Introduction

Head and neck squamous cell carcinoma (HNSCC) is a highly fatal malignancy arising from the mucosal epithelium of the tongue, mouth, nasopharynx, larynx, and throat, with an annual mortality rate of 40–50% [1, 2]. The disease is associated with significant clinical challenges, including a high incidence of distant metastases (10–30%) and tumor recurrence (30%-50%) [3]. HNSCC pathogenesis is driven by various factors such as tobacco exposure, betel nut consumption, alcohol consumption, consumption of spicy food, dental trauma, sunlight exposure, chronic inflammation, Human Papillomavirus infection, somatic genetic mutations, genetic predisposition, and alterations in the microbiome [1, 4].

Despite substantial technological breakthroughs in HNSCC therapy, the mortality rate remains high [5]. HNSCC management has traditionally relied on surgery, radiation, and systemic chemotherapy, either as monotherapy or in combination [6]. Concurrent chemo/radiotherapy, particularly with cisplatin, is a promising therapeutic option, significantly improving survival outcomes for patients with inoperable tumors [7]. However, the genetic complexity of HNSCC plays a crucial role in dictating patient outcomes, with loco-regional failure representing a significant challenge. The use of cisplatin as a radio-sensitizer has been associated with significant systemic toxicity, limiting its applicability in immunocompromised or frail HNSCC patients [5]. Therefore, there is a growing interest in developing innovative therapeutic strategies that maximize disease control while minimizing treatment-related morbidity.

Over the last decade, the field of cancer immunotherapy has rapidly advanced and has been successfully used to treat various malignancies such as melanoma, breast, colorectal, and lung cancer [8]. The use of immunotherapy in the treatment of HNSCC has also revolutionized the field by improving survival rates and decreasing side effects associated with traditional therapies [8]. With the discovery of immunotherapeutic modalities such as oncolytic viruses, monoclonal antibodies, CAR-T cells, and therapeutic vaccines, the HNSCC treatment landscape is rapidly changing and expanding [9]. These innovative approaches hold tremendous potential for improving patient outcomes and quality of life.

Anti-PD-1 checkpoint inhibitors (CPI) have shown great promise in the treatment of recurrent or metastatic HNSCC [10]. Nivolumab, as a PD-1 CPI, is a commonly used FDA-approved treatment for patients with platinum-restricted recurrent or metastatic HNSCC. However, the success rate of this therapy is still modest, with only around 15% of patients responding to PD-1/PD-L1 inhibitors in previous trials [11]. While most studies on CPI resistance have focused on the immune microenvironment and the deactivation and exhaustion of T and B cells [12], less attention has been given to tumor cell-intrinsic mechanisms of immunotherapy resistance in HNSCC. Thus, to identify drivers of response and resistance to CPI, studying the tumor microenvironment (TME) is crucial.

The TME is a complex and dynamic ecosystem that includes various immune cells that migrate to the tumor site. Among these immune cells, T regulatory cells (T_regs) play a crucial role in promoting tumor growth by regulating immune system hyper-activation and maintaining tolerance [13]. T_regs can suppress anti-cancer responses and promote angiogenesis at the tumor site [14]. However, the prognostic effect of T_regs is controversial and varies according to the type of cancer, tumor stage, and treatment [15]. In some cancers, such as melanoma, hepatocellular, cervical, renal, and breast cancer, FOXP3⁺ T_regs infiltration has been associated with decreased overall and disease-free survival [16]. However, the effect of Tregs on HNSCC is still controversial. Some studies have suggested that increased FOXP3⁺ T_regs infiltration leads to improved survival [17–19], while others have reported that it causes decreased survival [20–22]. Therefore, further studies are required to elucidate the role of T_regs in HNSCC.

The development of single-cell RNA sequencing (scRNA-seq) has provided a comprehensive examination of the transcriptome profiles of specific cell populations, which has significantly assisted the study of the TME [23, 24]. This approach has also been utilized to investigate the TME of HNSCC, and the results have given researchers a deeper understanding of how cells interact and change at the tumor site [23, 25]. The aim of this study is to investigate the effect of Nivolumab on the function and cellular communication of T_regs using scRNA-seq data analysis. By studying the effect of Nivolumab on T_regs, we hope to gain a better understanding of the mechanisms underlying the response and resistance to anti-PD-1 CPI in HNSCC. This study has the potential to contribute to the development of more effective immunotherapeutic strategies for the treatment of HNSCC.

