Single‑cell and bulk RNA sequencing identifes T cell marker genes to predict the prognosis of ovrian caner

Single‑cell and bulk RNA sequencing identifes T cell marker genes to predict the prognosis of ovrian caner | 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 Help Center Sign In Submit a Preprint Cite Share Download PDF Article Single‑cell and bulk RNA sequencing identifes T cell marker genes to predict the prognosis of ovrian caner Hengzi Sun, Xiao Huo, Shuhong Li, Liyuan Guo This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4721266/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 21 You are reading this latest preprint version Abstract Background: Ovarian cancer, with high mortality and often late diagnosis, shows high recurrence despite treatment. The variable effectiveness of immunotherapy highlights the urgent need for personalized, advanced therapeutic strategies. Methods: To investigate T-cell marker genes, single-cell RNA-sequencing (scRNA-seq) data were sourced from the Gene Expression Omnibus (GEO) database. Additionally, bulk RNA-sequencing data along with clinical information from ovarian cancer patients were retrieved from the Cancer Genome Atlas (TCGA) database to establish a prognostic signature. This study involved survival analysis to evaluate associations between different risk groups, and explored cellular communication and relevant pathway analyses, including metabolic pathways. Results: We identified 41 genes showing varied expression between two T-cell subclusters, marking subcluster 0 with CCL5 and GZMA, and attributing the rest to subcluster 1. These markers delineate four prognostic groups within the TCGA OV dataset, with T-cluster 2 exhibiting the poorest survival, in contrast to T-cluster 3, which shows the best. Analysis suggests subcluster 1 T-cells might be dysfunctional, potentially exacerbating ovarian cancer progression. We also developed a T-cell scoring model using eight significant genes, showing improved survival in the low-score group. Moreover, cellular and metabolic pathway analyses underscored the importance of CCL, IL2 and MGMT pathways in these subclusters. Conclusions: The study identifies CCL-5 as a biomarker for T-cell subtypes in ovarian cancer using scRNA-seq and bulk RNA-seq data. A T-cell scoring model based on eight genes predicts survival and progression rates, independent of clinical features. This model could be a prognostic indicator and CCL-5 a potential immunotherapy target in ovarian cancer. Health sciences/Oncology/Cancer Health sciences/Biomarkers/Prognostic markers ovarian cancer T-cell marker genes Immunotherapy Single-cell RNA-sequencing Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction Ovarian cancer, known for its poor prognosis and high mortality rate, remains one of the most challenging gynecological cancers to treat effectively. The majority of patients are diagnosed at advanced stages, primarily due to the non-specific symptoms associated with the disease 1 . Current treatment protocols involve a combination of surgery and chemotherapy, which, while effective for some, do not significantly improve long-term survival for many patients. Recurrence rates are high, and the five-year survival rate remains under 50% for advanced-stage ovarian cancer patients. This grim outlook underscores the urgent need for improved prognostic tools and personalized treatment strategies 2 . Immunotherapy has revolutionized the treatment landscape for various cancers, highlighting the crucial role of the immune system in controlling tumor growth. Strategies such as checkpoint inhibitors, CAR-T cell therapy, and vaccines aim to enhance the immune system's ability to detect and destroy cancer cells 3 . The efficacy of these therapies, however, varies significantly among cancer types due to differences in immune infiltration and the tumor microenvironment. Research has increasingly focused on understanding the immunological underpinnings of each cancer to tailor immunotherapeutic approaches. For instance, the presence of immune checkpoints, which tumors exploit to evade immune surveillance, has led to the development of inhibitors that can reactivate T cells 4 . Additionally, the role of T cells in mediating anti-tumor responses has been a major focus, with efforts to characterize and modify T cell functionality to improve therapeutic efficacy 5 . This research underscores the potential of immunological strategies to provide durable responses and highlights the ongoing need to integrate immunotherapy with traditional treatments. Recent advances in understanding the immune evasion mechanisms employed by ovarian tumors have paved the way for novel immunotherapeutic approaches. It is now recognized that the tumor microenvironment (TME) of ovarian cancer is not only complex but also immunosuppressive, limiting the efficacy of conventional therapies 6 . Studies have shown that certain immunomodulatory molecules and T cells within the TME are crucial in determining the tumor's progression and response to treatments. Immunotherapy, particularly the use of immune checkpoint inhibitors that target PD-1 and CTLA-4, has shown promise in several clinical trials, suggesting a pivotal role of the immune system in controlling this malignancy 7 . However, the response rates vary, and a significant proportion of patients do not benefit from these therapies, highlighting the need for precise biomarkers that can predict immunotherapeutic response. Single-cell RNA sequencing (scRNA-seq) and bulk RNA sequencing have become indispensable tools in cancer research, offering profound insights into the molecular and cellular heterogeneity of tumors. scRNA-seq allows for the detailed analysis of individual cells within a tumor, revealing the existence of various cell types and their states, which are crucial for understanding the complexity of the tumor microenvironment. This level of detail is essential for identifying specific cell populations that may influence prognosis or response to therapy, such as different subsets of T cells 8 . Bulk RNA sequencing, on the other hand, provides a broader overview of gene expression within the entire tumor mass, useful for identifying overarching expression patterns and biomarkers 9 . Together, these technologies enable the development of comprehensive gene expression profiles that can predict disease outcomes and guide treatment decisions, thereby enhancing the precision of cancer therapy and prognostication. The present study aims to explore the correlation between the prognosis of ovarian cancer and its immune T cell markers through the analysis of single-cell and bulk RNA-seq data. Materials and Methods Data Preparation and Preprocessing Single-cell sequencing data sets GSE158722, GSE147082, and GSE130000, including expression matrices and associated clinical feature information, were retrieved from the NCBI Gene Expression Omnibus (GEO) database ( https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi? ). Specifically, the GSE158722 dataset comprised single-cell data from samples collected at three postoperative time points from five ovarian cancer (OV) patients, retaining only data processed by the 10x Genomics platform. Additionally, high-grade serous ovarian cancer RNA-seq data and corresponding clinical features were downloaded from The Cancer Genome Atlas (TCGA). After excluding samples with missing survival times and those with metastatic disease, a total of 373 samples with prognostic information were secured for further analysis. Cell Clustering Analysis Initially, single-cell data were filtered to include only genes expressed in at least five cells and cells that expressed at least 500 genes. The mitochondrial and ribosomal RNA content percentages were calculated using the Percentage Feature Set function, ensuring that each cell expressed over 500 genes and had mitochondrial content below 30%. For multi-sample analyses, data from different samples were merged and subsequently normalized using log-normalization. Highly variable genes were identified using the Find Variable Features function, based on variance-stabilizing transformation ("vst"). Data scaling was performed on all genes using the Scale Data function, followed by dimension reduction through principal component analysis (PCA) to identify anchors, selecting the top 12 principal components. Cells were clustered using the Find Neighbors and Find Clusters functions with a resolution setting of 0.2, resulting in 17 distinct subgroups. Further dimension reduction for visualization of these 17 clusters was achieved using UMAP, utilizing the same top 12 principal components. Annotation of Cell Types Cell type annotation was performed using the SingleR package, which annotates cell identities by comparing the test dataset with reference datasets based on similarity to known cell type labels. SingleR includes seven reference datasets, of which five are human and two are mouse datasets. For our analysis, we utilized the Human Primary Cell Atlas Data and Data base Immune Cell Expression Data reference sets. The former was employed for annotating all cell clusters, while the latter provided a more detailed annotation specifically for T cell subgroups. Cellular Trajectory Analysis Developmental trajectories of various cell subgroups were predicted using the Monocle package, which assesses each cell's degree of differentiation by calculating pseudotime. The emergence of trajectory branches often reflects cells following different gene expression programs. As cells make fate decisions during development, branches in the trajectory evolve, leading to the bifurcation of developmental lineages along distinct paths. To accurately infer cellular trajectories, only marker genes with an average expression greater than 0.1 were retained. Cells expressing fewer than 200 genes were also excluded from the analysis. Analysis of Cellular Communication To explore how the cellular communication networks among different cell types facilitate normal and stable physiological processes, the CellChat package was employed for systematic analysis. CellChat models the probability of cellular communication by integrating the expression of signaling ligands, receptors, and their cofactors with prior knowledge of their interactions. The "Secreted Signaling" category from the CellChatDB database was selected as the reference dataset. Only signaling pathways with interactions involving more than ten cells between two groups were retained for analysis. Metabolic Pathway Analysis The analysis of metabolic pathways among different cell subpopulations was conducted using the ReactomeGSA package. ReactomeGSA is designed based on the evolutionary conservation in life sciences, automatically mapping data from various species onto a common pathway space. The analyse_sc_clusters method evaluates the activity score of each cell subgroup on different metabolic pathways based on the expression levels of pathway-specific genes, essentially performing gene set enrichment analysis for each cell subgroup. Results Temporal Variation of Immune Infiltration in Ovarian Cancer Using the GSE158722 dataset, which comprises single-cell data from five ovarian cancer (OV) patients at three postoperative time points retained exclusively in 10x format, we first conducted sub-group analysis of the single-cell data at these time points and annotated the cell types for each subgroup. The results revealed that cells were divided into 17 subgroups (Fig. 1 A), corresponding to four cell types: Epithelial cells, Fibroblasts, CD8 + T-cells, and Macrophages (Fig. 1 B). Subgroup 14 was identified as CD8 + T-cells. Annotation indicated that more than 90% of the cells were Epithelial cells (91.99%), while Fibroblasts accounted for 5.72%, and CD8 + T-cells and Macrophages were relatively rare, comprising only 1.48% and 0.80%, respectively. Differential expression analysis identified 2,251 genes across these subgroups (Supplementary S1). Further, using this dataset, we analyzed the potential developmental paths of these cells, referred to as cell trajectories. Cell trajectory analysis, which constructs a pseudo time sequence, can infer the differentiation trajectories or evolutionary processes of cell subtypes, and is frequently utilized in developmental studies. The analysis showed that CD8 + T-cells and Macrophages clustered into one branch and appeared at later time points (Fig. 1 C). Epithelial and Fibroblast cells exhibited a higher degree of differentiation as they were present at all three time points, reflecting their high heterogeneity. Adjusting for root time point also revealed that the temporal trajectories of CD8 + T-cells and Macrophages were longer than those of other cell types (Fig. 1 D), indicating a later emergence of immune cells. Data related to this analysis is available in Supplementary S2. Validation was carried out using datasets GSE147082 and GSE130000. The GSE130000 dataset included eight ovarian cancer samples, comprising four primary tumors, two peritoneal metastases, and two recurrent tumors. Employing the same methodology for sub-group analysis on the single-cell data from GSE147082 and GSE130000, results showed