Skip to main content
Download PDF

Prognostic and therapeutic potential of CXCR6 expression on CD8 + T cells in gastric cancer: a retrospective cohort study

BMC Gastroenterology volume 25, Article number: 139 (2025) Cite this article

Abstract

Background

Gastric cancer (GC) is a pressing global health concern, with prognosis intricately linked to the tumour stage and tumour microenvironment, especially, the presence of immune cells. Notably, CD8 + T cells play a pivotal role in the anti-tumour immune response, prompting investigations into their correlation with GC survival. This study aimed to investigate the intricate interplay between CD8 + T cells, particularly within the context of CXCR6, and survival outcomes in patients with GC.

Methods

Utilising datasets from The Cancer Genome Atlas, Gene Expression Omnibus, and Tumor Immune Dysfunction and Exclusion, the study employed xCell and Weighted Gene Co-expression Network Analysis to assess CD8 + T cell infiltration and identify key gene clusters. The prognostic significance of CXCR6 was evaluated via immunohistochemical staining of a GC tissue microarray.

Results

High CD8 + T cell infiltration correlated with improved survival in patients with GC. CXCR6 was identified as a prognostic gene and its expression was predominantly observed in CD8 + T cells. CXCR6 expression positively correlated with improved overall and disease-free survival. Furthermore, CXCR6 expression was associated with an immunoreactive microenvironment.

Conclusion

This study established that high CD8 + T-cell infiltration is related to CXCR6 expression, making it a key factor in predicting a favorable GC prognosis. The role of CXCR6 in shaping the tumour microenvironment and its potential utility in immunotherapy response prediction highlights its significance in GC management.

Peer Review reports

Background

Gastric cancer (GC) is the fifth most common cancer and the third leading cause of cancer-related deaths worldwide and poses considerable challenges [1, 2]. National cancer screening programs have substantially improved early diagnosis; however, the survival rate for patients with advanced-stage GC remains low at approximately 10% [2]. This is largely due to most patients being diagnosed at advanced stages of the disease. Treatment options, including surgery, can be restrictive and less effective at advanced stages of the disease, despite the evolution of surgical methods and chemotherapy regimens [1, 2]. These observations highlight the critical need for novel therapeutic approaches and a deeper understanding of the underlying mechanisms.

The recent progress in immunotherapy has revolutionised cancer treatment. Immune checkpoint blockade (ICB) therapies targeting molecules, such as PD-1 and CTLA-4, have gained attention [3]. These therapies aim to activate CD8 + T cells to enhance anti-tumour immune responses [4]. In Europe, the USA, and Asia, PD-(L)1 inhibitors like pembrolizumab, nivolumab, and sintilimab have been approved for treating advanced GC, both as first- and third-line treatments [5]. However, patient responses to these therapies are varied, with some showing remarkable improvements and others showing minimal or no benefits [6]. This variability underscores the need to better understand the immune landscape within the GC tumour microenvironment (TME) and factors influencing therapeutic outcomes.

Tumour-infiltrating lymphocytes (TILs), particularly CD8 + cytotoxic T lymphocytes, are pivotal for the success of immunotherapy [4]. These cells are the key components of the immune system that target cancer cells. Individual T cells can eliminate several target cells either sequentially or simultaneously [7]. Tumours responsive to ICB therapy typically exhibit increased CD8 + T-cell infiltration and rapid proliferation within the peripheral blood [8]. However, an overactivated CD8 + T-cell response can lead to tissue damage and autoimmunity. Thus, CD8 + T cells transiently express immune inhibitory receptors to maintain self-tolerance and control activation, which paradoxically contributes to tumour development through immunoediting and immunosurveillance [9, 10]. Low infiltration and exhaustion of CD8 + T cells are associated with poor response to immunotherapy [11]. Moreover, current standards such as PD-L1 expression and tumour mutational burden (TMB) have limitations in predicting immune status. Therefore, novel biomarkers are urgently required to predict therapeutic responses and provide insights into the density and functionality of CD8 + T cells in the TME.