Materials and methods

The study overall design and flow process are presented in Fig 1.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Schematic visualization of the study.

Figure made by Biorender. Reprinted from [86] under a CC BY license, with permission from Elsevier, original copyright 2023.

https://doi.org/10.1371/journal.pone.0295863.g001

Results

HNSCC TME cell fractions

We reanalyzed a published scRNA-seq dataset from eight HNSCC tissues in order to comprehend the heterogeneity in the patient response to CPI therapy. After removing cells that failed Quality Control (QC), a total of 53,730 cells remained for downstream analysis (pre-Nivolumab therapy: n = 27223, post-Nivolumab therapy: n = 26507) (Fig 2A and 2B). Cells were represented as twelve different clusters using uniform manifold approximation and projection (UMAP) and unsupervised graph-based clustering (Fig 2C).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. Characterization of T_regs in HNSCC TME in pre and post Nivolumab therapy.

A) UMAP visualization of cells in two statuses; B) UMAP visualization of cells according to their originated samples; C) UMAP visualization of HNSCC TME clusters based on Leiden clustering; D) UMAP visualization of T_regs markers and related expression pattern of them in TCGA tumor bulk dataset; The box plots display the median mRNA expression levels of Treg markers in HNSCC tumors (orange) compared to normal tissues (red). The values on the axes are presented in Log2 (TPM+1), where TPM (Transcripts Per Million) quantifies gene expression while considering transcript length and sequencing depth. Log2 transformation makes the data more interpretable and robust for visual comparisons. The "+1" avoids issues with zero TPM values, ensuring all values are positive. The * red indicates a p-value less than 0.01, implying statistical significance. The abbreviations T and N denote tumor and normal tissues, respectively. E) Histogram representation of T_regs cell count in pre (blue) and post (orange) Nivolumab therapy; paired t-test with p-value < 0.05, Error bars indicate standard deviation; F) Histogram representation of T_regs cell number in different donors in pre (blue) and post (orange) Nivolumab therapy; G) UMAP visualization of T_regs in two different statues.

https://doi.org/10.1371/journal.pone.0295863.g002

Based on the expression of canonical gene markers including: CD3D, FOXP3, TIGIT, and FANK1, cluster four was identified as the T_reg population (Fig 2D). Interestingly, examination of the The Cancer Genome Atlas (TCGA) cohort indicated that FOXP3 and TIGIT were significantly up-regulated in HNSCC samples compared to healthy (Fig 2E).

We found that tumor cells showed a notable increase in T_reg cluster compared to pretreatment tumors (Fig 2F and 2G).

DEG analysis of T_regs

The gene expression level in T_regs before and after treatment with Nivolumab was determined by differentially expressed gene (DEG) analysis (Fig 3A). Pseudogenes, mitochondrial encoded genes, and ribosomal genes were eliminated, and 4089 Differentially expressed genes (DEGs) (adjusted p-value <0.05), including 284 up-regulated genes and 3805 down-regulated genes, were identified (Fig 3A).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 3. DEGs, enrichment, and cell cycle analysis of T_regs in HNSCC TME.

A) Volcano plot of the Treg’s DEGs after Nivolumab therapy. Genes regulated with a fold change >|1| and Adj. p.value < 0.05 are highlighted in red showing the indicated fold changes derived from t-test statistic; B) GO Biological Processes of Treg’s DEGs using ORA. The figure contains bar charts showing the results of the enrichment analysis of GO developed using Enrichr. For each term, the x-axis indicates the -log10 (p.value); C) GSEA analysis of T_regs DEGs. GSEA analysis using hallmark gene sets from the molecular signature database for the transcriptional difference between pre and post Nivolumab therapy. NES = normalized enrichment score; D) Bar plot of the percentage of T_regs in different phases (G1: Growth 1 phase, S: DNA Synthesis phase, G2M: Checkpoint, Mitosis phase) and statues.

https://doi.org/10.1371/journal.pone.0295863.g003

Based on previous studies [35], heterogeneity within T_reg cells has been characterized by a bimodal distribution of TNFRSF9, a known marker for T_reg activation. To investigate whether this heterogeneity is associated with response to treatment, we further performed DEG analysis of TME cells. Compared to pre-treatment, we observed that tumor cells in TME of post-treatment samples consistently expressed TNFRSF9 at higher level. Also, the result of DEG analysis of T_reg cells between two statuses indicated that genes associated with immunosuppressive functions including TNFRSF4, ENTPD1, REL and LAYN were downregulated in post treatment T_regs. Notabely, we discovered numerous immediate early genes among T_reg DEGs, including NR4A2, DUSP1, FOSB, FOS, JUN, and JUNB. These genes are quickly activated in response to stimuli, with no or little nascent protein synthesis [36].

