Construction of a Novel T Cell Subtype with Immunotherapeutic Value in Bladder Cancer and a Neural Network-Based Prognostic Model

Construction of a Novel T Cell Subtype with Immunotherapeutic Value in Bladder Cancer and a Neural Network-Based Prognostic Model | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Construction of a Novel T Cell Subtype with Immunotherapeutic Value in Bladder Cancer and a Neural Network-Based Prognostic Model Xiaoliang Dou, Chaochao Cui, Yaodong zhang, Dali He, Tianci Mao, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8657603/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 7 You are reading this latest preprint version Abstract Bladder cancer (BLCA) remains a significant global health challenge, characterized by high heterogeneity and suboptimal responses to existing immunotherapies. This study identifies TNFRSF4/CD4+ T cells as a novel immune cell subtype with crucial prognostic and therapeutic relevance in BLCA. Leveraging single-cell RNA sequencing and bulk transcriptome data from TCGA-BLCA and GEO databases, we conducted multi-omics analyses to elucidate the immune landscape and its implications for clinical outcomes. Our findings reveal that higher levels of TNFRSF4/CD4+ T cells are associated with improved patient survival and robust immune activity, emphasizing their potential as biomarkers and therapeutic targets. Gene expression and pathway enrichment analyses highlight their involvement in immune-regulatory mechanisms, including the TCR signaling pathway and TNF family signaling. Additionally, we identified 11 core genes associated with TNFRSF4/CD4+ T cells, such as SEPTIN1 and FCMR, which significantly impact prognosis and immune function. Multiplex immunofluorescence demonstrated a strong positive correlation between the prognosis of bladder cancer patients and the expression of SEPTIN1 and FCMR within TNFRSF4/CD4+ T cells. These genes served as the foundation for constructing a neural network-based prognostic model, which demonstrated high predictive accuracy and stratified patients into distinct risk categories. This study underscores the critical role of TNFRSF4/CD4+ T cells in BLCA immunity and highlights their potential in advancing personalized treatment strategies, offering novel insights into improving immunotherapy efficacy in bladder cancer. TNFRSF4/CD4+ T cell Bladder cancer SEPTIN1 FCMR Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Introduction Bladder cancer ranks as the second most prevalent urological malignancy worldwide, with approximately 549,000 new cases and nearly 200,000 deaths reported annually[ 1 ]. Early detection, especially before muscle invasion, allows for effective treatment with minimal impact on survival. The disease manifests in a broad spectrum, from recurrent noninvasive tumors to aggressive, advanced stages requiring multimodal treatment approaches. This significant burden, both in terms of public health and socioeconomic impact, underscores the critical importance of early detection, diagnosis, and treatment[ 2 ]. And building prognostic models for bladder cancer play a key role in personalizing treatment by stratifying patients based on risk, optimizing therapeutic strategies, and ultimately improving clinical outcomes. T cells play a pivotal role in the immune response against bladder cancer, particularly in tumor recognition and elimination. CD8 + cytotoxic T cells are central to the anti-tumor immune response, as they directly kill tumor cells by recognizing tumor antigens and releasing cytolytic molecules such as perforin and granzyme[ 3 ]. Their presence in the tumor microenvironment has been correlated with better responses to immune checkpoint inhibitors (CPIs) targeting PD-1 or PD-L1, making them a key focus in immunotherapy research[ 4 ]. However, in certain cancers like bladder cancer, CPIs as stand-alone treatments have shown limited effectiveness, with only about 20% of patients experiencing a positive response[ 5 ]. This limited success may be partly attributed to the heterogeneous nature of tumor-infiltrating lymphocytes (TILs) and their varying abilities to mount an effective immune response. Research has primarily concentrated on cytotoxic CD8 + T cells to investigate how immunotherapies, particularly checkpoint inhibitors (CPIs), stimulate anti-tumor immune responses. In melanoma, specific gene expression patterns and chromatin state markers linked to cytotoxic activity and T cell exhaustion have shown a strong correlation with patient outcomes following PD-1-targeted therapies[ 6 ]. Additionally, the presence of CD8 + T cells at the invasive margins of tumors before treatment has been associated with improved therapeutic responses[ 4 ]. However, in metastatic bladder cancers, response rates to PD-1 inhibitors remain modest, with approximately 15%–20% of patients who have progressed on platinum-based chemotherapy responding, and slightly higher rates seen in patients ineligible for platinum therapy[ 7 ]. Predictive biomarkers such as PD-L1 expression have yet to provide consistent clarity in these cases[ 8 ]. Recent studies using bulk RNA sequencing have uncovered pre-treatment gene signatures—higher CD8 + gene expression, increased tumor mutational burden, and lower transforming growth factor-beta (TGF-b) signatures, particularly in immune-excluded tumors—that are associated with better responses to anti-PD-L1 therapy, such as atezolizumab[ 9 ]. While these findings shed light on potential mechanisms of response, more research is required to deepen our understanding of CD8 + T cell involvement in CPI efficacy, as well as the roles of other immune cell subsets beyond the conventional cytotoxic and exhausted CD8 + populations. Conversely, the role of CD4 + T cells in bladder cancer immunity is far less understood. Studies have highlighted the crucial role of distinguishing CD4 + T cell subtypes in immunotherapy, as they exhibit multiple tumor-specific states, particularly distinct subsets of regulatory T cells (Tregs). Notably, the research also uncovered cytotoxic CD4 + T cells with the ability to kill tumors through MHC class II-dependent mechanisms[ 10 ]. These cytotoxic CD4 + T cells undergo clonal expansion in bladder tumors, likely driven by their recognition of specific tumor antigens. In vitro experiments confirmed their capacity to kill autologous tumors, and their anti-tumor activity is further enhanced by the secretion of cytokines such as TNF-α and IFN-γ[ 11 ]. These findings highlight the complexity and diversity of CD4 + T cell populations within tumors and suggest that leveraging this heterogeneity could enhance the effectiveness of immunotherapies in bladder cancer. TNFRSF4 (OX40/CD134), a member of the tumor necrosis factor receptor superfamily (TNFRSF), plays a key role in T cell activation, survival, and memory formation, particularly within the tumor microenvironment. Following T cell receptor (TCR) signaling, TNFRSF4 is upregulated on CD4 + and CD8 + T cells, where it enhances their function by promoting sustained activation and preventing apoptosis[ 12 ]. This receptor, upon binding with its ligand OX40L, amplifies T cell responses, which is especially critical in chronic infections and cancers, where T cell exhaustion often diminishes immune efficacy[ 13 ]. Additionally, TNFRSF4 modulates the activity of regulatory T cells (Tregs), balancing immune suppression and activation depending on the context[ 13 ]. Despite TNFRSF4's well-established role in enhancing anti-tumor immunity in other cancers, its specific role in bladder cancer has yet to be investigated. This lack of research creates a significant gap in our understanding of its therapeutic potential in this particular cancer type. In this study, we employed a comprehensive multi-omics approach to investigate the immune landscape of bladder cancer, with a particular focus on TNFRSF4/CD4 + T cells. Utilizing single-cell RNA sequencing data from bladder cancer patients and bulk transcriptome data from the TCGA-BLCA cohort, we conducted differential gene expression analysis, Bayesian deconvolution, and survival analysis to identify significant associations between TNFRSF4/CD4 + T cells and bladder cancer prognosis. Furthermore, we explored the immunotherapeutic relevance of TNFRSF4/CD4 + T cells by analyzing immune-related gene sets and key immune checkpoint markers. Additionally, we constructed a neural network-based prognostic model incorporating 11 core genes identified through WGCNA and survival analysis. The model demonstrated high predictive accuracy, with GZMM being the most important gene contributing to the model’s performance. This neural network model provided valuable insights into patient risk stratification and could aid in clinical decision-making for bladder cancer treatment. Methods Data availability statement The bulk transcriptome data of bladder cancer (BLCA) patients were retrieved from the TCGA database. The TCGAbiolinks package facilitated the acquisition of gene count daa, simple nucleotide variation (SNV) data, and clinical information for TCGA-BLCA. Additionally, single-cell transcriptome data of BLCA (bladder urothelial carcinoma) patients were obtained from the GEO database (GSE222315). Genome-wide association study (GWAS) data for BLCA patients were accessed through the IEU Open GWAS Project (Dataset: ieu-b-4874). Single-cell and bulk transcriptomics analysis integrated with GWAS Data The analytical procedures in this research were executed using R and R Studio software. The Seurat and SingleR packages were utilized for the preprocessing and clustering of single-cell data, whereas the DESeq2 package facilitated the differential gene expression analysis for TCGA-BLCA bulk transcriptome data. Bayesian deconvolution, which involved the integration of single-cell and bulk transcriptome data, was carried out using the BayesPrism package. For processing GWAS data and pinpointing disease-associated genes and relevant cell populations at the single-cell level, the gwasvcf and scPagwas packages were predominantly employed. Particularly, the scPagwas package was applied to identify genes and cell populations related to specific traits. Analysis of convolutional cell correlations The survival analysis of convolutional cells was predominantly carried out using the survival package. Immune infiltration and differential analyses between groups were largely performed employing the tidyestimate and DESeq2 packages. The investigation of immune-related gene sets within convolutional cells was primarily executed utilizing the limma package. Immunotherapy analysis was chiefly accomplished through the Tidepy package (which necessitates Python environment configuration for use within R Studio). Drug response prediction was primarily achieved using the oncoPredict package. SNV analysis was largely conducted using the Maftools package, and WGCNA analysis was implemented using the WGCNA package. Construction of a prognostic model and analysis of core genes The Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses of core genes were predominantly conducted utilizing the clusterProfiler package, while prognostic assessments were performed through the survival and survminer packages. The Summary-data-based Mendelian Randomization (SMR) analysis was executed using the smr-1.3.1 software. The development