that Epithelial cells represented the majority in both datasets, accounting for 27.51% and 51.30%, respectively. T-cells and Macrophages were each classified into separate subgroups, with T-cells constituting 13.25% and 2.99%, respectively, and occupying a small proportion of the cell population (Fig. 1 E, F). Further Subgrouping of T-cell Subsets Further analysis of the T-cell subgroups from datasets GSE147082 and GSE130000 suggested potential finer subdivisions within T cells, such as CD8 + and CD4 + T cells. Cell trajectory analysis confirmed the distribution of T cells across two branches (Fig. 1 G, results from the GSE130000 dataset), prompting further subgrouping of the T cell clusters. Upon further classification of the cells annotated as T cells, significant clustering into two distinct categories was observed, designated as sub-cluster0 and sub-cluster1 (Fig. 1 H). These two sub-clusters exhibited significant differences in pseudotime, with sub-cluster0 showing notably greater pseudotime than sub-cluster1 (Fig. 1 I, J). This indicates a dynamic trend in T cell variation during the development of ovarian cancer. Further identification of differential marker genes between these two T cell types will aid in understanding the underlying mechanisms driving these dynamics at the gene expression level. Identification of Marker Genes for Immune T Cells In the GSE130000 dataset, T cells were categorized into two groups: sub-cluster0 and sub-cluster1. We identified genes differentially expressed between these T cell subgroups, selecting those with a fold change greater than or equal to 2 and an FDR less than 0.05 as significant. A total of 41 genes were identified (Fig. 2 A, Supplementary S3), among which CCL5 and GZMA were recognized as marker genes for sub-cluster0, while the remaining 39 genes were markers for sub-cluster1. The average expression levels of these 41 genes were computed across cell types, revealing that CCL5 and GZMA were predominantly expressed in sub-cluster0, with minimal expression in other cell types (Fig. 2 B). Interestingly, sub-cluster1 exhibited a similar expression profile to the macrophage cell group, and both were clustered in the same branch. Figure 2 C displays the top 10 significantly different genes, encompassing a total of 12. Using data from the Cell Marker website, preliminary annotations were made for two T cell sub-clusters. Based on the overlap of marker genes, sub-cluster0 is likely a CD4 + cytotoxic T cell type, while sub-cluster1 might be either a Natural Killer T (NKT) cell or a Microglial cell. Furthermore, GO and KEGG pathway enrichment analyses were performed for sub-cluster1, as sub-cluster0, with only two marker genes, was not amenable to enrichment analysis. The analysis indicated that the marker genes for sub-cluster1 are primarily associated with protein folding and processing. Significant enrichment was also noted in the MAPK signaling pathway (Fig. 2 D). Therefore, sub-cluster0 can be defined as a CCL5-GZMA activated type T cell, and sub-cluster1 as a protein processing type T cell. After incorporating metastatic and recurrent samples from the GSE130000 dataset previously omitted from the analysis, the same clustering method was applied, and T cell clusters were selected for further subdivision. Following parameter optimization, it was possible to distinguish 11 subtypes. A comparison of marker genes identified for sub-cluster0 and sub-cluster1 with these 11 subtypes revealed that subtype 4 was most similar to sub-cluster0 (sharing a similar expression pattern of marker genes), while the expression profiles of other subtypes were most similar to sub-cluster1, as shown in Fig. 2 E. Subsequently, we quantified these T cell types across primary, metastatic, and relapse samples. We found that subtype 4 was most prevalent in primary samples, whereas other T cell types were more abundant in metastatic samples, as depicted in Fig. 2 F. Association Between T Cell Marker Genes and Prognosis in Ovarian Cancer In the TCGA dataset, we analyzed 41 significantly different marker genes to cluster the ovarian cancer (OV) samples, resulting in four distinct groups (Fig. 3 A). These were defined as T-cluster1 (n = 100), T-cluster2 (n = 98), T-cluster3 (n = 154), and T-cluster4 (n = 21). T-cluster2 exhibited the lowest overall survival rate, while T-cluster3 demonstrated the highest (Fig. 3 B). Similarly, the disease-free progression survival for T-cluster2 was significantly shorter than that for T-cluster3 (Fig. 3 C). Notably, eight genes were most highly expressed in T-cluster2 (highlighted in the red box in Fig. 3 A): DUSP2, ATF3, NR4A1, HSPH1, CLK1, RGS2, RGS1, and TNFAIP3. This observation leads to the hypothesis that the T cells of sub-cluster1, identified in single-cell groups, might represent a dysfunctional subset contributing to the progression of ovarian cancer. This dysfunction could explain the significant upregulation of sub-cluster1 marker genes in T-cluster2 samples, which are associated with the poorest prognosis. Assessment of T Cell Scores in Ovarian Cancer Samples From the previous analysis, eight differentially expressed genes were identified as marker genes for T cell sub-cluster1. We attempted to construct an ovarian cancer sample T cell scoring model based on the expression profiles of these eight genes. Initially, we compared the expression of these genes across the four TCGA ovarian cancer subgroups. Six of the genes showed significant differential expression, all highest in T-cluster2 (Fig. 3 D). The expression of these eight genes was not significantly correlated with overall survival times of ovarian cancer patients (Fig. 3 E). However, three genes—RGS2, DUSP2, and RGS1—showed a significant association with disease-free progression time (log-rank P < 0.05) and were identified as adverse prognostic factors (Fig. 3 E). Excluding HSPH1 and CLK1, which did not show significant differences, the average expression levels of the remaining six genes were calculated as the T cell score for each sample. We observed that patients with lower scores (L1) exhibited better overall survival and disease-free progression rates compared to those with higher scores (L2, Fig. 3 F). Relationship Between T Cell Scores and Clinical Features in Ovarian Cancer Patients Utilizing the TCGA ovarian cancer dataset, which includes information on stage, grade, subdivision, lymphatic invasion, new events, tumor residuals, venous invasion, and therapy outcomes, we compared the T cell scores across these clinical features. No significant differences were observed in the T cell scores among the various samples under these clinical categories (Fig. 3 G). This indicates that there is no apparent correlation between T cell scores and these clinical characteristics of ovarian cancer patients. Cellular Communication and Related Pathway Analysis Using the GSE130000 dataset, we explored potential communication links between different cell types. The CellChat package, which provides tools for studying intercellular communication, models the probability of cell communication by integrating the expression of signaling ligands, receptors, and their cofactors with prior knowledge of their interactions. After inferring the intercellular communication network, CellChat facilitates further data exploration, analysis, and visualization. Figure 4 A displays the interaction frequency of each cell type with other groups, showing that epithelial cells, endothelial cells, and smooth muscle cells have the highest frequency of interaction (Supplementary S4). By assessing the weights between cell groups, the interactions between epithelial cells and neurons, as well as induced pluripotent stem (iPS) cells, were found to have the highest interaction strength (Fig. 4 B, Supplementary S5). Figure 4 C illustrates the interaction relationships of each cell type with other cell groups, highlighting that both T cell subgroups, sub-cluster0 and sub-cluster1, exhibited the highest interaction frequency with smooth muscle cells. Further analysis identified 44 signaling pathways contributing to intercellular interactions (Supplementary S6). We focused particularly on the pathways related to two T cell subgroups, sub-cluster0 and sub-cluster1, involving 11 pathways in total (Fig. 4 D). Among these, the CCL and IL2 signaling pathways contributed most significantly to the interactions between sub-cluster0 and sub-cluster1. Additionally, the CCLX signaling pathway was a major contributor to interactions between sub-cluster0, sub-cluster1, and macrophages. Gene expression analysis related to these signaling pathways confirmed these findings; high expression levels of genes such as CCL5, CCL3, and CCL4 were observed exclusively in sub-cluster0, sub-cluster1, and macrophages. Genes involved in the IL2 signaling pathway were predominantly expressed in sub-cluster0 and sub-cluster1 (Fig. 4 E). Given that the CCL5 gene is a marker gene for T cell sub-cluster0, these results suggest that aberrant expression of CCL5 may be closely linked to the differentiation of T cells into the sub-cluster0 and sub-cluster1 subgroups. Analysis of Metabolic Pathways To further investigate the functional characteristics of cell subgroups, single-cell data were analyzed using the Reactome database. Leveraging the conservation inherent in life sciences, data from different species were automatically mapped to a common pathway space, and enrichment scores for each pathway were calculated. We then assessed the differences in these pathway enrichment scores among various cell subgroups (Supplementary S7). The top 40 pathways with the greatest disparities were selected for further study, and a heatmap of these top 40 pathways was created (Fig. 4 F). Notably, the FGFR1c and Klotho ligand binding and activation pathways showed the lowest enrichment scores in the two T cell subgroups, sub-cluster0 and sub-cluster1, while the MGMT-mediated DNA damage reversal pathway was specifically enriched at the lowest level uniquely in sub-cluster0 (note that deeper red indicates lower scores). Figures 4 G and 4 H display the enrichment scores for these pathways across eight cell subgroups. Discussion Ovarian cancer remains a principal cause of mortality among patients with gynecological malignancies. Despite advancements in cytoreductive surgery and the administration of platinum-based chemotherapy, the prognosis for ovarian cancer patients continues to be dismal due to frequent dissemination and relapse 10 . This underscores the critical need for the development of novel targeted therapies. Recent studies underscore the significant role of the tumor microenvironment in the pathogenesis of ovarian cancer. Insights derived from comprehensive genomic, transcriptomic, and proteomic analyses have highlighted the tumor microenvironment not only as a key player in oncogenesis but also as a promising target for therapeutic intervention 11 . Addressing this component could pave the way for breakthroughs in the treatment modalities for ovarian cancer. Although tumor-infiltrating lymphocytes (TILs), particularly CD8 + T cells, are recognized as a positive prognostic marker in numerous solid tumors, their inability to effectively eradicate cancer cells persists 12,13 . One underlying reason for this immunological control failure is the extensive presence of immunosuppressive mechanisms within the TME, which impair the effector functions of infiltrating T cells 14 . A primary hallmark of T cell dysfunction in this context is the upregulation of PD-1 on T cells. The altered functional state of PD-1 + T cells, referred to as T cell exhaustion, has been extensively documented and studied in human infections and cancers 15 . The successful revitalization of T cell function through the blockade of PD-1 or its ligand PD-L1 underscores the significance of the PD-1/PD-L1 axis in T cell dysfunction. However, the majority of patients still do not respond or fail to achieve sustained responses. Specifically, it has been observed that a subset of T cells co-expresses various inhibitory receptors, known as immune checkpoints, and this expression escalates with the progression of dysfunction 16,17 . Given the exposure of the intratumoral T cell repertoire to a plethora of different immunosuppressive mechanisms, a wide spectrum of T cell dysfunction states can be anticipated. The functional states of intratumoral T cells exhibit variability across different tumors, yet even within a single tumor, T cell infiltration demonstrates significant heterogeneity. This heterogeneity is evident in at least three aspects: ( 1 ) the heterogeneity of traditional T cell subsets, including CD8+, CD4 + Th1, Th2, Treg cells, and γδ T cells; ( 2 ) the diversity in the degree and type of T cell dysfunction within given subsets; ( 3 ) the heterogeneous capacity of individual TCR to recognize tumor antigens 18 . Due to this heterogeneity, accurately capturing the range of T cell states within tumors requires analysis at the single-cell level. Recent studies utilizing such techniques have provided detailed descriptions of the tumor ecosystem across various cancer types. For instance, single-cell sequencing of 11 breast cancer