CD8+ T cell activation in endometrial cancer: prognostic implications and potential for personalized therapy

Background: As an important component in preventing the progression of endometrial cancer, CD8 T cells play a crucial role in this process and are important targets for immunotherapy. However, the status of CD8+ T cells in endometrial cancer and the key genes influencing their activation still remain to be elucidated.

Methods: Genes associated with the activation of CD8+ T cells were identified through differential analysis and weighted gene co-expression network analysis (WGCNA). A risk score model was constructed using the least absolute shrinkage and selection operator (LASSO) and multivariate Cox regression. The clinical characteristics and differences between the high-risk group and the low-risk group were explored, and the applicability of the model to chemotherapy, poly (ADP-ribose) polymerase (PARP) inhibitors, and immune checkpoint inhibitors was evaluated. The characteristics of the model at the single-cell level were studied, and the tumor-suppressive effect of ASB2 was verified through experiments on endometrial cancer cells.

Results: A risk model based on genes related to the activation of CD8+ T cells was constructed, and the prognostic differences were verified using the Kaplan-Meier curve. A nomogram was designed to predict the survival probability. Pathway analysis showed that it was related to metabolism and DNA repair. There were significant differences between the high-risk and low-risk groups in terms of tumor mutational burden (TMB), checkpoint molecules, and major histocompatibility complex (MHC) class I molecules, and they had different sensitivities to different therapies. The tumor-suppressive effect of ASB2 was confirmed in experiments on cell proliferation, invasion, and migration.

Conclusion: This study provides a predictive tool for endometrial cancer. The classification based on the status of CD8+ T cells can distinguish the prognosis and treatment response, highlighting the potential of this model in personalized treatment.

1 Background

Endometrial cancer (EC) is the second most common malignant tumor of female reproductive system. According to the data of GLOBOCAN 2018, there were 382069 new cases and 89929 deaths in 2018 (1), and the incidence of EC is still rapidly rising in China (2). In contrast to other common cancers, the incidence and annual mortality rates of endometrial cancer are increasing (3, 4). According to the pathological characteristics, EC can be divided into two types: endometrioid endometrial cancer (EEC) and uterine serous carcinoma (USC). It is generally believed that EEC contains most EC and has a good prognosis; However, USC contained more aggressive histologies and had worse prognosis (5–7).

Most endometrial cancer can be detected by endometrial biopsy in the early stage, which greatly improves the prognosis and survival rate of patients. However, there are still a considerable number of patients who have distant metastasis when they are found, and another part of patients are not sensitive to chemotherapy, which lead to the increased risk of cancer progression or recurrence, and reduce the survival rate and quality of life of patients (8). Maybe the higher risk can be explained by inter- tumoral and intra-tumoral heterogeneity (9). Therefore, it is important to adopt updated techniques to discuss possible cell subpopulations in EC.

The tumor microenvironment (TME) refers to the cellular environment in which tumor cells exist (10). The tumor microenvironment plays a crucial role in the occurrence, maintenance, and progression of endometrial cancer (11, 12). In the normal endometrium, T cells occupy the predominant subset of immune cells, accounting for 40-80% of CD45+ immune cells (13, 14). For patients with endometrial cancer, tumor-infiltrating T cells(TIL)are also the predominant component of the endometrial cancer microenvironment. Studies have demonstrated that CD8+ T cells are essential in controlling tumor progression. They achieve this by recognizing antigens and directly killing cancer cells via perforin, granzymes, and other cytokines (15). Conversely, CD4+ T cells coordinate various immune responses, either enhancing or suppressing the immune defense against cancer cells (16).

Multiple research efforts have underscored the importance of cytotoxic CD8+ T cells in controlling tumor progression (17, 18). Pathological studies have confirmed that the presence of CD8+ tumor-infiltrating lymphocytes in endometrial cancer can predict a better prognosis (19–21). Furthermore, relevant pathological studies have also demonstrated an inverse correlation between the presence of CD8+ T cells and histological grade, myometrial invasion, and lymph node metastasis. This strongly suggests the resistance role of CD8+ T cells against tumor progression (22). Additionally, the phenomenon of lower T cell density in advanced endometrial cancer further supports the notion that tumors are more likely to progress to later stages when cytotoxic T cell infiltration decreases (23, 24). In endometrial cancer, CD8+ T cells often exhibit a state of functional impairment, which may be exacerbated by various inhibitory signals within the tumor microenvironment (16). This dysfunction could also be related to the tendency of the endometrium to favor a regulatory immune environment in its healthy state (25). Furthermore, in postmenopausal women, while the total number of CD8+ T cells increases in endometrial adenocarcinoma, their cytotoxic activity significantly decreases (26). Considering that most women are at risk of developing endometrial cancer during or after menopause (27), the risk of tumor progression is significantly increased. Additionally, the activity of CD8+ T cells may be significantly reduced due to bystander effects or interaction-induced exhaustion with malignant cells, leading to the emergence of an inhibitory tumor microenvironment (28, 29).

Simply understanding the presence of CD8+ T cells is not sufficient; it is also essential to delve deeper into their internal differentiation and gene activation status. Traditional staining methods are limited to cell identification and cannot provide information about the intracellular state. Therefore, methods that can better measure the transcriptome of specific T cell subsets should be prioritized (30). Simultaneously, this necessitates the adoption of higher-resolution techniques to study the tumor microenvironment and CD8+ T cells in endometrial cancer.

Single-cell RNA sequencing (scRNA-seq) allows for the simultaneous assessment of the full gene expression profiles of thousands of individual cells within tumor tissue. Based on characteristic genes, these cells can be classified into specific subgroups at single-cell resolution. Recent studies have evaluated prognostic factors in tumors based on single-cell features (31–33). The tumor microenvironment in endometrial cancer may play a significant role in prognosis and treatment resistance (34, 35), Single-cell technologies will help us further understand the intricacies within. As for T cell subgroups, some studies recommend assessing the presence based on transcriptomic activity (30, 36). Some molecular biomarkers used for immunotherapy have been developed and validated in clinical practice, including PD-1 or PD-L1 expression levels (37), tumor mutational burden (TMB) (38, 39) and TILs (40).

However, these methods lack certain standards in practical use and show fluctuations in predictive efficacy. The effectiveness of immunotherapy often depends on the degree of immune escape caused by infiltrating T cells expressing inhibitory receptors (41). Therefore, in this study, we aimed to elucidate the immune infiltration characteristics of CD8+ T cells in endometrial cancer. To identify CD8+ T cell-related genes with high clinical value, we integrated bulk-seq and scRNA-seq data from endometrial cancer samples. This approach aimed to elucidate the fundamental mechanisms of CD8+ T cells in tumor progression, develop new molecular stratification and prognostic features, and assist in determining the benefits of different treatment modalities for patients to achieve maximum clinical benefit.

2 Method

2.1 Data download and processing

We downloaded RNA sequencing data for 578 uterine corpora endometrial carcinoma (UCEC) cases, along with 542 mutation datasets and the patients’ Masked Copy Number Segment data from the TCGA database (https://portal.gdc.cancer.gov/). Additionally, we accessed the UCSC Xena website (http://xena.ucsc.edu/) to download clinicopathological characteristics, including age, grade, TNM stage, and MSI value, as well as clinical outcomes of the corresponding UCEC patients.

We utilized GEO datasets GSE21882 to validate the prognostic value of the immune signature, and GSE78220, PRJEB23709, and IMvigor210 to assess the immune signature’s predictive value for immunotherapy response. Datasets E-MTAB-11552, GSE120490, GSE216872, and GSE205209 were employed to explore the relationship between CD8T cells and the development of endometrial cancer.

Five primary endometrial cancer patients from GSE173682 were selected for analysis using their scRNA-seq data to investigate the role of CD8T cells in endometrial cancer.

2.2 Evaluation of immune cell infiltration

The distribution of immune cells in each endometrial cancer tissue sample is assessed using the CIBERSORT algorithm. Additional tumor microenvironment components, including matrix scores and tumor purity, are evaluated with the ESTIMATE package (42). To quantify the relative abundance of Tumor-Infiltrating Cells in UCEC patients, the Single-sample Gene Set Enrichment Analysis (ssGSEA) algorithm was used. This algorithm, featured in Bindea et al.’s study, provided 29 gene sets that identify various TIC types (43). Concurrently, a comprehensive evaluation of the pan-cancer tumor microenvironment immune function was conducted, incorporating 29 features. These include angiogenesis, anti-tumor cytokines, co-stimulatory ligands and receptors, various trafficking characteristics (e.g., granulocyte, myeloid, effector cell), M1 characteristics, myeloid-derived immunosuppression, MHC class I and II molecules, tumor proliferation rate, and more. It also considers interactions within the microenvironment and involves cellular components like cancer-associated fibroblasts, B cells, effector cells, endothelial cells, macrophages, dendritic cells, NK cells, T cells, and regulatory T cells (44). Additionally, T cell subpopulations were identified using ImmuCellAI, providing a detailed view of their role within the tumor microenvironment (45). Kaplan-Meier (KM) curves were plotted to explore the correlations of immune cell infiltration level with overall survival (OS) and disease-free survival (DFS).

Due to the dual nature of immune function, which can both promote and inhibit tumor progression, we utilized enrichment scores from 29 pan-cancer tumor microenvironment immune functions. Hierarchical clustering was employed to categorize the samples into two groups: one characterized by high expression of pro-inflammatory markers and the other by high expression of inhibitory markers. Within these groups, samples with enrichment scores exceeding the group mean were classified as immune-activated in the pro-inflammatory group, and conversely, those below the mean in the inhibitory group were considered immune-suppressed.

We conducted differential gene expression analysis on TCGA UCEC RNAseq data between the activated and suppressed groups (relative to tumor-adjacent tissue) using the R limma package. The selection criteria for differentially expressed genes (DEGs) were set as an absolute log fold change (logFC) greater than 2 and an adjusted P-value of less than 0.05.

2.3 Weighted gene coexpression network analysis

Weighted Gene Coexpression Network Analysis (WGCNA) is a systematic biological method used to explore patterns of genetic association across various samples, aiming to identify clusters of genes with high co-variation. Initially, we performed a scale-free topology analysis to determine an appropriate soft thresholding power for network construction. Using this power, we constructed an adjacency network matrix among genes.

Subsequently, we calculated the topological overlap matrix and its corresponding dissimilarity from the adjacency matrix. The dissimilarity-based clustering of the topology matrix led to the generation of different gene modules through adaptive branch pruning of hierarchical clustering dendrograms, maintaining a specified minimum number of modules.

Finally, we assessed the correlations between these gene modules and clinical traits using Pearson correlation analysis. We focused on and selected the gene module that exhibited the strongest correlation with CD8+ T cell presence, potentially linking it to key immunological behaviors in the study context.

2.4 Establishment of risk model for the prognosis of EC

We utilized the clinical data of patients with endometrial cancer from The Cancer Genome Atlas (TCGA) database. Univariate Cox regression analysis was applied, and resampling was conducted 1000 times within the dataset to identify genes associated with prognosis. Subsequently, Least Absolute Shrinkage and Selection Operator (LASSO) Cox analysis and multivariate Cox regression analysis were further carried out. The regularization parameter was determined through the cross-validation method. After 10 rounds of cross-validation, a curve was plotted between the logarithm of the regularization parameter λ and the partial likelihood deviation. The λ value corresponding to the minimum partial likelihood deviation was determined as the optimal regularization parameter, and the corresponding regression coefficients were extracted. Genes corresponding to non-zero regression coefficients were obtained, and these genes were used to construct a multivariate Cox proportional hazards model. The stepwise regression method was employed to optimize the model, with variables selected bidirectionally. These computations were implemented using the R package glmnet.

The formula for calculating the risk score is as follows:

Risk Score=coef1×exp1+coef2×exp2+…+coefn×expn

where “coef” represents the coefficient, “coefn” is the coefficient related to the survival of the gene, and “expn” denotes the expression level of the gene. Patients were categorized into low-risk and high-risk groups based on the median value of the risk score among endometrial cancer patients. Kaplan-Meier curve analysis, time-dependent receiver operating characteristic (ROC) curves, and the area under the curve (AUC) were used to evaluate the sensitivity and specificity of the model for survival prediction (46).

2.5 Predictive nomogram establision

We incorporated clinical characteristics of endometrial cancer patients, including age, stage, and grade, into the risk model for both univariate and multivariate Cox regression analysis to determine if the risk model is an independent variable.

Additionally, we utilized clinical characteristics and the risk model to construct a nomogram. The Hosmer-Lemeshow test was used to generate a curve supporting the consistency between predicted results and actual outcomes, demonstrating the predictive ability of the nomogram (47).

