Mendelian randomization analysis reveals potential association between allergic rhinitis and nasopharyngeal carcinoma

Abstract

Background

Nasopharyngeal carcinoma (NPC) is characterized by complex interactions within its tumor microenvironment. Understanding the immune landscape and gene expression patterns is crucial for developing effective therapeutic strategies.

Methods

We employed multiple analytical approaches including Mendelian randomization analysis, single-cell sequencing, gene expression profiling, and spatiotemporal analysis. The study investigated associations between NPC and comorbidities, characterized immune cell populations, and analyzed gene expression patterns. Cytokine profiles and their effects on disease risk were also examined.

Results

The analysis revealed potential associations between NPC and both allergic rhinitis and high myopia. Single-cell sequencing identified distinct cellular populations, including three unique B cell subpopulations (M1, M2, M3) with specific molecular signatures and spatial distributions. Temporal analysis showed dynamic changes in immune cell composition: endothelial and epithelial cells dominated the early phase, B cells peaked in the middle phase, and dendritic and T cells increased in the late phase. Gene expression clustered into four main patterns, with significant roles for immune-related genes, particularly the TNFRSF family and HLA-related genes. Cytokine analysis identified IL-6 as a significant risk factor (31.4% increased risk) while IL-10 showed protective effects (8% risk reduction).

Conclusions

This comprehensive analysis provides detailed insights into the complex immune microenvironment and molecular mechanisms of NPC. The identification of distinct cellular populations, temporal patterns, and key molecular players offers valuable information for understanding disease progression and developing targeted therapeutic strategies.

Introduction

Nasopharyngeal carcinoma (NPC) is a complex malignancy characterized by distinct regional prevalence and unique pathological features. Understanding its underlying molecular mechanisms and immune microenvironment is crucial for developing effective therapeutic strategies. Recent advances in molecular and cellular analysis techniques have provided unprecedented opportunities to explore the intricate relationships between various components within the tumor microenvironment [1–4].

The immune system plays a pivotal role in NPC development and progression. Previous studies have suggested potential associations between NPC and immune-related conditions, such as allergic rhinitis, highlighting the importance of immune regulation in disease pathogenesis. The tumor microenvironment of NPC contains diverse immune cell populations, including T cells, B cells, macrophages, and dendritic cells, each contributing to the complex cellular interactions that influence disease outcomes [5–8].

Gene expression patterns in NPC demonstrate significant spatiotemporal dynamics, reflecting the evolving nature of the disease. The expression of immune-related genes, particularly those involved in immune regulation and cellular interactions, shows distinct patterns across different disease stages. Understanding these patterns and their implications is essential for identifying potential therapeutic targets and developing more effective treatment strategies [9–12].

Recent technological advances, particularly in single-cell sequencing and spatial transcriptomics, have enabled more detailed characterization of the cellular and molecular landscapes in NPC. These approaches have revealed previously unknown cellular subpopulations and their specific roles in disease progression. Additionally, the analysis of cytokine profiles and their effects on disease risk has provided valuable insights into the immune mechanisms underlying NPC development [13–15].

In this study, we aimed to comprehensively analyze the immune microenvironment and gene expression patterns in NPC using multiple analytical approaches. By integrating Mendelian randomization analysis, single-cell sequencing, spatiotemporal analysis, and cytokine profiling, we sought to better understand the complex interactions within the tumor microenvironment and their implications for disease progression and treatment.

Materials and methods

Study design

To conduct a comprehensive investigation into the causal associations between immune cells and nasopharyngeal carcinoma (NPC), we employed an innovative two-sample Mendelian randomization (TSMR) analysis method. This approach allowed us to systematically explore potential causal links by leveraging genetic variants as instrumental variables. Building upon the robust findings from the TSMR analysis, we further conducted a Mendelian randomization Bayesian model averaging (MR-BMA) analysis on results that met specific criteria. Specifically, we selected results with TSMR p-values of less than 0.05 and those that passed sensitivity analysis, ensuring the reliability and validity of the identified associations. This multi-layered, rigorous research design not only enhanced the scientific validity of our study but also provided a systematic paradigm for uncovering the molecular mechanisms underlying nasopharyngeal carcinoma.