The present study focused on the factors that activate the function of CD8 + T cells in GC. Specifically, we investigated CXCR6, a chemokine receptor expressed on CD8 + T cells known for its role in T cell homing, survival, and proliferation within the TME. The role of CXCR6 in GC, especially in relation to CD8 + T cell-mediated anti-tumour immunity and response to immunotherapy, has not been extensively explored. By examining the expression and functional implications of CXCR6 in GC, we aimed to highlight its potential as a prognostic marker and target for immunotherapeutic strategies.

Materials and methods

Data collection

We used a repository containing the RNA sequencing data and clinical details of 420 patients with GC from The Cancer Genome Atlas (TCGA) (available at http://cancergenome.nih.gov). An additional dataset (GSE265254, encompassing 257 subjects) was retrieved from the Gene Expression Omnibus (GEO) database (available at http://www.ncbi.nlm.nih.gov/geo). Furthermore, RNA expression data and clinical outcomes associated with anti-PD-1 treatment in the PRJEB25780 cohort were acquired from the Tumor Immune Dysfunction and Exclusion (TIDE) database (http://tide.dfci.harvard.edu/). These datasets are publicly accessible on their respective websites.

Evaluation of immune cell infiltration

The xCell algorithm can assess the infiltration levels of 64 immune cell types by analyzing a comprehensive set of 10,808 genes. This approach facilitates the identification of CD8 + T cells [12]. Through the implementation of the R package “maxstat” (which leverages maximum selection rank statistics alongside multiple p-value approximation methods; version 0.7–25), we established the optimal threshold for CD8 + T cell presence, modulating the sample sizes to ensure subgroups with less than 75% and more than 25% representation. Using these criteria, patient samples were stratified into high or low CD8 + T-cell infiltration cohorts, and these classifications were integrated with the survival analysis. The CIBERSORT algorithm, utilising support vector regression, was applied to quantitatively dissect the complexity of immune cell populations infiltrating the TME, delivering a precise deconvolution of the cellular compositions within the mix [13].

Identification of key genes related to CD8+ T cell infiltration

The infiltration level of CD8 + T cells, quantified using the xCell tool, facilitated the classification of patients with GC into groups with high and low expression, as determined by the 75th and 25th percentile values, respectively. The “limma” software package was then employed to identify differentially expressed genes (DEGs) between the low CD8 + T cell (CD8L) and high CD8 + T cell (CD8H) expression groups, with the significance criteria set at a false discovery rate of < 0.01 and a log fold change > 2 [14]. To identify critical gene clusters correlated with CD8 + T cell infiltration and strongly associated with GC, we used Weighted Gene Co-expression Network Analysis. Subsequently, univariate Cox regression analysis was conducted to evaluate the prognostic implications of these central hub genes, with a significance threshold of P < 0.05.

Analysis of single-cell RNA sequencing data for CXCR6 expression

To investigate the cellular sources of CXCR6 expression within the TME at single-cell resolution, we used the Tumor Immune Single-cell Hub (TISCH), a dedicated scRNA-sequencing database for TME studies. Specifically, we analysed two GC datasets, STAD_GSE134520 and STAD_GSE167297, available within TISCH, to elucidate the expression profile of CXCR6 across individual GC cells.

Immunotherapy response

The expression level of CXCR6 in mouse samples from in vivo ICB studies was detected using TISMO, a database for investigating and visualising gene expression, pathway enrichment, and immune cell infiltration levels in syngeneic mouse models across different ICB treatment and response groups in 23 cancer types [15]. RNA expression and clinical outcome data pertinent to anti-PD-1 treatment in the PRJEB25780 cohort were acquired from the TIDE database.