2.6 Functional analyses based on the risk model

The DEGs were first identified between 2 groups with different scores. GSEA, a method designed to assess the concerted behavior of functionally related genes forming a set, was carried out on the DEGs to reveal underlying pathways involved in tumor biology behavior using the R “ClusterProfiler” package.

2.7 Response to immunotherapy and chemotherapy in clusters

We utilized the maftools software package to process Mutation Annotation Format (MAF) data of endometrial cancer patients, as mutation burden and microsatellite instability are important indicators related to tumor immunity.

Furthermore, we obtained the Immune Phenotype Score (IPS) of endometrial cancer patient tumor samples from The Cancer Immunome Atlas (TCIA) (https://tcia.at/home) to infer patient response to immunotherapy by comparing these indicators between two subgroups (48, 49). We obtained scores related to immune dysfunction and exhaustion from the Cancer Immunome Dysfunction and Exclusion (TIDE) tool (http://tide.dfci.harvard.edu) (50). Using the R package Easier, we obtained scores predicting patient response to immunotherapy and compared them between two subgroups, along with scores from a series of previous studies (51). We downloaded datasets from the GEO database containing immunotherapy efficacy and survival data, computed model scores within this dataset, and compared the scores with patient response and survival. Finally, we obtained the activity intensity of different steps in immune activation in TCGA samples from the Tumor Immune Profiling (TIP) website, estimating the relationship between scores and different immune activation steps (52).

2.8 Anticancer drug sensitivity analysis

The Genomics of Drug Sensitivity in Cancer (GDSC) database (https://www.cancerrxgene.org/) and The Cancer Therapeutics Response Portal(CTRP) (http://portals.broadinstitute.org/ctrp/) was accessed, and using these data to calculated the predicted half maximal inhibitory concentration (IC50) values of commonly used anticancer drugs. The IC50 was compared between the two groups (53).

2.9 PARP inhibitor response prediction

According to the literature, genetic and epigenetic alterations of ARID1A, ATM, ATR, BAP1, BARD1, BLM, BRIP1, CHEK1/2, FANCA/B/C/D2/E/F/G, MRE11A, NBN, PALB2, RAD50, RAD51, RAD51B, and WRN play a crucial role in the Homologous Recombination (HR) mechanism. Clinical trials have suggested that PARP inhibitors are more active in BRCA-mutated patients, followed by HR-deficient (HRD) and HR-proficient (HRP) subgroups. We then evaluated the mutation frequencies of HRD-related genes between different groups.

2.10 Single-cell analysis

The raw single-cell expression matrix underwent quality control filtering by removing cells with mitochondrial gene percentage > 20% and total expressed gene counts < 300, UMI < 500, and UMI > 50,000. Batch effects between samples were eliminated using the R package harmony. Dimensionality reduction was performed using the “RunUMAP” function in the R package Seurat (version 4.1.0) with parameters reduction = “harmony” and dims = 1:50. For the remaining cells, global scaling normalization method “LogNormalize” was applied, which normalizes feature expression for each cell based on total expression, multiplying by a scale factor (default is 10,000), and log-transforming the result. Principal component analysis was conducted on the top 2000 variable genes. Shared nearest neighbor modularity optimization clustering algorithm was utilized with a resolution of 1.2, using the “FindClusters” function to identify cell clusters. Non-linear dimensionality reduction using Uniform Manifold Approximation and Projection (UMAP) was employed for visualization and filtering of clustered single-cell RNA sequencing data. Differential expression genes were identified using the FindAllMarkers function in Seurat, and Wilcoxon algorithm was applied to compare differential genes between cell clusters. Marker genes were selected based on adjusted p-value (Wilcoxon rank-sum test) < 0.05 and |logFC| > 0.25 (54, 55).

Gene set enrichment analysis: The clusterProfiler R package was used to perform gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis on the differentially expressed genes between cell clusters (56).

We utilized SCENIC to infer the gene regulatory network (GRN) of T cells. Using the human hg38 reference genome, we searched for DNA motifs within a range of 10kb upstream and downstream of the transcription start sites. This step aimed to define the scope of the GRN establishment. These algorithms predict regulatory relationships between genes based on gene motifs and gene expression data. The resulting data were standardized (z-score) across cell clusters and overlaid onto a diffusion map to reveal the GRN structure in different cell types and clusters. The proportion of significantly upregulated genes (or regulons) from a given set of genes was evaluated using AUC scores (57).

CellChat utilizes network analysis and pattern recognition methods to predict cell signaling transmission. In order to identify potential interacting cell types, the CellChat method was employed. The database was set to “ChatDB.human”, and default parameters were used (FDR-adjusted p-value < 0.05). The “computeCommunProb” function was executed to infer the probability and strength of cell-cell communication. The “mergeCellChat” function was used to combine CellChat objects for each stage (58).

Initially, a Monocle object was created using the original UMI count gene-cell matrix via the newCellDataSet function. The lowerDetectionLimit was set to 0.1, while the expressionFamily was configured to negative binomial size. Subsequently, dimensionality reduction was carried out using the reduceDimension function, employing the DDRTree method with max_components set to 2. Following dimensionality reduction, differential gene testing was conducted to identify genes exhibiting significant expression differences either temporally or across different cell states. Additionally, cytotrace was utilized to assess the differentiation hierarchy among various subtypes of CD8+T cells (59).

2.11 Cell culture

EC cell lines (HEC‐1A and Ishikawa) were obtained from the China Center for Type Culture Collection (CCTCC). HEC‐1A and Ishikawa cells were cultured with Dulbecco’s modified Eagle’s medium (DMEM; Gibco) supplemented with 10% (v/v) fetal bovine serum (FBS; Gibco), and 1% penicillin-streptomycin (MRC, Jintan). Cells were cultured in a thermostatic incubator at 37°C and 5% CO2. Cells were harvested for further experiments when they reached 70%–80% confluence.

Lipofectamine2000 (Invitrogen) was used to transfect the Plasmids into cells. Lentivirus containing small hairpin RNA (shRNA) targeting human ASB2 and control shRNA lentivirus were purchased from Shanghai Genechem Co.Ltd. Cells were cultured in 6-well cell culture plates (2 × 10⁵ cells per well).Plasmids were transfected into Ishikawa while shRNA were transfected into HEC‐1A according to the manufacturer’s protocol. The effect of transfection was assessed by qPCR analysis and Western blotting.

2.12 Cell viability assays

HEC‐1A and Ishikawa were seeded into 96-well microplates at a density of 2000 cells per well. Additionally, 200 μL of PBS was used around the edges of the 96-well plate to prevent edge effects. After culturing the cells for 24, 48, and 72 hours, the CCK8 reagent (Vazyme) was diluted in serum-free medium at a ratio of 1:10. Care was taken to protect it from light. The mixture was then incubated in a CO2 incubator for 2 hours, after which the absorbance at 450 nm was measured using an enzyme marker.

2.13 Wound healing assay

Cells were seeded into a 6-well plate at a density of 1×10^5 cells per well and cultured to 90% confluence for the scratch assay. A scratch was made in the middle of each well using a 200 μl pipette tip guided by a ruler. Subsequently, each well was washed with 1 mL PBS to remove detached cells and debris, and the initial scratch width was photographed and recorded under a fluorescence inverted microscope. Following this, cells were cultured in serum-free medium, and images were captured with an Olympus microscope after 24, 48, and 72 hours of incubation. The scratch width was measured by drawing straight lines along the healing edge of the cells using PS software, and the scratch healing rate was calculated. The scratch healing rate is proportional to the cell migration ability.

2.14 Transwell assay

An 8µm pore size 24-well Transwell chamber (Corning) was used, coated with Matrigel matrix (Corning). 1×10^5 EC cells were seeded into the upper chamber in serum-free medium, while the lower chamber contained medium with 10% FBS as a chemoattractant. After 48 hours of incubation, cells in the lower chamber were fixed in 4% paraformaldehyde and stained with 0.1% crystal violet. Cells penetrating the lower chamber were randomly photographed and counted under a microscope.

2.15 RNA Isolation and qPCR

Total RNA was extracted using the RNAsimple Total RNA Kit (TIANGEN, Beijing, China) and reverse-transcribed into cDNA using the HiScript SuperMix (Vazyme). Subsequently, RT-qPCR was performed using the ChamQ Universal SYBR qPCR Master Mix (Vazyme) on a CFX96™ instrument (Bio-Rad Laboratories, Inc.). mRNA levels were calculated using the 2^−ΔCt method. The primer sequences used were as follows:

ASB2-F 5′-GCGACCGCTCAGAGTTACTG-3′

ASB2-R 5′-TCTCGCCTGTGATGACTCAG-3′

CCL5-F 5′-GAGTATTTCTACACCAGTGGCAAG-3′

CCL5-R 5′-TCCCGAACCCATTTCTTCTCT-3′

CD3G-F 5′-GGGATGTATCAGTGTAAAGG-3′

CD3G-R 5′-CAGCAATGAAGTAGACCC-3′

BATF-F 5′-AAAGCGAGCGACATGTCCCT-3′

BATF-R 5′-TTTTCTTTAAAGCATTTATT-3′

KIAA1755-F: 5′-ATGTCTCTCGCCGTCTCCAG-3′

KIAA1755-R: 5′-CGGATGCTGTTGCTATGGCC-3′

GAPDH-F: 5′-ACCCGCCCTATCTCAACTACC-3′

GAPDH -R: 5′-AGGACACCATAATGACAGCC-3′

2.16 Western blotting

Cells or tissues were lysed with RIPA lysis buffer (Servicebio) and mixed with 5× loading buffer, followed by boiling for approximately 10 minutes in a water bath at 100°C. Prepared protein samples were separated by 10% SDS-PAGE at constant voltage and transferred onto PVDF membranes. After blocking with 5% skim milk (Servicebio) for about 1 hour at room temperature, the membranes were incubated with primary antibodies overnight at 4°C. Subsequently, the membranes were incubated with horseradish peroxidase-conjugated secondary antibodies (Thermo Fisher Scientific, 1:10,000) for 1 hour at room temperature. After repeated washing with TBST, protein bands were detected using the ECL chemiluminescence ultra-sensitive chromogenic reagent (Vazyme). The primary antibodies used were those against ASB2 (PA5-29476, Invitrogen, diluted 1:1500) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH; 10494-1-AP, proteintech, diluted 1:20,000).

2.17 Statistical analysis

The Student’s t - test and F - test were used to determine the significant differences and variances between groups respectively. Pearson’s correlation analysis was used to conduct the correlation analysis. The comparison of multiple and paired non - parametric continuous variables was achieved by the Kruskal - Wallis test and the Wilcoxon test respectively. The Kaplan - Meier (KM) method was used to plot the survival curves. The bootstrap cross - validation was applied to evaluate the predictive power of various regression modeling strategies. A calibration curve was plotted in the bootstrap samples and then tested in the subjects not included in the bootstrap samples. A P - value less than 0.05 was considered statistically significant. All statistical analyses were performed using the statistical software R version 4.0.5.

3 Result

3.1 EC TME immune cells infiltration analysis

TME is pivotal in the onset and progression of cancer. We analyzed the characteristics of the tumor immune microenvironment in each sample of endometrial cancer using various methodologies.

Employing the CIBERSORT algorithm and feature genes of 29 types of cells, we estimated the distribution of immune cells in each sample. Kaplan-Meier survival analysis confirmed the anti-tumor effect of CD8+ T cell infiltration, indicating that patients with higher levels of CD8+ T cell infiltration had significantly longer OS and PFS (P < 0.05) (Figures 1A–D).

Figure 1

Figure 1. EC TME immune cells infiltration analysis. (A–D) Kaplan-Meier survival analysis of patient survival based on CD8+ T cell infiltration. A two-sided log-rank test is used to compare patient survival between the two groups. (A) The infiltration level of CD8+ T cells was calculated using the CIBERSORT algorithm, and the survival data were represented by OS. (B) The infiltration level of CD8+ T cells was calculated using the CIBERSORT algorithm, and the survival data were represented by PFS. (C) The infiltration level of CD8+ T cells was calculated using the ssGSEA algorithm, and the survival data were represented by OS. (D) The infiltration level of CD8+ T cells was calculated using the ImmuCellAI algorithm, and the survival data were represented by OS. (E) Density plot of CD8+ T cells between TCGA type. (F) Immune scores, stromal scores, and tumor purity between samples with high and low levels of CD8+ T cell infiltration. (G) A heatmap was used to display the anti - tumor and pro - tumor immune characteristics of the tumor microenvironment. The grouping was based on a comprehensive consideration of anti - tumor and pro - tumor factors.