Data source

For the data source, RNA expression profiles and clinical information for lung adenocarcinoma patients were sourced from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases. These datasets provided a rich resource for understanding the genetic and transcriptional landscapes of lung adenocarcinoma. Additionally, to specifically investigate the immune cell landscape in NPC, we obtained single-cell RNA sequencing (scRNA-seq) data for nasopharyngeal carcinoma tissues from the GEO dataset [16, 17]. This scRNA-seq data enabled us to perform detailed single-cell analysis, allowing us to identify and characterize distinct immune cell subgroups within NPC tissues. By integrating these diverse datasets, we were able to construct a comprehensive framework to explore the complex interactions between immune cells and NPC development.

Selection of core hub genes

To identify core hub genes, we first determined the intersection of disease-related genes and visualized this intersection using a Venn diagram. Protein–protein interaction (PPI) network analysis for these intersecting genes was performed using the STRING database, with results further analyzed in Cytoscape to identify six central hub genes. Functional enrichment analysis of these core hub genes was conducted using the Metascape database.

Establishment of the model

Using patient survival status and survival time, we analyzed differentially expressed genes related to prognosis. Patients were classified into groups via unsupervised clustering methods for further analysis. A scoring system was then constructed using principal component analysis (PCA), specifically selecting the first two principal components for this scoring system.

Immune infiltration analysis

After grouping the main variables, the data was statistically analyzed to determine the distribution of each group within each category. The ggplot2 package was used to visualize the statistical data with overlaid bar charts. Using the core algorithm of CIBERSORT (CIBERSORT.R script analysis) and markers for 22 immune cells provided by the CIBERSORTx website (https://cibersortx.stanford.edu/), we calculated the immune infiltration status of the uploaded data. The stromal and immune scores for colorectal cancer patients from TCGA were calculated using the R package"estimate"[18, 19].

Single-cell level validation

To analyze the single-cell RNA sequencing (scRNA-seq) data, we employed the'Seurat'package in R, a powerful tool designed for the analysis of single-cell data. Our analysis pipeline began with data quality assessment, where we excluded cells with fewer than 200 features (genes) and those with mitochondrial content greater than 20%. These criteria were chosen to remove low-quality or empty droplets and to exclude cells undergoing stress or apoptosis, which can introduce noise into the analysis. We then integrated single-cell data from different samples to ensure that our analysis was not confounded by batch effects. To further mitigate batch effects, we applied specific correction methods within the'Seurat'package, ensuring that any observed differences in cell populations were due to biological variation rather than technical artifacts.

Following data integration, we normalized the data using the'LogNormalization'method, which log-transforms the data and scales it to account for differences in sequencing depth and other technical variations. We then performed unsupervised clustering of the cells based on their gene expression profiles. This step grouped cells into distinct populations, allowing us to identify different cell types within the dataset. To visualize the clustering results, we utilized Principal Component Analysis (PCA) and t-Distributed Stochastic Neighbor Embedding (t-SNE). These dimensionality reduction techniques helped us visualize high-dimensional data in two or three dimensions, making it easier to interpret the cell clusters. To annotate the cell types in each cluster, we used the'SingleR'package, which employs a reference-based approach to assign cell types by comparing the gene expression profiles of the clusters to known cell types from reference datasets. Finally, to identify marker genes that are differentially expressed across different cell types, we used the'FindAllMarkers'function from the'Seurat'package. This function identified genes that were significantly upregulated in one cluster compared to others, providing insights into the unique characteristics of each cell population. By following this comprehensive and multi-step analysis pipeline, we systematically processed and analyzed the scRNA-seq data, leading to the identification of distinct immune cell subgroups and their potential roles in nasopharyngeal carcinoma. This detailed approach ensures the robustness and reproducibility of our findings, providing a solid foundation for further investigation into the molecular mechanisms underlying this disease [20, 21].

Statistics

To ensure the robustness and reliability of our findings, we conducted all statistical analyses using the R programming language (Version 4.0.3). Our approach to statistical significance was stringent, with a p-value threshold of less than 0.05 considered statistically significant unless otherwise specified. This threshold was applied consistently across all analyses, including the two-sample Mendelian randomization (TSMR) and Mendelian randomization Bayesian model averaging (MR-BMA) analyses.