of the neural network-based prognostic model was primarily achieved by employing the reticulate package to invoke Python. Within the Python environment, the Pytorch package was essential for model construction, while the sharp package was integral for providing interpretability analysis of the model. The produced onnx model file was employed to visualize the model structure (NETRON: https://netron.app/ ). Multiplex immunofluorescence staining Following routine processing of fixation, embedding, and sectioning, sequential multiplex immunofluorescence staining was performed according to standard protocols. The staining was carried out using the following fluorophores: FGNA for TNFRSF4, CY5 for CD4, FBNA for SEPTIN1/KLRB1/TENT5C/FCMR, and DAPI for nuclei. Ethics statement This study involving human tumor tissue samples was conducted in accordance with the ethical principles of the Declaration of Helsinki. The collection and use of all samples from bladder cancer (BLCA) patients were approved by the Hospital Ethics Committee (Approval Reference: ChiCTR2500112525). Written informed consent was obtained from all participants prior to sample collection. Results TNFRSF4/CD4+ T cells are significantly associated with BLCA To explore immune cells in BLCA at the single-cell level, single-cell sequencing data from 9 BLCA patients were analyzed. Following preprocessing, principal component analysis and clustering were applied to the single-cell data (Figure 1A), resulting in the identification of 27 distinct cell subgroups. These subgroups were annotated using the singleR package, and the proportions, as well as expression levels of markers for recognized cell types, were determined within these 29 subgroups (Figure 1B). Ultimately, 11 key cell subgroups were identified (Figure 1C). Notably, CD4+ T and CD8+ T cells were further categorized into three distinct types based on the high expression levels of specific genes (min.pct = 0.25, logfc.threshold = 0.25): TNFRSF4/CD4+ T cell, MT-ATP6/CD4+ T cell, and CCL4L2/CD8+ T cell. A trait-related (BLCA) scoring of these 11 subgroups was performed, yielding trait-related scores (TRS) and identifying trait-associated genes (Figures 1D and E). The findings demonstrated that TNFRSF4/CD4+ T cells were significantly associated with the trait ( p < 0.001). In preparation for Bayesian deconvolution using bulk transcriptome data, a separate analysis of bulk transcriptome data was conducted. Differential gene expression analysis identified 14,932 genes with significant differences (padj < 0.05) between the BLCA group and the normal control group, with 9,691 genes upregulated and 5,241 downregulated (Figures 1F and G). During Bayesian deconvolution, protein-coding genes were determined to be the most consistent gene category across both datasets (R = 0.636, mean squared error = 3.29, Figure 1H). Consequently, deconvolution focused exclusively on protein-coding genes, which proved more valuable and expedited the process. The posterior mean of the cell type fraction theta was extracted from bulk transcriptome data. Prognostic analysis revealed significant survival differences between patients with high and low levels of TNFRSF4/CD4+ T cells ( p < 0.001), with longer survival observed in the high-level group. Additionally, notable survival differences were detected between patients with high and low levels of B cells and smooth muscle cells ( p < 0.01, Figures 1I-K). TNFRSF4/CD4+ T cells possess therapeutic value in immunotherapy To further assess the immunotherapeutic potential of TNFRSF4/CD4+ T cells, a differential gene expression analysis was initially performed between groups with high and low levels of the specified convolutional cells (TNFRSF4/CD4+ T cells). This investigation identified 10,631 differentially expressed genes (DEGs, padj < 0.05, Figure 1L), alongside notable differences across various cell types, especially in immune cells such as NK cells and B cells. Furthermore, the immune infiltration scores (estimate scores) between the two groups exhibited significant variation, particularly in the immune score (Figure 1M). In the analysis of immune-related gene sets, numerous significant discrepancies were observed between high and low TNFRSF4/CD4+ T cell levels, with prominent differences in the T cell receptor (TCR) signaling pathway, tumor necrosis factor (TNF) family members, and their corresponding receptors (Figure 2A-C). Similarly, immunotherapy analysis unveiled significant differences in the expression levels of various immunotherapy targets, including CAF, CD8, CD274, and IFNG, between high and low TNFRSF4/CD4+ T cell levels (Figure 2D). Additionally, drug sensitivity analysis highlighted significant differences in the response levels to multiple tumor treatment drugs between the two groups (Figure 3). SNVs analysis of TNFRSF4/CD4+ T cells To further investigate the gene mutations in TNFRSF4/CD4+ T cells, a global overview of the SNVs for the top 30 mutated genes was generated, leading to the construction of a gene waterfall plot (Figure 4A). TP53 was identified as having the highest mutation rate, with missense mutations representing the largest proportion. A statistical examination of transitions and transversions indicated that C > T pyrimidine transitions occurred most frequently (Figure 4B). Furthermore, an analysis comparing differentially mutated genes between high and low TNFRSF4/CD4+ T cell levels revealed that E300 exhibited the most significant differential mutation rate, with missense mutations being the predominant type of alteration. Selection of core genes via weighted gene co-expression network analysis (WGCNA) To identify core genes associated with BLCA and the specific convolutional cell (TNFRSF4/CD4+ T cell), WGCNA was utilized to select 169 genes from the MEpink module. Subsequently, the overlap of significant DEGs identified in bulk transcriptome analysis, marker genes of TNFRSF4/CD4+ T cells from single-cell transcriptome analysis, and trait-associated genes detected by scPagwas was determined, yielding 11 core genes (RGS1, KLRB1, CXCR4, GYPC, GBP5, FCMR, IL16, SEPTIN1, TENT5C, GZMM, LTB). Functional enrichment analyses, including GO and KEGG, were conducted on these core genes. The GO analysis highlighted functions such as immune system processes and CD4 receptor binding as the most significant, while KEGG analysis revealed that pathways, including the NF-kappa B signaling pathway, were most prominent. Additionally, survival analysis was performed on the core genes, demonstrating a significant difference in patient survival between high and low expression groups of SEPTIN1 (surdiff.P < 0.05), with longer survival observed in the high-expression group. Furthermore, SEPTIN1, KLRB1, TENT5C, and FCMR were significant in the Cox proportional hazards model (Cox.P < 0.05), where higher expression levels indicated a better prognosis (HR < 1). Validation of core genes’ association using SMR To further validate the association between core genes and BLCA, SMR was employed to assess whether significant causal associations existed between these genes and the disease. The findings revealed that within a 1,000 Kb window centered on the core gene GYPC, a gene significantly associated with BLCA, MYO7B (pSMR < 0.05), was identified. Construction of a prognosis model based on neural networks To maximize the prognostic utility of core genes, a neural network-based BLCA prognosis model was constructed. Initially, 11 core genes were incorporated, and various model architectures were evaluated. Following multiple parameter optimizations, an optimal structure was identified (Figure 7A). The influence and significance of model features, particularly gene expression levels, were assessed, with GZMM showing the highest level of importance (Figure 7B). Patients were then stratified into two groups according to the median RiskScore derived from the model. A marked difference in survival times was observed between the high and low RiskScore groups ( p < 0.001), with shorter survival periods noted in the high RiskScore group (Figure 7C). These findings indicate that this prognostic model provides valuable insight for clinical decision-making. SEPTIN1 and KLRB1 in TNFRSF4/CD4+ T Cells as Prognostic Biomarkers in BLCA To investigate the association between the infiltration of identified core genes within TNFRSF4/CD4+ T cells in BLCA tissues and patient prognosis, multiplex immunofluorescence staining was performed on tumor specimens from 13 BLCA patients. The expression levels of SEPTIN1, KLRB1, TENT5C, and FCMR were evaluated specifically in TNFRSF4/CD4+ T cells. The results revealed that in tumor tissues from five patients with favorable prognosis, SEPTIN1 and KLRB1 were highly expressed in TNFRSF4/CD4+ T cells ( Figure 8A, B), while TENT5C and FCMR showed moderate expression levels (Figure 9A, B). In contrast, in the eight patients with poorer prognosis, all four genes exhibited low expression within the same cell population. These findings indicate that high expression of the core genes SEPTIN1 and KLRB1 in TNFRSF4/CD4+ T cells is significantly associated with improved prognosis in BLCA patients. Discussion In this study, we identified TNFRSF4/CD4 + T cells as a novel immune cell subtype significantly associated with bladder cancer, showing strong prognostic value. Using single-cell RNA sequencing and bulk transcriptome data, we revealed that higher levels of TNFRSF4/CD4 + T cells were linked to improved survival, emphasizing their potential as a therapeutic target in immunotherapy. Additionally, we explored immune-related gene sets and found significant differences in pathways such as the TCR signaling pathway and TNF family members. Furthermore, through WGCNA analysis, we identified 11 core genes closely associated with TNFRSF4/CD4 + T cells and bladder cancer prognosis. These genes include RGS1, KLRB1, CXCR4, GYPC, GBP5, FCMR, IL16, SEPTIN1, TENT5C, GZMM, and LTB. These core genes were enriched in key immune functions, such as immune system processes and CD4 receptor binding, and formed the basis for constructing a neural network-based prognostic model. The model demonstrated high accuracy in predicting patient survival outcomes, with GZMM identified as the most significant contributor to the model. This prognostic model highlights the potential for integrating TNFRSF4/CD4 + T cells and the core genes into personalized treatment strategies for bladder cancer, offering new opportunities for improving clinical outcomes. Our analysis of single-cell RNA sequencing (scRNA-seq) data revealed that TNFRSF4/CD4 + T cells are significantly correlated with bladder cancer, suggesting their potential role in tumor progression and immune modulation. The identification of TNFRSF4/CD4 + T cells as a distinct subtype, alongside other T cell populations such as MT-ATP6/CD4 + T cells and CCL4L2/CD8 + T cells, highlights the heterogeneity within the tumor immune microenvironment. TNFRSF4 (OX40) is known to enhance T cell activation, survival, and memory formation through its interaction with its ligand OX40L. This signaling pathway likely contributes to the prolonged survival and sustained activation of CD4 + T cells within the tumor microenvironment, facilitating a more robust immune response against tumor cells[ 13 ]. Our findings suggest that TNFRSF4/CD4 + T cells may