Interestingly, analysis across multiple GEO datasets showed that CD8+ T cell infiltration did not significantly change in metastatic and recurrent patients. However, in the GSE120490 dataset, metastatic patients exhibited significantly higher levels of CD8+ T cells compared to non-metastatic patients (Supplementary Figure 1).

Regarding the molecular subtypes of endometrial cancer, the density plot of CD8+ T cells from TCGA data indicated significant infiltration only in the POLE mutation group, while the other three subtypes showed no significant differences in CD8+ T cell infiltration (Figure 1E).

Using ESTIMATE to assess stromal and immune scores, as well as calculate tumor purity for each sample, we observed significant differences in immune scores, stromal scores, and tumor purity between samples with high and low levels of CD8+ T cell infiltration (Figure 1F).

We evaluated the enrichment scores of 29 descriptors of the tumor microenvironment. Through heatmap analysis, a positive correlation was found between CD8+ T cell infiltration levels and other anti-tumor immune activities, such as antitumor cytokines, co-stimulatory ligands, and co-stimulatory receptors. In contrast, factors associated with tumor progression like angiogenesis, cancer-associated fibroblasts, and the matrix showed no correlation or a negative correlation with CD8+ T cell infiltration (Figure 1G).

3.2 Identification of CD8+ T cell activation-related gene

Using enrichment scores that described 29 tumor microenvironment features, subgroups at different hierarchical levels were delineated through hierarchical clustering. These subgroups were then analyzed against the average pathway activity within each to distinguish between an immune activation group and an immune suppression group. Differential expression analysis between these groups led to the identification of 1,268 DEGs listed in the Supplementary Material.

Subsequently, through the ImmuCellAI website, we determined the proportions of various T cell subtypes in endometrial cancer patients from the TCGA database. Leveraging these DEGs, key modules most correlated with CD8+ T cells in the TCGA cohort were identified. During the construction of the co-expression network, a soft thresholding power of 3 was selected, achieving a scale-free topology fitting index of 0.9. This process culminated in the formation of five modules through dynamic adaptive branch pruning, with the blue module being the most significantly correlated with CD8+ T cells (r=0.26, P<0.05) (Figure 2A).

Figure 2

Figure 2. Identification and establishment of CD8+ T Cell Activation Model. (A) WGCNA showed blue module being the most significantly correlated with CD8+ T cells. (B) Consensus matrix when k = 2. The grid where rows and columns intersect represents an independent sample. Consistency is represented by the color of the matrix, from dark blue representing 1 to white representing 0. (C, D) Kaplan-Meier survival analysis. (C) entire TCGA set. (D) GSE21882 set. (E) Time-related ROC analysis of TCGA. (F) Time-related ROC analysis of GSE21882.

3.3 NMF analysis of CD8+ T cell activation-related gene features

The gene features related to CD8+ T cell activation, identified earlier, were analyzed using Non-negative Matrix Factorization (NMF) clustering to explore their relationships with the clinical characteristics of endometrial cancer. At a clustering variable of (k = 2), the highest intra-group correlation was observed, suggesting that this represents the most stable classification. This finding was further validated by t-SNE analysis, which confirmed significant differences between the two groups.

Clinically, the differences between these two groups were profound, with patients in Type 1 exhibiting significantly lower overall survival and disease-free survival compared to those in Type 2. This indicates the existence of two distinct states of CD8+ T cell activation-related genes: one that promotes activation and is favorable for survival, and another that has the opposite effect (Figure 2B, Supplementary Figure 2).

3.4 Establishment and external validation of CD8+ T cell activation model

The LASSO Cox analysis is helpful for improving the prediction accuracy of statistical models while reducing the risk of overfitting. Therefore, we first applied univariate Cox proportional hazards regression to identify independent prognostic factors. Subsequently, a LASSO Cox analysis with ten-fold cross-validation was carried out to calculate the regression coefficients of the prognostic factors and further determine the genes to be included in the model. Finally, a multivariate Cox proportional hazards model was constructed using the included genes, and the model was optimized by the stepwise regression method. In addition to evaluating this model using a separate external dataset, we also improved the reliability of the prognostic model by repeated sampling within the group during the process of model construction. The predictive prognostic model was constructed, comprising CD8+ T cell activation-related genes—ASB2, BATF, CD3G, KIAA1755, and CCL5—using lasso regression analysis. The linear prediction model was established based on the weighted regression coefficients of these five prognostic-related genes:

Risk Score=(−1.5989×ASB2)+(−2.7822×BATF)+(−0.5837×CD3G)+(−1.2417×KIAA1755)+(0.9590×CCL5).

In this model, CCL5 was the only gene showing a significant positive correlation with the risk score, whereas ASB2, BATF, CD3G, and KIAA1755 exhibited significant negative correlations with the risk coefficient.

To confirm the stability and reliability of our model, we used the TCGA dataset as the internal training set and GSE21882 as the external validation set. Patients were divided into low-risk and high-risk groups based on the median value of the risk score. The predictive ability of the model was validated using Kaplan-Meier survival analysis across the EEC set, SEC set, and the entire TCGA dataset (Figures 2C, D, Supplementary Figure 2). Results showed significantly longer OS in the low-risk group compared to the high-risk group. Time-related ROC analysis confirmed the consistent prognostic accuracy of the risk score, demonstrating the model’s robust predictive performance (Figures 2E, F).

3.5 Validation of the CD8+ T cell activation model as an independent prognostic indicator and nomogram construction

In this study, 548 endometrial cancer patients with clinical indicators such as age, tumor staging, and grading were analyzed. We conducted univariate and multivariate Cox regression analyses to assess whether the CD8+ T cell activation model could serve as an independent prognostic factor. The hazard ratio (HR) was 1.289 with a 95% confidence interval (CI) of 1.181-1.407 in the univariate Cox regression analysis, and an HR of 1.217 with a 95% CI of 1.112-1.331 in the multivariate Cox regression analysis, both with p-values < 0.001. These results indicate that the CD8+ T cell activation model is independent of clinical features such as age, tumor staging, and grading, and qualifies as an independent prognostic indicator (Figures 3A, B). We also evaluated the independence of the risk model using GEO dataset analyses, which similarly demonstrated that the risk coefficient is independent of pathological staging and tumor grading (Supplementary Figure 3).

Figure 3

Figure 3. Validation of the CD8+ T cell activation model as an independent prognostic indicator and nomogram construction. (A) univariate Cox regression analyses of CD8+ T cell activation model. (B) multivariate Cox regression analyses of CD8+ T cell activation model. (C) A nomogram shows the score correspondence table. The predictive factors include age, disease stage, tumor grade, and risk score. (D) Calibration curves for 1-year,3-year and 5-year overall survival.

A nomogram was successfully constructed that incorporates risk scores and clinical features, including age, tumor grading, and staging (Figure 3C). Calibration curves for 3-year and 5-year OS showed high consistency between the predicted mortality rates and actual mortality rates, underscoring the nomogram’s excellent predictive ability (Figure 3D).

3.6 Multi-omics features of the CD8+ T cell activation model

Using the TCGA database, we visualized and analyzed somatic mutation data for high-risk and low-risk groups using the maftools package. The visualization displays the top 20 driver genes with the highest mutation frequencies in high and low subgroups. We identified five genes showing discrepancies between the two groups, with differences in mutation probabilities for the same genes, indicating significant divergence in tumor mechanisms between the subtypes. The three genes with the highest mutation frequencies in the high-risk (HR) group were PTEN (58%), TP53 (44%), and PIK3CA (47%). In the low-risk (LR) group, the most frequently mutated genes were PTEN (72%), ARID1A (52%), and PIK3CA (52%). This also explains the concentrated appearance of SEC in the HR group (Figures 4A, B).

Figure 4

Figure 4. Multi-omics features of the CD8+ T cell activation model. (A, B) Genomic changes associated with CD8+ T cell activation model in EC. (A) Waterfall Plot displays the top 20 driver genes with the highest mutation frequencies in high groups. (B) Waterfall Plot displays the top 20 driver genes with the highest mutation frequencies in low groups. (C) Heatmap shows the collinearity analysis of model gene mutations with PTEN, ARID1A, and PIK3CA. Green represents the co - occurrence of mutations, and dark yellow represents mutual exclusivity. (D) Bar chart shows the changes in gene copy numbers in the CD8+ T cell activation model. (E) Waterfall Plot displays the changes in mutations of genes from the CD8+ T cell activation model. (F) Kaplan-Meier analysis of patient survival based on protein prediction model, the levels of ERALPHA, PAX8, P16INK4A, and PR were obtained from the TCGA database. (G) Kaplan-Meier analysis of patient survival for combining protein model and the CD8+ T cell activation model. (H) Volcano plot showed differences in methylation between the high and low CD8+ T cell groups. (I) Volcano plot showed differences in methylation between the high and low-risk groups of the CD8+ T cell activation mode. *, P < 0.01, proving that the correlation between the two mutations is reliable.

Further analysis of the mutations in genes included in our model revealed that, except for KIAA1755, the mutation probabilities of other genes were relatively low, suggesting their differential expression might primarily occur at transcriptional and post-transcriptional stages. Co-linearity analysis confirmed positive correlations between mutations in model genes and mutations in PTEN, ARID1A, and PIK3CA, but these were not significantly associated with TP53 mutations. These findings indicate that model genes are relatively stable in endometrial cancers driven by TP53 mutations (Figure 4C).

Using the GISTIC 2.0 data from UCEC, we demonstrated the changes in gene mutations in the CD8+T cell activation model. KIAA1755 and ASB2 are the two most frequently mutated genes, but their mutation rates are only 8% and 4%, respectively. Similar to mutations, copy number increases or decreases are also rare, indicating that the chromosomes of these genes are relatively stable (Figures 4D, E).

ER, PR, PAX8, and P16 expression are commonly used in the pathological examination of endometrial cancer to guide treatment selection. We examined whether the combined use of the CD8+ T cell activation model with the expression of these proteins could better predict patient survival. Proteomics studies confirmed that all three proteins were associated with prognosis. Consequently, we employed LASSO Cox analysis to construct a protein prediction model using the TCGA database, combining it with the CD8+ T cell activation model. We found that survival outcomes for the high-risk group in both the protein model and the RNA high-risk score group were significantly worse than the other three groups, while the survival of the low-risk group in both models was the best, with no significant differences observed between the other two groups. This indicates that the combined use of both models can more effectively predict patient prognosis (Figures 4F, G).

Lastly, based on methylation data from TCGA, we analyzed the differences in methylated sites between the high and low CD8+ T cell groups, as well as between the high and low-risk groups of the CD8+ T cell activation model. The results showed that differences in methylation were more significant between the two groups of the CD8+ T cell activation model, indicating that the model can better reflect the heterogeneity among endometrial cancer patients and distinguish the origins of transcriptional pre-regulation that may influence differences in immune activity (Figures 4H, I).

3.7 Function enrichment analysis of the CD8+ T cell activation model

To further delineate the functional disparities between the two subgroups of endometrial cancer patients, we employed Gene Set Variation Analysis (GSVA) to identify genes that were upregulated and downregulated between the groups. This analysis was pivotal in uncovering the underlying biological processes differentiating the subgroups.

Following GSVA, we conducted immune pathway analysis to gain insights into the affected biochemical pathways. The results indicated that the model genes are strongly linked to several critical signaling pathways, including cell adhesion molecules, chemokine signaling pathways, B cell receptor signaling pathways, and T cell receptor signaling pathways (Figure 5A). These pathways are integral to the modulation of immune responses and cellular interactions within the tumor microenvironment, highlighting their potential role in tumor progression and patient prognosis.

Figure 5

Figure 5. Function enrichment analysis of the CD8+ T cell activation model. (A) GSVA of immune pathway analysis between HR and LR. (B) GSEA analysis between HR and LR.

Further analysis revealed distinct pathway enrichments between the two risk groups. Samples from the high-risk group showed significant enrichment in pathways involved in alpha-linolenic acid metabolism and the cell cycle. These pathways are often associated with rapid tumor growth and aggressive cancer behavior. In contrast, pathways related to spliceosomes and RNA degradation were more actively engaged in the low-risk group (Figure 5B).

3.8 Predicting and validating the efficacy of immunotherapy

Immune checkpoint blockade (ICB) therapy, a critical category of immunotherapy, operates by inhibiting signals that suppress T-cell activation, thereby facilitating tumor-reactive T-cells to effectively combat tumors (60).In-depth analyses were performed to explore differences in the immune microenvironment among subtypes of the CD8+ T-cell activation model. We found that both TMB and Microsatellite Instability (MSI) were elevated in the high-risk group compared to the low-risk group, suggesting a greater likelihood of response to immunotherapy in the high-risk group (Figures 6A, B).