Biological Process (BP) analysis of T_regs in post treatment centered around the neutrophil degranulation, neutrophil immunity, neutrophil activation, and the cytokine-mediated signaling pathway (Fig 3B). Furthermore, (Molecular Signatures Database) MSigDB analysis showed the negative regulation of KRAS signaling, Notch signaling, bile acid metabolism, coagulation, and angiogenesis in T_regs after Nivolumab therapy (Fig 3C).

We found no significant increase in the Treg cluster’s S.Score (module score of genes associated with the S phase of the cell cycle) or G2M.Score (module score of genes associated with the G2M phase of the cell cycle) after treatment with Nivolumab, indicating that this unique transcriptional profile was not due to active cell cycling (Fig 3D).

Pseudo-time trajectory analysis of T_reg dynamic changes

Given the heterogeneity of the TME, we performed a pseudo-temporal reconstruction using scFates to determine the lineage structures and pseudo-temporal variables of the HNSCC TME. There were eight unique nodes found based on transcriptional alterations in pseudo-time trajectory analysis, which node number five mapped on T_reg cluster. The top 10 up-regulated genes associated with this node included DUSP4, TRBC2, CD7, GZMA, CD2, and TRAC. Also, CCL5 as an important chemokine was among these specific genes that indicating stimulation of T_reg cells presumably by interferons (Fig 4A). Furthermore, enhanced production of CD3E/D, may reveal epigenetic change in the T_reg cells. IL-32 was another gene with strong expression in the T_reg cluster. Thanks to mRNA alternative splicing, this cytokine has nine distinct isoforms and is now recognized as a key pro-inflammatory factor.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 4. Pseudo-temporal analysis of HNSCC TME.

A) Top ten up-regulated genes of T_regs during the pseudo-time; UMAP presentation illustrating the dynamics of the HNSCC TME and its transition states as they progress through pseudo-time. The lineage tracking initiates at the 0 branch and concludes at the 1 branch. The heatmap depicts genes that experience upregulation within the Treg branch as pseudo-time advances. The highlighted genes comprise the top ten upregulated genes specific to the Treg branch compared to other branches during pseudo-time analysis. B) Venny plot of the top ten up-regulated genes of T_regs and T_regs DEGs; C) The expression pattern of four similar genes between the top ten up-regulated genes of T_regs and T_regs DEGs along the pseudo-time.

https://doi.org/10.1371/journal.pone.0295863.g004

When the list of top 10 up-regulated genes associated to T_regs node were compared with the T_reg DEG list between two statuses, we found that four genes including GZMA, CD2, IL-32, and TRAC were also dys-regulated significantly in T_regs before and after treatment (Fig 4B). In this regard, post-treatment T_regs demonstrated a markedly decrease in IL-32, CD2, and TRAC expression while GZMA was up-regulated in response to Nivolumab (Fig 4B).

To determine if these alteration in expression of GZMA, CD2, IL-32, and TRAC is a distinct feature of T_reg cells, we further assessed the expression of these four genes in all clusters across the pseudo-time. Although these genes were expressed slightly in certain clusters, the T_reg cluster exhibited higher levels of expression than the other clusters (Fig 4C).

IIC expression pattern in T_regs

Considering the importance of IICs in suppressing immune responses, we compared the expression of 14 common IICs in T_regs before and after Nivolumab therapy (Fig 5A). Among the IICs examined, only five were significantly altered between two statuses (Fig 5B). TIGIT, ENTPD1, and CD276 and LY9 levels were significantly lower in T_regs after therapy, while LAG-3 showed an increased expression level (Fig 5C). However, the expression of PDCD1 (PD-1), as the target of Nivolumab, did not significantly dys-regulate before and after the treatment (Fig 5C).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 5. Characterization of expression pattern of IICs.