Functional enrichment analysis

To elucidate the potential biological mechanisms and signaling pathways involving DEGs, we conducted gene ontology enrichment analyses using the "ClusterProfiler" package in R [16]. We identified significantly enriched GO terms and signaling pathways, with P < 0.05 considered significant and a false discovery rate < 0.1. The findings were then visualised using the "ggplot2" package in R.

Validation of CXCR6 expression via immunohistochemical analysis

Protein expression levels of CXCR6 were verified by immunohistochemical analysis. This study was approved by the Institutional Review Board of the Dong-A University Hospital (approval no. DAUHIRB-22–007) and the requirement for informed consent was waived. Tissue samples were obtained from 107 patients diagnosed with stage II-III GC who underwent radical gastrectomy at the Dong-A University Hospital between 2015 and 2016, excluding patients who underwent neoadjuvant therapy. The clinical characteristics of this study are summarised in Supplementary Table 1. The most indicated regions were first identified on hematoxylin and eosin (H&E) stained slides, followed by extraction from the corresponding formalin-fixed, paraffin-embedded tissue blocks for assembly of tissue microarrays. Depending on the tumour size, one to three 3-mm-diameter cores were extracted from each GC specimen and organised into tissue microarray blocks using a trephine apparatus (Superbiochips Laboratories, Seoul, Korea).

Immunohistochemical staining of CXCR6 was performed using an automated BenchMark XT immunostainer (Ventana Medical Systems, Inc., Tucson, AZ, USA). This process included deparaffinisation, rehydration, antigen retrieval, and incubation with primary anti-CXCR6 rabbit monoclonal antibodies (NBP1-76,871; Novus, Beverly, CO, USA; 1:300 dilution). CXCR6 expression in immune cells was assessed by two pathologists (SHH and MKP) with no prior knowledge of clinicopathological results and classified as follows: low, any staining < 10%; high, strong staining ≥ 10%.

Statistical analysis

Statistical analyses were conducted using SPSS (version 21.0; IBM Corp., Armonk, NY, USA). Continuous variables were compared between the two groups using the t-test, whereas categorical variables were evaluated using either the chi-square test or Fisher's exact test, depending on the suitability of the data. Spearman's correlation coefficient was used to examine bivariate quantitative associations. Survival data were analysed using Kaplan–Meier curves paired with the log-rank test for significance. Hazard ratios (HRs) and 95% confidence intervals (CIs) were computed using a Cox proportional hazards model. A P-value of < 0.05 was considered significant, with the levels of significance denoted by *P < 0.05, P < 0.01, and P < 0.001.

Results

Impact of high CD8 + T cell infiltration on survival in patients with GC

Each GC sample in TCGA dataset was assessed for immune infiltration of CD8 + T cells using the xCell methodology. Subsequently, the GC samples were categorised into high (CD8H) and low (CD8L) CD8 + T-cell infiltration groups based on the optimal cutoff value for CD8 + T-cell infiltration, and overall survival (OS) in the two groups were compared. Based on the survival analysis (Fig. 1a), the CD8H group had a significantly better prognosis than the CD8L group (HR 0.61; 0.39–0.94, P = 0.02). Overall, the results of this study suggest that the immune infiltration of CD8 + T cells into the GC sample is crucial for predicting improved OS.

Fig. 1
figure 1

CD8 + T Cell-Related Gene Expression in gastric cancer (GC). a Kaplan–Meier plot showing overall survival based on CD8 + T cell infiltration in GC. b Differentially expressed genes between CD8H and CD8L groups from TCGA-STAD dataset. c Heatmap of top 30 CD8 + T cell-related genes in CD8H vs. CD8L groups. d, e Gene Ontology functional annotation: biological function (d) and molecular function (e)

Full size image

Genomic features, molecular functions, and mechanisms between the CD8H and CD8L groups

To assess the differences in molecular functions between the CD8H and CD8L groups, we analysed the DEGs. Transcriptional data analysis revealed 597 DEGs between the CD8H and CD8L groups, of which 495 were upregulated and 102 were downregulated (Fig. 1b). The heat map in Fig. 1b shows the top 30 DEGs in each category. We employed weighted gene co-expression network analysis (WGCNA) to identify gene co-expression patterns linked to GC prognosis and CD8 + T cell infiltration [17].