Figure 6

Figure 6. Predicting and validating the efficacy of immunotherapy. (A) Box plot of TMB between high-risk group and low-risk group. (B) Bar plot of MSI between high-risk group and low-risk group. (C) Box plot of TIDE score between high-risk group and low-risk group. (D–F) Kaplan-Meier analysis of patient survival based on CD8+ T cell activation model from three patient cohorts undergoing immunotherapy. (D) GSE78200. (E) imv210. (F) PRJEB23709. (G) Heatmap shows the differences in TIP scores between the high and low groups. The TIP score represents a series of cell recruitment steps. (H) Heatmap displays the differences in the expression of immune checkpoints between the high and low groups. (I) Box plot illustrates the differences in the expression of MHC class I molecules between the high and low groups. ***, P < 0.001; ****, P < 0.0001, vs High risk group.

Using the TCIA, we calculated IPS to gauge the responsiveness of endometrial cancer patients to immunotherapy. Despite similar levels of PD-L1 and CTLA4 positivity, the low-risk group had higher IPS, indicating potential sensitivity to immune checkpoint therapy. And another scoring method TIDE also proves that the low-risk group has a higher probability of responding (Figure 6C, Supplementary Figure 4).

To further validate the utility of risk scoring in predicting immunotherapy outcomes, three patient cohorts undergoing immunotherapy were analyzed. Consistently, the high-risk group showed lower survival rates, while the low-risk group had higher probabilities of clinical remission and complete remission, underscoring a correlation between lower risk scores and favorable responses to immunotherapy (Figures 6D–F, Supplementary Figure 4).

In addition, we compared the immunological activities among different groups using the TIP score. The low-risk group showed increased activities related to the recruitment of NK cells and CD8+ T cells, as well as enhanced functions of the chemokine system, indicating a greater potential in mobilizing tumor-infiltrating immune cells. The evaluation of CD8+ T cell infiltration among different groups revealed a significant increase in CD8+ T cell infiltration in the low-risk group, strongly confirming the enhanced recruitment (Figure 6G, Supplementary Figure 4). However, further studies are needed to assess the possible differences in tissue-resident immune cells among different groups.

The expression analysis of immune checkpoints and major histocompatibility complex (MHC) class I molecules revealed significant differences: Although the high-risk group generally had higher expression levels of various immune checkpoints, the expression of MHC class I molecules was lower, suggesting a changed immune landscape. In contrast, the low-risk group had higher expression levels of identified immune checkpoint targets, such as PDCD1, (TIGIT), HAVCR2, ICOS, and CTLA4, implying that there might be both immune activation and immune exhaustion in this group (Figures 6H, I). Scoring the T cell cytotoxicity and IFN-γ in the high-risk and low-risk groups, the results showed that the scores in the low-risk group increased, supporting the enhanced anti-tumor effect and the occurrence of immune activation in the low-risk group (Supplementary Figure 5). By deconvoluting the TCGA data according to the annotated single-cell data of EC, it was found that the tissue-resident memory (TRM) cells in the low-risk group increased significantly (Supplementary Figure 5). Meanwhile, within the single-cell data, the TRM cells had the lowest risk scores. Considering that TRM cells mainly play an anti-tumor role in EC (61, 62), this may partly explain why the low-risk score group has a better prognosis.

3.9 Evaluation of the CD8+ T cell activation model for chemotherapy sensitivity and DNA repair pathway mutations

Chemotherapy remains a pivotal treatment for endometrial cancer. To ascertain the clinical utility of the CD8+ T cell activation model in guiding chemotherapy, we leveraged data from GDSC and CTRP. Using the Oncopredict algorithm, we calculated the half-maximal inhibitory concentration (IC50) of common chemotherapy drugs for each patient sample, which helped predict their sensitivity to these drugs.

The analyses revealed that patients in the low-risk group generally exhibited greater sensitivity to most common chemotherapy drugs, with the notable exception of docetaxel, compared to those in the high-risk group (Figures 7A–F).

Figure 7

Figure 7. Evaluation of the CD8+ T cell activation model for chemotherapy sensitivity and DNA repair pathway mutations. (A–F) IC50 of common chemotherapy drugs for each patient sample, using box plot compare the sensitivity between two groups. (A) Cisplatin. (B) Docetaxel. (C) Doxorubicin. (D) Gemcitabine. (E) Paclitaxel. (F) Rapamycin. (G) The bar plot demonstrates the KEGG pathway enrichment of HR and mismatch repair pathways between HR and LR. (H, I) Genomic changes associated with CD8+ T cell activation model in EC. (H) mutations in genes associated with HR repair in the high-risk group. (I) mutations in genes associated with HR repair in the low-risk group. *, P < 0.05; ****, P < 0.0001, vs High risk group.

Further supporting personalized treatment strategies, research indicates that tumors with BRCA1/2 mutations or those characterized by homologous recombination deficiency may benefit more from treatment with PARP inhibitors (63).

In our study, we extensively analyzed the activation and mutation status of the homologous recombination (HR) and mismatch repair (MMR) pathways, crucial for DNA repair mechanisms, across various risk groups in endometrial cancer. We found significant activation of both HR and MMR pathways in the high-risk group (Figure 7G). Additionally, we detected frequent mutations in genes linked to HR repair, such as ARAD1A, ATR, and ATM, in both risk groups. Notably, the gene FANCD2 showed a higher mutation frequency in the high-risk group, while BRIP1 was more commonly mutated in the low-risk group (Figures 7H, I). Despite these differences, the overall mutation spectra between the two groups remained quite similar. Interestingly, the low-risk group, characterized by a greater number of mutations and less active homologous recombination pathways, appears more likely to benefit from PARP inhibitors clinically. However, it’s important to consider that drugs targeting various pathways may still provide effective treatment options for both groups. The varied response to different therapeutic agents underscores the complexity of endometrial cancer and highlights the necessity of a personalized approach in the treatment regimen.

3.10 scRNA-seq analysis of endometrial cancer reveals CD8+ T cell subtypes and their functional states

In our study, we analyzed 10x scRNA-seq data from five endometrial cancer samples retrieved from the GSE173682 dataset. We initially focused on the top 2000 highly variable genes, which were visualized in the analysis. Dimensionality reduction through UMAP analysis enabled the identification of 31 distinct cell subtypes within these samples.

Further annotation and visual clustering of these dimensionality-reduced cell types using known cell subtype molecular markers allowed us to discern 8 major cell types, including myeloid cells, endothelial cells, smooth muscle cells, epithelial cells, B cells, MAST cells, stromal cells, as well as T and NK cells (Figures 8A, B).

Figure 8

Figure 8. scRNA-seq analysis of endometrial cancer reveals CD8+ T cell subtypes and their functional states. (A) Tsne plot of cells from GSE173682 including five primary EC cases, colored by eight major cell types. (B) Heatmap of the top marker genes in each major cell types. (C) Tsne plot of T cells from GSE173682 including five primary EC cases, colored by nine major cell types. (D) Cellchat shows the signal intensity of MHC-I related molecules between tumor cells and tissue-resident memory T cells (TRM). Tumor cells and TRM calculate the risk coefficients according to the risk model. (E) Scenic calculates the activity of transcription factors in different CD8+ T cell subsets. Different colors represent different CD8+ T cell subsets. (F) Monocle simulates the developmental trajectory of CD8+ T cells. Different colors represent different CD8+ T cell subsets. (G) Cytotrace calculates the cellular stemness of CD8+ T cells. Different colors represent different CD8+ T cell subsets.

For T lymphocytes, detailed clustering revealed subsets such as CD8+ and CD4+ T cells. We further characterized the states of these T cells:

- CD8+ T cells were subdivided into:

- CD8+TEM (T effector memory): Expressing GZMA, GZMB, GZMH, GZMK, and NKG7.

- CD8+TRM (T resident memory): Marked by ITGAE, CCL4, and XCL1.

- CD8+TCM (T central memory): Characterized by CD28 and IL7R.

- CD8+Tex (T exhausted): Identified by expression of CTLA4, PDCD1, and LAG3.

- MAIT (Mucosal-associated invariant T): Defined by ZBTB16 and KLRB1.

- CD4+ T cells were categorized into:

- CD4+TNaïve: Positive for TCF7, IL7R, and CCR7.

- CD4+TEM: Expressing CTLA4, PDCD1, and HAVCR2.

- CD4+Treg (T regulatory): Defined by FOXP3 and TNFRSF4 (Figure 9C).

Additionally, we inferred cell-cell communication networks, identifying that the MHC-I signaling pathway played a crucial role, particularly in relation to T cell interactions (Figure 8D). The presence of MHC-I is vital for antigen presentation to T lymphocytes, with HLA-B and HLA-C showing strong activity. Notably, the activity of these molecules was enhanced in T cells corresponding to lower CD8+ T cell activation model scores, suggesting a direct influence of these scores on the tumor-killing function of CD8+ T cells.

Transcription factor activity was analyzed using SCENIC in different CD8+ T cell subgroups:

TEM and Tex subgroups showed consistent activity in certain transcription factors, with REST, FOSB, RAD21, UBTF, and MXI1 activated in the TEM subgroup, and CEBPB, MAFF, CREM, IRF7, and KLF2 in the Tex subgroup. Heterogeneity within the TRM subgroup suggested diverse origins from different precursor cells post-activation. Interestingly, TCM and MAIT subgroups displayed similar activation patterns, hinting at a deeper relationship between these cells (Figure 8E).

Finally, pseudo-temporal trajectories calculated using Monocle2 demonstrated progression of CD8+ T cells from TEM to both TRM and Tex subtypes, indicating that effector cells do not solely progress towards exhaustion but maintain substantial anti-tumor effects, aligning with contemporary research insights. This comprehensive analysis underscores the complexity of immune responses within endometrial cancer and highlights the importance of detailed cellular profiling in understanding tumor-immune dynamics (64).

Using Cytotrace, we calculated the stemness across different cellular subtypes within the CD8+ T cell population in endometrial cancer. The T central memory (TCM) subtype demonstrated lower stemness compared to other groups, indicative of its limited proliferation capacity due to its terminally differentiated state. This aspect was further validated by Monocle2 trajectory analysis, which confirmed the progression of CD8+ T cells from T effector memory (TEM) to both T resident memory (TRM) and T exhausted (Tex) subtypes in early-stage endometrial cancer (Figures 8F, G).

In terms of gene expression: BATF, CD3G, KIAA1755, and CCL5 are predominantly expressed in activated T cell subtypes, suggesting their crucial roles in T cell activation and function.

ASB2, on the other hand, does not show significant differences in expression among different immune cells, yet interestingly, its expression appears elevated in tumor cells from patients with a higher proportion of activated T cells.

3.11 The expression of model genes in endometrial cancer cells and patient tissues

This study employed bioinformatics methods to analyze the expression of model genes in cancer (UCEC). The results indicate that apart from BATF, which is highly expressed in UCEC, the expression levels of ASB2 and KIAA1755 in tumors are relatively low, while the remaining genes did not show significant differences (Figure 9A). To validate these bioinformatics findings, we further examined the mRNA levels of these genes in endometrial cancer tissues and adjacent non-cancerous tissues using real-time quantitative PCR (RT-qPCR). The results were consistent with the bioinformatics analysis, confirming that the mRNA levels of BATF are significantly higher in the cancer tissues compared to the adjacent tissues, while the expressions of CCL5, ASB2, and KIAA1755 are lower, and CD3G did not show significant differences (Figure 9B). Additionally, we measured the expression of ASB2 in the EC cell lines Ishikawa and HEC-1A using RT-qPCR. The results demonstrated that ASB2 expression is significantly higher in the HEC-1A cell line compared to Ishikawa. Protein testing in cell lines also validated the RT-qPCR results, with higher ASB2 protein expression in the HEC-1A cell line compared to Ishikawa (Figure 9C).

Figure 9

Figure 9. The expression of model genes in endometrial cancer cells and patient tissues. (A) Expression differences of the model gene between tumor and normal tissues in the TCGA database (***, P < 0.001 vs. normal). (B) RT-qPCR validation of model gene expression differences between tumor tissues and adjacent normal tissues in patients(***, P < 0.001 vs. normal). (C) RT-qPCR and WB validation of ASB2 expression in endometrial cancer cell lines Ishikawa and HEC-1A (***, P < 0.001 vs. HEC-1A). (D) WB validation of ASB2 expression in tumor and adjacent normal tissues in patients (***, P < 0.001 vs. normal). NS: nonsense.