A) Dot plot of the expression pattern of IICs in T_regs before and after Nivolumab therapy; B) Venny plot of the significant IICs in T_regs; C) The log₂FC and Adj P.val of IICs in T_regs between two statuses. D) The expression pattern of significant IICs along the pseudo-time; Yellow branches shows highest expression of the gene across the pseudo-time. E) The expression pattern of significant IICs in TCGA tumor bulk dataset, The box plots display the median mRNA expression levels of T_reg markers in HNSCC tumors (represented by the orange plots) and the corresponding normal tissues (represented by the red plot). The values on the axes are represented in units of Log₂ (TPM+1). The * red indicates a p-value less than 0.01, implying statistical significance. The abbreviations T and N denote tumor and normal tissues, respectively.

https://doi.org/10.1371/journal.pone.0295863.g005

UMAP analysis indicated that among the five mentioned IICs, TIGIT and LAG-3 have a higher expression level in T_regs than other clusters (Fig 5D). Nevertheless, according to the UMAP results, LAG-3 expression has increased and TIGIT expression has decreased in T_regs following treatment with Nivolumab. Based on TCGA HNSCC tumor bulk dataset, patients with HNSCC had higher TIGIT, LAG3, and CD276 expression levels than healthy individuals (Fig 5E).

The cell-cell interaction of T_regs

We next identified ligand-receptor pairs between other cells and T_regs before and after Nivolumab treatment using Squidpy algorithm. This python-based tool includes a database of ligand-receptor interactions as well as a statistical model for identifying relationships between two cell types that are enriched from single-cell transcriptomics data. Results of the analysis demonstrated a dense communication network among cluster two and T_reg cells both before and after Nivolumab therapy (Fig 6A). The MARCO, CD14, FCGR3A, and CD163 genes, which are unique to tumor-associated macrophages (TAMs), were expressed in cluster two (Fig 6B). Among the interactions, five pairs SPP1/CD44, HLA-E/KLRC2, HLA-E/KLRK1, ANXA1/FPR3, and CXCL9/FCGR2A had shown notable changes after the therapy (Fig 6C). We identified hallmark associations for these ligands and receptors using MSigDB, focusing on the top 5 pairs (10 genes). The analysis was conducted using ORA and revealed connections to various cellular pathways, including those related to inflammation response, apoptosis, glycolysis, angiogenesis, and signaling pathways such as IL-6/JAK/STAT3 signaling, IL2/STAT5 signaling, and TNFα signaling via NF-κB (Fig 6D).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 6. Cell-cell interaction of T_regs in HNSCC TME.

A) Bar plot of the interaction scores of T_regs in Pre and Post treatment. Interaction score: sum of all expression of all ligand and receptors which were involved between Treg and others; B) UMAP visualization of TAMs markers; C) Ligand/Receptors which are significant between T_regs and TAMs. Interaction score: Top highly expressed interactions between Treg and TAMs; D) MSigDB analysis of the ligand/receptors pathways. The figure contains bar charts showing the results of the enrichment analysis of MSigDB developed using Enrichr. For each term, the x-axis indicates the -log10 (p.value).

https://doi.org/10.1371/journal.pone.0295863.g006

Discussion

Not all T cells participate in the development of anti-cancer responses. The TME’s development is significantly influenced by a specific subset of T cells that express CD4, CD25, and FOXP3 markers and is known as the "T_reg" subset [37]. These cells, which in normal circumstances assist in preventing the development of chronic inflammation and autoimmune diseases, promote the growth and development of tumor cells [38]. T_regs, in addition to inhibiting CD4⁺ and CD8⁺ T cells from becoming activated, enhance angiogenesis to improve the delivery of oxygen and nutrients to the tumor site [39]. As a result, numerous studies have indicated that targeting T_reg immunosuppressive mechanisms or eliminating these cells is an essential cancer treatment target [40–42].

T_regs infiltrating TME, on the other hand, had inconclusive results in some cancers, such as HNSCC [43]. Some studies have shown the poor prognosis of HNSCC patients with a high FOXP3⁺ T_reg cell infiltration [20, 21]. In contrast, others have highlighted the correlation between the high recruitment of these cells and better overall survival and local control [18, 44].

In this study, we re-analyzed a scRNA-seq data of HNSCC patients treated with Nivolumab to examine the cellular and molecular dynamics of T_regs in the TME.

