered have been demonstrated to play a vital role in preventing recurrence. The question of whether to preserve all negative lymph nodes or just some regional lymph nodes remain open to discussion (Figure 3). Clearly, the scope of surgery should not rely solely on the clinician’s surgical experience. More preclinical and clinical evidence is needed to address this issue and establish a series of standards to advance neoadjuvant immunotherapy. Optimizing treatment for HNSCC could provoke a more extensive and enduring anti-tumor response in vivo, thereby enhancing overall survival.

6.
Conclusion
As the standard treatment strategy, combined and sequential therapy including surgery, radiotherapy, and chemotherapy has not significantly improved overall rate of HNSCC patients over the past few decades. Immunotherapy is now rapidly changing the treatment landscape of HNSCC. Regrettably, a single immune checkpoint inhibitor is still ineffective for a large proportion of HNSCC patients. The synergistic effect of combined immunotherapy provides a solid theoretical foundation for improving the survival of HNSCC patients. Regarding combination immunotherapy, the administration of multiple immune checkpoint inhibitors, while capable of enhancing antitumor responses, may induce compensatory upregulation of alternative immune checkpoints, resulting in transient therapeutic efficacy. In the setting of radiotherapy combination, proton therapy exhibits superior immune-protective properties relative to conventional photon-based approaches. The potential synergy between proton therapy and immunotherapy may elicit a more robust abscopal effect, thereby improving therapeutic outcomes. Notably, the neoadjuvant immunotherapy domain has witnessed substantial advances, with recent clinical evidence demonstrating that the incorporation of neoadjuvant immunotherapy into standard treatment regimens significantly improves event-free survival in locally advanced HNSCC patients, representing a notable evolution in treatment paradigms. However, it must be acknowledged that combination immunotherapy also presents several drawbacks and challenges. Firstly, combination immunotherapy may result in more severe side effects, such as anemia, hair loss, and fatigue. Secondl, as combination immunotherapy involves multiple therapies, the overall treatment cost is relatively high, potentially imposing a financial burden on patients and their families. Additionally, the complexity of the treatment regimen increases, and patients’ responses to these regimens vary, necessitating the development of personalized treatment plans for each patient. Therefore, while combination immunotherapy can enhance treatment efficacy, these challenges must be carefully managed to optimize patient outcomes and minimize the adverse impacts on patients’ health and finances.