patients revealed three immune cell clusters characterized by T cells exhibiting either an exhausted or regulatory phenotype, B cells, and predominantly M2-like macrophages 19 . Another study extensively analyzed immune infiltration in 73 patients with clear cell renal carcinoma (RCC), identifying 22 T cell phenotypes and 17 tumor-associated macrophage phenotypes. Within the T cell clusters, seven distinct clusters of PD-1-expressing CD8 + T cells were observed. Notably, the co-expression of other inhibitory receptors, including Tim-3 and CTLA-4, as well as the expression of co-stimulatory or proliferation markers, varied greatly among these clusters, suggesting that they may reflect different states of dysfunction 20 . Ovarian cancer exhibits significant immunogenic potential, evidenced by its expression of numerous tumor antigens and the correlation between TIL accumulation and improved patient outcomes 21,22 . Despite this, the response of ovarian cancer to endogenous immune modulation via immune checkpoint inhibitors has been modest 23,24 . The identification and characterization of these tumor-reactive T cells represent a promising strategy for the development and optimization of cellular therapies. TILs, particularly subsets of these cells, have been identified as predictive biomarkers for outcomes in patients with epithelial ovarian cancer 25 . In a multicenter observational trial involving high-grade serous ovarian cancer, a notable finding was the heightened infiltration of CD8 + T cells compared to other ovarian cancer subtypes. Detailed analysis revealed a dose-response relationship between the levels of CD8 + TILs and patient survival rates. Specifically, patients with high CD8 + TIL levels exhibited a median survival of 5.1 years, significantly longer than the 2.8 to 3.8 years observed in patients with low to moderate levels of these cells 26 . Notably, the survival benefit associated with CD8 + TILs was prominent in patients harboring BRCA1 mutations but not in those with BRCA2 mutations 27 . Despite the known role of CD4 + helper T cells in augmenting CD8 + T cell responses, no direct correlation between CD4 + T cell infiltration and survival outcomes was observed. Conversely, the presence of immunosuppressive Tregs consistently predicted poorer survival outcomes 28 . Interestingly, only a small fraction of the TIL population is tumor-reactive, with the majority being inactive bystanders. Activation markers such as CD25, CD38, and CD137, as well as exhaustion markers like TIM3, PD-1, TIGIT, and CTLA4—often upregulated due to chronic antigen exposure—were prevalent among TILs. The presence of these markers serves as a surrogate for the activity of immunoreactive TILs, and increased populations of PD-1 + TILs have been linked with improved survival outcomes in epithelial ovarian cancer 29,30 . This study further analyzed the data from ovarian cancer single-cell and bulk RNA sequencing available in the GSE158722 dataset, wherein the ovarian cancer cells were annotated into four cell types: Epithelial cells, Fibroblasts, CD8 + T-cells, and Macrophages. A vast majority of the cells were identified as Epithelial cells and Fibroblasts, with immune-related cells like T-cells comprising a small proportion. This phenomenon was also observed in two other independent datasets. In pseudotime sequence analysis, after time adjustment, the pseudotime results indicated that CD8 + T cells and Macrophages appeared later than other cell types. Additionally, T cells exhibited bifurcation, reflecting their potential differentiation states. Utilizing the GSE130000 dataset, we identified differentially expressed genes between two T cell subgroups, totaling 41 genes. Among these, CCL5 and GZMA were marker genes for sub-cluster 0, while the remaining 39 were marker genes for sub-cluster 1. These 41 marker genes differentiated the TCGA OV dataset into four classes, where samples in T-cluster 2 exhibited the lowest overall survival, and those in T-cluster 3, the highest. These findings suggest that T cells in sub-cluster 1 may represent a dysfunctional subgroup contributing to the progression of ovarian cancer. Based on the eight genes (DUSP2、ATF3、NR4A1、HSPH1、CLK1、RGS2、RGS1 and TNFAIP3) exhibiting the highest expression levels in samples from T-cluster 2, we constructed a T cell scoring model for ovarian cancer samples. We observed that patients in the low score group (L1) had better overall survival and disease-free progression rates compared to those in the high score group (L2). Moreover, there was no significant association between T cell scores and these clinical features. Cell communication analysis indicated that the CCL and IL2 signaling pathways contributed most significantly to communication between sub-cluster 0 and sub-cluster 1 T cells. Given that the CCL5 gene is a marker gene for T cell sub-cluster 0, these results suggest that abnormal expression of CCL5 may be closely associated with the differentiation of T cells into sub-cluster 0 and sub-cluster 1. Pathway analysis revealed that the FGFR1c and Klotho ligand binding and activation pathways had the lowest enrichment scores in both T cell subgroups, whereas the MGMT-mediated DNA damage reversal pathway was uniquely least enriched in sub-cluster 0. CCL5, also known as RANTES, is involved in several signaling pathways, primarily related to immune function and inflammation. It acts through the CCR5 receptor among others, playing a critical role in various cellular processes including chemotaxis, immune surveillance, and inflammation ​ 31 . CCL5 interacts with multiple signaling pathways. One of its primary roles is in the chemokine signaling pathway, particularly involving the CCL5/CCR5 axis, which is crucial for the regulation of immune cell migration and activation. This axis is also significant in cancer progression, influencing tumor microenvironment by promoting tumor cell invasion, metastasis, and immune evasion ​ 32 ​. Additionally, CCL5 is involved in the MAPK signaling pathway, which plays a role in cell proliferation, survival, and differentiation 31 . In the study on Hirschsprung's disease, Pan et al. utilized RNA-sequencing to identify several genes, including ATF3, NOS1, CCL5, DUSP1, CXCL3, VIP, and FOSB, that might be involved in the development of HSCR through their effects on the nervous system. Additionally, their analysis of the protein-protein interaction (PPI) network revealed potential interactions among ATF3, CCL5, and TNFAIP3, suggesting a complex interplay that could underlie the pathogenesis of the disease 33 . Although scRNA-seq and bulk RNA sequencing analyses offer significant advantages and have identified differentially expressed factors in present study, the specific interactions between CCL-5 and the differentially expressed genes and related signaling pathways within sub-cluster1 should be further validated in human tissues and cell lines. The search for reliable biomarkers predictive of immune therapy outcomes in ovarian cancer remains a challenging and formidable task. Conclusion In conclusion, the present study identifies differential expression of CCL-5 as a biomarker for distinct T-cell subtypes in ovarian cancer by integrating scRNA-seq and bulk RNA-seq data. We further discovered that a T-cell scoring model, constructed using eight genes significantly associated with dysfunctional T-cell subtypes, predicts lower overall survival and disease-free progression rates in ovarian cancer patients with lower scores, which appeared independent of clinical features. These findings suggest that the T-cell scoring model could serve as a reliable prognostic indicator for patients with ovarian cancer. Additionally, CCL-5 may represent a significant target for immunotherapy in ovarian cancer. Declarations Conflicts of Interest The authors declare that they have no conflicts of interest related to this work. Author Contributions SHZ and HX: Study design, data analysis. GLY: Study design and data analysis and manuscript writing. LSH: data analysis. All authors have read, edited and approved the final version of the manuscript. Funding Not applicable. Data availability Sequence data that support the findings of this study have been deposited in https://www.ncbi.nlm.ni-h.gov/geo/query/acc.cgi? Acknowledgments Ethical approval: All procedures in studies involving human participants were performed in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration and its later amendments. (Name and affiliation of the approving ethics committee: Institutional Ethics Committee of Peking University Third Hospital; No. A2023035). References Torre, L. A. et al. Ovarian cancer statistics, 2018. CA Cancer J Clin , doi:10.3322/caac.21456 (2018). Roberts, C. M., Cardenas, C. & Tedja, R. The role of intra-tumoral heterogeneity and its clinical relevance in epithelial ovarian cancer recurrence and metastasis. Cancers 11 , 1083 (2019). Zhang, Y. & Zhang, Z. The history and advances in cancer immunotherapy: understanding the characteristics of tumor-infiltrating immune cells and their therapeutic implications. Cellular & molecular immunology 17 , 807-821, doi:10.1038/s41423-020-0488-6 (2020). Ribas, A. & Wolchok, J. D. Cancer immunotherapy using checkpoint blockade. Science 359 , 1350-1355, doi:10.1126/science.aar4060 (2018). Tanaka, A. & Sakaguchi, S. Regulatory T cells in cancer immunotherapy. Cell research 27 , 109-118, doi:10.1038/cr.2016.151 (2017). Nowak, M. & Klink, M. The Role of Tumor-Associated Macrophages in the Progression and Chemoresistance of Ovarian Cancer. Cells 9 , doi:10.3390/cells9051299 (2020). Gitto, S. B., Ihewulezi, C. J. N. & Powell, D. J., Jr. Adoptive T cell therapy for ovarian cancer. Gynecol Oncol 186 , 77-84, doi:10.1016/j.ygyno.2024.04.001 (2024). Moncada, R. et al. Integrating microarray-based spatial transcriptomics and single-cell RNA-seq reveals tissue architecture in pancreatic ductal adenocarcinomas. Nat Biotechnol 38 , 333-342, doi:10.1038/s41587-019-0392-8 (2020). Kuksin, M. et al. Applications of single-cell and bulk RNA sequencing in onco-immunology. European journal of cancer (Oxford, England : 1990) 149 , 193-210, doi:10.1016/j.ejca.2021.03.005 (2021). Konstantinopoulos, P. A. & Matulonis, U. A. Clinical and translational advances in ovarian cancer therapy. Nature cancer 4 , 1239-1257, doi:10.1038/s43018-023-00617-9 (2023). Yang, Y., Yang, Y., Yang, J., Zhao, X. & Wei, X. Tumor Microenvironment in Ovarian Cancer: Function and Therapeutic Strategy. Frontiers in cell and developmental biology 8 , 758, doi:10.3389/fcell.2020.00758 (2020). Fridman, W. H., Pagès, F., Sautès-Fridman, C. & Galon, J. The immune contexture in human tumours: impact on clinical outcome. Nature reviews. Cancer 12 , 298-306, doi:10.1038/nrc3245 (2012). Boon, T., Coulie, P. G., Van den Eynde, B. J. & van der Bruggen, P. Human T cell responses against melanoma. Annual review of immunology 24 , 175-208, doi:10.1146/annurev.immunol.24.021605.090733 (2006). Chen, D. S. & Mellman, I. Oncology meets immunology: the cancer-immunity cycle. Immunity 39 , 1-10, doi:10.1016/j.immuni.2013.07.012 (2013). Baitsch, L. et al. Exhaustion of tumor-specific CD8⁺ T cells in metastases from melanoma patients. J Clin Invest 121 , 2350-2360, doi:10.1172/jci46102 (2011). Thommen, D. S. et al. Progression of Lung Cancer Is Associated with Increased Dysfunction of T Cells Defined by Coexpression of Multiple Inhibitory Receptors. Cancer immunology research 3 , 1344-1355, doi:10.1158/2326-6066.Cir-15-0097 (2015). Wherry, E. J. T cell exhaustion. Nature immunology 12 , 492-499, doi:10.1038/ni.2035 (2011). Thommen, D. S. & Schumacher, T. N. T Cell Dysfunction in Cancer. Cancer Cell 33 , 547-562, doi:10.1016/j.ccell.2018.03.012 (2018). Chung, W. et al. Single-cell RNA-seq enables comprehensive tumour and immune cell profiling in primary breast cancer. Nat Commun 8 , 15081, doi:10.1038/ncomms15081 (2017). Chevrier, S. et al. An Immune Atlas of Clear Cell Renal Cell Carcinoma. Cell 169 , 736-749.e718, doi:10.1016/j.cell.2017.04.016 (2017). Rodriguez-Garcia, A., Minutolo, N. G., Robinson, J. M. & Powell, D. J. T-cell target antigens across major gynecologic cancers. Gynecol Oncol 145 , 426-435, doi:10.1016/j.ygyno.2017.03.510 (2017). Zhang, L. et al. Intratumoral T cells, recurrence, and survival in epithelial ovarian cancer. The New England journal of medicine 348 , 203-213, doi:10.1056/NEJMoa020177 (2003). Disis, M. L. et al. Efficacy and Safety of Avelumab for Patients With Recurrent or Refractory Ovarian Cancer: Phase 1b Results From the JAVELIN Solid Tumor Trial. JAMA Oncol 5 , 393-401, doi:10.1001/jamaoncol.2018.6258 (2019). Matulonis, U. A. et al. Antitumor activity and safety of pembrolizumab in patients with advanced recurrent ovarian cancer: results from the phase II KEYNOTE-100 study. Annals of oncology : official journal of the European Society for Medical Oncology 30 , 1080-1087, doi:10.1093/annonc/mdz135 (2019). James, F. R. et al. Association between tumour infiltrating lymphocytes, histotype and clinical outcome in epithelial ovarian cancer. BMC Cancer 17 , 657, doi:10.1186/s12885-017-3585-x (2017). Goode, E. L. et al. Dose-Response Association of CD8+ Tumor-Infiltrating Lymphocytes