play a crucial role in modulating the immune response in bladder cancer through several mechanisms. First, the OX40-OX40L interaction is known to prevent T cell apoptosis and enhance their proliferation, which could contribute to sustained anti-tumor activity, particularly in chronic tumor settings where T cell exhaustion typically diminishes immune effectiveness[ 14 ]. The presence of these cells may help maintain the immune system’s capacity to recognize and attack bladder cancer cells over time, reducing the likelihood of immune evasion by the tumor. Our study demonstrates the crucial role of TNFRSF4/CD4 + T cells in bladder cancer and their potential value in immunotherapy. We identified over 10,631 differentially expressed genes between high and low TNFRSF4/CD4 + T cell groups, indicating a profound impact on the tumor immune environment. Patients with higher levels of these cells exhibited increased infiltration of immune cells, such as NK cells and B cells, alongside higher overall immune infiltration scores, suggesting a more robust immune presence. Additionally, TNFRSF4 interacts with OX40L to prevent T cell apoptosis and sustain activation, activating downstream pathways like NF-κB and MAPK that are essential for T cell proliferation and immune memory formation. This prolonged activation helps prevent T cell exhaustion, a major challenge in cancer immunotherapy, and enhances the balance between effector T cells and regulatory T cells (Tregs), reducing immune suppression within the tumor microenvironment[ 15 , 16 ]. Furthermore, significant differences in the expression of key immune pathways, such as the TCR signaling pathway and TNF family members, as well as immunotherapy targets like PD-L1 and IFNG, underscore the potential for TNFRSF4/CD4 + T cells to improve responses to immune checkpoint inhibitors. The distinct differences in drug sensitivity also highlight the potential of these cells as biomarkers for predicting treatment outcomes, offering valuable insights for more personalized therapeutic approaches in bladder cancer. These findings collectively reinforce the importance of TNFRSF4/CD4 + T cells as critical modulators of bladder cancer immunity and as promising targets for enhancing the efficacy of immunotherapy. However, the heterogeneity of T cells may contribute to inconsistent immunotherapy outcomes in some patients, and addressing this complexity remains an area for future research. In our genetic mutation analysis of TNFRSF4/CD4 + T cells, TP53 was identified as the most frequently mutated gene, with a significant proportion of missense mutations. TP53 mutations are well-documented in bladder cancer and are known to disrupt its tumor-suppressive function, potentially contributing to uncontrolled cell growth and immune evasion. In bladder cancer specifically, TP53 mutations are associated with higher tumor grade, increased invasiveness, and poorer prognosis[ 17 , 18 ]. The frequent mutation of TP53 in TNFRSF4/CD4 + T cells suggests a possible link between this genetic alteration and impaired immune function, as TP53 plays a role in regulating the immune response by influencing apoptosis and cell cycle control[ 19 ]. This disruption could lead to dysfunctional T cell activity, potentially hindering anti-tumor immunity[ 20 ]. The analysis also revealed a prevalence of C > T transitions, a common mutational signature in cancers, which could point to oxidative damage or other mutagenic processes affecting T cell functionality in the tumor microenvironment[ 21 ]. While these findings suggest that TP53 mutations may influence the function of TNFRSF4/CD4 + T cells in bladder cancer, we did not directly examine their impact on T cell behavior or immune evasion. Future research should focus on understanding how these mutations affect immune modulation and contribute to therapeutic resistance in bladder cancer. We utilized WGCNA to identify core genes associated with bladder cancer and the specified TNFRSF4/CD4 + T cells. By analyzing the MEpink module, we selected 169 genes, which were then intersected with differentially expressed genes from bulk transcriptome data, marker genes from single-cell RNA sequencing of TNFRSF4/CD4 + T cells, and phenotype-associated genes identified by scPagwas. This approach yielded 11 core genes: RGS1, KLRB1, CXCR4, GYPC, GBP5, FCMR, IL16, SEPTIN1, TENT5C, GZMM, and LTB. These genes share common roles in immune regulation, cell signaling, and migration, which are essential for immune surveillance and tumor progression. Many, such as RGS1 and KLRB1, are involved in T cell activation and migration[ 22 , 23 ], while CXCR4 is specifically known for its role in cancer metastasis[ 24 ]. GZMM plays a key role in cytotoxicity[ 25 ], and GBP5 contributes to antiviral responses[ 26 ]. Additionally, SEPTIN1 is linked to cellular structure and tumor stability, making these genes collectively significant in both immune response and bladder cancer progression[ 27 ]. These core genes were enriched in biological processes like the immune system process and CD4 receptor binding, and pathways such as NF-kappa B signaling in the KEGG analysis, further highlighting their role in immune regulation[ 28 ]. Survival analysis of these core genes revealed that SEPTIN1 was significantly associated with patient survival (p < 0.05), where higher expression was linked to longer survival. Additionally, KLRB1, TENT5C, and FCMR were significant in the COX proportional hazards model, suggesting their potential as prognostic markers, with higher expression correlating with better outcomes (HR < 1). Supporting these findings, multiplex immunofluorescence analysis revealed high expression of both SEPTIN1 and FCMR within TNFRSF4/CD4 + T cells in bladder cancer patients with favorable prognosis, underscoring the importance of these two genes as prognostic biomarkers. Furthermore, we utilized SMR analysis to examine the association between core genes and bladder cancer. The analysis identified a significant association between MYO7B and bladder cancer (pSMR < 0.05) within a 1000Kb window centered around the core gene GYPC. This suggests that while GYPC was highlighted as a core gene, MYO7B may also play a relevant role in the genetic landscape of bladder cancer within this region. The finding underscores the need for further investigation into MYO7B to explore its specific contribution to bladder cancer development, as it may share regulatory elements or be influenced by genetic variants affecting the broader region. This highlights the importance of considering the surrounding genetic context when evaluating disease-associated genes. Finally, we developed a neural network-based prognostic model for bladder cancer, incorporating 11 core genes identified from our earlier analyses. After testing various model architectures and fine-tuning parameters, the final model achieved strong performance, with GZMM identified as the most influential gene in predicting patient outcomes. The model-generated RiskScore effectively stratified patients into high- and low-risk groups, with significant differences in survival between the two groups (p < 0.001). Patients in the high-risk group exhibited notably shorter survival times, emphasizing the model’s potential to serve as a valuable tool for clinical decision-making.The ability to classify patients based on genetic profiles and predict their survival outcomes with high accuracy can significantly improve the individualization of treatment strategies. This could allow clinicians to identify high-risk patients who may benefit from more aggressive therapies, while sparing low-risk patients from overtreatment. The antitumor immune functions of CD4 + T cells have attracted growing attention in recent years, as exemplified by the 2025 Nobel Prize-awarded research on regulatory T cells (Tregs)[ 29 ]. The TNFRSF4/CD4 + T-cell subset identified in this study likely contains antitumor-active effector T cells or functionally specialized Tregs, thus providing a novel direction for understanding the functional diversity of CD4 + T cells in tumor immunity. Declarations Conflict of interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Contributorship statement D.X and C.C conceived research ideas and designed and performed experiments, analyzed data, and wrote the manuscript; C.C and Y.Z preformed sequencing data analysis; T.L, L.G, T.M, and D.H, assisted in completing the experiment; D.X and C.C provided clinical consultation, experimental design and guidance; J.F conceptualized and guided the research, interpreted the results, revised and edited the manuscript. All authors have read and approved the article. Funding and acknowledgments This research was supported by the following funding bodies: [Shaanxi Province key research and development plan project, with grant numbers [No.S2024-YF-YBSF-1796]. References Compérat, E., et al., Current best practice for bladder cancer: a narrative review of diagnostics and treatments. The Lancet, 2022. 400 (10364): p. 1712-1721. Hashem, M., et al., Non-coding RNA transcripts, incredible modulators of cisplatin chemo-resistance in bladder cancer through operating a broad spectrum of cellular processes and signaling mechanism. Non-coding RNA Research, 2024. 9 (2): p. 560-582. van der Leun, A.M., D.S. Thommen, and T.N. Schumacher, CD8(+) T cell states in human cancer: insights from single-cell analysis. Nat Rev Cancer, 2020. 20 (4): p. 218-232. Kornepati, A.V.R., R.K. Vadlamudi, and T.J. Curiel, Programmed death ligand 1 signals in cancer cells. Nat Rev Cancer, 2022. 22 (3): p. 174-189. Salome, B., et al., NKG2A and HLA-E define an alternative immune checkpoint axis in bladder cancer. Cancer Cell, 2022. 40 (9): p. 1027-1043 e9. Ayers, M., et al., IFN-gamma-related mRNA profile predicts clinical response to PD-1 blockade. J Clin Invest, 2017. 127 (8): p. 2930-2940. Nadal, R. and J. Bellmunt, Management of metastatic bladder cancer. Cancer Treat Rev, 2019. 76 : p. 10-21. Koshkin, V.S. and P. Grivas, Emerging Role of Immunotherapy in Advanced Urothelial Carcinoma. Curr Oncol Rep, 2018. 20 (6): p. 48. Mariathasan, S., et al., TGFbeta attenuates tumour response to PD-L1 blockade by contributing to exclusion of T cells. Nature, 2018. 554 (7693): p. 544-548. Haabeth, O.A.W., et al., CD4(+) T-cell-Mediated Rejection of MHC Class II-Positive Tumor Cells Is Dependent on Antigen Secretion and Indirect Presentation on Host APCs. Cancer Res, 2018. 78 (16): p. 4573-4585. Mortara, L., et al., Anti-cancer Therapies Employing IL-2 Cytokine Tumor Targeting: Contribution of Innate, Adaptive and Immunosuppressive Cells in the Anti-tumor Efficacy. Front Immunol, 2018. 9 : p. 2905. Thapa, B., et al., OX40/OX40 ligand and its role in precision immune oncology. Cancer and Metastasis Reviews, 2024. 43 (3): p. 1001-1013. Croft, M., et al., The significance of OX40 and OX40L to T-cell biology and immune disease. Immunol Rev, 2009. 229 (1): p. 173-91. Fu, Y., et al., Therapeutic strategies for the costimulatory molecule OX40 in T-cell-mediated immunity. Acta Pharm Sin B, 2020. 10 (3): p. 414-433. Múnera-Rodríguez, A.M., et al., Sulforaphane-mediated immune regulation through inhibition of NF-kB and MAPK signaling pathways in human dendritic cells. Biomedicine & Pharmacotherapy, 2024. 