Previous studies have already established several markers for T_reg cells such as CD3D, FOXP3, FANK1, and TIGIT using single cell sequencing [29, 45], which we also employed in our study to accurately validate T_reg cell annotation. In addition, we discovered that FOXP3 and TIGIT are expressed at higher levels in HNSCC patients by analyzing the TCGA tumor bulk dataset (Fig 2D). TIGIT⁺ T_regs are highly immunosuppressive, persistent, and concentrated in cancers, according to Julien et al [46]. Additionally, FOXP3⁺ T_regs are capable of inhibiting the proliferation of autologous CD4⁺ CD25^- T cells [47].

Our results demonstrate that both collectively and individually, T_reg numbers increased following treatment with Nivolumab (Fig 2E and 2F). According to a comparison of the cell cycle of T_regs before and after treatment, evidence showed that T_reg proliferation increased dramatically following Nivolumab treatments (Fig 3D). The increase of T_regs after treatment with anti-PD-1 has also been confirmed in the study of Kamada et al [48]. In addition, Xiong et al. showed that Nivolumab treatment increases the number of FOXP3⁺ CD4⁺ T-cells in peripheral blood [49].

There are some crumbs of evidence to support the cross-talk between neutrophils and T_regs. According to studies, T_regs facilitate neutrophil migration into the TME by generating CXCL8 [50]. Furthermore, T_reg-secreted IL-10 and TGF-β can induce neutrophil polarization and conversion to tumor-associated neutrophils (TANs) [51]. T_reg also suppress the phagocytosis ability of neutrophils [52]. Following the discovery by Eruslanov et al. that TANs are not immunosuppressive in the early stages of cancer but rather stimulate T cell responses, the significance of neutrophil-T_reg cross-talk in TME doubles [53]. Our results showed that the mechanisms of T_reg involved in the regulation of neutrophil activity, such as degranulation and mediating immunity, had changed after Nivolumab therapy (Fig 3B).

According to GSEA analysis, KRAS signaling, Notch signaling, and angiogenesis activity were all reduced in T_regs following Nivolumab treatment (Fig 3C). Abundant molecules in the KRAS signaling pathway participate in intracellular signaling and regulate cell differentiation, proliferation, growth, and apoptosis [54]. Uncontrolled proliferation and an elevated risk of malignancy can result from mutations in any of the genes responsible for producing these molecules [55]. Numerous studies have also shown that RAS signaling pathway alterations impact T cell function [56–58]. Mor et al. found that inhibiting Ras increases the level of FOXP3 in pre-existing T_regs and improves the conversion of CD4⁺ CD25^- T cells to CD4⁺ CD25⁺ FOXP3⁺ T cells [59]. Additionally, this work demonstrated that RAS inhibition improves T_regs’ immunosuppressive function in a FOXP3-dependent manner, which allowed RAS-deficient T_regs to prevent the development of diabetes in a mouse model of the disease by up to 70% [59]. Our research showed that Nivolumab therapy reduced KRAS signaling in T_regs, which could cause an increase in the expression of FOXP3 and enhance the immunosuppressive ability of T_regs.

The Notch signaling pathway is one of the most evolutionarily conserved pathways and is activated only by cell-to-cell communication [60]. Rong et al. demonstrated that the suppressive activity of T_regs is adversely affected by the activation of the Notch signaling pathway [61]. Activation of the Notch signaling pathway, in other words, causes T_regs to express less TGF-β, IL-27b (a component of IL-35), and PD-1 [61]. T_regs treated with Notch ligands produced more IFN-γ and were less capable of inhibiting T-effector proliferation and preventing the release of TNF-α and IFN-γ [61]. Our results demonstrated that Nivolumab treatment decreased the Notch signaling pathway in T_regs, which may have increased their immunosuppressive ability.