Results

Mendelian randomization analysis reveals potential association between allergic rhinitis and nasopharyngeal carcinoma

A study using Mendelian randomization (MR) methods has thoroughly investigated the causal relationship between allergic rhinitis and nasopharyngeal carcinoma through systematic analyses presented in four complementary visualizations (Fig. 1A–D). The research began with a detailed forest plot (Fig. 1A) demonstrating the independent effects of each SNP and their 95% confidence intervals, distributed around a central null effect line in red, providing a foundational view of individual genetic contributions. To further explore these relationships, a scatter plot (Fig. 1B) was constructed, plotting SNP effects on allergic rhinitis against their effects on nasopharyngeal cancer risk. This visualization revealed the pattern of associations, helping researchers assess the potential strength and direction of causal relationships through the distribution of data points across the effect space. The research team employed rigorous sensitivity analyses, including a comprehensive leave-one-out analysis (Fig. 1C), which systematically evaluated result robustness by sequentially excluding individual SNPs. This approach ensured that no single genetic variant disproportionately influenced the overall conclusions, with horizontal lines clearly depicting effect estimates under various scenarios. The analysis culminated in a funnel plot (Fig. 1D), which synthesized multiple MR analytical methods. This plot contrasted the weighted median method (shown in blue) against other MR approaches (shown in red), providing a comprehensive view of the relationship's consistency across different analytical frameworks and strengthening the credibility of the findings through methodological triangulation.

Fig. 1.

Mendelian randomization analysis reveals potential association between allergic rhinitis and nasopharyngeal carcinoma. The forest plot (A) shows independent effects of individual SNPs; the scatter plot (B) demonstrates the association pattern of SNPs between both diseases; the leave-one-out analysis (C) validates result robustness by excluding individual SNPs; and the funnel plot (D) integrates multiple analytical methods, confirming the reliability of research findings

The association between allergic rhinitis and nasopharyngeal carcinoma

Forest plot (Fig. 2A) shows most SNPs'effect estimates and their 95% confidence intervals crossing the red null effect line, suggesting relatively small individual SNP effects. Scatter plot (Fig. 2B) data point distribution suggests a possible weak association between allergic rhinitis and nasopharyngeal carcinoma. Leave-one-out analysis (Fig. 2C) confirms result robustness, showing minimal impact of removing any single SNP. Funnel plot (Fig. 2D) shows consistency between different MR methods (blue and red lines), enhancing result credibility.

Fig. 2.

The association between allergic rhinitis and nasopharyngeal carcinoma. The forest plot (A) shows SNP effect estimates and confidence intervals mostly crossing the null effect line, indicating small individual SNP effects; the scatter plot (B) distribution suggests a possible weak correlation between the two diseases; leave-one-out analysis (C) confirms result robustness, with minimal impact from removing single SNPs; the funnel plot (D) demonstrates consistency between different methods (blue and red lines), enhancing study credibility

Mendelian randomization analysis reveals potential association between high myopia and nasopharyngeal carcinoma

Forest plot (Fig. 3A) shows SNP effect estimates and their 95% confidence intervals, mostly distributed around the null effect line (red), indicating small individual SNP effects. Scatter plot (Fig. 3B) distribution pattern demonstrates possible association between the two conditions, with Y-axis representing nasopharyngeal carcinoma effects and X-axis showing high myopia effects. Leave-one-out analysis (Fig. 3C) evaluates result robustness by systematically excluding individual SNPs, showing minimal impact on overall conclusions. Funnel plot (Fig. 3D) combines results from different MR methods, including weighted median (blue line) and other approaches (red line), showing consistency across methods and enhancing the credibility of findings.

Fig. 3.