177 . Pan, Y., et al., The protective and pathogenic role of Th17 cell plasticity and function in the tumor microenvironment. Front Immunol, 2023. 14 : p. 1192303. Vang, R., et al., Molecular Alterations of TP53 are a Defining Feature of Ovarian High-Grade Serous Carcinoma: A Rereview of Cases Lacking TP53 Mutations in The Cancer Genome Atlas Ovarian Study. Int J Gynecol Pathol, 2016. 35 (1): p. 48-55. Zhang, Z., et al., TP53 Mutation Infers a Poor Prognosis and Is Correlated to Immunocytes Infiltration in Breast Cancer. Front Cell Dev Biol, 2021. 9 : p. 759154. Zhang, H., et al., Novel Role of p53 in Septic Immunosuppression: Involvement in Loss and Dysfunction of CD4+ T Lymphocytes. Cell Physiol Biochem, 2018. 51 (1): p. 452-469. Guo, G., et al., Local Activation of p53 in the Tumor Microenvironment Overcomes Immune Suppression and Enhances Antitumor Immunity. Cancer Res, 2017. 77 (9): p. 2292-2305. Helleday, T., S. Eshtad, and S. Nik-Zainal, Mechanisms underlying mutational signatures in human cancers. Nat Rev Genet, 2014. 15 (9): p. 585-98. Urano, D., et al., Endocytosis of the seven-transmembrane RGS1 protein activates G-protein-coupled signalling in Arabidopsis. Nature Cell Biology, 2012. 14 (10): p. 1079-1088. Cheng, X., et al., Systematic Pan-Cancer Analysis of KLRB1 with Prognostic Value and Immunological Activity across Human Tumors. J Immunol Res, 2022. 2022 : p. 5254911. Zhou, W., et al., Targeting CXCL12/CXCR4 Axis in Tumor Immunotherapy. Current Medicinal Chemistry, 2019. 26 (17): p. 3026-3041. Chaves-Pozo, E., et al., Fish Granzyme A Shows a Greater Role Than Granzyme B in Fish Innate Cell-Mediated Cytotoxicity. Front Immunol, 2019. 10 : p. 2579. Shenoy, A.R., et al., GBP5 Promotes NLRP3 Inflammasome Assembly and Immunity in Mammals. Science, 2012. 336 (6080): p. 481-485. Connolly, D., et al., Septin roles in tumorigenesis. Biol Chem, 2011. 392 (8-9): p. 725-38. Skundric, D.S., et al., Emerging role of IL-16 in cytokine-mediated regulation of multiple sclerosis. Cytokine, 2015. 75 (2): p. 234-48. Sakaguchi, Shimon et al., Regulatory T Cells and Human Disease . Annual review of immunology. 2020(38): p. 541-566. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 10 Mar, 2026 Reviews received at journal 28 Feb, 2026 Reviewers agreed at journal 25 Feb, 2026 Reviewers invited by journal 30 Jan, 2026 Editor assigned by journal 28 Jan, 2026 Submission checks completed at journal 28 Jan, 2026 First submitted to journal 21 Jan, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8657603","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":583098856,"identity":"4628aa1a-a5b4-4f7a-8739-de9adc591dd1","order_by":0,"name":"Xiaoliang Dou","email":"","orcid":"","institution":"Xi'an Jiaotong University","correspondingAuthor":false,"prefix":"","firstName":"Xiaoliang","middleName":"","lastName":"Dou","suffix":""},{"id":583098857,"identity":"1eed0135-7f44-4904-a75f-406fd92263a0","order_by":1,"name":"Chaochao Cui","email":"","orcid":"","institution":"Northwest University","correspondingAuthor":false,"prefix":"","firstName":"Chaochao","middleName":"","lastName":"Cui","suffix":""},{"id":583098858,"identity":"78bfba5f-0adb-4a08-aca4-2e5fd16dc1fa","order_by":2,"name":"Yaodong zhang","email":"","orcid":"","institution":"Xi'an Jiaotong University","correspondingAuthor":false,"prefix":"","firstName":"Yaodong","middleName":"","lastName":"zhang","suffix":""},{"id":583098859,"identity":"3abcd1d0-e015-4f1d-814e-551038a7020b","order_by":3,"name":"Dali He","email":"","orcid":"","institution":"Xi'an Jiaotong University","correspondingAuthor":false,"prefix":"","firstName":"Dali","middleName":"","lastName":"He","suffix":""},{"id":583098860,"identity":"04792a53-64cb-4a5d-8c75-aec14ea1fb9b","order_by":4,"name":"Tianci Mao","email":"","orcid":"","institution":"Northwest University","correspondingAuthor":false,"prefix":"","firstName":"Tianci","middleName":"","lastName":"Mao","suffix":""},{"id":583098861,"identity":"134264ff-d2b4-44c4-8a19-5b66fdb1ca0c","order_by":5,"name":"Tao Li","email":"","orcid":"","institution":"Northwest University","correspondingAuthor":false,"prefix":"","firstName":"Tao","middleName":"","lastName":"Li","suffix":""},{"id":583098862,"identity":"50900ff8-945b-4189-b0a0-d5a6632e12b9","order_by":6,"name":"Long Gao","email":"","orcid":"","institution":"Northwest University","correspondingAuthor":false,"prefix":"","firstName":"Long","middleName":"","lastName":"Gao","suffix":""},{"id":583098863,"identity":"c322d9b0-3a95-4c36-8846-b5c1e78f1eed","order_by":7,"name":"Jinhai Fan","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAzUlEQVRIiWNgGAWjYBACNvb+BwcSKmp42NibDxCnhY/nDOODB2eOyfDxHEsgToucRA6z4cM2Zhsgw4BIhzHkHpNIOMPGwyaR8/HGGwY7Od0GglrOpUkkVMjwsPG83Ww5hyHZ2OwAIS2MDWYQW9hzt0nzMBxI3EZQCzODmURiGzMPG0POMyK1sPEYG4C1cOSwEamFhy3xQcKZY0C/HDO2nGNAhF/k5z8+cPBHRY29fHvzwxtvKuzkCGpBARI8REYNshZSdYyCUTAKRsGIAAD7mz1vjeDiOQAAAABJRU5ErkJggg==","orcid":"","institution":"Xi'an Jiaotong University","correspondingAuthor":true,"prefix":"","firstName":"Jinhai","middleName":"","lastName":"Fan","suffix":""}],"badges":[],"createdAt":"2026-01-21 09:20:11","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8657603/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8657603/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":101789060,"identity":"b6bf3e89-3786-4ab7-ad33-418a99f2cbf6","added_by":"auto","created_at":"2026-02-03 15:56:02","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":435340,"visible":true,"origin":"","legend":"\u003cp\u003eSingle-cell, bulk transcriptome, and convolutional cell survival and grouping analyses of BLCA patients. A: Dimensionality reduction and clustering of BLCA single-cell data. B: Expression levels of known cell type markers in various cell categories. C: Final classification of cell subgroups. D and E: scPagwas results indicating a significant correlation between TNFRSF4/CD4+ T cells and BLCA. F-H: Convolutional cells obtained through gene differential analysis of bulk transcriptome data followed by Bayesian deconvolution. I-K: Prognostic analysis of convolutional cells. L: Gene differential analysis between high and low levels of TNFRSF4/CD4+ T cells. M: Analysis of cellular composition and immune infiltration between high and low levels of TNFRSF4/CD4+ T cells.\u003c/p\u003e","description":"","filename":"Picture1.png","url":"https://assets-eu.researchsquare.com/files/rs-8657603/v1/509720018fac5cf78ff8f258.png"},{"id":101789082,"identity":"b1801a02-49ca-4869-a6e7-df28b41df6f6","added_by":"auto","created_at":"2026-02-03 15:56:07","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":404734,"visible":true,"origin":"","legend":"\u003cp\u003eAnalysis of immune-related gene sets and immunotherapy targets. A-C: Significant differences were observed in the expression levels of genes related to the TCR signaling pathway, TNF family members, and TNF family members receptors between high and low TNFRSF4/CD4+ T cell groups. D: Expression levels of multiple immunotherapy targets were significantly different between high and low TNFRSF4/CD4+ T cell groups.\u003c/p\u003e","description":"","filename":"Picture2.png","url":"https://assets-eu.researchsquare.com/files/rs-8657603/v1/469439cb79e962d1eb886329.png"},{"id":101789050,"identity":"0d70db85-4783-4b52-9ab8-c62957014e49","added_by":"auto","created_at":"2026-02-03 15:55:59","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":241343,"visible":true,"origin":"","legend":"\u003cp\u003eDrug sensitivity analysis. Significant differences in response levels to multiple tumor treatment drugs were observed between high and low levels of TNFRSF4/CD4+ T cells.\u003c/p\u003e","description":"","filename":"Picture3.png","url":"https://assets-eu.researchsquare.com/files/rs-8657603/v1/2a1f9168ebed531d9955726f.png"},{"id":101789171,"identity":"62b8829c-2422-4192-a467-cc36766c673d","added_by":"auto","created_at":"2026-02-03 15:56:23","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":346524,"visible":true,"origin":"","legend":"\u003cp\u003eSNV analysis of TNFRSF4/CD4+ T cells. A: Global visualization of SNVs for the top 30 mutated genes in TNFRSF4/CD4+ T cells. B: Summary of mutation types in TNFRSF4/CD4+ T cells. C: Differential analysis of mutated genes between high and low TNFRSF4/CD4+ T cell level groups.\u003c/p\u003e","description":"","filename":"Picture4.png","url":"https://assets-eu.researchsquare.com/files/rs-8657603/v1/926b6d834ff9b2a6f2c422d2.png"},{"id":101789054,"identity":"a5a5daa9-cdc7-4229-a43e-0421676afa30","added_by":"auto","created_at":"2026-02-03 15:56:01","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":454415,"visible":true,"origin":"","legend":"\u003cp\u003eWGCNA analysis and core gene selection. A: Module gene selection through WGCNA analysis. B: Identification of 11 core genes by intersecting four gene sets. C and D: GO and KEGG analyses of the core genes. E: Survival analysis of the core genes.\u003c/p\u003e","description":"","filename":"Picture5.png","url":"https://assets-eu.researchsquare.com/files/rs-8657603/v1/8c361b93147b8f0f5db60b55.png"},{"id":101789077,"identity":"fa96f1a3-c861-4a00-92d7-93cf367f1b19","added_by":"auto","created_at":"2026-02-03 15:56:05","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":222921,"visible":true,"origin":"","legend":"\u003cp\u003eValidation of the association between core genes and disease using SMR\u003c/p\u003e","description":"","filename":"Picture6.png","url":"https://assets-eu.researchsquare.com/files/rs-8657603/v1/42fc02c78477f503739a2814.png"},{"id":101789081,"identity":"ea07f1f4-ffbe-4348-b491-59a148ba30e4","added_by":"auto","created_at":"2026-02-03 15:56:06","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":192672,"visible":true,"origin":"","legend":"\u003cp\u003eConstruction of a neural network-based prognostic model. A: Visualisation of the model structure. B: SHAP analysis reveals that GZMM has the highest importance for the prognostic model. C: Kaplan-Meier curve analysis demonstrates a significant difference in survival time between the high and low RiskScore groups, indicating that the high RiskScore group has a shorter survival period.\u003c/p\u003e","description":"","filename":"Picture7.png","url":"https://assets-eu.researchsquare.com/files/rs-8657603/v1/716c68ef72148ccbd2fd5f5a.png"},{"id":101789065,"identity":"1a01b1dd-d959-4743-a9a6-f1592ecb089a","added_by":"auto","created_at":"2026-02-03 15:56:02","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":305350,"visible":true,"origin":"","legend":"\u003cp\u003eElevated SEPTIN1 and KLRB1 expression in TNFRSF4/CD4+ T cells of BLCA patients with favorable prognosis detected by multiplex immunofluorescence (nuclei: blue, TNFRSF4: red, CD4: purple, SEPTIN1/KLRB1: green).\u003c/p\u003e","description":"","filename":"Picture8.png","url":"https://assets-eu.researchsquare.com/files/rs-8657603/v1/9a5e3a491ee411e42762e211.png"},{"id":101789084,"identity":"f4f26e83-22ca-4393-b938-79550db1d352","added_by":"auto","created_at":"2026-02-03 15:56:07","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":268560,"visible":true,"origin":"","legend":"\u003cp\u003eTENT5C and FCMR expression in TNFRSF4/CD4+ T cells of BLCA patients with favorable prognosis detected by multiplex immunofluorescence (nuclei: blue, TNFRSF4: red, CD4: purple, TENT5C/FCMR: green).