Furthermore, we studied the protein level expression of ASB2, finding that the protein content of ASB2 is lower in cancer tissues compared to adjacent non-cancerous tissues (Figure 9D).

3.12 Knocking down ASB2 promoted cancer cells proliferation, migration, and EMT

Based on the findings that ASB2 is underexpressed in EC tissues and cells, we hypothesized that the reduction of ASB2 might play a significant role in the progression of EC. Considering that ASB2 is expressed at the highest level in HEC-1A cells and at a lower level in Ishikawa cells, our study chose to knock down the ASB2 in HEC-1A cells and to overexpress it in Ishikawa cells.

Using liposome-based transfection and Lentivirus infection, we created HEC-1A cell lines with knocked-down ASB2 and Ishikawa cell lines with overexpressed ASB2. The efficacy of the knockdown and overexpression was validated through RT-qPCR and WB (Figure 10A).

Figure 10

Figure 10. Knocking down ASB2 promoted cancer cells proliferation, migration, and EMT. (A) WB validation of ASB2 knockdown in HEC-1A cell line and ASB2 overexpression in Ishikawa cell line (***P < 0.001 shRNA vs. control, ***P < 0.001 shRNA vs. vector, ***P < 0.001 ASB2 vs. control, ***P < 0.001 ASB2 vs. vector). (B) RT-qPCR validation of ASB2 knockdown in HEC-1A cell line and ASB2 overexpression in Ishikawa cell line (***P < 0.001 shRNA vs. control, ***P < 0.001 shRNA vs. vector, ***P < 0.001 ASB2 vs. control, ***P < 0.001 ASB2 vs. vector). (C) CCK8 assay shows that ASB2 knockdown in HEC-1A cell line enhances proliferation, while ASB2 overexpression in Ishikawa cell line reduces proliferation (***P < 0.001 HEC-ASB2 vs. HEC, ***P < 0.001 ISK-KO vs. ISK). (D) Wound healing assay shows that ASB2 knockdown in HEC-1A cell line enhances migration, while ASB2 overexpression in Ishikawa cell line reduces migration (Scale bar: 200 μm; ***P < 0.001 HEC-ASB2 vs. HEC, ***P < 0.001 ISK-KO vs. ISK). (E) Transwell assay shows that ASB2 knockdown in HEC-1A cell line enhances invasion, while ASB2 overexpression in Ishikawa cell line reduces invasion (Scale bar: 100 μm; ***P < 0.001 HEC-ASB2 vs. HEC, ***P < 0.001 ISK-KO vs. ISK).

A CCK-8 assay demonstrated that silencing ASB2 significantly reduced the proliferative ability of HEC-1A cells, while the overexpression construct showed the opposite effect, indicating that ASB2 expression can significantly inhibit cancer cell proliferation. We conducted scratch healing and Transwell invasion assays related to the cell migration phenotype. Results from the scratch healing assays showed that cells in the overexpression group had lower healing and migration capabilities compared to the control group, while the knockdown group significantly enhanced cell migration and invasion (Figures 10B–E).

4 Discussion

The tumor microenvironment is a complex system composed of cancer-associated fibroblasts, endothelial cells, various immune cells, extracellular matrix, and cytokines. It has a significant impact on tumor progression, leading to the emergence of drug resistance and metastasis. Immunotherapy targeting immune escape caused by the tumor microenvironment, such as immune checkpoint blockade therapy, is one of the new strategies for treating endometrial cancer. Numerous clinical trials are currently underway to investigate the benefits of various immunotherapies in endometrial cancer, aiming to address the challenges posed by the tumor microenvironment and enhance treatment outcomes (65). Traditional tumor staging does not reflect the changes in the tumor microenvironment and immunity that are crucial for assessing the efficacy of immunotherapy. The demand for predicting the sensitivity to immunotherapy in EC necessitates the exploration of updated molecular markers (66). As one of the primary factors contributing to anti-tumor activity within the tumor microenvironment, CD8+ T cells have come into focus. CD8+ T cells exert direct cytotoxic effects on cell surfaces through “ligand-receptor” interactions, and they participate in cellular immunity against tumor cells by producing cytokines, tumor necrosis factor, and interferon-alpha. However, continuous exposure of T cells to antigens and/or inflammatory signals within cancer can lead to T cell exhaustion, wherein T cells lose their ability to eliminate tumors (67, 68). The interaction between PD-1/PD-L1 can exert a potent inhibitory effect on the function of CD8+ T cells, including their cytotoxicity capabilities (69). In endometrial cancer, PD-1 expression is observed in 61% of cases among TILs, and PD-L1 expression is detected in 80% of cases. Additionally, 100% of metastatic endometrial cancer cases express PD1 (70). It appears that the phenomenon of T cell exhaustion due to the action of the PD-1/PD-L1 axis is quite common in patients with endometrial cancer. Additionally, according to previous research, there is a decrease in the proportion of cytotoxic CD8+ T cells in endometrial cancer compared to normal endometrial tissue (71). This suggests that, compared to a larger number of CD8+ T cells, maintaining an activated immune state of CD8+ T cells may play a more significant role in preventing tumor progression. Therefore, we performed ssGSEA scoring on 29 immune pathways in the tumor microenvironment based on the obtained responses (44). Subsequently, through unsupervised clustering, we divided them into two groups: strong and weak immune activity, and compared the differential genes between the two groups to identify the differential immune activation genes in EC. Next, we conducted WGCNA to identify key gene clusters closely associated with EC CD8+ T cell infiltration. Subsequently, using Lasso Cox regression, we selected 5 immune-related prognostic genes. These scores not only have good prognostic predictive value with AUC > 0.6, but also stratify into two subgroups: a high-risk group with poorer prognosis associated with hypertension and more pregnancies, and a low-risk group with better prognosis, younger age, and lower clinical stage and grade. Regarding molecular subtypes of endometrial cancer, patients in the low-risk group are mainly composed of CN-L and MSI types.

EC, being a hypermutated cancer, has a higher probability of MSI, which makes it highly promising for the application of ICIs (72, 73). We aim to validate the predictive system’s value in immunotherapy by comparing the scores with features strongly associated with ICI responsiveness, including TMB, PD-L1/PD-1 expression levels, and leukocyte infiltration (74). In endometrial cancer, the POLEmut and MMRd subtypes exhibit higher PD-L1 expression levels (75). This suggests that these subtypes may be particularly responsive to immunotherapy, further emphasizing the potential utility of immune checkpoint inhibitors in the treatment of endometrial cancer.

The elevated expression of PDCD1 in the high-risk group in our study further validates its predictive value. Furthermore, in endometrial cancer, other immune checkpoint molecules should also receive more attention. CTLA-4 blockade has been shown to improve prognosis in endometrial cancer patients (76). Additionally, molecules such as VISTA, TIM-3, and LAG3 are expressed in endometrial cancer and may serve as markers for T cell exhaustion (77–79). These molecules also exhibit differential expression, with higher expression observed in the low-risk group. As another predictive biomarker for immunotherapy, TMB is also a positive prognostic factor, negatively correlated with the score (80). Additionally, MSI has been utilized for immunotherapy prediction (81). Considering that MSI tumors often have higher levels of CD3+ and CD8+ T cells (82), the two groups based on differential activation of CD8+ T cells also exhibit differences in MSI status. Tumor neoantigens more accurately predict outcomes in melanoma patients treated with first-line anti-PD-1 or anti-CTLA-4 therapy compared to other biomarkers such as PD-L1 levels and mutational burden (50). We observed different mutational landscapes in the low-risk group, which may generate more tumor neoantigens compared to the high-risk group, thereby activating cytotoxic CD8+ T cells. MHC class I molecules presenting tumor antigens are necessary for the anti-tumor function of CD8+ T cells (83, 84). However, loss of MHC class I molecules can occur in up to 42% of endometrial cancer patients (85), potentially leading to a further decrease in the proportion of cytotoxic CD8+ T cells (86). Therefore, this study also explored the relationship between MHC class I molecules and predictive models and stratification. The results revealed that MHC class I molecules are less active in the high-risk group, potentially contributing to the decreased cytotoxicity of CD8+ T cells.

Based on existing research (74), an analysis of the composition of immune subtypes in the two types revealed that LR is primarily composed of wound healing and IFN-γ dominant subtypes, while HR includes wound healing, IFN-γ dominant, lymphocyte depleted, and inflammatory subtypes. Interestingly, within the LR immune response-active subtypes, particularly the inflammatory subtype, the majority of deceased endometrial cancer patients were concentrated. This contradicts the viewpoint in some literature that the inflammatory subtype achieving immune balance has a better prognosis. To further validate the potential of risk scoring in predicting the outcomes of immunotherapy, we selected three patient cohorts undergoing immunotherapy. The results across the three datasets were consistent, indicating that patients in the high-risk scoring group had lower survival rates and significant differences, while those in the low-risk group had higher probabilities of clinical remission and complete remission. This suggests that patients who respond to immunotherapy tend to have lower risk scores. This difference reflects the advantage of the model proposed in this study compared with the traditional pan-cancer model, as it can better reflect the anti-tumor ability of CD8 T cells in EC, rather than simply focusing on the situation of infiltration. Although macroscopically, the high-risk group has lower expression of immune checkpoint molecules, this is not reflected in the Tex cells. The high-risk group has a proportion of Tex cells similar to that of the low-risk group. In EC, lower infiltration of CD8+ T cells implies a higher possibility of non-response to immunotherapy (87).

This difference reflects the advantage of the model proposed in this study compared with the traditional pan-cancer model, as it can better reflect the anti-tumor ability of CD8 T cells in EC, rather than simply focusing on the situation of infiltration. Although macroscopically, the high-risk group has lower expression of immune checkpoint molecules, this is not reflected in the Tex cells. The high-risk group has a proportion of Tex cells similar to that of the low-risk group. In EC, lower infiltration of CD8+ T cells implies a higher possibility of non-response to immunotherapy. Although it is not possible to conclude that high-risk patients are less suitable for immunotherapy without using data from endometrial cancer patient cohorts, the differences in survival outcomes remind us that high-risk patients may have unique immune response mechanisms, and traditional immunotherapy might fail during disease progression.

ER (estrogen receptor) and PR (progesterone receptor) also play a role in the tumor immune microenvironment, which may be why combining proteomics and transcriptomics can improve prediction accuracy. Elevated levels of E2 (estradiol) in patients with endometrial cancer may directly impact CD8+ T cells and suppress their cytotoxic abilities (88). However, at the same time, the expression of ER and PR in epithelial cells brings these cells closer to CD8+ T cells, implying that they are more likely to be influenced by the anti-tumor effects of CD8+ T cells (89, 90). This suggests that common pathological markers may have broader utility in the future, but their regulatory role in the immune response of endometrial cancer remains to be further elucidated.

Chemotherapy-induced apoptosis may activate CD8+ T lymphocytes through various pathways, including depleting immune response inhibitory cells and inducing the emergence of tumor-specific CD8+ T lymphocytes (91–94). This has also been confirmed in gynecologic tumors, where CD8+ T cell infiltration remains stable even after receiving neoadjuvant platinum-based chemotherapy in ovarian cancer (95, 96). Although there is currently a lack of evidence specifically for EC, for those considered to have an immunogenic or “hot” phenotype of EC, there is a relatively greater likelihood of increased anti-tumor response following chemotherapy. Consequently, we also investigated the commonly used chemotherapy regimens in EC patients to discuss the possibility of combining chemotherapy with immunotherapy. Through the GDSC database, we validated the commonly used chemotherapy regimens in EC patients. The results revealed that low-risk group patients were more sensitive to paclitaxel, docetaxel, cisplatin, doxorubicin, gemcitabine, and rapamycin. This suggests a potential guiding role of this classification in the chemotherapy and immunotherapy of patients. Chemotherapeutic drugs used in endometrial cancer often achieve efficacy by increasing DNA damage in cancer cells. Comparisons between groups indicate that in high-risk patients, homologous recombination and mismatch repair processes are more active, suggesting that we can adopt corresponding strategies targeting this pathway.