A Mendelian randomization analysis reveals potential association between high myopia and nasopharyngeal carcinoma. Forest plot (A) shows SNP effects distributed around the null line; scatter plot (B) distribution suggests possible association between the two diseases; leave-one-out analysis (C) confirms result robustness; funnel plot (D) enhances study credibility through consistency between different methods (blue and red lines)

Gene expression analysis of nasopharyngeal carcinoma data

In the RNA count distribution analysis (Fig. 4A), we observed two distinct correlation patterns: one with a correlation coefficient of 0.08 showing low association, and another with a correlation coefficient of 0.92 indicating significant expression synergy. The red dots in the scatter plots represent tumor samples, clearly demonstrating the distribution characteristics of RNA expression levels. Further differential expression analysis (Fig. 4B) used volcano plots to show gene expression patterns. Purple dots indicate differentially expressed genes, with some key genes specifically marked in red, which may play crucial roles in nasopharyngeal carcinoma development. Finally, the gene correlation analysis (Fig. 4C) revealed associations between two groups of genes, with the distribution of blue dots reflecting the strength of gene correlations. This correlation analysis helps understand the functional connections between genes, providing important clues for uncovering the molecular mechanisms of nasopharyngeal carcinoma.

Fig. 4.

Gene expression analysis of nasopharyngeal carcinoma data. RNA count distribution analysis (A) showed two distinct correlation patterns (0.08 and 0.92); differential expression analysis (B) identified key differential genes through volcano plots; and gene correlation analysis (C) revealed functional connections between genes. These findings provide important insights into the molecular mechanisms of nasopharyngeal carcinoma

Single-cell sequencing analysis of nasopharyngeal carcinoma

This study utilized single-cell sequencing technology to reveal the cellular composition characteristics of nasopharyngeal carcinoma. The heatmap analysis (Fig. 5A) demonstrates gene expression clustering patterns, with a blue to red gradient intuitively reflecting expression levels of different genes, while the dendrogram shows gene similarity relationships. To better understand the cellular composition of the tumor microenvironment, the study employed two dimensionality reduction methods, UMAP and tSNE (Fig. 5B and C), for cell visualization analysis. Through cell type annotation (Fig. 5D and E), multiple cell types were successfully identified, including T cells, B cells, macrophages, tumor cells, and epithelial cells. Cell populations marked with different colors clearly demonstrate the distribution characteristics of various cells in the tumor microenvironment.

Fig. 5.

Single-cell sequencing analysis of nasopharyngeal carcinoma. Heatmap (A) shows gene expression clustering patterns; UMAP and tSNE dimensionality reduction (B, C) and cell type annotation (D, E) identified multiple cell types (T cells, B cells, macrophages, etc.), revealing cellular heterogeneity in the tumor microenvironment and providing new insights into nasopharyngeal carcinoma pathogenesis

The statistical analysis of cytokines and allergic rhinitis risk

The research examined how different immune system components affect the risk of developing allergic rhinitis. Among pro-inflammatory factors, IL-6 emerged as the most significant risk factor, with a 31.4% increased risk, while IL-12B and DNER showed moderate risk increases of 12.7% and 8.5–10.7% respectively. In contrast, anti-inflammatory factors demonstrated protective effects against allergic rhinitis. IL-10 reduced risk by 8%, while TNFB_LTA showed a similar protective effect with a 6.5% risk reduction. These findings highlight the balancing role of anti-inflammatory factors in allergic responses. For example, IL-12B_IL12B has an OR of 1.127 with a 95% CI of 1.044–1.216, indicating a positive association with the outcome and that this association is statistically significant. In contrast, TNFB_LTA has an OR of 0.935 with a 95% CI of 0.889–0.983, suggesting a potential negative association with the outcome, which is also statistically significant (Fig. 6).

Fig. 6.

The statistical analysis of cytokines and allergic rhinitis risk. Pro-inflammatory factors (IL-6, IL-12B, DNER) increase allergic rhinitis risk, with IL-6 showing the strongest effect, increasing risk by approximately 31.4%. Anti-inflammatory factors (IL-10, TNFB_LTA) demonstrate protective effects, reducing allergic rhinitis risk by 6.5–8%. CCL19 and CCL13 increase disease risk (8–10%). CXCL9 shows protective effects, reducing risk by about 7.6%. Growth factors (FGF-19, LIF-R) both show trends of increasing risk, with increases ranging from 7.7% to 18.3%