\u003c/p\u003e","description":"","filename":"Picture9.png","url":"https://assets-eu.researchsquare.com/files/rs-8657603/v1/184bf7b8e0398ae46eb8a58e.png"},{"id":101789237,"identity":"b5dbca57-5ea6-464e-a168-71c735aa94a8","added_by":"auto","created_at":"2026-02-03 15:56:36","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3604132,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8657603/v1/a136748e-7b8e-47d1-b77b-8827de8e9db0.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Construction of a Novel T Cell Subtype with Immunotherapeutic Value in Bladder Cancer and a Neural Network-Based Prognostic Model","fulltext":[{"header":"Introduction","content":"\u003cp\u003eBladder cancer ranks as the second most prevalent urological malignancy worldwide, with approximately 549,000 new cases and nearly 200,000 deaths reported annually[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. Early detection, especially before muscle invasion, allows for effective treatment with minimal impact on survival. The disease manifests in a broad spectrum, from recurrent noninvasive tumors to aggressive, advanced stages requiring multimodal treatment approaches. This significant burden, both in terms of public health and socioeconomic impact, underscores the critical importance of early detection, diagnosis, and treatment[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. And building prognostic models for bladder cancer play a key role in personalizing treatment by stratifying patients based on risk, optimizing therapeutic strategies, and ultimately improving clinical outcomes.\u003c/p\u003e \u003cp\u003eT cells play a pivotal role in the immune response against bladder cancer, particularly in tumor recognition and elimination. CD8\u0026thinsp;+\u0026thinsp;cytotoxic T cells are central to the anti-tumor immune response, as they directly kill tumor cells by recognizing tumor antigens and releasing cytolytic molecules such as perforin and granzyme[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Their presence in the tumor microenvironment has been correlated with better responses to immune checkpoint inhibitors (CPIs) targeting PD-1 or PD-L1, making them a key focus in immunotherapy research[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. However, in certain cancers like bladder cancer, CPIs as stand-alone treatments have shown limited effectiveness, with only about 20% of patients experiencing a positive response[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. This limited success may be partly attributed to the heterogeneous nature of tumor-infiltrating lymphocytes (TILs) and their varying abilities to mount an effective immune response. Research has primarily concentrated on cytotoxic CD8\u0026thinsp;+\u0026thinsp;T cells to investigate how immunotherapies, particularly checkpoint inhibitors (CPIs), stimulate anti-tumor immune responses. In melanoma, specific gene expression patterns and chromatin state markers linked to cytotoxic activity and T cell exhaustion have shown a strong correlation with patient outcomes following PD-1-targeted therapies[\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Additionally, the presence of CD8\u0026thinsp;+\u0026thinsp;T cells at the invasive margins of tumors before treatment has been associated with improved therapeutic responses[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. However, in metastatic bladder cancers, response rates to PD-1 inhibitors remain modest, with approximately 15%\u0026ndash;20% of patients who have progressed on platinum-based chemotherapy responding, and slightly higher rates seen in patients ineligible for platinum therapy[\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Predictive biomarkers such as PD-L1 expression have yet to provide consistent clarity in these cases[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Recent studies using bulk RNA sequencing have uncovered pre-treatment gene signatures\u0026mdash;higher CD8\u0026thinsp;+\u0026thinsp;gene expression, increased tumor mutational burden, and lower transforming growth factor-beta (TGF-b) signatures, particularly in immune-excluded tumors\u0026mdash;that are associated with better responses to anti-PD-L1 therapy, such as atezolizumab[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. While these findings shed light on potential mechanisms of response, more research is required to deepen our understanding of CD8\u0026thinsp;+\u0026thinsp;T cell involvement in CPI efficacy, as well as the roles of other immune cell subsets beyond the conventional cytotoxic and exhausted CD8\u0026thinsp;+\u0026thinsp;populations.\u003c/p\u003e \u003cp\u003eConversely, the role of CD4\u0026thinsp;+\u0026thinsp;T cells in bladder cancer immunity is far less understood. Studies have highlighted the crucial role of distinguishing CD4\u0026thinsp;+\u0026thinsp;T cell subtypes in immunotherapy, as they exhibit multiple tumor-specific states, particularly distinct subsets of regulatory T cells (Tregs). Notably, the research also uncovered cytotoxic CD4\u0026thinsp;+\u0026thinsp;T cells with the ability to kill tumors through MHC class II-dependent mechanisms[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. These cytotoxic CD4\u0026thinsp;+\u0026thinsp;T cells undergo clonal expansion in bladder tumors, likely driven by their recognition of specific tumor antigens. In vitro experiments confirmed their capacity to kill autologous tumors, and their anti-tumor activity is further enhanced by the secretion of cytokines such as TNF-α and IFN-γ[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. These findings highlight the complexity and diversity of CD4\u0026thinsp;+\u0026thinsp;T cell populations within tumors and suggest that leveraging this heterogeneity could enhance the effectiveness of immunotherapies in bladder cancer.\u003c/p\u003e \u003cp\u003eTNFRSF4 (OX40/CD134), a member of the tumor necrosis factor receptor superfamily (TNFRSF), plays a key role in T cell activation, survival, and memory formation, particularly within the tumor microenvironment. Following T cell receptor (TCR) signaling, TNFRSF4 is upregulated on CD4\u0026thinsp;+\u0026thinsp;and CD8\u0026thinsp;+\u0026thinsp;T cells, where it enhances their function by promoting sustained activation and preventing apoptosis[\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. This receptor, upon binding with its ligand OX40L, amplifies T cell responses, which is especially critical in chronic infections and cancers, where T cell exhaustion often diminishes immune efficacy[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. Additionally, TNFRSF4 modulates the activity of regulatory T cells (Tregs), balancing immune suppression and activation depending on the context[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. Despite TNFRSF4's well-established role in enhancing anti-tumor immunity in other cancers, its specific role in bladder cancer has yet to be investigated. This lack of research creates a significant gap in our understanding of its therapeutic potential in this particular cancer type.\u003c/p\u003e \u003cp\u003eIn this study, we employed a comprehensive multi-omics approach to investigate the immune landscape of bladder cancer, with a particular focus on TNFRSF4/CD4\u0026thinsp;+\u0026thinsp;T cells. Utilizing single-cell RNA sequencing data from bladder cancer patients and bulk transcriptome data from the TCGA-BLCA cohort, we conducted differential gene expression analysis, Bayesian deconvolution, and survival analysis to identify significant associations between TNFRSF4/CD4\u0026thinsp;+\u0026thinsp;T cells and bladder cancer prognosis. Furthermore, we explored the immunotherapeutic relevance of TNFRSF4/CD4\u0026thinsp;+\u0026thinsp;T cells by analyzing immune-related gene sets and key immune checkpoint markers. Additionally, we constructed a neural network-based prognostic model incorporating 11 core genes identified through WGCNA and survival analysis. The model demonstrated high predictive accuracy, with GZMM being the most important gene contributing to the model\u0026rsquo;s performance. This neural network model provided valuable insights into patient risk stratification and could aid in clinical decision-making for bladder cancer treatment.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eData availability statement\u003c/h2\u003e \u003cp\u003eThe bulk transcriptome data of bladder cancer (BLCA) patients were retrieved from the TCGA database. The TCGAbiolinks package facilitated the acquisition of gene count daa, simple nucleotide variation (SNV) data, and clinical information for TCGA-BLCA. Additionally, single-cell transcriptome data of BLCA (bladder urothelial carcinoma) patients were obtained from the GEO database (GSE222315). Genome-wide association study (GWAS) data for BLCA patients were accessed through the IEU Open GWAS Project (Dataset: ieu-b-4874).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eSingle-cell and bulk transcriptomics analysis integrated with GWAS Data\u003c/h3\u003e\n\u003cp\u003eThe analytical procedures in this research were executed using R and R Studio software. The Seurat and SingleR packages were utilized for the preprocessing and clustering of single-cell data, whereas the DESeq2 package facilitated the differential gene expression analysis for TCGA-BLCA bulk transcriptome data. Bayesian deconvolution, which involved the integration of single-cell and bulk transcriptome data, was carried out using the BayesPrism package. For processing GWAS data and pinpointing disease-associated genes and relevant cell populations at the single-cell level, the gwasvcf and scPagwas packages were predominantly employed. Particularly, the scPagwas package was applied to identify genes and cell populations related to specific traits.\u003c/p\u003e\n\u003ch3\u003eAnalysis of convolutional cell correlations\u003c/h3\u003e\n\u003cp\u003eThe survival analysis of convolutional cells was predominantly carried out using the survival package. Immune infiltration and differential analyses between groups were largely performed employing the tidyestimate and DESeq2 packages. The investigation of immune-related gene sets within convolutional cells was primarily executed utilizing the limma package. Immunotherapy analysis was chiefly accomplished through the Tidepy package (which necessitates Python environment configuration for use within R Studio). Drug response prediction was primarily achieved using the oncoPredict package. SNV analysis was largely conducted using the Maftools package, and WGCNA analysis was implemented using the WGCNA package.\u003c/p\u003e\n\u003ch3\u003eConstruction of a prognostic model and analysis of core genes\u003c/h3\u003e\n\u003cp\u003eThe Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses of core genes were predominantly conducted utilizing the clusterProfiler package, while prognostic assessments were performed through the survival and survminer packages. The Summary-data-based Mendelian Randomization (SMR) analysis was executed using the smr-1.3.1 software. The development of the neural network-based prognostic model was primarily achieved by employing the reticulate package to invoke Python. Within the Python environment, the Pytorch package was essential for model construction, while the sharp package was integral for providing interpretability analysis of the model. The produced onnx model file was employed to visualize the model structure (NETRON: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://netron.app/\u003c/span\u003e\u003cspan address=\"https://netron.app/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e\n\u003ch3\u003eMultiplex immunofluorescence staining\u003c/h3\u003e\n\u003cp\u003eFollowing routine processing of fixation, embedding, and sectioning, sequential multiplex immunofluorescence staining was performed according to standard protocols. The staining was carried out using the following fluorophores: FGNA for TNFRSF4, CY5 for CD4, FBNA for SEPTIN1/KLRB1/TENT5C/FCMR, and DAPI for nuclei.