and Survival Time in High-Grade Serous Ovarian Cancer. JAMA Oncol 3 , e173290, doi:10.1001/jamaoncol.2017.3290 (2017). Clarke, B. et al. Intraepithelial T cells and prognosis in ovarian carcinoma: novel associations with stage, tumor type, and BRCA1 loss. Modern pathology : an official journal of the United States and Canadian Academy of Pathology, Inc 22 , 393-402, doi:10.1038/modpathol.2008.191 (2009). Sato, E. et al. Intraepithelial CD8+ tumor-infiltrating lymphocytes and a high CD8+/regulatory T cell ratio are associated with favorable prognosis in ovarian cancer. Proceedings of the National Academy of Sciences of the United States of America 102 , 18538-18543, doi:10.1073/pnas.0509182102 (2005). Santoiemma, P. P. et al. Systematic evaluation of multiple immune markers reveals prognostic factors in ovarian cancer. Gynecol Oncol 143 , 120-127, doi:10.1016/j.ygyno.2016.07.105 (2016). Webb, J. R., Milne, K., Kroeger, D. R. & Nelson, B. H. PD-L1 expression is associated with tumor-infiltrating T cells and favorable prognosis in high-grade serous ovarian cancer. Gynecol Oncol 141 , 293-302, doi:10.1016/j.ygyno.2016.03.008 (2016). Zhu, G. D. et al. Identification of differentially expressed genes and signaling pathways with Candida infection by bioinformatics analysis. European journal of medical research 27 , 43, doi:10.1186/s40001-022-00651-w (2022). Aldinucci, D., Borghese, C. & Casagrande, N. The CCL5/CCR5 Axis in Cancer Progression. Cancers 12 , doi:10.3390/cancers12071765 (2020). Pan, W. K. et al. Identifying key genes associated with Hirschsprung's disease based on bioinformatics analysis of RNA-sequencing data. World journal of pediatrics : WJP 13 , 267-273, doi:10.1007/s12519-017-0002-0 (2017). Additional Declarations No competing interests reported. Supplementary Files SupplementaryS1.xlsx SupplementaryS2.xlsx SupplementaryS3.xlsx SupplementaryS4.xlsx SupplementaryS5.xlsx SupplementaryS6.xlsx SupplementaryS7.xlsx Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 06 Sep, 2026 Reviews received at journal 06 Sep, 2026 Reviewers agreed at journal 05 Sep, 2026 Reviews received at journal 05 Sep, 2026 Reviewers agreed at journal 04 Sep, 2026 Reviewers agreed at journal 04 Sep, 2026 Reviewers agreed at journal 04 Sep, 2026 Reviewers agreed at journal 03 Sep, 2026 Reviewers agreed at journal 03 Sep, 2026 Reviews received at journal 03 Sep, 2026 Reviewers agreed at journal 03 Sep, 2026 Reviews received at journal 03 Sep, 2026 Reviewers agreed at journal 03 Sep, 2026 Reviewers agreed at journal 03 Sep, 2026 Reviewers agreed at journal 03 Sep, 2026 Reviewers agreed at journal 03 Sep, 2026 Reviewers invited by journal 28 Jul, 2024 Editor assigned by journal 22 Jul, 2024 Editor invited by journal 16 Jul, 2024 Submission checks completed at journal 15 Jul, 2024 First submitted to journal 10 Jul, 2024 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 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-4721266","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":335202529,"identity":"3e7e5094-de56-4f86-a20e-bff9d602db9a","order_by":0,"name":"Hengzi Sun","email":"","orcid":"","institution":"Beijing Chao-Yang Hospital","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Hengzi","middleName":"","lastName":"Sun","suffix":""},{"id":335202531,"identity":"a6c47c13-b819-4435-ba1c-6161bb908540","order_by":1,"name":"Xiao Huo","email":"","orcid":"","institution":"Peking University Third Hospital","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Xiao","middleName":"","lastName":"Huo","suffix":""},{"id":335202533,"identity":"8f9da314-d3c2-480b-a790-413f9f220136","order_by":2,"name":"Shuhong Li","email":"","orcid":"","institution":"Beijing Chao-Yang Hospital","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Shuhong","middleName":"","lastName":"Li","suffix":""},{"id":335202536,"identity":"eae9ffaf-eaea-4b90-90ad-ca324aedf420","order_by":3,"name":"Liyuan Guo","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA7UlEQVRIiWNgGAWjYHACxgMJIIq9seFAQoWEHD8xeiBaeA4ffPDhjIWxZAMxWsCkRFqy4cy2isQNhLTIRyQfOPCw7bC8OUOOmTTvPAnGDQzMDx/dwKPF8EZawoHEtsOGOxvOALVsk2A2Z2AzNs7Bp2VGjgFIC+OGgz1gLWyWDTxs0sRosd9wmAeoZY4Ej8EBAlrkJSBaEjccYwN6v0FCgqAWA55nCQcSzqUnbzjDDAzkYxIGks0E/CLfnnzw4Y8ya9sN9x8Co7Kmrr6fvfnhY7y2HAASjGzIQsx4lINtaQCRfwioGgWjYBSMgpENACsSVXD8Iml7AAAAAElFTkSuQmCC","orcid":"","institution":"Beijing Chao-Yang Hospital","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Liyuan","middleName":"","lastName":"Guo","suffix":""}],"badges":[],"createdAt":"2024-07-11 02:37:45","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4721266/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4721266/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":62222176,"identity":"e8fc83f4-a6e5-4743-9c79-cec96f873903","added_by":"auto","created_at":"2024-08-11 12:35:47","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1066099,"visible":true,"origin":"","legend":"\u003cp\u003escRNA-seq and bulk RNA-seq analysis to identify different cell populations in the TME of ovarian cancer. (A-B) UMAP was adopted to identify different clusters and cell types in scRNA-seq analysis of data GSE158722. (C-D) Cell trajectory and pseudo-time analysis of four cells subtypes in data GSE158722. (E-F) UMAP was adopted to identify different clusters and cell types in bulk RNA-seq analysis of data GSE147082 and GSE130000. (G) Pseudo-time analysis of different cells subtypes in data GSE130000. (H-J) The UMAP and Pseudo-time analysis divide the T cell clusters into two sub-cluster.\u003c/p\u003e","description":"","filename":"Picture1.png","url":"https://assets-eu.researchsquare.com/files/rs-4721266/v1/90b22b157f43b15e47899d82.png"},{"id":62222177,"identity":"ac5c53ab-de9e-48ab-99cd-04ef0818e3a2","added_by":"auto","created_at":"2024-08-11 12:35:47","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1767880,"visible":true,"origin":"","legend":"\u003cp\u003eIdentification of marker genes for immune T cells. (A) Heatmap to display the expression pattern of TMGs in different clusters. (B-C) Clustering analysis of different cell populations (B) and significant differential gene expression (C). (D) GO and KEGG pathway enrichment analysis of marker genes in sub-cluster1 cells. (E-F) Clustering analysis after adding the sample dataset from GSE130000.\u003c/p\u003e","description":"","filename":"Picture2.png","url":"https://assets-eu.researchsquare.com/files/rs-4721266/v1/5892e4f519e2a7628373434c.png"},{"id":62222181,"identity":"5e2a875d-b588-4837-894b-ded226f88836","added_by":"auto","created_at":"2024-08-11 12:35:47","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1285219,"visible":true,"origin":"","legend":"\u003cp\u003eAssociation between T cell marker genes and prognosis in ovarian cancer. (A) Clustering analysis of marker genes in OC samples on the TCGA dataset. (B-C) the OS (B) and PFS (C) between different T-cluster. (D) The expression of eight marker genes across four ovarian cancer subgroups in the TCGA dataset. (E) Analysis of the correlation between marker genes and ovarian cancer OS and PFS. (F) The OS and PFS between different T cell scores. (G) The correlation analysis between T cell score and clinical characteristics of ovarian cancer patients.\u003c/p\u003e","description":"","filename":"Picture3.png","url":"https://assets-eu.researchsquare.com/files/rs-4721266/v1/af4ab21e18e461c5d857873a.png"},{"id":62222185,"identity":"46026d8b-8b96-4b75-91df-17e151181ef3","added_by":"auto","created_at":"2024-08-11 12:35:48","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1608352,"visible":true,"origin":"","legend":"\u003cp\u003eCellular communication and related pathways analysis. (A-C) Cell–cell communication network between different cell types. (D) Analysis of the contributions of different signaling pathways to cell-cell communication. (E) Analysis of the contributions of related genes in different signaling pathways to cell-cell communication. (F) Heatmap of the top 40 metabolic pathways. (G-H) Enrichment scores of different pathways across eight cell clusters.\u003c/p\u003e","description":"","filename":"Picture4.png","url":"https://assets-eu.researchsquare.com/files/rs-4721266/v1/e711b669ff735a5802bac8ad.png"},{"id":62223234,"identity":"49883afe-109a-4c9b-a0de-8e118327c611","added_by":"auto","created_at":"2024-08-11 12:43:55","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":6857530,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4721266/v1/7b92b6f2-8ca1-405d-b193-bd43a1636239.pdf"},{"id":62222179,"identity":"0095bac1-41c6-465a-a3f1-ef1e11ed0fa3","added_by":"auto","created_at":"2024-08-11 12:35:47","extension":"xlsx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":210341,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryS1.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-4721266/v1/3e24d5a3edf22192f9e2a62d.xlsx"},{"id":62222184,"identity":"fed32a67-bece-4337-854d-016ee99a2158","added_by":"auto","created_at":"2024-08-11 12:35:48","extension":"xlsx","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":2865708,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryS2.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-4721266/v1/a715567d6d8e731cd6c2b0ac.xlsx"},{"id":62223230,"identity":"71ade347-77d6-4c22-be8a-244d9080c8aa","added_by":"auto","created_at":"2024-08-11 12:43:47","extension":"xlsx","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":12088,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryS3.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-4721266/v1/44f5027c9379c0983e6df633.xlsx"},{"id":62222180,"identity":"7899d412-9daf-4059-93f3-0f3fab88db5d","added_by":"auto","created_at":"2024-08-11 12:35:47","extension":"xlsx","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":9036,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryS4.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-4721266/v1/3333e747bd3fa1c212ac52a9.xlsx"},{"id":62223231,"identity":"87708760-e0f7-4514-9efe-cbe4a83402b0","added_by":"auto","created_at":"2024-08-11 12:43:48","extension":"xlsx","order_by":6,"title":"","display":"","copyAsset":false,"role":"supplement","size":9955,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryS5.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-4721266/v1/40e7c9c12d611b775063e2c8.xlsx"},{"id":62222182,"identity":"4d39177e-be65-43b8-bb0b-6cb0c7ea288e","added_by":"auto","created_at":"2024-08-11 12:35:47","extension":"xlsx","order_by":7,"title":"","display":"","copyAsset":false,"role":"supplement","size":44558,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryS6.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-4721266/v1/3aac4344f0bbdb83a3766abe.xlsx"},{"id":62222183,"identity":"3b9dfdb5-09dd-4f1f-b137-ef5c1cce5a47","added_by":"auto","created_at":"2024-08-11 12:35:47","extension":"xlsx","order_by":8,"title":"","display":"","copyAsset":false,"role":"supplement","size":323525,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryS7.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-4721266/v1/82602e177aa28f190af35299.xlsx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Single‑cell and bulk RNA sequencing identifes T cell marker genes to predict the prognosis of ovrian caner","fulltext":[{"header":"Introduction","content":"\u003cp\u003eOvarian cancer, known for its poor prognosis and high mortality rate, remains one of the most challenging gynecological cancers to treat effectively. The majority of patients are diagnosed at advanced stages, primarily due to the non-specific symptoms associated with the disease \u003csup\u003e1\u003c/sup\u003e. Current treatment protocols involve a combination of surgery and chemotherapy, which, while effective for some, do not significantly improve long-term survival for many patients. Recurrence rates are high, and the five-year survival rate remains under 50% for advanced-stage ovarian cancer patients. This grim outlook underscores the urgent need for improved prognostic tools and personalized treatment strategies \u003csup\u003e2\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eImmunotherapy has revolutionized the treatment landscape for various cancers, highlighting the crucial role of the immune system in controlling tumor growth. Strategies such as checkpoint inhibitors, CAR-T cell therapy, and vaccines aim to enhance the immune system's ability to detect and destroy cancer cells \u003csup\u003e3\u003c/sup\u003e. The efficacy of these therapies, however, varies significantly among cancer types due to differences in immune infiltration and the tumor microenvironment. Research has increasingly focused on understanding the immunological underpinnings of each cancer to tailor immunotherapeutic approaches. For instance, the presence of immune checkpoints, which tumors exploit to evade immune surveillance, has led to the development of inhibitors that can reactivate T cells \u003csup\u003e4\u003c/sup\u003e. Additionally, the role of T cells in mediating anti-tumor responses has been a major focus, with efforts to characterize and modify T cell functionality to improve therapeutic efficacy \u003csup\u003e5\u003c/sup\u003e. This research underscores the potential of immunological strategies to provide durable responses and highlights the ongoing need to integrate immunotherapy with traditional treatments.