Analysis of the dynamic immune microenvironment changes during the tumorigenic process of EEC showed tendencies of decreased proportion of cytotoxic and naive CD8+ lymphocyte population and increased proportion of CD4 Treg population, indicating immune escape during endometrial tumorigenesis (71). Our study observed that CD8+ T cells, which continue to exert their effects in early-stage endometrial cancer patients, still constitute the majority. Simply attributing this to a highly immunosuppressive microenvironment seems too hasty; further discussion is warranted on the transcriptional activation and evolutionary trends of CD8+ T cells. This study is the first to explore the prognostic significance and immune activation roles of BATF, CD3G, KIAA1755 and CCL5 in endometrial cancer. Although previous studies have found that the expression level of ASB2 in EC is lower than that in normal tissues, which implies that upregulating the expression of ASB2 is a potential therapeutic strategy for endometrial cancer (97). Through experiments, it discussed for the first time the expression and anti-cancer effects of ASB2 in endometrial cancer. Previous research has concentrated on the role of ASB2 in immune cells in forming and maintaining cell scaffolding and immature morphology but lacked studies on its potential role in tumors. As the expression of ASB2 significantly decreases during the progression from normal tissue to endometrial cancer, it is difficult to attribute the drastic decrease solely to the consumption of immune cells. The study preliminarily confirmed the anti-cancer effect of ASB2 in endometrial cancer cell lines through experiments and suggested its correlation with being a protective prognostic factor. Additionally, although some studies have indicated a negative correlation between the presence of CD8+ T cells and clinical pathological features such as histological grade, muscular layer infiltration, and lymph node metastasis (18, 22). However, in this study, the infiltration status of CD8+ T cells obtained through transcriptomic data using methods such as CIBERSORT showed no significant relationship with recurrence, metastasis, etc. On the other hand, ASB2, as identified in the corresponding transcriptomic data, could play a role in distinguishing and indicating the occurrence of recurrence and metastasis. This suggests that in transcriptomic data, ASB2 can replace complex methods, making transcriptomic data reflect potential prognosis of patients more simply and clearly. Multivariable Cox regression analysis demonstrated that this model is an independent risk factor for overall survival (OS), and ROC curves also showed that this model outperforms traditional methods such as clinical stage and grade in predicting risk. Additionally, the forest plots based on risk coefficients for predicting 3-year and 5-year OS closely matched the observed values, indicating the great potential of this new predictive system in guiding risk stratification and clinical decisions.

However, this study has some limitations. Firstly, the model was constructed using only TCGA UCEC dataset, validated using a single independent GEO dataset and a small-scale clinical sample, without validation from larger datasets. The stability of the predictive model and forest plots also requires confirmation through large-scale clinical trials. Moreover, there are similar issues in predicting immunotherapy sensitivity, as validation was only performed using data from other tumors, with limited reference value for endometrial cancer. Therefore, we plan to further collect clinical specimens and conduct more basic experiments to elucidate their specific roles, while simultaneously paying attention to the emergence of possible new external data for model validation.

5 Conclusion

In summary, our study provides potential predictive tools for endometrial cancer patients and contributes to further exploring the mechanisms of interaction between CD8+ T cell activation genes in endometrial cancer. Endometrial cancer can be divided into two distinct groups based on the activation status of CD8+ T cells, which have markedly different prognoses and tumor microenvironments. This suggests the potential of CD8+ T cell activation in personalized therapy. Thus, CD8+ T cell activation can serve as an important basis for prognosis prediction and treatment decisions in endometrial cancer.

Data availability statement

The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding author.

Author contributions

HG: Data curation, Investigation, Visualization, Writing – original draft. JX: Data curation, Supervision, Writing – original draft. QX: Data curation, Formal analysis, Writing – original draft. YL: Data curation, Investigation, Writing – original draft. WZ: Supervision, Writing – review & editing.

Funding

The author(s) declare that financial support was received for the research and/or publication of this article. This work was supported by the Foundation of Wuhan University (1606-413000153). The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, and in the decision to publish the results.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declare that no Generative AI was used in the creation of this manuscript.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fimmu.2025.1542669/full#supplementary-material

Glossary

UCEC: Uterine Corpus Endometrial Carcinoma

EC: Endometrial Cancer

EEC: Endometrioid Endometrial Cancer

USC: Uterine Serous Carcinoma

TCGA: The Cancer Genome Atlas

GEO: Gene Expression Omnibus

PRJEB: Project European Nucleotide Archive (ENA) Browser

GSEA: Gene Set Enrichment Analysis

IMvigor: (Refers to a clinical study, not an abbreviation)

RNA-seq: RNA Sequencing

scRNA-seq: Single-cell RNA Sequencing

NMF: Non-negative Matrix Factorization

t-SNE: t-distributed Stochastic Neighbor Embedding

UMAP: Uniform Manifold Approximation and Projection

WGCNA: Weighted Gene Co-expression Network Analysis

LASSO: Least Absolute Shrinkage and Selection Operator

ROC: Receiver Operating Characteristic

AUC: Area Under the Curve

HR: Hazard Ratio

CI: Confidence Interval

DEG: Differentially Expressed Gene

logFC: Log Fold Change

CIBERSORT: (A method, not an abbreviation)

ESTIMATE: Estimation of Stromal and Immune cells in Malignant Tumor tissues using Expression data

ssGSEA: Single-sample Gene Set Enrichment Analysis

KM: Kaplan-Meier

OS: Overall Survival

DFS: Disease-Free Survival

TME: Tumor Microenvironment

MHC: Major Histocompatibility Complex

NK: Natural Killer

PD-1/PD-L1: Programmed Death-1/Programmed Death-Ligand 1

MSI: Microsatellite Instability

TMB: Tumor Mutational Burden

TIDE: Tumor Immune Dysfunction and Exclusion

IC50: Half Maximal Inhibitory Concentration

GDSC: Genomics of Drug Sensitivity in Cancer

CTRP: Cancer Therapeutics Response Portal

PARP: Poly ADP Ribose Polymerase

HRD/HRP: Homologous Recombination Deficient/Proficient

UMI: Unique Molecular Identifier

TCIA: The Cancer Imaging Archive

IPS: Immune Phenotype Scores

TIP: Tumor Immune Phenotype

ICB: Immune Checkpoint Blockade

MAIT: Mucosal-Associated Invariant T cells

EMT: Epithelial-Mesenchymal Transition

RT-qPCR: Real-Time Quantitative Polymerase Chain Reaction

WB: Western Blot

CCK-8: Cell Counting Kit-8

ICIs: Immune Checkpoint Inhibitors

TILs: Tumor-Infiltrating Lymphocytes

CN-L: Copy Number Low

POLEmut: Polymerase Epsilon Mutant

MMRd: Mismatch Repair Deficient

CTLA-4: Cytotoxic T-Lymphocyte-Associated Protein 4

VISTA: V-domain Ig Suppressor of T cell Activation

TIM-3: T cell Immunoglobulin and Mucin-domain containing-3

LAG3: Lymphocyte Activation Gene-3

IFN-γ: Interferon Gamma

ER: Estrogen Receptor

PR: Progesterone Receptor

E2: Estradiol

References

1. Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA: Cancer J Clin. (2021) 71:209–49. doi: 10.3322/caac.21660

PubMed Abstract | Crossref Full Text | Google Scholar

2. Zheng R, Zhang S, Zeng H, Wang S, Sun K, Chen R, et al. Cancer incidence and mortality in China, 2016. J Natl Cancer Center. (2022) 2:1–9. doi: 10.1016/j.jncc.2022.02.002

PubMed Abstract | Crossref Full Text | Google Scholar

3. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2018. CA: Cancer J Clin. (2018) 68:7–30. doi: 10.3322/caac.21442

PubMed Abstract | Crossref Full Text | Google Scholar

4. Jemal A, Ward EM, Johnson CJ, Cronin KA, Ma J, Ryerson AB, et al. Annual report to the nation on the status of cancer, 1975–2014, featuring survival. JNCI: J Natl Cancer Institute. (2017) 109:djx030. doi: 10.1093/jnci/djx030

PubMed Abstract | Crossref Full Text | Google Scholar

5. Llobet D, Pallarés J, Yeramian A, Santacana M, Eritja N, Velasco A, et al. Molecular pathology of endometrial carcinoma: practical aspects from the diagnostic and therapeutic viewpoints. J Clin Pathol. (2009) 62:777–85. doi: 10.1136/jcp.2008.056101

PubMed Abstract | Crossref Full Text | Google Scholar

6. Matias-Guiu X, Prat J. Molecular pathology of endometrial carcinoma. Histopathology. (2013) 62:111–23. doi: 10.1111/his.12053

PubMed Abstract | Crossref Full Text | Google Scholar

7. Yeramian A, Moreno-Bueno G, Dolcet X, Catasus L, Abal M, Colas E, et al. Endometrial carcinoma: molecular alterations involved in tumor development and progression. Oncogene. (2013) 32:403–13. doi: 10.1038/onc.2012.76

PubMed Abstract | Crossref Full Text | Google Scholar

8. Huvila J, Pors J, Thompson EF, Gilks CB. Endometrial carcinoma: molecular subtypes, precursors and the role of pathology in early diagnosis. J Pathol. (2021) 253:355–65. doi: 10.1002/path.5608

PubMed Abstract | Crossref Full Text | Google Scholar

9. Gatius S, Cuevas D, Fernández C, Roman-Canal B, Adamoli V, Piulats JM, et al. Tumor heterogeneity in endometrial carcinoma: practical consequences. Pathobiology. (2018) 85:35–40. doi: 10.1159/000475529

PubMed Abstract | Crossref Full Text | Google Scholar

10. Arneth B. Tumor microenvironment. Medicina. (2019) 56:15. doi: 10.3390/medicina56010015

PubMed Abstract | Crossref Full Text | Google Scholar

11. Vanderstraeten A, Tuyaerts S, Amant F. The immune system in the normal endometrium and implications for endometrial cancer development. J Reprod Immunol. (2015) 109:7–16. doi: 10.1016/j.jri.2014.12.006

PubMed Abstract | Crossref Full Text | Google Scholar

12. Blaisdell A, Crequer A, Columbus D, Daikoku T, Mittal K, Dey SK, et al. Neutrophils oppose uterine epithelial carcinogenesis via debridement of hypoxic tumor cells. Cancer Cell. (2015) 28:785–99. doi: 10.1016/j.ccell.2015.11.005

PubMed Abstract | Crossref Full Text | Google Scholar

13. Eidukaite A, Tamosiunas V. Endometrial and peritoneal macrophages: expression of activation and adhesion molecules. Am J Reprod Immunol. (2004) 52:113–7. doi: 10.1111/j.1600-0897.2004.00201.x

PubMed Abstract | Crossref Full Text | Google Scholar

14. Lin B, Du L, Li H, Zhu X, Cui L, Li X. Tumor-infiltrating lymphocytes: Warriors fight against tumors powerfully. Biomedicine Pharmacotherapy. (2020) 132:110873. doi: 10.1016/j.biopha.2020.110873

PubMed Abstract | Crossref Full Text | Google Scholar

15. Halle S, Halle O, Förster R. Mechanisms and dynamics of T cell-mediated cytotoxicity. vivo. Trends Immunol. (2017) 38:432–43. doi: 10.1016/j.it.2017.04.002

PubMed Abstract | Crossref Full Text | Google Scholar

16. Ren X, Zhang L, Zhang Y, Li Z, Siemers N, Zhang Z. Insights gained from single-cell analysis of immune cells in the tumor microenvironment. Annu Rev Immunol. (2021) 39:583–609. doi: 10.1146/annurev-immunol-110519-071134

PubMed Abstract | Crossref Full Text | Google Scholar

17. Raskov H, Orhan A, Christensen JP, Gögenur I. Cytotoxic CD8+ T cells in cancer and cancer immunotherapy. Br J Cancer. (2021) 124:359–67. doi: 10.1038/s41416-020-01048-4

PubMed Abstract | Crossref Full Text | Google Scholar

18. Paul MS, Ohashi PS. The roles of CD8+ T cell subsets in antitumor immunity. Trends Cell Biol. (2020) 30:695–704. doi: 10.1016/j.tcb.2020.06.003

PubMed Abstract | Crossref Full Text | Google Scholar

19. Guo F, Dong Y, Tan Q, Kong J, Yu B. Tissue infiltrating immune cells as prognostic biomarkers in endometrial cancer: a meta-analysis. Dis Markers. (2020) 2020. doi: 10.1155/2020/1805764

PubMed Abstract | Crossref Full Text | Google Scholar

20. Suemori T, Susumu N, Iwata T, Banno K, Yamagami W, Hirasawa A, et al. Intratumoral CD8+ lymphocyte infiltration as a prognostic factor and its relationship with cyclooxygenase 2 expression and microsatellite instability in endometrial cancer. Int J Gynecologic Cancer. (2015) 25:1165–72. doi: 10.1097/IGC.0000000000000482

PubMed Abstract | Crossref Full Text | Google Scholar

21. Palomero J, Panisello C, Lozano-Rabella M, Tirtakasuma R, Díaz-Gómez J, Grases D, et al. Biomarkers of tumor-reactive CD4+ and CD8+ TILs associate with improved prognosis in endometrial cancer. J immunotherapy Cancer. (2022) 10:e005443. doi: 10.1136/jitc-2022-005443

PubMed Abstract | Crossref Full Text | Google Scholar

22. Jung IK, Kim SS, Suh DS, Kim KH, Lee CH, Yoon MS. Tumor-infiltration of T-lymphocytes is inversely correlated with clinicopathologic factors in endometrial adenocarcinoma. Obstetrics Gynecology Sci. (2014) 57:266. doi: 10.5468/ogs.2014.57.4.266

PubMed Abstract | Crossref Full Text | Google Scholar

23. Čermáková P, Melichar B, Tomšová M, Zoul Z, Kalabova H, Špaček J, et al. Prognostic significance of CD3+ tumor-infiltrating lymphocytes in patients with endometrial carcinoma. Anticancer Res. (2014) 34:5555–61.