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eEthics statement\u003c/h2\u003e \u003cp\u003eThis study involving human tumor tissue samples was conducted in accordance with the ethical principles of the Declaration of Helsinki. The collection and use of all samples from bladder cancer (BLCA) patients were approved by the Hospital Ethics Committee (Approval Reference: ChiCTR2500112525). Written informed consent was obtained from all participants prior to sample collection.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003ch3\u003eTNFRSF4/CD4+ T cells are significantly associated with BLCA\u003c/h3\u003e\n\u003cp\u003eTo explore immune cells in BLCA at the single-cell level, single-cell sequencing data from 9 BLCA patients were analyzed. Following preprocessing, principal component analysis and clustering were applied to the single-cell data (Figure 1A), resulting in the identification of 27 distinct cell subgroups. These subgroups were annotated using the singleR package, and the proportions, as well as expression levels of markers for recognized cell types, were determined within these 29 subgroups (Figure 1B). Ultimately, 11 key cell subgroups were identified (Figure 1C). Notably, CD4+ T and CD8+ T cells were further categorized into three distinct types based on the high expression levels of specific genes (min.pct = 0.25, logfc.threshold = 0.25): TNFRSF4/CD4+ T cell, MT-ATP6/CD4+ T cell, and CCL4L2/CD8+ T cell. A trait-related (BLCA) scoring of these 11 subgroups was performed, yielding trait-related scores (TRS) and identifying trait-associated genes (Figures 1D and E). The findings demonstrated that TNFRSF4/CD4+ T cells were significantly associated with the trait (\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001). In preparation for Bayesian deconvolution using bulk transcriptome data, a separate analysis of bulk transcriptome data was conducted. Differential gene expression analysis identified 14,932 genes with significant differences (padj \u0026lt; 0.05) between the BLCA group and the normal control group, with 9,691 genes upregulated and 5,241 downregulated (Figures 1F and G). During Bayesian deconvolution, protein-coding genes were determined to be the most consistent gene category across both datasets (R = 0.636, mean squared error = 3.29, Figure 1H). Consequently, deconvolution focused exclusively on protein-coding genes, which proved more valuable and expedited the process. The posterior mean of the cell type fraction theta was extracted from bulk transcriptome data. Prognostic analysis revealed significant survival differences between patients with high and low levels of TNFRSF4/CD4+ T cells (\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001), with longer survival observed in the high-level group. Additionally, notable survival differences were detected between patients with high and low levels of B cells and smooth muscle cells (\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.01, Figures 1I-K).\u003c/p\u003e\n\u003ch3\u003eTNFRSF4/CD4+ T cells possess therapeutic value in immunotherapy\u003c/h3\u003e\n\u003cp\u003eTo further assess the immunotherapeutic potential of TNFRSF4/CD4+ T cells, a differential gene expression analysis was initially performed between groups with high and low levels of the specified convolutional cells (TNFRSF4/CD4+ T cells). This investigation identified 10,631 differentially expressed genes (DEGs, padj \u0026lt; 0.05, Figure 1L), alongside notable differences across various cell types, especially in immune cells such as NK cells and B cells. Furthermore, the immune infiltration scores (estimate scores) between the two groups exhibited significant variation, particularly in the immune score (Figure 1M). In the analysis of immune-related gene sets, numerous significant discrepancies were observed between high and low TNFRSF4/CD4+ T cell levels, with prominent differences in the T cell receptor (TCR) signaling pathway, tumor necrosis factor (TNF) family members, and their corresponding receptors (Figure 2A-C). Similarly, immunotherapy analysis unveiled significant differences in the expression levels of various immunotherapy targets, including CAF, CD8, CD274, and IFNG, between high and low TNFRSF4/CD4+ T cell levels (Figure 2D). Additionally, drug sensitivity analysis highlighted significant differences in the response levels to multiple tumor treatment drugs between the two groups (Figure 3).\u003c/p\u003e\n\u003ch3\u003eSNVs analysis of TNFRSF4/CD4+ T cells\u003c/h3\u003e\n\u003cp\u003eTo further investigate the gene mutations in TNFRSF4/CD4+ T cells, a global overview of the SNVs for the top 30 mutated genes was generated, leading to the construction of a gene waterfall plot (Figure 4A). TP53 was identified as having the highest mutation rate, with missense mutations representing the largest proportion. A statistical examination of transitions and transversions indicated that C \u0026gt; T pyrimidine transitions occurred most frequently (Figure 4B). Furthermore, an analysis comparing differentially mutated genes between high and low TNFRSF4/CD4+ T cell levels revealed that E300 exhibited the most significant differential mutation rate, with missense mutations being the predominant type of alteration.\u003c/p\u003e\n\u003ch3\u003eSelection of core genes via weighted gene co-expression network analysis (WGCNA)\u003c/h3\u003e\n\u003cp\u003eTo identify core genes associated with BLCA and the specific convolutional cell (TNFRSF4/CD4+ T cell), WGCNA was utilized to select 169 genes from the MEpink module. Subsequently, the overlap of significant DEGs identified in bulk transcriptome analysis, marker genes of TNFRSF4/CD4+ T cells from single-cell transcriptome analysis, and trait-associated genes detected by scPagwas was determined, yielding 11 core genes (RGS1, KLRB1, CXCR4, GYPC, GBP5, FCMR, IL16, SEPTIN1, TENT5C, GZMM, LTB). Functional enrichment analyses, including GO and KEGG, were conducted on these core genes. The GO analysis highlighted functions such as immune system processes and CD4 receptor binding as the most significant, while KEGG analysis revealed that pathways, including the NF-kappa B signaling pathway, were most prominent. Additionally, survival analysis was performed on the core genes, demonstrating a significant difference in patient survival between high and low expression groups of SEPTIN1 (surdiff.P \u0026lt; 0.05), with longer survival observed in the high-expression group. Furthermore, SEPTIN1, KLRB1, TENT5C, and FCMR were significant in the Cox proportional hazards model (Cox.P \u0026lt; 0.05), where higher expression levels indicated a better prognosis (HR \u0026lt; 1).\u003c/p\u003e\n\u003ch3\u003eValidation of core genes\u0026rsquo; association using SMR\u003c/h3\u003e\n\u003cp\u003eTo further validate the association between core genes and BLCA, SMR was employed to assess whether significant causal associations existed between these genes and the disease. The findings revealed that within a 1,000 Kb window centered on the core gene GYPC, a gene significantly associated with BLCA, MYO7B (pSMR \u0026lt; 0.05), was identified.\u003c/p\u003e\n\u003ch3\u003eConstruction of a prognosis model based on neural networks\u003c/h3\u003e\n\u003cp\u003eTo maximize the prognostic utility of core genes, a neural network-based BLCA prognosis model was constructed. Initially, 11 core genes were incorporated, and various model architectures were evaluated. Following multiple parameter optimizations, an optimal structure was identified (Figure 7A). The influence and significance of model features, particularly gene expression levels, were assessed, with GZMM showing the highest level of importance (Figure 7B). Patients were then stratified into two groups according to the median RiskScore derived from the model. A marked difference in survival times was observed between the high and low RiskScore groups (\u003cem\u003ep\u003c/em\u003e\u003cem\u003e\u0026nbsp;\u003c/em\u003e\u0026lt; 0.001), with shorter survival periods noted in the high RiskScore group (Figure 7C). These findings indicate that this prognostic model provides valuable insight for clinical decision-making.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSEPTIN1 and KLRB1 in TNFRSF4/CD4+ T Cells as Prognostic Biomarkers in BLCA\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo investigate the association between the infiltration of identified core genes within TNFRSF4/CD4+ T cells in BLCA tissues and patient prognosis, multiplex immunofluorescence staining was performed on tumor specimens from 13 BLCA patients. The expression levels of SEPTIN1, KLRB1, TENT5C, and FCMR were evaluated specifically in TNFRSF4/CD4+ T cells.\u003c/p\u003e\n\u003cp\u003eThe results revealed that in tumor tissues from five patients with favorable prognosis, SEPTIN1 and KLRB1 were highly expressed in TNFRSF4/CD4+ T cells ( Figure 8A, B), while TENT5C and FCMR showed moderate expression levels (Figure 9A, B). In contrast, in the eight patients with poorer prognosis, all four genes exhibited low expression within the same cell population. These findings indicate that high expression of the core genes SEPTIN1 and KLRB1 in TNFRSF4/CD4+ T cells is significantly associated with improved prognosis in BLCA patients.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this study, we identified TNFRSF4/CD4\u0026thinsp;+\u0026thinsp;T cells as a novel immune cell subtype significantly associated with bladder cancer, showing strong prognostic value. Using single-cell RNA sequencing and bulk transcriptome data, we revealed that higher levels of TNFRSF4/CD4\u0026thinsp;+\u0026thinsp;T cells were linked to improved survival, emphasizing their potential as a therapeutic target in immunotherapy. Additionally, we explored immune-related gene sets and found significant differences in pathways such as the TCR signaling pathway and TNF family members. Furthermore, through WGCNA analysis, we identified 11 core genes closely associated with TNFRSF4/CD4\u0026thinsp;+\u0026thinsp;T cells and bladder cancer prognosis. These genes include RGS1, KLRB1, CXCR4, GYPC, GBP5, FCMR, IL16, SEPTIN1, TENT5C, GZMM, and LTB. These core genes were enriched in key immune functions, such as immune system processes and CD4 receptor binding, and formed the basis for constructing a neural network-based prognostic model. The model demonstrated high accuracy in predicting patient survival outcomes, with GZMM identified as the most significant contributor to the model. This prognostic model highlights the potential for integrating TNFRSF4/CD4\u0026thinsp;+\u0026thinsp;T cells and the core genes into personalized treatment strategies for bladder cancer, offering new opportunities for improving clinical outcomes.