\u003c/p\u003e \u003cp\u003eRecent advances in understanding the immune evasion mechanisms employed by ovarian tumors have paved the way for novel immunotherapeutic approaches. It is now recognized that the tumor microenvironment (TME) of ovarian cancer is not only complex but also immunosuppressive, limiting the efficacy of conventional therapies \u003csup\u003e6\u003c/sup\u003e. Studies have shown that certain immunomodulatory molecules and T cells within the TME are crucial in determining the tumor's progression and response to treatments. Immunotherapy, particularly the use of immune checkpoint inhibitors that target PD-1 and CTLA-4, has shown promise in several clinical trials, suggesting a pivotal role of the immune system in controlling this malignancy \u003csup\u003e7\u003c/sup\u003e. However, the response rates vary, and a significant proportion of patients do not benefit from these therapies, highlighting the need for precise biomarkers that can predict immunotherapeutic response.\u003c/p\u003e \u003cp\u003eSingle-cell RNA sequencing (scRNA-seq) and bulk RNA sequencing have become indispensable tools in cancer research, offering profound insights into the molecular and cellular heterogeneity of tumors. scRNA-seq allows for the detailed analysis of individual cells within a tumor, revealing the existence of various cell types and their states, which are crucial for understanding the complexity of the tumor microenvironment. This level of detail is essential for identifying specific cell populations that may influence prognosis or response to therapy, such as different subsets of T cells \u003csup\u003e8\u003c/sup\u003e. Bulk RNA sequencing, on the other hand, provides a broader overview of gene expression within the entire tumor mass, useful for identifying overarching expression patterns and biomarkers \u003csup\u003e9\u003c/sup\u003e. Together, these technologies enable the development of comprehensive gene expression profiles that can predict disease outcomes and guide treatment decisions, thereby enhancing the precision of cancer therapy and prognostication. The present study aims to explore the correlation between the prognosis of ovarian cancer and its immune T cell markers through the analysis of single-cell and bulk RNA-seq data.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eData Preparation and Preprocessing\u003c/h2\u003e \u003cp\u003eSingle-cell sequencing data sets GSE158722, GSE147082, and GSE130000, including expression matrices and associated clinical feature information, were retrieved from the NCBI Gene Expression Omnibus (GEO) database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?\u003c/span\u003e\u003cspan address=\"https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Specifically, the GSE158722 dataset comprised single-cell data from samples collected at three postoperative time points from five ovarian cancer (OV) patients, retaining only data processed by the 10x Genomics platform. Additionally, high-grade serous ovarian cancer RNA-seq data and corresponding clinical features were downloaded from The Cancer Genome Atlas (TCGA). After excluding samples with missing survival times and those with metastatic disease, a total of 373 samples with prognostic information were secured for further analysis.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eCell Clustering Analysis\u003c/h3\u003e\n\u003cp\u003eInitially, single-cell data were filtered to include only genes expressed in at least five cells and cells that expressed at least 500 genes. The mitochondrial and ribosomal RNA content percentages were calculated using the Percentage Feature Set function, ensuring that each cell expressed over 500 genes and had mitochondrial content below 30%. For multi-sample analyses, data from different samples were merged and subsequently normalized using log-normalization. Highly variable genes were identified using the Find Variable Features function, based on variance-stabilizing transformation (\"vst\"). Data scaling was performed on all genes using the Scale Data function, followed by dimension reduction through principal component analysis (PCA) to identify anchors, selecting the top 12 principal components. Cells were clustered using the Find Neighbors and Find Clusters functions with a resolution setting of 0.2, resulting in 17 distinct subgroups. Further dimension reduction for visualization of these 17 clusters was achieved using UMAP, utilizing the same top 12 principal components.\u003c/p\u003e\n\u003ch3\u003eAnnotation of Cell Types\u003c/h3\u003e\n\u003cp\u003eCell type annotation was performed using the SingleR package, which annotates cell identities by comparing the test dataset with reference datasets based on similarity to known cell type labels. SingleR includes seven reference datasets, of which five are human and two are mouse datasets. For our analysis, we utilized the Human Primary Cell Atlas Data and Data base Immune Cell Expression Data reference sets. The former was employed for annotating all cell clusters, while the latter provided a more detailed annotation specifically for T cell subgroups.\u003c/p\u003e\n\u003ch3\u003eCellular Trajectory Analysis\u003c/h3\u003e\n\u003cp\u003eDevelopmental trajectories of various cell subgroups were predicted using the Monocle package, which assesses each cell's degree of differentiation by calculating pseudotime. The emergence of trajectory branches often reflects cells following different gene expression programs. As cells make fate decisions during development, branches in the trajectory evolve, leading to the bifurcation of developmental lineages along distinct paths. To accurately infer cellular trajectories, only marker genes with an average expression greater than 0.1 were retained. Cells expressing fewer than 200 genes were also excluded from the analysis.\u003c/p\u003e\n\u003ch3\u003eAnalysis of Cellular Communication\u003c/h3\u003e\n\u003cp\u003eTo explore how the cellular communication networks among different cell types facilitate normal and stable physiological processes, the CellChat package was employed for systematic analysis. CellChat models the probability of cellular communication by integrating the expression of signaling ligands, receptors, and their cofactors with prior knowledge of their interactions. The \"Secreted Signaling\" category from the CellChatDB database was selected as the reference dataset. Only signaling pathways with interactions involving more than ten cells between two groups were retained for analysis.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eMetabolic Pathway Analysis\u003c/h2\u003e \u003cp\u003eThe analysis of metabolic pathways among different cell subpopulations was conducted using the ReactomeGSA package. ReactomeGSA is designed based on the evolutionary conservation in life sciences, automatically mapping data from various species onto a common pathway space. The analyse_sc_clusters method evaluates the activity score of each cell subgroup on different metabolic pathways based on the expression levels of pathway-specific genes, essentially performing gene set enrichment analysis for each cell subgroup.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eTemporal Variation of Immune Infiltration in Ovarian Cancer\u003c/h2\u003e \u003cp\u003eUsing the GSE158722 dataset, which comprises single-cell data from five ovarian cancer (OV) patients at three postoperative time points retained exclusively in 10x format, we first conducted sub-group analysis of the single-cell data at these time points and annotated the cell types for each subgroup. The results revealed that cells were divided into 17 subgroups (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA), corresponding to four cell types: Epithelial cells, Fibroblasts, CD8\u0026thinsp;+\u0026thinsp;T-cells, and Macrophages (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB). Subgroup 14 was identified as CD8\u0026thinsp;+\u0026thinsp;T-cells. Annotation indicated that more than 90% of the cells were Epithelial cells (91.99%), while Fibroblasts accounted for 5.72%, and CD8\u0026thinsp;+\u0026thinsp;T-cells and Macrophages were relatively rare, comprising only 1.48% and 0.80%, respectively. Differential expression analysis identified 2,251 genes across these subgroups (Supplementary S1).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFurther, using this dataset, we analyzed the potential developmental paths of these cells, referred to as cell trajectories. Cell trajectory analysis, which constructs a pseudo time sequence, can infer the differentiation trajectories or evolutionary processes of cell subtypes, and is frequently utilized in developmental studies. The analysis showed that CD8\u0026thinsp;+\u0026thinsp;T-cells and Macrophages clustered into one branch and appeared at later time points (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC). Epithelial and Fibroblast cells exhibited a higher degree of differentiation as they were present at all three time points, reflecting their high heterogeneity. Adjusting for root time point also revealed that the temporal trajectories of CD8\u0026thinsp;+\u0026thinsp;T-cells and Macrophages were longer than those of other cell types (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eD), indicating a later emergence of immune cells. Data related to this analysis is available in Supplementary S2.\u003c/p\u003e \u003cp\u003eValidation was carried out using datasets GSE147082 and GSE130000. The GSE130000 dataset included eight ovarian cancer samples, comprising four primary tumors, two peritoneal metastases, and two recurrent tumors. Employing the same methodology for sub-group analysis on the single-cell data from GSE147082 and GSE130000, results showed that Epithelial cells represented the majority in both datasets, accounting for 27.51% and 51.30%, respectively. T-cells and Macrophages were each classified into separate subgroups, with T-cells constituting 13.25% and 2.99%, respectively, and occupying a small proportion of the cell population (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eE, F).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eFurther Subgrouping of T-cell Subsets\u003c/h2\u003e \u003cp\u003eFurther analysis of the T-cell subgroups from datasets GSE147082 and GSE130000 suggested potential finer subdivisions within T cells, such as CD8\u0026thinsp;+\u0026thinsp;and CD4\u0026thinsp;+\u0026thinsp;T cells. Cell trajectory analysis confirmed the distribution of T cells across two branches (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eG, results from the GSE130000 dataset), prompting further subgrouping of the T cell clusters. Upon further classification of the cells annotated as T cells, significant clustering into two distinct categories was observed, designated as sub-cluster0 and sub-cluster1 (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eH). These two sub-clusters exhibited significant differences in pseudotime, with sub-cluster0 showing notably greater pseudotime than sub-cluster1 (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eI, J). This indicates a dynamic trend in T cell variation during the development of ovarian cancer. Further identification of differential marker genes between these two T cell types will aid in understanding the underlying mechanisms driving these dynamics at the gene expression level.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eIdentification of Marker Genes for Immune T Cells\u003c/h2\u003e \u003cp\u003eIn the GSE130000 dataset, T cells were categorized into two groups: sub-cluster0 and sub-cluster1. We identified genes differentially expressed between these T cell subgroups, selecting those with a fold change greater than or equal to 2 and an FDR less than 0.05 as significant. A total of 41 genes were identified (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA, Supplementary S3), among which CCL5 and GZMA were recognized as marker genes for sub-cluster0, while the remaining 39 genes were markers for sub-cluster1. The average expression levels of these 41 genes were computed across cell types, revealing that CCL5 and GZMA were predominantly expressed in sub-cluster0, with minimal expression in other cell types (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB). Interestingly, sub-cluster1 exhibited a similar expression profile to the macrophage cell group, and both were clustered in the same branch. Figure\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC displays the top 10 significantly different genes, encompassing a total of 12.