Three co-expression modules were identified (Supplementary Fig. 1), and correlation analysis between these modules and CD8 + T cell infiltration was conducted. The grey and yellow modules demonstrated strong correlations with CD8 + T-cell infiltration (r = -0.68, P < 0.001; r = 0.63, P < 0.001), warranting further investigation. Particularly, the grey module showed a high correlation with the CD8H group (r = 0.65, P < 0.001), and 42 hub genes were identified within this module. The full list of analyzed genes and their corresponding statistics is provided in Supplementary Table 2.

Gene Ontology enrichment analysis highlighted several key processes, including the immune system, adaptive immune response," and "T-cell activation" (Fig. 1). These processes are critical for the immune response against cancer. Pathway analysis revealed significant involvement in the "T cell signaling pathway" and "antigen receptor-mediated signaling pathway," suggesting a role of T cell mechanisms in targeting malignancies (Fig. 1d).

In terms of molecular function (Fig. 1e), the DEGs are enriched in "signaling receptor activity," "molecular transducer activity," and "signaling receptor binding." Notably, specific functionalities like "MHC protein binding" and "T cell receptor binding" highlight their involvement in immune recognition and antigen presentation. These findings suggest that these genes not only modulate membrane structures but also play crucial roles in cellular signal transduction and immune responses.

CXCR6 as a marker for CD8 + T cell in gastric cancer

Using univariate Cox regression analysis, we identified CXCR6 as the only chemokine receptor among the 42 hub genes that showed a statistically significant association with prognosis in gastric cancer (p < 0.05), making it a strong candidate for further investigation. (Supplementary Fig. 2). Kaplan–Meier analysis of TCGA cohorts revealed a positive association between CXCR6 expression and improved OS and disease-free survival (DFS) in patients (Fig. 2a, b). Although not statistically significant, the GSE 62254 cohort data suggested higher survival rates in patients with GC with high CXCR6 expression (CXCR6H) than in those with low expression (CXCR6L) (Fig. 2c, d).

Fig. 2
figure 2

Impact of CXCR6 Expression on GC Prognosis. a, b TCGA-STAD dataset analysis of overall survival (a) and disease-free survival (b) in high vs. low CXCR6 expression. c, d GSE62254 dataset analysis of overall survival (c) and disease-free survival (d) in high vs. low CXCR6 expression

Full size image

To explore the response of the TME to CXCR6 expression, we analysed single-cell RNA-seq data from two GC datasets (STAD GSE134520 and GSE167297) from the TISCH database. CXCR6 expression was predominantly observed in CD8 + T cells (Fig. 3a). T-distributed neighbor embedding visualisation of CXCR6 expression after clustering in the TISCH database showed distinct expression patterns in each cluster (Fig. 3b), reaffirming its prevalence in CD8 + T cells.

Fig. 3
figure 3

CXCR6 Expression and Immune Infiltration in GC. a Heatmap of CXCR6 expression using single-cell sequencing datasets (STAD_GSE134520 and STAD_GSE167297). b Single-cell cluster maps showing CXCR6 distribution in datasets. c Correlation of CXCR6 with CD8 + T cells. d Heatmap showing the relationship of CXCR6 with 23 tumour-infiltrating lymphocytes in GC

Full size image

Spearman correlation analysis of TCGA dataset confirmed the positive correlation of CXCR6 with CD8 + T cell infiltration and its strong association with Granzyme A (GZMA) and Perforin-1 (PRF1) expression (P < 0.001, r = 0.84, and r = 0.86, respectively), indicating CD8 + T cell cytotoxicity (Fig. 3c). Additionally, in the GSE62254 dataset, CXCR6 expression correlated positively with gamma delta (γδ) T cells and CD8 + T cell infiltration in GC among Asian patients but negatively with T regulatory cells (Fig. 3d).