\u003c/p\u003e \u003cp\u003eOur analysis of single-cell RNA sequencing (scRNA-seq) data revealed that TNFRSF4/CD4\u0026thinsp;+\u0026thinsp;T cells are significantly correlated with bladder cancer, suggesting their potential role in tumor progression and immune modulation. The identification of TNFRSF4/CD4\u0026thinsp;+\u0026thinsp;T cells as a distinct subtype, alongside other T cell populations such as MT-ATP6/CD4\u0026thinsp;+\u0026thinsp;T cells and CCL4L2/CD8\u0026thinsp;+\u0026thinsp;T cells, highlights the heterogeneity within the tumor immune microenvironment. TNFRSF4 (OX40) is known to enhance T cell activation, survival, and memory formation through its interaction with its ligand OX40L. This signaling pathway likely contributes to the prolonged survival and sustained activation of CD4\u0026thinsp;+\u0026thinsp;T cells within the tumor microenvironment, facilitating a more robust immune response against tumor cells[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. Our findings suggest that TNFRSF4/CD4\u0026thinsp;+\u0026thinsp;T cells may play a crucial role in modulating the immune response in bladder cancer through several mechanisms. First, the OX40-OX40L interaction is known to prevent T cell apoptosis and enhance their proliferation, which could contribute to sustained anti-tumor activity, particularly in chronic tumor settings where T cell exhaustion typically diminishes immune effectiveness[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. The presence of these cells may help maintain the immune system\u0026rsquo;s capacity to recognize and attack bladder cancer cells over time, reducing the likelihood of immune evasion by the tumor.\u003c/p\u003e \u003cp\u003eOur study demonstrates the crucial role of TNFRSF4/CD4\u0026thinsp;+\u0026thinsp;T cells in bladder cancer and their potential value in immunotherapy. We identified over 10,631 differentially expressed genes between high and low TNFRSF4/CD4\u0026thinsp;+\u0026thinsp;T cell groups, indicating a profound impact on the tumor immune environment. Patients with higher levels of these cells exhibited increased infiltration of immune cells, such as NK cells and B cells, alongside higher overall immune infiltration scores, suggesting a more robust immune presence. Additionally, TNFRSF4 interacts with OX40L to prevent T cell apoptosis and sustain activation, activating downstream pathways like NF-κB and MAPK that are essential for T cell proliferation and immune memory formation. This prolonged activation helps prevent T cell exhaustion, a major challenge in cancer immunotherapy, and enhances the balance between effector T cells and regulatory T cells (Tregs), reducing immune suppression within the tumor microenvironment[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Furthermore, significant differences in the expression of key immune pathways, such as the TCR signaling pathway and TNF family members, as well as immunotherapy targets like PD-L1 and IFNG, underscore the potential for TNFRSF4/CD4\u0026thinsp;+\u0026thinsp;T cells to improve responses to immune checkpoint inhibitors. The distinct differences in drug sensitivity also highlight the potential of these cells as biomarkers for predicting treatment outcomes, offering valuable insights for more personalized therapeutic approaches in bladder cancer. These findings collectively reinforce the importance of TNFRSF4/CD4\u0026thinsp;+\u0026thinsp;T cells as critical modulators of bladder cancer immunity and as promising targets for enhancing the efficacy of immunotherapy. However, the heterogeneity of T cells may contribute to inconsistent immunotherapy outcomes in some patients, and addressing this complexity remains an area for future research.\u003c/p\u003e \u003cp\u003eIn our genetic mutation analysis of TNFRSF4/CD4\u0026thinsp;+\u0026thinsp;T cells, TP53 was identified as the most frequently mutated gene, with a significant proportion of missense mutations. TP53 mutations are well-documented in bladder cancer and are known to disrupt its tumor-suppressive function, potentially contributing to uncontrolled cell growth and immune evasion. In bladder cancer specifically, TP53 mutations are associated with higher tumor grade, increased invasiveness, and poorer prognosis[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. The frequent mutation of TP53 in TNFRSF4/CD4\u0026thinsp;+\u0026thinsp;T cells suggests a possible link between this genetic alteration and impaired immune function, as TP53 plays a role in regulating the immune response by influencing apoptosis and cell cycle control[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. This disruption could lead to dysfunctional T cell activity, potentially hindering anti-tumor immunity[\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. The analysis also revealed a prevalence of C\u0026thinsp;\u0026gt;\u0026thinsp;T transitions, a common mutational signature in cancers, which could point to oxidative damage or other mutagenic processes affecting T cell functionality in the tumor microenvironment[\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. While these findings suggest that TP53 mutations may influence the function of TNFRSF4/CD4\u0026thinsp;+\u0026thinsp;T cells in bladder cancer, we did not directly examine their impact on T cell behavior or immune evasion. Future research should focus on understanding how these mutations affect immune modulation and contribute to therapeutic resistance in bladder cancer.\u003c/p\u003e \u003cp\u003eWe utilized WGCNA to identify core genes associated with bladder cancer and the specified TNFRSF4/CD4\u0026thinsp;+\u0026thinsp;T cells. By analyzing the MEpink module, we selected 169 genes, which were then intersected with differentially expressed genes from bulk transcriptome data, marker genes from single-cell RNA sequencing of TNFRSF4/CD4\u0026thinsp;+\u0026thinsp;T cells, and phenotype-associated genes identified by scPagwas. This approach yielded 11 core genes: RGS1, KLRB1, CXCR4, GYPC, GBP5, FCMR, IL16, SEPTIN1, TENT5C, GZMM, and LTB. These genes share common roles in immune regulation, cell signaling, and migration, which are essential for immune surveillance and tumor progression. Many, such as RGS1 and KLRB1, are involved in T cell activation and migration[\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e], while CXCR4 is specifically known for its role in cancer metastasis[\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. GZMM plays a key role in cytotoxicity[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e], and GBP5 contributes to antiviral responses[\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. Additionally, SEPTIN1 is linked to cellular structure and tumor stability, making these genes collectively significant in both immune response and bladder cancer progression[\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. These core genes were enriched in biological processes like the immune system process and CD4 receptor binding, and pathways such as NF-kappa B signaling in the KEGG analysis, further highlighting their role in immune regulation[\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. Survival analysis of these core genes revealed that SEPTIN1 was significantly associated with patient survival (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05), where higher expression was linked to longer survival. Additionally, KLRB1, TENT5C, and FCMR were significant in the COX proportional hazards model, suggesting their potential as prognostic markers, with higher expression correlating with better outcomes (HR\u0026thinsp;\u0026lt;\u0026thinsp;1). Supporting these findings, multiplex immunofluorescence analysis revealed high expression of both SEPTIN1 and FCMR within TNFRSF4/CD4\u0026thinsp;+\u0026thinsp;T cells in bladder cancer patients with favorable prognosis, underscoring the importance of these two genes as prognostic biomarkers.\u003c/p\u003e \u003cp\u003eFurthermore, we utilized SMR analysis to examine the association between core genes and bladder cancer. The analysis identified a significant association between MYO7B and bladder cancer (pSMR\u0026thinsp;\u0026lt;\u0026thinsp;0.05) within a 1000Kb window centered around the core gene GYPC. This suggests that while GYPC was highlighted as a core gene, MYO7B may also play a relevant role in the genetic landscape of bladder cancer within this region. The finding underscores the need for further investigation into MYO7B to explore its specific contribution to bladder cancer development, as it may share regulatory elements or be influenced by genetic variants affecting the broader region. This highlights the importance of considering the surrounding genetic context when evaluating disease-associated genes.\u003c/p\u003e \u003cp\u003eFinally, we developed a neural network-based prognostic model for bladder cancer, incorporating 11 core genes identified from our earlier analyses. After testing various model architectures and fine-tuning parameters, the final model achieved strong performance, with GZMM identified as the most influential gene in predicting patient outcomes. The model-generated RiskScore effectively stratified patients into high- and low-risk groups, with significant differences in survival between the two groups (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Patients in the high-risk group exhibited notably shorter survival times, emphasizing the model\u0026rsquo;s potential to serve as a valuable tool for clinical decision-making.The ability to classify patients based on genetic profiles and predict their survival outcomes with high accuracy can significantly improve the individualization of treatment strategies. This could allow clinicians to identify high-risk patients who may benefit from more aggressive therapies, while sparing low-risk patients from overtreatment.\u003c/p\u003e \u003cp\u003eThe antitumor immune functions of CD4\u0026thinsp;+\u0026thinsp;T cells have attracted growing attention in recent years, as exemplified by the 2025 Nobel Prize-awarded research on regulatory T cells (Tregs)[\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. The TNFRSF4/CD4\u0026thinsp;+\u0026thinsp;T-cell subset identified in this study likely contains antitumor-active effector T cells or functionally specialized Tregs, thus providing a novel direction for understanding the functional diversity of CD4\u0026thinsp;+\u0026thinsp;T cells in tumor immunity.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eConflict of interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eContributorship statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eD.X and C.C conceived research ideas and designed and performed experiments, analyzed data, and wrote the manuscript; C.C and Y.Z preformed sequencing data analysis; T.L, L.G, T.M, and D.H, assisted in completing the experiment; D.X and C.C provided clinical consultation, experimental design and guidance; J.F conceptualized and guided the research, interpreted the results, revised and edited the manuscript. All authors have read and approved the article.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding and acknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research was supported by the following funding bodies: [Shaanxi Province key research and development plan project, with grant numbers [No.S2024-YF-YBSF-1796].