Analysis of immune cells and biomarkers in nasopharyngeal cancer

First, looking at feature expression patterns, the study examines both immune cell distributions and key biomarkers. For immune cells, it includes terminally differentiated CD4-CD8- T cells, monocytes, IgD- CD24- B cells, CD3 on CD45RA+ CD4+ T cells, CD14+ CD16- monocytes, and BAFF-R on transitional B cells. These are analyzed alongside crucial biomarkers such as DNER, IL12B, CCL19, CXCL9, FGF19, IL10, IL6, LIFR, and CCL13. The soft threshold analysis provides crucial insights into network construction. By examining both the scale-free topology model fit and mean connectivity across different soft power thresholds, researchers can determine the optimal threshold for subsequent network analysis (Fig. 7A). The hierarchical clustering analysis of B cells reveals distinct subgroup patterns. The dendrogram, spanning heights from 0.93 to 0.99, uses different colors to represent various modules, effectively illustrating the hierarchical structure of B cell subgroups (Fig. 7B). Finally, the correlation analysis among B cell subgroups (M1, M2, M3) employs a color gradient from − 1 to 1, with green indicating positive correlations and purple showing negative correlations. This visualization effectively demonstrates the interrelationships between different B cell subgroups (Fig. 7C and D).

Fig. 7.

Analysis of immune cells and biomarkers in nasopharyngeal cancer. The study investigates immune cell distributions and biomarker patterns through several analytical methods. A Presents an analysis of various immune cells (including CD4-CD8- T cells, monocytes) and key biomarkers (such as DNER, IL12B), along with a soft threshold analysis for network optimization. The research then examines B cell characteristics, as shown in B using a hierarchical clustering approach (height range 0.93–0.99), with different modules represented by distinct colors. Finally, C and D visualize the relationships between three B cell subgroups (M1, M2, M3) using a color gradient system, where green indicates positive correlations and purple shows negative correlations

B cell subgroup analysis in nasopharyngeal carcinoma

In our analysis of the nasopharyngeal carcinoma microenvironment, we uncovered three distinct B cell subpopulations, each with its own molecular and spatial signatures. The M1 subgroup is marked by the expression of PRDX2, LYZ, and BPIFA1, appearing in cyan on spatial maps with a dispersed distribution pattern. These cells show extensive interaction with the innate immune compartment, particularly engaging with macrophages, and also form significant connections with stromal fibroblasts. The M2 subgroup presents a different profile, characterized by high levels of cytoskeletal and housekeeping genes including ACTB, PPIB, and ACTG1. These cells cluster more tightly in the tissue, appearing as dark blue regions in our spatial analysis. Their distribution pattern suggests a specialized niche, with notable connections to adaptive immune components, especially T cells and dendritic cells. The third subgroup, M3, shows a distinctive ribosomal signature, with elevated expression of RPS18, RPL13, and RPL3 A, suggesting high protein synthesis activity. These cells, marked in red, are distributed uniformly throughout the tissue and show unique interactions with proliferating cells and neuroendocrine populations. This pattern suggests they may play a role in supporting cellular growth and neuroendocrine signaling within the tumor microenvironment (Fig. 8A–C).

Fig. 8.

B cell subgroup analysis in nasopharyngeal carcinoma. A–C The M1 subgroup expresses genes such as PRDX2, LYZ, and BPIFA1, shows a dispersed cyan distribution, and primarily interacts with macrophages and fibroblasts. The M2 subgroup highly expresses cytoskeleton-related genes ACTB, PPIB, and ACTG1, appears as dark blue clusters in the tissue, and mainly interacts with T cells and dendritic cells. The M3 subgroup specifically expresses ribosome-related genes RPS18, RPL13, and RPL3 A, shows a uniform red distribution, and has close associations with proliferating cells and neuroendocrine cells, potentially participating in the regulation of tumor growth and neuroendocrine signaling

Spatial distribution and temporal evolution analysis of immune cells in nasopharyngeal carcinoma

Analysis of immune cell distribution and evolution in the nasopharyngeal carcinoma microenvironment reveals complex spatiotemporal dynamics. Spatially, the tissue contains various immune cell types, including B cells, T cells, endothelial cells, and macrophages, which form distinct spatial organization patterns and cellular interaction networks. Temporal evolution analysis demonstrates dynamic changes in cellular composition: the early phase is dominated by endothelial and epithelial cells, followed by peak levels of B cells and proliferating cells in the middle phase, and finally a significant increase in dendritic cells and T cells in the late phase. At the gene expression level, the study tracked six key genes: CPSF3L maintains stable low expression, ISG15 shows consistently high expression, particularly in the early phase; SDF4 and UBE2 J2 display similar expression patterns, while immune regulation-related genes TNFRSF18 and TNFRSF4 show significant upregulation in the late phase (Fig. 9A–C).