Overall, these findings suggest that CXCR6 mRNA expression is linked to anti-tumour immune activation in GC, correlating with CD8 + T cell infiltration levels and cytotoxic function, indicating its potential as a regulatory gene in the immune microenvironment of GC.

CXCR6-related immune microenvironment signatures

GC is categorised into four subtypes: Epstein–Barr virus (EBV)-associated, microsatellite instability (MSI), genomically stable (GS), and chromosomally unstable (CIN). This classification aids in evaluating tumour immunogenicity, particularly MSI status and EBV infection. In this context, we observed higher CXCR6 expression in EBV-associated and MSI-H GC than in GS and CIN subtypes (Fig. 4a). This aligns with our finding of a significant positive correlation between CXCR6 expression and CD8 + T cell infiltration in GC. However, our analysis revealed no significant correlation between CXCR6 expression and TMB levels in GC (Fig. 4b).

Fig. 4
figure 4

CXCR6 and GC Microenvironment. a CXCR6 expression across GC molecular subtypes. b Tumour Mutation Burden in top and bottom quartiles of CXCR6 expression. c HLA Class I molecule expression pattern by CXCR6 groups. d Correlation of CXCR6 with mature dendritic cell markers. e Pathways significantly enriched in relation to CXCR6

Full size image

CD8 + T-cell surveillance requires antigenic epitope recognition in tumour cells via HLA class I molecules. Naïve CD8 + T cells become functional effector cells upon epitope recognition with costimulatory signals provided by professional antigen-presenting cells such as dendritic cells (DCs). The expression of HLA class I molecule (HLA-A, HLA-B, HLA-C, and B2M) was significantly higher in the CXCR6-high group than in the low group. In contrast, HLA-G showed higher expression in the CXCR6-high group but was relatively lower compared to the other HLA class I molecules (Fig. 4c).

Additionally, CXCR6 demonstrated a significant positive correlation with the expression levels of the co-stimulatory molecules CD28, CD80, and CD86, indicating that activated DCs prime antigen-specific CD8 + T cells (Fig. 4d). This suggests that higher expression of MHC class I molecules in cancer cells influences pro-tumoural immunity.

To further explore the TME, we conducted a differential pathway analysis using GSVA on 50 MSigDB hallmark gene sets, comparing the CXCR6 high and low groups in GC. Notably, the interferon-alpha and interferon-gamma pathways were upregulated (Fig. 4e). These findings imply that CXCR6 expression in GC is correlated with an immunoactive environment, potentially influencing tumour responsiveness to immune-mediated interventions.

Association of CXCR6 expression with response to immune checkpoint blockade therapies

Given that CXCR6 expression in bulk RNA-seq analysis reflects the contributions of infiltrating CD8 + T cells, we investigated whether CXCR6 expression could predict the response to ICB treatment. To test this, we analysed the correlation between the expression of immune checkpoints such as PD1, PD-L1, CTLA4, and CXCR6. The results showed that PD1, PD-L1, and CTLA4 were significantly and positively correlated with CXCR6 expression (Fig. 5a).

Fig. 5
figure 5

CXCR6 Expression and Immunotherapy Response in GC. a Correlation of immune checkpoint molecules with CXCR6 levels. b Correlation of CXCR6 with PD-L1 immunotherapy intensity and response in PRJEB25780 dataset. c CXCR6 expression and response to PD-1 and anti-CTLA-4 immunotherapy in TISMO datasets

Full size image

Furthermore, we used the TISMO website to evaluate gene and immune cell infiltration in the context of ICB treatment to confirm the hypothesis of an immunotherapeutic response. These results demonstrate how different ICB treatments stimulate CXCR6 expression in various models. For both anti-PD-1 (pembrolizumab) and anti-CTLA-4 (ipilimumab), CXCR6 expression levels in the YTN16 models were significantly upregulated in ICB responders but not in non-responders (Fig. 5b).