Accurate study of diverse gene expression in scRNA-seq utilizing pseudotime can aid in better understanding the specificity of any therapy, leading to the identification of novel genes as critical player targets. Accordingly, we found that the expression levels of four genes GZMA, CD2, IL32, and TRAC were altered in T_regs after Nivolumab treatment (Fig 4B). The expression level of GZMA increased while the expression of other genes decreased. It has long been known that tumor and virus-infected cells can be killed by (Natural Killers) NK and cytotoxic CD8⁺ T cells using the granzyme/perforin mechanism [62]. Today, it is evident that this mechanism is not just restricted to NK and TCD8+, and T_regs also employ this to induce cell cytolysis in TCD8⁺, NK, and B effector cells in TME [63–65]. According to the study of Grossman et al., CD4⁺CD25⁺ natural T_regs express more GZMA and less GZMB, whereas adaptive T_regs express these molecules differently [66]. Additionally, they demonstrated that both T_reg subtypes had a strong capacity for killing DC, T cells, and CD14⁺ monocytes to suppress immune responses [66]. Our results also indicated that GZMA expression levels in T_regs had increased following treatment, which may be a sign of their increased capacity for cell cytolysis of TME cells.

IL-32 is a pro-inflammatory cytokine produced by different immune cells, including NK cells, T cells, and monocytes [67]. This cytokine has been shown to play a role in the induction of other pro-inflammatory cytokines like IL-6, IL-8, TNF-α, and macrophage inflammatory protein-2 (MIP-2) [67]. The decrease in the expression of IL-32 in T_regs after treatment can be an inhibitory mechanism of this cell to prevent the formation of inflammation in TME.

The importance of immunosuppressive TME and dysfunctional expression of IICs in reducing anti-tumoral immune responses has been underlined by increasing evidence [11]. Based on our findings, five of the IICs had significant expression changes when the IIC panel in T_regs was compared before and after treatment (Fig 5C). According to these findings, PD-1⁺ T_regs significantly increased LAG-3 expression after Nivolumab treatment, while TIGIT, ENTPD1, CD276, and LY9 expression decreased significantly. LAG-3 is a CD4-dependent molecule [68]. This molecule is produced on the surface of T_regs after activation, binds to MHC-II, and inhibits DCs’ and effector T cells’ functions [68]. In addition, it has been demonstrated that LAG-3⁺ T_regs can control humoral immune responses [69]. The production of high amounts of IL-10 and TGF-β by LAG-3⁺ T_regs inhibits the function of T_fh, prevents the formation of the germinal center (GC), and finally suppresses the production of antibodies [70]. The increase of LAG-3 expression in T_reg after Nivolumab injection can increase their ability to suppress cellular and humoral immune responses. It should be emphasized that after the treatment, CTLA-4 and PDCD-1 expression levels also increased; however, this increase was not significant.

TIGIT is one of the most important IICs on the surface of T_regs, capable of suppressing effector T cells, decreasing NK cytolysis activity, and inhibiting antibody formation [71]. Preclinical research suggests that blocking the PD-1/PD-L1 pathway and TIGIT simultaneously promotes tumor rejection and has promise for patients with solid tumors [71]. Our findings demonstrate that anti-PD-1 (Nivolumab) therapy alone also reduces TIGIT expression in T_regs.

Understanding T_regs immunosuppressive actions is an appealing therapeutic option for promoting anti-tumor immune responses. However, the distinct cell-cell interaction patterns, notify for further investigation. In this case, comparing the cell-cell contact before and after Nivolumab administration showed a reduction in T_reg communication with most clusters after treatment (Fig 6A). Additionally, the most significant ligand-receptor pairs between T_regs and TAMs were marked in Fig 5C. After receiving Nivolumab, the interaction of five pairs of receptors and ligands (SPP1/CD44, HLA-E/KLRC2, HLA-E/KLRK1, ANXA1/FPR3, and CXCL9/FCGR2A) significantly decreased (Fig 5C). According to Cheng et al., the CD44-SSP1 axis is a control mechanism that prevents the proliferation of effector T cells in the TME [72]. Additionally, they demonstrated that once T_regs were presented in the TME, the level of CD44 expression in those cells was reduced, increasing the ability of T_regs to proliferate [72]. Additionally, they demonstrated that once T_regs were presented in the TME, the level of CD44 expression in those cells was reduced, and their proliferation ability increased [72]. Our study revealed that Nivolumab therapy reduces CD44-SSP1 interaction, which may have increased T_reg proliferation potential.