PubMed Abstract | Google Scholar

24. Fusco MJ, West HJ, Walko CM. Tumor mutation burden and cancer treatment. JAMA Oncol. (2021) 7:316–6. doi: 10.1001/jamaoncol.2020.6371

PubMed Abstract | Crossref Full Text | Google Scholar

25. van der Woude H, Hally KE, Currie MJ, Gasser O, Henry CE. Importance of the endometrial immune environment in endometrial cancer and associated therapies. Front Oncol. (2022) 12:975201. doi: 10.3389/fonc.2022.975201

PubMed Abstract | Crossref Full Text | Google Scholar

26. Patel MV, Shen Z, Rodriguez-Garcia M. Endometrial cancer suppresses CD8+ T cell-mediated cytotoxicity in postmenopausal women. Front Immunol. (2021) 12:657326. doi: 10.3389/fimmu.2021.657326

PubMed Abstract | Crossref Full Text | Google Scholar

27. Sheikh MA, Althouse AD, Freese KE, Soisson S, Edwards RP, Welburn S, et al. USA endometrial cancer projections to 2030: should we be concerned? Future Oncol. (2014) 10:2561–8. doi: 10.2217/fon.14.192

PubMed Abstract | Crossref Full Text | Google Scholar

28. Scheper W, Kelderman S, Fanchi LF, Linnemann C, Bendle G, de Rooij MA, et al. Low and variable tumor reactivity of the intratumoral TCR repertoire in human cancers. Nat Med. (2019) 25:89–94. doi: 10.1038/s41591-018-0266-5

PubMed Abstract | Crossref Full Text | Google Scholar

29. Im SJ, Hashimoto M, Gerner MY, Lee J, Kissick HT, Burger MC, et al. Defining CD8+ T cells that provide the proliferative burst after PD-1 therapy. Nature. (2016) 537:417–21. doi: 10.1038/nature19330

PubMed Abstract | Crossref Full Text | Google Scholar

30. Finotello F, Trajanoski Z. Quantifying tumor-infiltrating immune cells from transcriptomics data. Cancer Immunology Immunotherapy. (2018) 67:1031–40. doi: 10.1007/s00262-018-2150-z

PubMed Abstract | Crossref Full Text | Google Scholar

31. Jackson HW, Fischer JR, Zanotelli VR, Ali HR, Mechera R, Soysal SD, et al. The single-cell pathology landscape of breast cancer. Nature. (2020) 578:615–20. doi: 10.1038/s41586-019-1876-x

PubMed Abstract | Crossref Full Text | Google Scholar

32. Xiao X, Guo Q, Cui C, Lin Y, Zhang L, Ding X, et al. Multiplexed imaging mass cytometry reveals distinct tumor-immune microenvironments linked to immunotherapy responses in melanoma. Commun Med. (2022) 2:131. doi: 10.1038/s43856-022-00197-2

PubMed Abstract | Crossref Full Text | Google Scholar

33. Launonen I-M, Lyytikäinen N, Casado J, Anttila EA, Szabó A, Haltia U-M, et al. Single-cell tumor-immune microenvironment of BRCA1/2 mutated high-grade serous ovarian cancer. Nat Commun. (2022) 13:835. doi: 10.1038/s41467-022-28389-3

PubMed Abstract | Crossref Full Text | Google Scholar

34. Li BL, Wan XP. Prognostic significance of immune landscape in tumour microenvironment of endometrial cancer. J Cell Mol Med. (2020) 24:7767–77. doi: 10.1111/jcmm.v24.14

Crossref Full Text | Google Scholar

35. Subramaniam KS, Tham ST, Mohamed Z, Woo YL, Mat Adenan NA, Chung I. Cancer-associated fibroblasts promote proliferation of endometrial cancer cells. PloS One. (2013) 8:e68923. doi: 10.1371/journal.pone.0068923

PubMed Abstract | Crossref Full Text | Google Scholar

36. Pauken KE, Wherry EJ. Overcoming T cell exhaustion in infection and cancer. Trends Immunol. (2015) 36:265–76. doi: 10.1016/j.it.2015.02.008

PubMed Abstract | Crossref Full Text | Google Scholar

37. Patel SP, Kurzrock R. PD-L1 expression as a predictive biomarker in cancer immunotherapy. Mol Cancer Ther. (2015) 14:847–56. doi: 10.1158/1535-7163.MCT-14-0983

PubMed Abstract | Crossref Full Text | Google Scholar

38. Hellmann MD, Nathanson T, Rizvi H, Creelan BC, Sanchez-Vega F, Ahuja A, et al. Genomic features of response to combination immunotherapy in patients with advanced non-small-cell lung cancer. Cancer Cell. (2018) 33:843–52. doi: 10.1016/j.ccell.2018.03.018

PubMed Abstract | Crossref Full Text | Google Scholar

39. Topalian SL, Taube JM, Anders RA, Pardoll DM. Mechanism-driven biomarkers to guide immune checkpoint blockade in cancer therapy. Nat Rev Cancer. (2016) 16:275–87. doi: 10.1038/nrc.2016.36

PubMed Abstract | Crossref Full Text | Google Scholar

40. Horeweg N, de Bruyn M, Nout RA, Stelloo E, Kedzierska K, León-Castillo A, et al. Prognostic integrated image-based immune and molecular profiling in early-stage endometrial cancer. Cancer Immunol Res. (2020) 8:1508–19. doi: 10.1158/2326-6066.CIR-20-0149

PubMed Abstract | Crossref Full Text | Google Scholar

41. McLane LM, Abdel-Hakeem MS, Wherry EJ. CD8 T cell exhaustion during chronic viral infection and cancer. Annu Rev Immunol. (2019) 37:457–95. doi: 10.1146/annurev-immunol-041015-055318

PubMed Abstract | Crossref Full Text | Google Scholar

42. Newman AM, Liu CL, Green MR, Gentles AJ, Feng W, Xu Y, et al. Robust enumeration of cell subsets from tissue expression profiles. Nat Methods. (2015) 12:453–7. doi: 10.1038/nmeth.3337

PubMed Abstract | Crossref Full Text | Google Scholar

43. Bindea G, Mlecnik B, Tosolini M, Kirilovsky A, Waldner M, Obenauf AC, et al. Spatiotemporal dynamics of intratumoral immune cells reveal the immune landscape in human cancer. Immunity. (2013) 39:782–95. doi: 10.1016/j.immuni.2013.10.003

PubMed Abstract | Crossref Full Text | Google Scholar

44. Bagaev A, Kotlov N, Nomie K, Svekolkin V, Gafurov A, Isaeva O, et al. Conserved pan-cancer microenvironment subtypes predict response to immunotherapy. Cancer Cell. (2021) 39:845–65. doi: 10.1016/j.ccell.2021.04.014

PubMed Abstract | Crossref Full Text | Google Scholar

45. Miao YR, Zhang Q, Lei Q, Luo M, Xie GY, Wang H, et al. ImmuCellAI: a unique method for comprehensive T-cell subsets abundance prediction and its application in cancer immunotherapy. Advanced Sci. (2020) 7:1902880. doi: 10.1002/advs.201902880

PubMed Abstract | Crossref Full Text | Google Scholar

46. Liu Z, Lu T, Li J, Wang L, Dang Q, Liu L, et al. Clinical significance and inflammatory landscape of anovel recurrence-associated immune signature in stage II/III colorectal cancer. Front Immunol. (2021) 12:702594. doi: 10.3389/fimmu.2021.702594

PubMed Abstract | Crossref Full Text | Google Scholar

47. Iasonos A, Schrag D, Raj GV, Panageas KS. How to build and interpret a nomogram for cancer prognosis. J Clin Oncol. (2008) 26:1364–70. doi: 10.1200/JCO.2007.12.9791

PubMed Abstract | Crossref Full Text | Google Scholar

48. Mayakonda A, Lin D-C, Assenov Y, Plass C, Koeffler HP. Maftools: efficient and comprehensive analysis of somatic variants in cancer. Genome Res. (2018) 28:1747–56. doi: 10.1101/gr.239244.118

PubMed Abstract | Crossref Full Text | Google Scholar

49. Charoentong P, Finotello F, Angelova M, Mayer C, Efremova M, Rieder D, et al. Pan-cancer immunogenomic analyses reveal genotype-immunophenotype relationships and predictors of response to checkpoint blockade. Cell Rep. (2017) 18:248–62. doi: 10.1016/j.celrep.2016.12.019

PubMed Abstract | Crossref Full Text | Google Scholar

50. Jiang P, Gu S, Pan D, Fu J, Sahu A, Hu X, et al. Signatures of T cell dysfunction and exclusion predict cancer immunotherapy response. Nat Med. (2018) 24:1550–8. doi: 10.1038/s41591-018-0136-1

PubMed Abstract | Crossref Full Text | Google Scholar

51. Lapuente-Santana Ó, van Genderen M, Hilbers PA, Finotello F, Eduati F. Interpretable systems biomarkers predict response to immune-checkpoint inhibitors. Patterns. (2021) 2. doi: 10.1101/2021.02.05.429977

Crossref Full Text | Google Scholar

52. Xu L, Deng C, Pang B, Zhang X, Liu W, Liao G, et al. TIP: a web server for resolving tumor immunophenotype profiling. Cancer Res. (2018) 78:6575–80. doi: 10.1158/0008-5472.CAN-18-0689

PubMed Abstract | Crossref Full Text | Google Scholar

53. Vanden Heuvel JP, Maddox E, Maalouf SW. Replication Study: Systematic identification of genomic markers of drug sensitivity in cancer cells. Elife. (2018) 7:e29747. doi: 10.7554/eLife.29747

PubMed Abstract | Crossref Full Text | Google Scholar

54. Hao Y, Hao S, Andersen-Nissen E, Mauck WM, Zheng S, Butler A, et al. Integrated analysis of multimodal single-cell data. Cell. (2021) 184:3573–87. doi: 10.1016/j.cell.2021.04.048

PubMed Abstract | Crossref Full Text | Google Scholar

55. Korsunsky I, Millard N, Fan J, Slowikowski K, Zhang F, Wei K, et al. Fast, sensitive and accurate integration of single-cell data with Harmony. Nat Methods. (2019) 16:1289–96. doi: 10.1038/s41592-019-0619-0

PubMed Abstract | Crossref Full Text | Google Scholar

56. Wu T, Hu E, Xu S, Chen M, Guo P, Dai Z, et al. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. Innovation. (2021) 2. doi: 10.1016/j.xinn.2021.100141

PubMed Abstract | Crossref Full Text | Google Scholar

57. Aibar S, González-Blas CB, Moerman T, Huynh-Thu VA, Imrichova H, Hulselmans G, et al. SCENIC: single-cell regulatory network inference and clustering. Nat Methods. (2017) 14:1083–6. doi: 10.1038/nmeth.4463

PubMed Abstract | Crossref Full Text | Google Scholar

58. Jin S, Guerrero-Juarez CF, Zhang L, Chang I, Ramos R, Kuan C-H, et al. Inference and analysis of cell-cell communication using CellChat. Nat Commun. (2021) 12:1088. doi: 10.1038/s41467-021-21246-9

PubMed Abstract | Crossref Full Text | Google Scholar

59. Qiu X, Mao Q, Tang Y, Wang L, Chawla R, Pliner HA, et al. Reversed graph embedding resolves complex single-cell trajectories. Nat Methods. (2017) 14:979–82. doi: 10.1038/nmeth.4402

PubMed Abstract | Crossref Full Text | Google Scholar

60. Wei SC, Duffy CR, Allison JP. Fundamental mechanisms of immune checkpoint blockade therapy. Cancer Discovery. (2018) 8:1069–86. doi: 10.1158/2159-8290.CD-18-0367