Finally, we evaluated the role of CXCR6 expression in GC of Asian patients receiving anti-PD1 therapy using RNA-seq data from the PRJEB25780 dataset. The response subtypes were divided into two groups: response (PR/CR) and non-response (SD/PD). The expression of CXCR6 was higher in the response group than in the non-response group. (Fig. 5c) Thus, GC cells with high CXCR6 expression may be more sensitive to immunotherapy compared to those with low CXCR6 expression.

Analysis of CXCL6 expression of immune cells in GC related to prognosis

Considering the significance of CXCR6 at the RNA level, we performed immunohistochemistry to determine its protein expression in tissues. CXCR6 expression in GC was predominantly observed in the cytoplasm of immune cells and to a lesser extent in the nuclei or membranes of tumour cells. The percentage of stained immune cells was assessed and the cells were divided into negative and positive groups. Representative staining results are shown in Fig. 6a–c. We recorded 42 (39.62%) positive and 65 (60.38%) negative cases. The positive expression of CXCR6 in immune cells was significantly correlated with a lower pT stage (P = 0.040) (Supplementary Table 3). In addition, we evaluated the clinical prognosis of patients with GC in relation to CXCR6 expression levels. Kaplan–Meier analysis revealed a significant difference in OS (P = 0.007) and DFS (P = 0.007) between the positive and negative CXCR6 groups (Fig. 6b, c). In the univariate and multivariate Cox analysis of DFS and OS, positive CXCR6 expression independently predicted favorable OS (HR, 0.33; 95% CI, 0.17 to 0.65; P = 0.001)) and good DFS (HR, 0.261; 95% CI, 0.083 to 0.825; P = 0.025) (Table 1).

Fig. 6
figure 6

CXCR6 in Immune Cells: TMA Cohort Analysis. a Immunostaining images of CXCR6 in GC, comparing negative and positive CXCR6 expression groups (× 400; Scale bar, 20 μm). b, c Kaplan–Meier analysis of overall survival (b) and disease-free survival (c) in CXCR6 positive vs. negative groups

Full size image
Table 1 Univariate and multivariate Cox regression analysis to predict progression in patients with GC
Full size table

Discussion

Identifying factors closely involved with anti-tumour immunity and therapeutic responses in patients with GC is crucial for understanding the mechanisms underlying current immunotherapies and for identifying novel strategies to optimise therapeutic outcomes by targeting appropriate cell subsets. In this study, we examined the expression and functional implications of CXCR6 in GC to assess its potential as a prognostic marker and a target for immunotherapeutic strategies. We found that a notable increase in CD8 + T cells correlated with improved OS in patients with GC. Moreover, high CXCR6 expression was robustly linked to cytolytic activity and the expansion and activation of T cell subsets integral to anti-tumour immune responses. Notably, our findings revealed a correlation between CXCR6 expression and the survival outcome of patients with GC, suggesting that CXCR6 may serve as an independent prognostic factor for GC.

Studies on TILs in various epithelial cancers have highlighted their significant prognostic value [18,19,20,21,22]. CD8 + cytotoxic T lymphocytes are often considered the most critical TIL subset, given their direct cancer cell-targeting ability [4, 23]. Elevated CD8 + T-cell densities correlate with favorable prognoses in several cancer types, including melanoma, colorectal cancer, hepatocellular cancer, triple-negative breast cancer, non-small cell lung cancer, and GC [18,19,20,21,22, 24]. However, this correlation does not apply to all types of cancer. Deng et al. observed that interstitial CD8 + T cells were correlated with poorer OS and disease-free survival in intrahepatic cholangiocarcinoma [25]. Zhang et al. found that in head and neck squamous cell carcinoma, the density of infiltrating CD8 + T cells in the TME serves as an independent prognostic factor, where higher densities are indicative of a better prognosis, whereas CD8 + T cell exhaustion in the same environment is associated with poorer outcomes [26]. Additionally, Kim et al. demonstrated that immune cell infiltrates differed from DCIS in terms of their evolution to invasive breast cancer, and high FOXP3 T-cell infiltrates and PD-L1 + immune cells were associated with recurrence [27]. Thus, evaluating both the functionality and density of tumour-reactive CD8 + T cells is essential to comprehensively understand immune responses in the TME. The present study showed that an increased count of CD8 + T cells was associated with high cytolytic activity and favorable clinical outcomes in patients with GC. CXCR6 expression was strongly correlated with the cytolytic activity of CD8 + T cell subsets. CXCR6 expression may be an independent prognostic factor in GC, as it is correlated with favorable clinical outcomes.