KLRC2 (NKG2C) and KLRK1 (NKG2D) are activating receptors mainly expressed on the surface of NK cells and T cells [73]. By binding these receptors to their ligands, these cells become more capable of cytotoxicity [73]. Though Protein Atlas has confirmed a small expression of KLRK1 on the surface of non-classical monocytes, we did not identify any studies that specifically addressed these two receptors on the surface of TAMs. Nevertheless, our findings support a decreased NKG2D/C-HLA-E interaction between TAMs and T_regs after treatment. Yang et al. showed that NKG2D⁺CD4⁺ T cells kill T_regs following the binding of NKG2D to its ligand [74]. Although NKG2D/C-dependent cytolysis in macrophages has yet to be confirmed, we do know that these cells can directly kill other cells after stimulation by releasing lethal mediators such as TNF-α, reactive nitrogen species, and reactive oxygen species (ROS) [75]. The reduction of NKG2D/C-HLA-E interaction after Nivolumab injection could be the mechanism by which T_regs escape death, although this data needs further study. Since the function of FPR3 has not yet been determined [76], more experiments are essential to justify the effect of the reduction of the FPR3-ANXA1 axis on T_reg behavior after treatment with Nivolumab.

Furthermore, the reduction in T_reg-TAM interaction following Nivolumab therapy has influenced several mechanisms, including the inflammatory response, apoptosis, IL-6/JAK/STAT3 signaling, IL2/STAT5 signaling, and TNF-α signaling via NF-κB, though our data do not show the nature of these changes.

Our results showed that the increase in the number of T_regs after Nivolumab treatment could be due to the reduction of the CD44-SSP1 axis, KRAS signaling pathway, and NKG2C/D-HLA-E axis, which respectively lead to increased T_regs proliferation, FOXP3 expression, T_regs differentiation, and the prevention of destruction by other cells (Fig 1).

The decrease in KRAS and Notch signaling pathway activity, which increases the expression of FOXP3, PD-1, and IL-35, can also contribute to the increased suppressive capacity of T_regs. These cells exhibit elevated CTLA-4 and LAG-3 expression on their surface in addition to PD-1 and TIGIT expression, which inhibit the activity of T effectors and dendritic cells (DCs), prevent the development of GC, and inhibit the production of antibodies. Cellular and humoral immune systems can be suppressed more effectively by these cells. T_regs produce more anti-inflammatory cytokines such as IL-10, TGF-β, and IL-35 while producing less pro-inflammatory cytokine IL-32. The potential of these cells to destroy other cells has increased along with GZMA synthesis. Finally, different biological processes, including those related to neutrophil degranulation and activation, changed in the T_regs of HNSCC patient’s post-Nivolumab therapy. Although it is unknown from our data what kind of modifications these are (Fig 1).

Although reducing the immunosuppressive mechanisms and restoring the antitumor power of immune cells in the TME are of the IIC therapy goals, our study provided completely different results. T_regs, the most effective cells at inducing TME, are regarded as one of the most important targets of IIC therapy. However, our results did not only not show a decrease in the number and suppressive function of T_regs, but all the evidence indicated an increase in the number and intensification of the suppressive function of T_regs.

Finally, by studying other articles, the probability of hyper progressive disease (HPD) in these four HNSCC patients after being treated with Nivolumab was strengthened. A subset of cancer patients receiving CPI experience enhanced tumor cell proliferation, rapid disease progression, and a poor prognosis [77–84]. This condition known as HPD is seen in 4–29% of cancer patients [85]. Here are still no specific criteria to predict the probability of HPD occurrence in patients. In addition, the mechanisms involved in HPD remain unclear. However, some potential HPD mechanisms have been proposed, including T_regs proliferation, up-regulation of CTLA-4 and LAG-3, an increase in ILC-3, etc. [48]. Further studies of these mechanisms can probably help clarify the limitations of IIC therapy success in HNSCC patients and identify effective therapeutic targets for HPD patients.

Conclusion

In conclusion, this study investigated the dynamic changes in the TME of HNSCC patients treated with Nivolumab, with a particular focus on T_regs. DEG analysis of T_regs revealed downregulation of genes associated with immunosuppressive functions, and upregulation of immediate early genes. Biological process and MSigDB analysis showed negative regulation of KRAS signaling, Notch signaling, and angiogenesis in T_regs after Nivolumab therapy. Additionally, the study found altered expression of certain IICs in T_regs before and after treatment. These findings provide new insights into the response to Nivolumab therapy in HNSCC patients, and suggest that targeting T_regs and IICs could potentially enhance the efficacy of immunotherapy.

Supporting information