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eUsing data from the Cell Marker website, preliminary annotations were made for two T cell sub-clusters. Based on the overlap of marker genes, sub-cluster0 is likely a CD4\u0026thinsp;+\u0026thinsp;cytotoxic T cell type, while sub-cluster1 might be either a Natural Killer T (NKT) cell or a Microglial cell. Furthermore, GO and KEGG pathway enrichment analyses were performed for sub-cluster1, as sub-cluster0, with only two marker genes, was not amenable to enrichment analysis. The analysis indicated that the marker genes for sub-cluster1 are primarily associated with protein folding and processing. Significant enrichment was also noted in the MAPK signaling pathway (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD). Therefore, sub-cluster0 can be defined as a CCL5-GZMA activated type T cell, and sub-cluster1 as a protein processing type T cell. After incorporating metastatic and recurrent samples from the GSE130000 dataset previously omitted from the analysis, the same clustering method was applied, and T cell clusters were selected for further subdivision. Following parameter optimization, it was possible to distinguish 11 subtypes. A comparison of marker genes identified for sub-cluster0 and sub-cluster1 with these 11 subtypes revealed that subtype 4 was most similar to sub-cluster0 (sharing a similar expression pattern of marker genes), while the expression profiles of other subtypes were most similar to sub-cluster1, as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE. Subsequently, we quantified these T cell types across primary, metastatic, and relapse samples. We found that subtype 4 was most prevalent in primary samples, whereas other T cell types were more abundant in metastatic samples, as depicted in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eF.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eAssociation Between T Cell Marker Genes and Prognosis in Ovarian Cancer\u003c/h2\u003e \u003cp\u003eIn the TCGA dataset, we analyzed 41 significantly different marker genes to cluster the ovarian cancer (OV) samples, resulting in four distinct groups (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). These were defined as T-cluster1 (n\u0026thinsp;=\u0026thinsp;100), T-cluster2 (n\u0026thinsp;=\u0026thinsp;98), T-cluster3 (n\u0026thinsp;=\u0026thinsp;154), and T-cluster4 (n\u0026thinsp;=\u0026thinsp;21). T-cluster2 exhibited the lowest overall survival rate, while T-cluster3 demonstrated the highest (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). Similarly, the disease-free progression survival for T-cluster2 was significantly shorter than that for T-cluster3 (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC). Notably, eight genes were most highly expressed in T-cluster2 (highlighted in the red box in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA): DUSP2, ATF3, NR4A1, HSPH1, CLK1, RGS2, RGS1, and TNFAIP3. This observation leads to the hypothesis that the T cells of sub-cluster1, identified in single-cell groups, might represent a dysfunctional subset contributing to the progression of ovarian cancer. This dysfunction could explain the significant upregulation of sub-cluster1 marker genes in T-cluster2 samples, which are associated with the poorest prognosis.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eAssessment of T Cell Scores in Ovarian Cancer Samples\u003c/h2\u003e \u003cp\u003eFrom the previous analysis, eight differentially expressed genes were identified as marker genes for T cell sub-cluster1. We attempted to construct an ovarian cancer sample T cell scoring model based on the expression profiles of these eight genes. Initially, we compared the expression of these genes across the four TCGA ovarian cancer subgroups. Six of the genes showed significant differential expression, all highest in T-cluster2 (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD). The expression of these eight genes was not significantly correlated with overall survival times of ovarian cancer patients (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE). However, three genes\u0026mdash;RGS2, DUSP2, and RGS1\u0026mdash;showed a significant association with disease-free progression time (log-rank P\u0026thinsp;\u0026lt;\u0026thinsp;0.05) and were identified as adverse prognostic factors (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE). Excluding HSPH1 and CLK1, which did not show significant differences, the average expression levels of the remaining six genes were calculated as the T cell score for each sample. We observed that patients with lower scores (L1) exhibited better overall survival and disease-free progression rates compared to those with higher scores (L2, Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eRelationship Between T Cell Scores and Clinical Features in Ovarian Cancer Patients\u003c/h2\u003e \u003cp\u003eUtilizing the TCGA ovarian cancer dataset, which includes information on stage, grade, subdivision, lymphatic invasion, new events, tumor residuals, venous invasion, and therapy outcomes, we compared the T cell scores across these clinical features. No significant differences were observed in the T cell scores among the various samples under these clinical categories (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eG). This indicates that there is no apparent correlation between T cell scores and these clinical characteristics of ovarian cancer patients.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eCellular Communication and Related Pathway Analysis\u003c/h2\u003e \u003cp\u003eUsing the GSE130000 dataset, we explored potential communication links between different cell types. The CellChat package, which provides tools for studying intercellular communication, models the probability of cell communication by integrating the expression of signaling ligands, receptors, and their cofactors with prior knowledge of their interactions. After inferring the intercellular communication network, CellChat facilitates further data exploration, analysis, and visualization. Figure\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA displays the interaction frequency of each cell type with other groups, showing that epithelial cells, endothelial cells, and smooth muscle cells have the highest frequency of interaction (Supplementary S4). By assessing the weights between cell groups, the interactions between epithelial cells and neurons, as well as induced pluripotent stem (iPS) cells, were found to have the highest interaction strength (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB, Supplementary S5). Figure\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC illustrates the interaction relationships of each cell type with other cell groups, highlighting that both T cell subgroups, sub-cluster0 and sub-cluster1, exhibited the highest interaction frequency with smooth muscle cells.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFurther analysis identified 44 signaling pathways contributing to intercellular interactions (Supplementary S6). We focused particularly on the pathways related to two T cell subgroups, sub-cluster0 and sub-cluster1, involving 11 pathways in total (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD). Among these, the CCL and IL2 signaling pathways contributed most significantly to the interactions between sub-cluster0 and sub-cluster1. Additionally, the CCLX signaling pathway was a major contributor to interactions between sub-cluster0, sub-cluster1, and macrophages. Gene expression analysis related to these signaling pathways confirmed these findings; high expression levels of genes such as CCL5, CCL3, and CCL4 were observed exclusively in sub-cluster0, sub-cluster1, and macrophages. Genes involved in the IL2 signaling pathway were predominantly expressed in sub-cluster0 and sub-cluster1 (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE). Given that the CCL5 gene is a marker gene for T cell sub-cluster0, these results suggest that aberrant expression of CCL5 may be closely linked to the differentiation of T cells into the sub-cluster0 and sub-cluster1 subgroups.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eAnalysis of Metabolic Pathways\u003c/h2\u003e \u003cp\u003eTo further investigate the functional characteristics of cell subgroups, single-cell data were analyzed using the Reactome database. Leveraging the conservation inherent in life sciences, data from different species were automatically mapped to a common pathway space, and enrichment scores for each pathway were calculated. We then assessed the differences in these pathway enrichment scores among various cell subgroups (Supplementary S7). The top 40 pathways with the greatest disparities were selected for further study, and a heatmap of these top 40 pathways was created (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eF). Notably, the FGFR1c and Klotho ligand binding and activation pathways showed the lowest enrichment scores in the two T cell subgroups, sub-cluster0 and sub-cluster1, while the MGMT-mediated DNA damage reversal pathway was specifically enriched at the lowest level uniquely in sub-cluster0 (note that deeper red indicates lower scores). Figures\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eG and \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eH display the enrichment scores for these pathways across eight cell subgroups.\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eOvarian cancer remains a principal cause of mortality among patients with gynecological malignancies. Despite advancements in cytoreductive surgery and the administration of platinum-based chemotherapy, the prognosis for ovarian cancer patients continues to be dismal due to frequent dissemination and relapse \u003csup\u003e10\u003c/sup\u003e. This underscores the critical need for the development of novel targeted therapies. Recent studies underscore the significant role of the tumor microenvironment in the pathogenesis of ovarian cancer. Insights derived from comprehensive genomic, transcriptomic, and proteomic analyses have highlighted the tumor microenvironment not only as a key player in oncogenesis but also as a promising target for therapeutic intervention \u003csup\u003e11\u003c/sup\u003e. Addressing this component could pave the way for breakthroughs in the treatment modalities for ovarian cancer.\u003c/p\u003e \u003cp\u003eAlthough tumor-infiltrating lymphocytes (TILs), particularly CD8\u0026thinsp;+\u0026thinsp;T cells, are recognized as a positive prognostic marker in numerous solid tumors, their inability to effectively eradicate cancer cells persists \u003csup\u003e12,13\u003c/sup\u003e. One underlying reason for this immunological control failure is the extensive presence of immunosuppressive mechanisms within the TME, which impair the effector functions of infiltrating T cells \u003csup\u003e14\u003c/sup\u003e. A primary hallmark of T cell dysfunction in this context is the upregulation of PD-1 on T cells. The altered functional state of PD-1\u0026thinsp;+\u0026thinsp;T cells, referred to as T cell exhaustion, has been extensively documented and studied in human infections and cancers \u003csup\u003e15\u003c/sup\u003e. The successful revitalization of T cell function through the blockade of PD-1 or its ligand PD-L1 underscores the significance of the PD-1/PD-L1 axis in T cell dysfunction. However, the majority of patients still do not respond or fail to achieve sustained responses. Specifically, it has been observed that a subset of T cells co-expresses various inhibitory receptors, known as immune checkpoints, and this expression escalates with the progression of dysfunction \u003csup\u003e16,17\u003c/sup\u003e. Given the exposure of the intratumoral T cell repertoire to a plethora of different immunosuppressive mechanisms, a wide spectrum of T cell dysfunction states can be anticipated. The functional states of intratumoral T cells exhibit variability across different tumors, yet even within a single tumor, T cell infiltration demonstrates significant heterogeneity. This heterogeneity is evident in at least three aspects: (\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) the heterogeneity of traditional T cell subsets, including CD8+, CD4\u0026thinsp;+\u0026thinsp;Th1, Th2, Treg cells, and γδ T cells; (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e) the diversity in the degree and type of T cell dysfunction within given subsets; (\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e) the heterogeneous capacity of individual TCR to recognize tumor antigens \u003csup\u003e18\u003c/sup\u003e. Due to this heterogeneity, accurately capturing the range of T cell states within tumors requires analysis at the single-cell level. Recent studies utilizing such techniques have provided detailed descriptions of the tumor ecosystem across various cancer types. For instance, single-cell sequencing of 11 breast cancer patients revealed three immune cell clusters characterized by T cells exhibiting either an exhausted or regulatory phenotype, B cells, and predominantly M2-like macrophages \u003csup\u003e19\u003c/sup\u003e. Another study extensively analyzed immune infiltration in 73 patients with clear cell renal carcinoma (RCC), identifying 22 T cell phenotypes and 17 tumor-associated macrophage phenotypes. Within the T cell clusters, seven distinct clusters of PD-1-expressing CD8\u0026thinsp;+\u0026thinsp;T cells were observed. Notably, the co-expression of other inhibitory receptors, including Tim-3 and CTLA-4, as well as the expression of co-stimulatory or proliferation markers, varied greatly among these clusters, suggesting that they may reflect different states of dysfunction \u003csup\u003e20\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eOvarian cancer exhibits significant immunogenic potential, evidenced by its expression of numerous tumor antigens and the correlation between TIL accumulation and improved patient outcomes \u003csup\u003e21,22\u003c/sup\u003e. Despite this, the response of ovarian cancer to endogenous immune modulation via immune checkpoint inhibitors has been modest \u003csup\u003e23,24\u003c/sup\u003e. The identification and characterization of these tumor-reactive T cells represent a promising strategy for the development and optimization of cellular therapies. TILs, particularly subsets of these cells, have been identified as predictive biomarkers for outcomes in patients with epithelial ovarian cancer \u003csup\u003e25\u003c/sup\u003e. In a multicenter observational trial involving high-grade serous ovarian cancer, a notable finding was the heightened infiltration of CD8\u0026thinsp;+\u0026thinsp;T cells compared to other ovarian cancer subtypes. Detailed analysis revealed a dose-response relationship between the levels of CD8\u0026thinsp;+\u0026thinsp;TILs and patient survival rates. Specifically, patients with high CD8\u0026thinsp;+\u0026thinsp;TIL levels exhibited a median survival of 5.1 years, significantly longer than the 2.8 to 3.8 years observed in patients with low to moderate levels of these cells \u003csup\u003e26\u003c/sup\u003e. Notably, the survival benefit associated with CD8\u0026thinsp;+\u0026thinsp;TILs was prominent in patients harboring BRCA1 mutations but not in those with BRCA2 mutations \u003csup\u003e27\u003c/sup\u003e. Despite the known role of CD4\u0026thinsp;+\u0026thinsp;helper T cells in augmenting CD8\u0026thinsp;+\u0026thinsp;T cell responses, no direct correlation between CD4\u0026thinsp;+\u0026thinsp;T cell infiltration and survival outcomes was observed. Conversely, the presence of immunosuppressive Tregs consistently predicted poorer survival outcomes \u003csup\u003e28\u003c/sup\u003e. Interestingly, only a small fraction of the TIL population is tumor-reactive, with the majority being inactive bystanders. Activation markers such as CD25, CD38, and CD137, as well as exhaustion markers like TIM3, PD-1, TIGIT, and CTLA4\u0026mdash;often upregulated due to chronic antigen exposure\u0026mdash;were prevalent among TILs. The presence of these markers serves as a surrogate for the activity of immunoreactive TILs, and increased populations of PD-1\u0026thinsp;+\u0026thinsp;TILs have been linked with improved survival outcomes in epithelial ovarian cancer \u003csup\u003e29,30\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThis study further analyzed the data from ovarian cancer single-cell and bulk RNA sequencing available in the GSE158722 dataset, wherein the ovarian cancer cells were annotated into four cell types: Epithelial cells, Fibroblasts, CD8\u0026thinsp;+\u0026thinsp;T-cells, and Macrophages. A vast majority of the cells were identified as Epithelial cells and Fibroblasts, with immune-related cells like T-cells comprising a small proportion. This phenomenon was also observed in two other independent datasets. In pseudotime sequence analysis, after time adjustment, the pseudotime results indicated that CD8\u0026thinsp;+\u0026thinsp;T cells and Macrophages appeared later than other cell types. Additionally, T cells exhibited bifurcation, reflecting their potential differentiation states. Utilizing the GSE130000 dataset, we identified differentially expressed genes between two T cell subgroups, totaling 41 genes. Among these, CCL5 and GZMA were marker genes for sub-cluster 0, while the remaining 39 were marker genes for sub-cluster 1. These 41 marker genes differentiated the TCGA OV dataset into four classes, where samples in T-cluster 2 exhibited the lowest overall survival, and those in T-cluster 3, the highest. These findings suggest that T cells in sub-cluster 1 may represent a dysfunctional subgroup contributing to the progression of ovarian cancer. Based on the eight genes (DUSP2、ATF3、NR4A1、HSPH1、CLK1、RGS2、RGS1 and TNFAIP3) exhibiting the highest expression levels in samples from T-cluster 2, we constructed a T cell scoring model for ovarian cancer samples. We observed that patients in the low score group (L1) had better overall survival and disease-free progression rates compared to those in the high score group (L2). Moreover, there was no significant association between T cell scores and these clinical features. Cell communication analysis indicated that the CCL and IL2 signaling pathways contributed most significantly to communication between sub-cluster 0 and sub-cluster 1 T cells. Given that the CCL5 gene is a marker gene for T cell sub-cluster 0, these results suggest that abnormal expression of CCL5 may be closely associated with the differentiation of T cells into sub-cluster 0 and sub-cluster 1. Pathway analysis revealed that the FGFR1c and Klotho ligand binding and activation pathways had the lowest enrichment scores in both T cell subgroups, whereas the MGMT-mediated DNA damage reversal pathway was uniquely least enriched in sub-cluster 0.\u003c/p\u003e \u003cp\u003eCCL5, also known as RANTES, is involved in several signaling pathways, primarily related to immune function and inflammation. It acts through the CCR5 receptor among others, playing a critical role in various cellular processes including chemotaxis, immune surveillance, and inflammation ​\u003csup\u003e31\u003c/sup\u003e. CCL5 interacts with multiple signaling pathways. One of its primary roles is in the chemokine signaling pathway, particularly involving the CCL5/CCR5 axis, which is crucial for the regulation of immune cell migration and activation. This axis is also significant in cancer progression, influencing tumor microenvironment by promoting tumor cell invasion, metastasis, and immune evasion ​\u003csup\u003e32\u003c/sup\u003e​. Additionally, CCL5 is involved in the MAPK signaling pathway, which plays a role in cell proliferation, survival, and differentiation \u003csup\u003e31\u003c/sup\u003e. In the study on Hirschsprung's disease, \u003cem\u003ePan et al.\u003c/em\u003e utilized RNA-sequencing to identify several genes, including ATF3, NOS1, CCL5, DUSP1, CXCL3, VIP, and FOSB, that might be involved in the development of HSCR through their effects on the nervous system. Additionally, their analysis of the protein-protein interaction (PPI) network revealed potential interactions among ATF3, CCL5, and TNFAIP3, suggesting a complex interplay that could underlie the pathogenesis of the disease \u003csup\u003e33\u003c/sup\u003e. Although scRNA-seq and bulk RNA sequencing analyses offer significant advantages and have identified differentially expressed factors in present study, the specific interactions between CCL-5 and the differentially expressed genes and related signaling pathways within sub-cluster1 should be further validated in human tissues and cell lines. The search for reliable biomarkers predictive of immune therapy outcomes in ovarian cancer remains a challenging and formidable task.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn conclusion, the present study identifies differential expression of CCL-5 as a biomarker for distinct T-cell subtypes in ovarian cancer by integrating scRNA-seq and bulk RNA-seq data. We further discovered that a T-cell scoring model, constructed using eight genes significantly associated with dysfunctional T-cell subtypes, predicts lower overall survival and disease-free progression rates in ovarian cancer patients with lower scores, which appeared independent of clinical features. These findings suggest that the T-cell scoring model could serve as a reliable prognostic indicator for patients with ovarian cancer. Additionally, CCL-5 may represent a significant target for immunotherapy in ovarian cancer.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch1\u003eConflicts of Interest\u003c/h1\u003e\n\u003cp\u003eThe authors declare that they have no conflicts of interest related to this work.\u003c/p\u003e\n\u003ch1\u003eAuthor Contributions\u003c/h1\u003e\n\u003cp\u003eSHZ and HX:\u0026nbsp;Study design,\u0026nbsp;data analysis.\u0026nbsp;GLY: Study design and data analysis and manuscript writing.\u0026nbsp;LSH: data analysis.\u0026nbsp;All authors have read, edited and approved the final version of the manuscript.\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003ch1\u003eFunding\u003c/h1\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003ch1\u003eData availability\u003c/h1\u003e\n\u003cp\u003eSequence data that support the findings of this study have been deposited in https://www.ncbi.nlm.ni-h.gov/geo/query/acc.cgi?\u003c/p\u003e\n\u003ch1\u003eAcknowledgments\u003c/h1\u003e\n\u003cp\u003eEthical approval: All procedures in studies involving human participants were performed in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration and its later amendments. (Name and affiliation of the approving ethics committee: Institutional Ethics Committee of Peking University Third Hospital; No. A2023035).\u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eTorre, L. A.\u003cem\u003e et al.\u003c/em\u003e Ovarian cancer statistics, 2018. \u003cem\u003eCA Cancer J Clin\u003c/em\u003e, doi:10.3322/caac.21456 (2018).\u003c/li\u003e\n\u003cli\u003eRoberts, C. M., Cardenas, C. \u0026amp; Tedja, R. The role of intra-tumoral heterogeneity and its clinical relevance in epithelial ovarian cancer recurrence and metastasis. \u003cem\u003eCancers\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 1083 (2019).\u003c/li\u003e\n\u003cli\u003eZhang, Y. \u0026amp; Zhang, Z. The history and advances in cancer immunotherapy: understanding the characteristics of tumor-infiltrating immune cells and their therapeutic implications. \u003cem\u003eCellular \u0026amp; molecular immunology\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 807-821, doi:10.1038/s41423-020-0488-6 (2020).\u003c/li\u003e\n\u003cli\u003eRibas, A. \u0026amp; Wolchok, J. D. Cancer immunotherapy using checkpoint blockade. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e359\u003c/strong\u003e, 1350-1355, doi:10.1126/science.aar4060 (2018).\u003c/li\u003e\n\u003cli\u003eTanaka, A. \u0026amp; Sakaguchi, S. Regulatory T cells in cancer immunotherapy. \u003cem\u003eCell research\u003c/em\u003e \u003cstrong\u003e27\u003c/strong\u003e, 109-118, doi:10.1038/cr.2016.151 (2017).\u003c/li\u003e\n\u003cli\u003eNowak, M. \u0026amp; Klink, M. The Role of Tumor-Associated Macrophages in the Progression and Chemoresistance of Ovarian Cancer. \u003cem\u003eCells\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, doi:10.3390/cells9051299 (2020).\u003c/li\u003e\n\u003cli\u003eGitto, S. B., Ihewulezi, C. J. N. \u0026amp; Powell, D. J., Jr. Adoptive T cell therapy for ovarian cancer. \u003cem\u003eGynecol Oncol\u003c/em\u003e \u003cstrong\u003e186\u003c/strong\u003e, 77-84, doi:10.1016/j.ygyno.2024.04.001 (2024).\u003c/li\u003e\n\u003cli\u003eMoncada, R.\u003cem\u003e et al.\u003c/em\u003e Integrating microarray-based spatial transcriptomics and single-cell RNA-seq reveals tissue architecture in pancreatic ductal adenocarcinomas. \u003cem\u003eNat Biotechnol\u003c/em\u003e \u003cstrong\u003e38\u003c/strong\u003e, 333-342, doi:10.1038/s41587-019-0392-8 (2020).\u003c/li\u003e\n\u003cli\u003eKuksin, M.\u003cem\u003e et al.\u003c/em\u003e Applications of single-cell and bulk RNA sequencing in onco-immunology. \u003cem\u003eEuropean journal of cancer (Oxford, England : 1990)\u003c/em\u003e \u003cstrong\u003e149\u003c/strong\u003e, 193-210, doi:10.1016/j.ejca.2021.03.005 (2021).\u003c/li\u003e\n\u003cli\u003eKonstantinopoulos, P. A. \u0026amp; Matulonis, U. A. Clinical and translational advances in ovarian cancer therapy. \u003cem\u003eNature cancer\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 1239-1257, doi:10.1038/s43018-023-00617-9 (2023).\u003c/li\u003e\n\u003cli\u003eYang, Y., Yang, Y., Yang, J., Zhao, X. \u0026amp; Wei, X. Tumor Microenvironment in Ovarian Cancer: Function and Therapeutic Strategy. \u003cem\u003eFrontiers in cell and developmental biology\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 758, doi:10.3389/fcell.2020.00758 (2020).\u003c/li\u003e\n\u003cli\u003eFridman, W. H., Pag\u0026egrave;s, F., Saut\u0026egrave;s-Fridman, C. \u0026amp; Galon, J. The immune contexture in human tumours: impact on clinical outcome. \u003cem\u003eNature reviews. Cancer\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 298-306, doi:10.1038/nrc3245 (2012).\u003c/li\u003e\n\u003cli\u003eBoon, T., Coulie, P. G., Van den Eynde, B. J. \u0026amp; van der Bruggen, P. Human T cell responses against melanoma. \u003cem\u003eAnnual review of immunology\u003c/em\u003e \u003cstrong\u003e24\u003c/strong\u003e, 175-208, doi:10.1146/annurev.immunol.24.021605.090733 (2006).\u003c/li\u003e\n\u003cli\u003eChen, D. S. \u0026amp; Mellman, I. Oncology meets immunology: the cancer-immunity cycle. \u003cem\u003eImmunity\u003c/em\u003e \u003cstrong\u003e39\u003c/strong\u003e, 1-10, doi:10.1016/j.immuni.2013.07.012 (2013).\u003c/li\u003e\n\u003cli\u003eBaitsch, L.\u003cem\u003e et al.