\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eComp\u0026eacute;rat, E., et al., \u003cem\u003eCurrent best practice for bladder cancer: a narrative review of diagnostics and treatments.\u003c/em\u003e The Lancet, 2022. \u003cstrong\u003e400\u003c/strong\u003e(10364): p. 1712-1721.\u003c/li\u003e\n\u003cli\u003eHashem, M., et al., \u003cem\u003eNon-coding RNA transcripts, incredible modulators of cisplatin chemo-resistance in bladder cancer through operating a broad spectrum of cellular processes and signaling mechanism.\u003c/em\u003e Non-coding RNA Research, 2024. \u003cstrong\u003e9\u003c/strong\u003e(2): p. 560-582.\u003c/li\u003e\n\u003cli\u003evan der Leun, A.M., D.S. Thommen, and T.N. Schumacher, \u003cem\u003eCD8(+) T cell states in human cancer: insights from single-cell analysis.\u003c/em\u003e Nat Rev Cancer, 2020. \u003cstrong\u003e20\u003c/strong\u003e(4): p. 218-232.\u003c/li\u003e\n\u003cli\u003eKornepati, A.V.R., R.K. Vadlamudi, and T.J. Curiel, \u003cem\u003eProgrammed death ligand 1 signals in cancer cells.\u003c/em\u003e Nat Rev Cancer, 2022. \u003cstrong\u003e22\u003c/strong\u003e(3): p. 174-189.\u003c/li\u003e\n\u003cli\u003eSalome, B., et al., \u003cem\u003eNKG2A and HLA-E define an alternative immune checkpoint axis in bladder cancer.\u003c/em\u003e Cancer Cell, 2022. \u003cstrong\u003e40\u003c/strong\u003e(9): p. 1027-1043 e9.\u003c/li\u003e\n\u003cli\u003eAyers, M., et al., \u003cem\u003eIFN-gamma-related mRNA profile predicts clinical response to PD-1 blockade.\u003c/em\u003e J Clin Invest, 2017. \u003cstrong\u003e127\u003c/strong\u003e(8): p. 2930-2940.\u003c/li\u003e\n\u003cli\u003eNadal, R. and J. Bellmunt, \u003cem\u003eManagement of metastatic bladder cancer.\u003c/em\u003e Cancer Treat Rev, 2019. \u003cstrong\u003e76\u003c/strong\u003e: p. 10-21.\u003c/li\u003e\n\u003cli\u003eKoshkin, V.S. and P. Grivas, \u003cem\u003eEmerging Role of Immunotherapy in Advanced Urothelial Carcinoma.\u003c/em\u003e Curr Oncol Rep, 2018. \u003cstrong\u003e20\u003c/strong\u003e(6): p. 48.\u003c/li\u003e\n\u003cli\u003eMariathasan, S., et al., \u003cem\u003eTGFbeta attenuates tumour response to PD-L1 blockade by contributing to exclusion of T cells.\u003c/em\u003e Nature, 2018. \u003cstrong\u003e554\u003c/strong\u003e(7693): p. 544-548.\u003c/li\u003e\n\u003cli\u003eHaabeth, O.A.W., et al., \u003cem\u003eCD4(+) T-cell-Mediated Rejection of MHC Class II-Positive Tumor Cells Is Dependent on Antigen Secretion and Indirect Presentation on Host APCs.\u003c/em\u003e Cancer Res, 2018. \u003cstrong\u003e78\u003c/strong\u003e(16): p. 4573-4585.\u003c/li\u003e\n\u003cli\u003eMortara, L., et al., \u003cem\u003eAnti-cancer Therapies Employing IL-2 Cytokine Tumor Targeting: Contribution of Innate, Adaptive and Immunosuppressive Cells in the Anti-tumor Efficacy.\u003c/em\u003e Front Immunol, 2018. \u003cstrong\u003e9\u003c/strong\u003e: p. 2905.\u003c/li\u003e\n\u003cli\u003eThapa, B., et al., \u003cem\u003eOX40/OX40 ligand and its role in precision immune oncology.\u003c/em\u003e Cancer and Metastasis Reviews, 2024. \u003cstrong\u003e43\u003c/strong\u003e(3): p. 1001-1013.\u003c/li\u003e\n\u003cli\u003eCroft, M., et al., \u003cem\u003eThe significance of OX40 and OX40L to T-cell biology and immune disease.\u003c/em\u003e Immunol Rev, 2009. \u003cstrong\u003e229\u003c/strong\u003e(1): p. 173-91.\u003c/li\u003e\n\u003cli\u003eFu, Y., et al., \u003cem\u003eTherapeutic strategies for the costimulatory molecule OX40 in T-cell-mediated immunity.\u003c/em\u003e Acta Pharm Sin B, 2020. \u003cstrong\u003e10\u003c/strong\u003e(3): p. 414-433.\u003c/li\u003e\n\u003cli\u003eM\u0026uacute;nera-Rodr\u0026iacute;guez, A.M., et al., \u003cem\u003eSulforaphane-mediated immune regulation through inhibition of NF-kB and MAPK signaling pathways in human dendritic cells.\u003c/em\u003e Biomedicine \u0026amp; Pharmacotherapy, 2024. \u003cstrong\u003e177\u003c/strong\u003e.\u003c/li\u003e\n\u003cli\u003ePan, Y., et al., \u003cem\u003eThe protective and pathogenic role of Th17 cell plasticity and function in the tumor microenvironment.\u003c/em\u003e Front Immunol, 2023. \u003cstrong\u003e14\u003c/strong\u003e: p. 1192303.\u003c/li\u003e\n\u003cli\u003eVang, R., et al., \u003cem\u003eMolecular Alterations of TP53 are a Defining Feature of Ovarian High-Grade Serous Carcinoma: A Rereview of Cases Lacking TP53 Mutations in The Cancer Genome Atlas Ovarian Study.\u003c/em\u003e Int J Gynecol Pathol, 2016. \u003cstrong\u003e35\u003c/strong\u003e(1): p. 48-55.\u003c/li\u003e\n\u003cli\u003eZhang, Z., et al., \u003cem\u003eTP53 Mutation Infers a Poor Prognosis and Is Correlated to Immunocytes Infiltration in Breast Cancer.\u003c/em\u003e Front Cell Dev Biol, 2021. \u003cstrong\u003e9\u003c/strong\u003e: p. 759154.\u003c/li\u003e\n\u003cli\u003eZhang, H., et al., \u003cem\u003eNovel Role of p53 in Septic Immunosuppression: Involvement in Loss and Dysfunction of CD4+ T Lymphocytes.\u003c/em\u003e Cell Physiol Biochem, 2018. \u003cstrong\u003e51\u003c/strong\u003e(1): p. 452-469.\u003c/li\u003e\n\u003cli\u003eGuo, G., et al., \u003cem\u003eLocal Activation of p53 in the Tumor Microenvironment Overcomes Immune Suppression and Enhances Antitumor Immunity.\u003c/em\u003e Cancer Res, 2017. \u003cstrong\u003e77\u003c/strong\u003e(9): p. 2292-2305.\u003c/li\u003e\n\u003cli\u003eHelleday, T., S. Eshtad, and S. Nik-Zainal, \u003cem\u003eMechanisms underlying mutational signatures in human cancers.\u003c/em\u003e Nat Rev Genet, 2014. \u003cstrong\u003e15\u003c/strong\u003e(9): p. 585-98.\u003c/li\u003e\n\u003cli\u003eUrano, D., et al., \u003cem\u003eEndocytosis of the seven-transmembrane RGS1 protein activates G-protein-coupled signalling in Arabidopsis.\u003c/em\u003e Nature Cell Biology, 2012. \u003cstrong\u003e14\u003c/strong\u003e(10): p. 1079-1088.\u003c/li\u003e\n\u003cli\u003eCheng, X., et al., \u003cem\u003eSystematic Pan-Cancer Analysis of KLRB1 with Prognostic Value and Immunological Activity across Human Tumors.\u003c/em\u003e J Immunol Res, 2022. \u003cstrong\u003e2022\u003c/strong\u003e: p. 5254911.\u003c/li\u003e\n\u003cli\u003eZhou, W., et al., \u003cem\u003eTargeting CXCL12/CXCR4 Axis in Tumor Immunotherapy.\u003c/em\u003e Current Medicinal Chemistry, 2019. \u003cstrong\u003e26\u003c/strong\u003e(17): p. 3026-3041.\u003c/li\u003e\n\u003cli\u003eChaves-Pozo, E., et al., \u003cem\u003eFish Granzyme A Shows a Greater Role Than Granzyme B in Fish Innate Cell-Mediated Cytotoxicity.\u003c/em\u003e Front Immunol, 2019. \u003cstrong\u003e10\u003c/strong\u003e: p. 2579.\u003c/li\u003e\n\u003cli\u003eShenoy, A.R., et al., \u003cem\u003eGBP5 Promotes NLRP3 Inflammasome Assembly and Immunity in Mammals.\u003c/em\u003e Science, 2012. \u003cstrong\u003e336\u003c/strong\u003e(6080): p. 481-485.\u003c/li\u003e\n\u003cli\u003eConnolly, D., et al., \u003cem\u003eSeptin roles in tumorigenesis.\u003c/em\u003e Biol Chem, 2011. \u003cstrong\u003e392\u003c/strong\u003e(8-9): p. 725-38.\u003c/li\u003e\n\u003cli\u003eSkundric, D.S., et al., \u003cem\u003eEmerging role of IL-16 in cytokine-mediated regulation of multiple sclerosis.\u003c/em\u003e Cytokine, 2015. \u003cstrong\u003e75\u003c/strong\u003e(2): p. 234-48.\u003c/li\u003e\n\u003cli\u003eSakaguchi, Shimon et al., \u003cem\u003eRegulatory T Cells and Human Disease\u003c/em\u003e. Annual review of immunology. 2020(38): p. 541-566. \u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"cancer-immunology-immunotherapy","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"ciim","sideBox":"Learn more about [Cancer Immunology, Immunotherapy](http://link.springer.com/journal/262)","snPcode":"262","submissionUrl":"https://submission.nature.com/new-submission/262/3","title":"Cancer Immunology, Immunotherapy","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"TNFRSF4/CD4+, T cell, Bladder cancer, SEPTIN1, FCMR","lastPublishedDoi":"10.21203/rs.3.rs-8657603/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8657603/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eBladder cancer (BLCA) remains a significant global health challenge, characterized by high heterogeneity and suboptimal responses to existing immunotherapies. This study identifies TNFRSF4/CD4+ T cells as a novel immune cell subtype with crucial prognostic and therapeutic relevance in BLCA. Leveraging single-cell RNA sequencing and bulk transcriptome data from TCGA-BLCA and GEO databases, we conducted multi-omics analyses to elucidate the immune landscape and its implications for clinical outcomes. Our findings reveal that higher levels of TNFRSF4/CD4+ T cells are associated with improved patient survival and robust immune activity, emphasizing their potential as biomarkers and therapeutic targets. Gene expression and pathway enrichment analyses highlight their involvement in immune-regulatory mechanisms, including the TCR signaling pathway and TNF family signaling. Additionally, we identified 11 core genes associated with TNFRSF4/CD4+ T cells, such as SEPTIN1 and FCMR, which significantly impact prognosis and immune function. Multiplex immunofluorescence demonstrated a strong positive correlation between the prognosis of bladder cancer patients and the expression of SEPTIN1 and FCMR within TNFRSF4/CD4+ T cells. These genes served as the foundation for constructing a neural network-based prognostic model, which demonstrated high predictive accuracy and stratified patients into distinct risk categories. This study underscores the critical role of TNFRSF4/CD4+ T cells in BLCA immunity and highlights their potential in advancing personalized treatment strategies, offering novel insights into improving immunotherapy efficacy in bladder cancer.\u003c/p\u003e","manuscriptTitle":"Construction of a Novel T Cell Subtype with Immunotherapeutic Value in Bladder Cancer and a Neural Network-Based Prognostic Model","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-03 15:54:45","doi":"10.21203/rs.3.rs-8657603/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-03-11T00:44:01+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-01T03:31:07+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"25852028694636506803632102863022094153","date":"2026-02-26T03:31:22+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-01-30T06:12:37+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-01-29T03:57:46+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-01-29T03:54:45+00:00","index":"","fulltext":""},{"type":"submitted","content":"Cancer Immunology, Immunotherapy","date":"2026-01-21T09:01:58+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"cancer-immunology-immunotherapy","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"ciim","sideBox":"Learn more about [Cancer Immunology, Immunotherapy](http://link.springer.com/journal/262)","snPcode":"262","submissionUrl":"https://submission.nature.com/new-submission/262/3","title":"Cancer Immunology, Immunotherapy","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"18bacbbc-f377-412b-98fb-75ef6aeefe86","owner":[],"postedDate":"February 3rd, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-05-06T04:38:50+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-03 15:54:45","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8657603","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8657603","identity":"rs-8657603","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}