PubMed Abstract | Crossref Full Text | Google Scholar

61. Jiang F, Jiao Y, Yang K, Mao M, Yu M, Cao D, et al. Single-cell profiling of the immune atlas of tumor-infiltrating lymphocytes in endometrial carcinoma. Cancers. (2022) 14:4311. doi: 10.3390/cancers14174311

PubMed Abstract | Crossref Full Text | Google Scholar

62. Jiang F, Mao M, Jiang S, Jiao Y, Cao D, Xiang Y. PD-1 and TIGIT coexpressing CD8+ CD103+ tissue-resident memory cells in endometrial cancer as potential targets for immunotherapy. Int Immunopharmacol. (2024) 127:111381. doi: 10.1016/j.intimp.2023.111381

PubMed Abstract | Crossref Full Text | Google Scholar

63. Post CC, Westermann AM, Bosse T, Creutzberg CL, Kroep JR. PARP and PD-1/PD-L1 checkpoint inhibition in recurrent or metastatic endometrial cancer. Crit Rev oncology/hematology. (2020) 152:102973. doi: 10.1016/j.critrevonc.2020.102973

PubMed Abstract | Crossref Full Text | Google Scholar

64. Chu Y, Dai E, Li Y, Han G, Pei G, Ingram DR, et al. Pan-cancer T cell atlas links a cellular stress response state to immunotherapy resistance. Nat Med. (2023) 29:1550–62. doi: 10.1038/s41591-023-02371-y

PubMed Abstract | Crossref Full Text | Google Scholar

65. Polastro L, Closset C, Kerger J. Immunotherapy in gynecological cancers: where are we? Curr Opin Oncol. (2020) 32:459–70. doi: 10.1097/CCO.0000000000000661

PubMed Abstract | Crossref Full Text | Google Scholar

66. Wei L, Owen D, Rosen B, Guo X, Cuneo K, Lawrence TS, et al. A deep survival interpretable radiomics model of hepatocellular carcinoma patients. Physica Med. (2021) 82:295–305. doi: 10.1016/j.ejmp.2021.02.013

PubMed Abstract | Crossref Full Text | Google Scholar

67. Wherry EJ. T cell exhaustion. Nat Immunol. (2011) 12:492–9. doi: 10.1038/ni.2035

PubMed Abstract | Crossref Full Text | Google Scholar

68. Nguyen LT, Ohashi PS. Clinical blockade of PD1 and LAG3—potential mechanisms of action. Nat Rev Immunol. (2015) 15:45–56. doi: 10.1038/nri3790

PubMed Abstract | Crossref Full Text | Google Scholar

69. Pardoll DM. The blockade of immune checkpoints in cancer immunotherapy. Nat Rev Cancer. (2012) 12:252–64. doi: 10.1038/nrc3239

PubMed Abstract | Crossref Full Text | Google Scholar

70. Mo Z, Liu J, Zhang Q, Chen Z, Mei J, Liu L, et al. Expression of PD−1, PD−L1 and PD−L2 is associated with differentiation status and histological type of endometrial cancer. Oncol Lett. (2016) 12:944–50. doi: 10.3892/ol.2016.4744

PubMed Abstract | Crossref Full Text | Google Scholar

71. Ren X, Liang J, Zhang Y, Jiang N, Xu Y, Qiu M, et al. Single-cell transcriptomic analysis highlights origin and pathological process of human endometrioid endometrial carcinoma. Nat Commun. (2022) 13:6300. doi: 10.1038/s41467-022-33982-7

PubMed Abstract | Crossref Full Text | Google Scholar

72. Haraldsdottir S, Hampel H, Tomsic J, Frankel WL, Pearlman R, de la Chapelle A, et al. Colon and endometrial cancers with mismatch repair deficiency can arise from somatic, rather than germline, mutations. Gastroenterology. (2014) 147:1308–16. doi: 10.1053/j.gastro.2014.08.041

PubMed Abstract | Crossref Full Text | Google Scholar

73. Samstein RM, Lee C-H, Shoushtari AN, Hellmann MD, Shen R, Janjigian YY, et al. Tumor mutational load predicts survival after immunotherapy across multiple cancer types. Nat Genet. (2019) 51:202–6. doi: 10.1038/s41588-018-0312-8

PubMed Abstract | Crossref Full Text | Google Scholar

74. Thorsson V, Gibbs DL, Brown SD, Wolf D, Bortone DS, Yang T-HO, et al. The immune landscape of cancer. Immunity. (2018) 48:812–30. doi: 10.1016/j.immuni.2018.03.023

PubMed Abstract | Crossref Full Text | Google Scholar

75. Victoor J, Vander Borght S, Spans L, Lehnert S, Brems H, Laenen A, et al. Comprehensive immunomolecular profiling of endometrial carcinoma: a tertiary retrospective study. Gynecologic Oncol. (2021) 162:694–701. doi: 10.1016/j.ygyno.2021.06.030

PubMed Abstract | Crossref Full Text | Google Scholar

76. Oh MS, Chae YK. Deep and durable response with combination CTLA-4 and PD-1 blockade in mismatch repair (MMR)-proficient endometrial cancer. J Immunotherapy. (2019) 42:51–4. doi: 10.1097/CJI.0000000000000244

PubMed Abstract | Crossref Full Text | Google Scholar

77. Moore M, Ring KL, Mills AM. TIM-3 in endometrial carcinomas: an immunotherapeutic target expressed by mismatch repair-deficient and intact cancers. Modern Pathol. (2019) 32:1168–79. doi: 10.1038/s41379-019-0251-7

PubMed Abstract | Crossref Full Text | Google Scholar

78. Zhang Y, Yang R, Xu C, Zhang Y, Deng M, Wu D, et al. Analysis of the immune checkpoint lymphocyte activation gene-3 (LAG-3) in endometrial cancer: An emerging target for immunotherapy. Pathology-Research Pract. (2022) 236:153990. doi: 10.1016/j.prp.2022.153990

PubMed Abstract | Crossref Full Text | Google Scholar

79. Zong L, Mo S, Sun Z, Lu Z, Yu S, Chen J, et al. Analysis of the immune checkpoint V-domain Ig-containing suppressor of T-cell activation (VISTA) in endometrial cancer. Modern Pathol. (2022) 35:266–73. doi: 10.1038/s41379-021-00901-y

PubMed Abstract | Crossref Full Text | Google Scholar

80. Levine DA. Mike 1 CGARNGscBIGGGSBCKLESASCL, Heather WUiSLKCDDFRFLK-VJMMDOLMS, California UoS, 22 Laird Peter W. 22 Shen Hui 22 JHBSBBMSLPHTJTJVDBDJWDJ, 23 IfSBRSMSI: Integrated genomic characterization of endometrial carcinoma. Nature. (2013) 497:67–73. doi: 10.1038/nature12113

PubMed Abstract | Crossref Full Text | Google Scholar

81. Le DT, Durham JN, Smith KN, Wang H, Bartlett BR, Aulakh LK, et al. Mismatch repair deficiency predicts response of solid tumors to PD-1 blockade. Science. (2017) 357:409–13. doi: 10.1126/science.aan6733

PubMed Abstract | Crossref Full Text | Google Scholar

82. van Gool IC, Eggink FA, Freeman-Mills L, Stelloo E, Marchi E, de Bruyn M, et al. POLE proofreading mutations elicit an antitumor immune response in endometrial cancer. Clin Cancer Res. (2015) 21:3347–55. doi: 10.1158/1078-0432.CCR-15-0057

PubMed Abstract | Crossref Full Text | Google Scholar

83. Neefjes J, Jongsma ML, Paul P, Bakke O. Towards a systems understanding of MHC class I and MHC class II antigen presentation. Nat Rev Immunol. (2011) 11:823–36. doi: 10.1038/nri3084

PubMed Abstract | Crossref Full Text | Google Scholar

84. Mittica G, Genta S, Aglietta M, Valabrega G. Immune checkpoint inhibitors: a new opportunity in the treatment of ovarian cancer? Int J Mol Sci. (2016) 17:1169. doi: 10.3390/ijms17071169

PubMed Abstract | Crossref Full Text | Google Scholar

85. Friedman LA, Bullock TN, Sloan EA, Ring KL, Mills AM. MHC class I loss in endometrial carcinoma: a potential resistance mechanism to immune checkpoint inhibition. Modern Pathol. (2021) 34:627–36. doi: 10.1038/s41379-020-00682-w

PubMed Abstract | Crossref Full Text | Google Scholar

86. de Jong RA, Boerma A, Boezen HM, Mourits MJ, Hollema H, Nijman HW. Loss of HLA class I and mismatch repair protein expression in sporadic endometrioid endometrial carcinomas. Int J Cancer. (2012) 131:1828–36. doi: 10.1002/ijc.v131.8

PubMed Abstract | Crossref Full Text | Google Scholar

87. Chen J, Song Y, Huang J, Wan X, Li Y. Integrated single-cell RNA sequencing and spatial transcriptomics analysis reveals the tumour microenvironment in patients with endometrial cancer responding to anti-PD-1 treatment. Clin Trans Med. (2024) 14:e1668. doi: 10.1002/ctm2.v14.4

PubMed Abstract | Crossref Full Text | Google Scholar

88. Shen Z, Rodriguez-Garcia M, Patel M, Wira C. Direct and Indirect endocrine-mediated suppression of human endometrial CD8+ T cell cytotoxicity. Sci Rep. (2021) 11:1773. doi: 10.1038/s41598-021-81380-8

PubMed Abstract | Crossref Full Text | Google Scholar

89. Lien HE, Berg HF, Halle MK, Trovik J, Haldorsen IS, Akslen LA, et al. Single-cell profiling of low-stage endometrial cancers identifies low epithelial vimentin expression as a marker of recurrent disease. EBioMedicine. (2023) 92. doi: 10.1016/j.ebiom.2023.104595

PubMed Abstract | Crossref Full Text | Google Scholar

90. Farhood B, Najafi M, Mortezaee K. CD8+ cytotoxic T lymphocytes in cancer immunotherapy: A review. J Cell Physiol. (2019) 234:8509–21. doi: 10.1002/jcp.v234.6

Crossref Full Text | Google Scholar

91. Sakaguchi S, Mikami N, Wing JB, Tanaka A, Ichiyama K, Ohkura N. Regulatory T cells and human disease. Annu Rev Immunol. (2020) 38:541–66. doi: 10.1146/annurev-immunol-042718-041717

PubMed Abstract | Crossref Full Text | Google Scholar

92. Gabrilovich DI. Myeloid-derived suppressor cells. Cancer Immunol Res. (2017) 5:3–8. doi: 10.1158/2326-6066.CIR-16-0297

PubMed Abstract | Crossref Full Text | Google Scholar

93. Buhtoiarov IN, Sondel PM, Wigginton JM, Buhtoiarova TN, Yanke EM, Mahvi DA, et al. Anti-tumour synergy of cytotoxic chemotherapy and anti-CD40 plus CpG-ODN immunotherapy through repolarization of tumour-associated macrophages. Immunology. (2011) 132:226–39. doi: 10.1111/j.1365-2567.2010.03357.x

PubMed Abstract | Crossref Full Text | Google Scholar

94. Gattinoni L, Finkelstein SE, Klebanoff CA, Antony PA, Palmer DC, Spiess PJ, et al. Removal of homeostatic cytokine sinks by lymphodepletion enhances the efficacy of adoptively transferred tumor-specific CD8+ T cells. J Exp Med. (2005) 202:907–12. doi: 10.1084/jem.20050732

PubMed Abstract | Crossref Full Text | Google Scholar

95. Böhm S, Montfort A, Pearce OM, Topping J, Chakravarty P, Everitt GL, et al. Neoadjuvant chemotherapy modulates the immune microenvironment in metastases of tubo-ovarian high-grade serous carcinoma. Clin Cancer Res. (2016) 22:3025–36. doi: 10.1158/1078-0432.CCR-15-2657

PubMed Abstract | Crossref Full Text | Google Scholar

96. Lo CS, Sanii S, Kroeger DR, Milne K, Talhouk A, Chiu DS, et al. Neoadjuvant chemotherapy of ovarian cancer results in three patterns of tumor-infiltrating lymphocyte response with distinct implications for immunotherapy. Clin Cancer Res. (2017) 23:925–34. doi: 10.1158/1078-0432.CCR-16-1433

PubMed Abstract | Crossref Full Text | Google Scholar

97. Wang Z, Cheng S, Liu Y, Zhao R, Zhang J, Zhou X, et al. Gene signature and prognostic value of ubiquitination-related genes in endometrial cancer. World J Surg Oncol. (2023) 21:3. doi: 10.1186/s12957-022-02875-w

PubMed Abstract | Crossref Full Text | Google Scholar