Fig. 9.

Spatial distribution and temporal evolution analysis of immune cells in nasopharyngeal carcinoma. A–C Various immune cells (such as B cells, T cells) form specific distribution patterns. Temporally, three phases emerge: early phase dominated by endothelial and epithelial cells, middle phase showing peak B cell levels, and late phase marked by increased dendritic and T cells. Gene expression analysis shows low CPSF3L expression, sustained high ISG15 expression, and significant upregulation of immune regulatory genes TNFRSF18 and TNFRSF4 in the late phase

Gene expression dynamics and spatial distribution in nasopharyngeal carcinoma

Temporal analysis shows gene expression clustering into four main groups: some genes show progressive increase from low to high expression, others transition from low to high expression regions, some decrease from high expression, and others shift from moderate to low expression. Notably, HLA-related genes show significantly low expression in late stages, while immune regulation-related genes (such as the TNFRSF family) display unique expression patterns across different phases, and cell cycle-related genes show high expression during the middle phase(Fig. 10A). Spatial distribution analysis demonstrates distinct regional differences, with certain genes showing concentrated expression in specific areas, forming unique gene co-expression modules and tissue-specific expression patterns. These findings not only reveal the complex regulatory network of gene expression in nasopharyngeal carcinoma but also highlight the crucial role of immune-related genes in disease progression, providing important insights for understanding disease mechanisms and developing therapeutic strategies (Fig. 10B).

Fig. 10.

Gene expression dynamics and spatial distribution in nasopharyngeal carcinoma. A Temporal analysis revealed four gene expression patterns: progressively increasing, low-to-high transition, high-to-low transition, and moderate-to-low shift. HLA-related genes showed low expression in late stages, immune regulatory genes displayed unique patterns across phases, and cell cycle genes peaked in the middle phase. B Spatial analysis revealed region-specific gene expression forming distinct co-expression modules, revealing the regulatory network characteristics of nasopharyngeal carcinoma

Discussion

NPC is a unique malignancy characterized by its strong association with Epstein-Barr virus (EBV) and high levels of immune infiltration. This cancer is endemic in certain regions, such as East and Southeast Asia and North Africa, indicating a significant influence of epidemiological patterns and genetic susceptibility. The intense immune infiltration and commonly low degree of differentiation are unique characteristics that distinguish NPC from other cancers [22–24]. The presence of EBV significantly shapes the NPC microenvironment through chronic immune activation, which ultimately affects tumor progression and therapeutic outcomes. Despite the use of chemoradiotherapy as the first-line treatment, it often fails to achieve optimal results in patients with advanced or metastatic tumors. Therefore, understanding the immune microenvironment and gene expression patterns in NPC is crucial for developing effective therapeutic strategies.

The comprehensive analysis of the immune microenvironment and gene expression patterns in NPC presented in this study provides a detailed and multi-faceted understanding of the disease mechanisms and potential therapeutic targets. The findings highlight the intricate interplay between immune cells, cytokines, and gene expression dynamics, which collectively contribute to the progression of NPC [25–28].

The Mendelian randomization analysis in this study revealed potential causal associations between NPC and both allergic rhinitis and high myopia. These findings suggest that immune-related conditions may play a role in the development of NPC, possibly through chronic inflammation or altered immune surveillance mechanisms. The identification of IL-6 as a significant risk factor and IL-10 as a protective factor further underscores the importance of cytokine balance in disease pathogenesis. These insights could guide the development of targeted immunotherapies aimed at modulating the immune response in NPC patients.

The single-cell sequencing analysis identified distinct B cell subpopulations (M1, M2, M3) with unique molecular signatures and spatial distributions. The M1 subgroup's interaction with innate immune cells and stromal fibroblasts suggests a role in early immune responses and tissue remodeling. In contrast, the M2 subgroup's association with adaptive immune components highlights its potential involvement in later stages of immune regulation. The M3 subgroup's high protein synthesis activity and interactions with proliferating cells and neuroendocrine populations imply a supportive role in tumor growth and signaling. These findings emphasize the heterogeneity of B cell populations in NPC and their differential contributions to disease progression.