\u003c/em\u003e Exhaustion of tumor-specific CD8⁺ T cells in metastases from melanoma patients. \u003cem\u003eJ Clin Invest\u003c/em\u003e \u003cstrong\u003e121\u003c/strong\u003e, 2350-2360, doi:10.1172/jci46102 (2011).\u003c/li\u003e\n\u003cli\u003eThommen, D. S.\u003cem\u003e et al.\u003c/em\u003e Progression of Lung Cancer Is Associated with Increased Dysfunction of T Cells Defined by Coexpression of Multiple Inhibitory Receptors. \u003cem\u003eCancer immunology research\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, 1344-1355, doi:10.1158/2326-6066.Cir-15-0097 (2015).\u003c/li\u003e\n\u003cli\u003eWherry, E. J. T cell exhaustion. \u003cem\u003eNature immunology\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 492-499, doi:10.1038/ni.2035 (2011).\u003c/li\u003e\n\u003cli\u003eThommen, D. S. \u0026amp; Schumacher, T. N. T Cell Dysfunction in Cancer. \u003cem\u003eCancer Cell\u003c/em\u003e \u003cstrong\u003e33\u003c/strong\u003e, 547-562, doi:10.1016/j.ccell.2018.03.012 (2018).\u003c/li\u003e\n\u003cli\u003eChung, W.\u003cem\u003e et al.\u003c/em\u003e Single-cell RNA-seq enables comprehensive tumour and immune cell profiling in primary breast cancer. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 15081, doi:10.1038/ncomms15081 (2017).\u003c/li\u003e\n\u003cli\u003eChevrier, S.\u003cem\u003e et al.\u003c/em\u003e An Immune Atlas of Clear Cell Renal Cell Carcinoma. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e169\u003c/strong\u003e, 736-749.e718, doi:10.1016/j.cell.2017.04.016 (2017).\u003c/li\u003e\n\u003cli\u003eRodriguez-Garcia, A., Minutolo, N. G., Robinson, J. M. \u0026amp; Powell, D. J. T-cell target antigens across major gynecologic cancers. \u003cem\u003eGynecol Oncol\u003c/em\u003e \u003cstrong\u003e145\u003c/strong\u003e, 426-435, doi:10.1016/j.ygyno.2017.03.510 (2017).\u003c/li\u003e\n\u003cli\u003eZhang, L.\u003cem\u003e et al.\u003c/em\u003e Intratumoral T cells, recurrence, and survival in epithelial ovarian cancer. \u003cem\u003eThe New England journal of medicine\u003c/em\u003e \u003cstrong\u003e348\u003c/strong\u003e, 203-213, doi:10.1056/NEJMoa020177 (2003).\u003c/li\u003e\n\u003cli\u003eDisis, M. L.\u003cem\u003e et al.\u003c/em\u003e Efficacy and Safety of Avelumab for Patients With Recurrent or Refractory Ovarian Cancer: Phase 1b Results From the JAVELIN Solid Tumor Trial. \u003cem\u003eJAMA Oncol\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 393-401, doi:10.1001/jamaoncol.2018.6258 (2019).\u003c/li\u003e\n\u003cli\u003eMatulonis, U. A.\u003cem\u003e et al.\u003c/em\u003e Antitumor activity and safety of pembrolizumab in patients with advanced recurrent ovarian cancer: results from the phase II KEYNOTE-100 study. \u003cem\u003eAnnals of oncology : official journal of the European Society for Medical Oncology\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, 1080-1087, doi:10.1093/annonc/mdz135 (2019).\u003c/li\u003e\n\u003cli\u003eJames, F. R.\u003cem\u003e et al.\u003c/em\u003e Association between tumour infiltrating lymphocytes, histotype and clinical outcome in epithelial ovarian cancer. \u003cem\u003eBMC Cancer\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 657, doi:10.1186/s12885-017-3585-x (2017).\u003c/li\u003e\n\u003cli\u003eGoode, E. L.\u003cem\u003e et al.\u003c/em\u003e Dose-Response Association of CD8+ Tumor-Infiltrating Lymphocytes and Survival Time in High-Grade Serous Ovarian Cancer. \u003cem\u003eJAMA Oncol\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, e173290, doi:10.1001/jamaoncol.2017.3290 (2017).\u003c/li\u003e\n\u003cli\u003eClarke, B.\u003cem\u003e et al.\u003c/em\u003e Intraepithelial T cells and prognosis in ovarian carcinoma: novel associations with stage, tumor type, and BRCA1 loss. \u003cem\u003eModern pathology : an official journal of the United States and Canadian Academy of Pathology, Inc\u003c/em\u003e \u003cstrong\u003e22\u003c/strong\u003e, 393-402, doi:10.1038/modpathol.2008.191 (2009).\u003c/li\u003e\n\u003cli\u003eSato, E.\u003cem\u003e et al.\u003c/em\u003e Intraepithelial CD8+ tumor-infiltrating lymphocytes and a high CD8+/regulatory T cell ratio are associated with favorable prognosis in ovarian cancer. \u003cem\u003eProceedings of the National Academy of Sciences of the United States of America\u003c/em\u003e \u003cstrong\u003e102\u003c/strong\u003e, 18538-18543, doi:10.1073/pnas.0509182102 (2005).\u003c/li\u003e\n\u003cli\u003eSantoiemma, P. P.\u003cem\u003e et al.\u003c/em\u003e Systematic evaluation of multiple immune markers reveals prognostic factors in ovarian cancer. \u003cem\u003eGynecol Oncol\u003c/em\u003e \u003cstrong\u003e143\u003c/strong\u003e, 120-127, doi:10.1016/j.ygyno.2016.07.105 (2016).\u003c/li\u003e\n\u003cli\u003eWebb, J. R., Milne, K., Kroeger, D. R. \u0026amp; Nelson, B. H. PD-L1 expression is associated with tumor-infiltrating T cells and favorable prognosis in high-grade serous ovarian cancer. \u003cem\u003eGynecol Oncol\u003c/em\u003e \u003cstrong\u003e141\u003c/strong\u003e, 293-302, doi:10.1016/j.ygyno.2016.03.008 (2016).\u003c/li\u003e\n\u003cli\u003eZhu, G. D.\u003cem\u003e et al.\u003c/em\u003e Identification of differentially expressed genes and signaling pathways with Candida infection by bioinformatics analysis. \u003cem\u003eEuropean journal of medical research\u003c/em\u003e \u003cstrong\u003e27\u003c/strong\u003e, 43, doi:10.1186/s40001-022-00651-w (2022).\u003c/li\u003e\n\u003cli\u003eAldinucci, D., Borghese, C. \u0026amp; Casagrande, N. The CCL5/CCR5 Axis in Cancer Progression. \u003cem\u003eCancers\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, doi:10.3390/cancers12071765 (2020).\u003c/li\u003e\n\u003cli\u003ePan, W. K.\u003cem\u003e et al.\u003c/em\u003e Identifying key genes associated with Hirschsprung\u0026apos;s disease based on bioinformatics analysis of RNA-sequencing data. \u003cem\u003eWorld journal of pediatrics : WJP\u003c/em\u003e\u003cstrong\u003e13\u003c/strong\u003e, 267-273, doi:10.1007/s12519-017-0002-0 (2017).\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":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"ovarian cancer, T-cell marker genes, Immunotherapy, Single-cell RNA-sequencing","lastPublishedDoi":"10.21203/rs.3.rs-4721266/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4721266/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground: \u003c/strong\u003eOvarian cancer, with high mortality and often late diagnosis, shows high recurrence despite treatment. The variable effectiveness of immunotherapy highlights the urgent need for personalized, advanced therapeutic strategies.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods:\u003c/strong\u003e To investigate T-cell marker genes, single-cell RNA-sequencing (scRNA-seq) data were sourced from the Gene Expression Omnibus (GEO) database. Additionally, bulk RNA-sequencing data along with clinical information from ovarian cancer patients were retrieved from the Cancer Genome Atlas (TCGA) database to establish a prognostic signature. This study involved survival analysis to evaluate associations between different risk groups, and explored cellular communication and relevant pathway analyses, including metabolic pathways.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults: \u003c/strong\u003eWe identified 41 genes showing varied expression between two T-cell subclusters, marking subcluster 0 with CCL5 and GZMA, and attributing the rest to subcluster 1. These markers delineate four prognostic groups within the TCGA OV dataset, with T-cluster 2 exhibiting the poorest survival, in contrast to T-cluster 3, which shows the best. Analysis suggests subcluster 1 T-cells might be dysfunctional, potentially exacerbating ovarian cancer progression. We also developed a T-cell scoring model using eight significant genes, showing improved survival in the low-score group. Moreover, cellular and metabolic pathway analyses underscored the importance of CCL, IL2 and MGMT pathways in these subclusters.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusions:\u003c/strong\u003e The study identifies CCL-5 as a biomarker for T-cell subtypes in ovarian cancer using scRNA-seq and bulk RNA-seq data. A T-cell scoring model based on eight genes predicts survival and progression rates, independent of clinical features. This model could be a prognostic indicator and CCL-5 a potential immunotherapy target in ovarian cancer.\u003c/p\u003e","manuscriptTitle":"Single‑cell and bulk RNA sequencing identifes T cell marker genes to predict the prognosis of ovrian caner","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-08-11 12:35:42","doi":"10.21203/rs.3.rs-4721266/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-09-06T12:20:21+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-09-06T09:13:52+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"205933708866165354210827498605684652692","date":"2026-09-05T09:08:29+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-09-05T07:30:14+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"184402437674383022962699019399313997509","date":"2026-09-04T09:10:52+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"53967920953419536370668947485363381323","date":"2026-09-04T05:54:20+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"121737439310707988016419893204578261679","date":"2026-09-04T05:31:21+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"290749889639839159294070202528710362988","date":"2026-09-04T01:55:04+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"268785563302404555957728248580894123782","date":"2026-09-03T15:46:50+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-09-03T14:09:31+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"187668254805636124964733041913571012336","date":"2026-09-03T13:56:30+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-09-03T13:03:20+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"210328813875657720924434310229483741929","date":"2026-09-03T11:42:37+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"204471320254726096934065697118766437790","date":"2026-09-03T09:19:37+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"187927536394796130463018480190554892317","date":"2026-09-03T08:55:47+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"232613047799515061306163541267706321824","date":"2026-09-03T08:30:14+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-07-28T11:44:49+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-07-22T14:11:01+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2024-07-16T04:47:31+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-07-15T06:03:33+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2024-07-11T02:36:24+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"634d3f1c-0ae8-4545-bf21-0324e996ab76","owner":[],"postedDate":"August 11th, 2024","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"","date":"2026-09-06T12:20:21+00:00","index":193,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-09-06T09:13:52+00:00","index":192,"fulltext":""},{"type":"reviewerAgreed","content":"205933708866165354210827498605684652692","date":"2026-09-05T09:08:29+00:00","index":190,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-09-05T07:30:14+00:00","index":189,"fulltext":""},{"type":"reviewerAgreed","content":"184402437674383022962699019399313997509","date":"2026-09-04T09:10:52+00:00","index":188,"fulltext":""},{"type":"reviewerAgreed","content":"53967920953419536370668947485363381323","date":"2026-09-04T05:54:20+00:00","index":187,"fulltext":""},{"type":"reviewerAgreed","content":"121737439310707988016419893204578261679","date":"2026-09-04T05:31:21+00:00","index":186,"fulltext":""},{"type":"reviewerAgreed","content":"290749889639839159294070202528710362988","date":"2026-09-04T01:55:04+00:00","index":185,"fulltext":""},{"type":"reviewerAgreed","content":"268785563302404555957728248580894123782","date":"2026-09-03T15:46:50+00:00","index":184,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-09-03T14:09:31+00:00","index":182,"fulltext":""},{"type":"reviewerAgreed","content":"187668254805636124964733041913571012336","date":"2026-09-03T13:56:30+00:00","index":181,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-09-03T13:03:20+00:00","index":180,"fulltext":""},{"type":"reviewerAgreed","content":"210328813875657720924434310229483741929","date":"2026-09-03T11:42:37+00:00","index":179,"fulltext":""},{"type":"reviewerAgreed","content":"204471320254726096934065697118766437790","date":"2026-09-03T09:19:37+00:00","index":178,"fulltext":""},{"type":"reviewerAgreed","content":"187927536394796130463018480190554892317","date":"2026-09-03T08:55:47+00:00","index":177,"fulltext":""},{"type":"reviewerAgreed","content":"232613047799515061306163541267706321824","date":"2026-09-03T08:30:14+00:00","index":176,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":35501404,"name":"Health sciences/Oncology/Cancer"},{"id":35501405,"name":"Health sciences/Biomarkers/Prognostic markers"}],"tags":[],"updatedAt":"2024-08-11T12:35:42+00:00","versionOfRecord":[],"versionCreatedAt":"2024-08-11 12:35:42","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4721266","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4721266","identity":"rs-4721266","version":["v1"]},"buildId":"CiT4i_kKBbxQbnFL0ufpk","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}