The temporal analysis of immune cell composition and gene expression patterns revealed dynamic changes across different phases of NPC. The early dominance of endothelial and epithelial cells, followed by a peak in B cells and proliferating cells in the middle phase, and the late increase in dendritic cells and T cells, reflect the evolving nature of the tumor microenvironment. The gene expression clustering into four main patterns, with significant roles for immune-related genes such as the TNFRSF family and HLA-related genes, further highlights the complexity of molecular mechanisms underlying NPC progression. The downregulation of HLA-related genes in the late stages suggests a potential escape mechanism from immune surveillance, while the upregulation of TNFRSF family genes indicates ongoing immune regulation.

Single-cell sequencing has proven to be an invaluable technique for understanding the diversity of cells and molecules within NPC. Recent research leveraging single-cell RNA sequencing (scRNA-seq) has identified unique immune cell subsets and their interactions within the NPC microenvironment. For example, scRNA-seq has uncovered various T cell subsets, including exhausted and regulatory T cells, each with unique functional profiles that contribute to immune evasion in NPC. Additionally, the discovery of distinct B cell subsets and their roles in immune regulation and tumor progression has shed new light on the NPC microenvironment. These insights underscore the potential of single-cell sequencing to reveal new therapeutic targets and biomarkers for NPC [26, 29, 30].

The thorough analysis of the immune microenvironment and gene expression in NPC provides crucial information for the development of targeted treatments. The discovery of specific B cell subsets and their roles suggests that therapies aimed at these subsets could alter the immune response and enhance treatment efficacy. Moreover, the changing dynamics of immune cell composition and gene expression over time underscore the significance of timing in treatment strategies. Treatments targeting early immune responses might be more effective during the disease's initial stages, whereas those designed to bolster adaptive immunity could be more advantageous in later stages.

Moreover, cytokine analysis offers potential biomarkers for evaluating disease risk and monitoring therapeutic responses. Focusing on pro-inflammatory cytokines like IL-6 while promoting the effects of anti-inflammatory cytokines such as IL-10 could help reestablish immune equilibrium and improve patient outcomes. The spatial distribution of immune cells and gene expression patterns also indicates the feasibility of spatially targeted therapies, including localized immunomodulation or gene therapy.

Limitations

One limitation may be the cross-sectional nature of the study, which restricts the understanding of the dynamics of immune cell composition and gene expression patterns over time. Due to the cross-sectional design, the study results may not fully reveal the evolution of immune cells and molecular mechanisms during the progression of the disease. Furthermore, while single-cell sequencing provides detailed information on cellular and molecular heterogeneity, it may not fully capture the complexity of cell-to-cell interactions and their dynamic changes within the tumor microenvironment.

Another limitation could be the sample size and representativeness. If the study sample size is small or lacks sufficient diversity, the results may not be generalizable to a broader population. Additionally, if there is bias in the sample selection, it could affect the accuracy and reliability of the findings.

Conclusion

In conclusion, this study's comprehensive analysis of the immune microenvironment and gene expression patterns in NPC provides a detailed roadmap for understanding disease mechanisms and developing targeted therapeutic strategies. The integration of multiple analytical approaches, including Mendelian randomization, single-cell sequencing, and spatiotemporal analysis, offers a robust framework for future research and clinical applications in NPC. Future studies should focus on validating these findings in larger cohorts and exploring the therapeutic potential of targeting specific immune cell populations and molecular pathways identified in this study.

Author contributions

Qingfu Bao: Conceptualization, methodology, formal analysis, writing original draft, supervision, and project administration. Jianjun Zou: Investigation, data curation, validation, and writing—review and editing. Changyang Wang: Software, data analysis, visualization, and methodology. Haiying Wang: Resources, validation, writing review and editing, and supervision. All authors have read and agreed to the published version of the manuscript.

Funding

2024 Zhejiang Province Traditional Chinese Medicine Science and Technology Plan (2024ZL694).

Data availability

RNA expression profiles and clinical information for lung adenocarcinoma patients were sourced from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

All authors agree to publish this article.

Competing interests

The authors declare no competing interests.