CXCR6 is selectively expressed on CD8 + and CD4 + T cells and natural killer cells [28, 29]. CXCR6 mediates CD8 + T-cell homing to perivascular niches within the tumour stroma, promotes interactions with CXCL16 + dendritic cells, and transduces IL-15, a cytokine critical for T-cell survival and proliferation in the TME. However, CXCR6 expression can activate oncogenic pathways associated with cancer progression and metastasis [30, 31]. The prognostic significance of CXCR6 in malignancies varies widely. High CXCR6 expression correlates with increased survival probability in melanoma and breast cancer [28]. In contrast, studies have shown that CXCR6 is associated with adverse outcomes in hepatocellular carcinoma and clear cell renal cell carcinoma [32,33,34]. Furthermore, CXCR6 accelerated tumour development by promoting epithelial-mesenchymal transition and regulating AKT-mediated MMP-2/-9 expression, and inhibition of CXCR6 signaling may suppress GC invasion [34, 35]. These findings do not directly contradict the results of our study because they evaluated GC cells rather than immune cells. To the best of our knowledge, the present study is the first to report predominant CXCR6 expression in CD8 + T cells in GC, which may correlate with effective cytotoxic T lymphocytes and predict a favorable prognosis. This is in line with the findings of Di Pilato et al., who demonstrated the importance of CXCR6 in the survival and expansion of T cells in the tumour environment to maximise their anti-tumour activity prior to their dysfunction [28] Similar to our results, Yu et al. observed that CXCR6 and CD8 + T cells showed the highest correlation and were significantly associated with favorable OS in patients with GC [36].

Treatment response is a key factor for improving cancer survival. Recently, pembrolizumab was approved by the Food and Drug Administration for use in combination with trastuzumab and first-line chemotherapy in patients with HER2-positive AGC based on interim results from KEYNOTE-811 [37]. Few ICB biomarkers are currently available for predicting response to ICB in GC. The first clinically validated and currently the most widely used biomarker in ICB therapy is immunohistochemistry for detecting PD-L1 expression in tumour cells or tumour-infiltrating immune cells [38]. However, patients with immunohistochemistry PD-L1 (-) may still benefit from ICB in some clinical trials because the absence of PD-L1 expression is considered a poor negative predictor of immunotherapy response [39]. TMB, a measure of total tumour mutations, is another biomarker [40]. TMB detection in clinical practice is challenging because it requires sophisticated technology, complex data, and bioinformatics expertise, and next-generation sequencing technology is expensive [41]. Therefore, the focus of this study was to identify reliable prognostic and predictive biomarkers that can help establish personalised treatment strategies for these patients. One feature that distinguishes cancer patients susceptible to immunotherapy from those who are not in the presence or absence of tumour-specific T cells [42, 43]. As previously noted, CXCR6 is expressed by a subset of activated T cells, γδ T cells, natural killer cells, and natural killer T cells [28, 29]. In tumours, the expression of CXCR6 is essential for CD8 + T cells to mount an appropriate anti-tumour immune response and respond to ICB treatment [44]. Recently, CXCR6-mediated tumour-specific T cells were shown to be able to migrate, adhere, and better recognise CXCL16-producing pancreatic cancer cells, thus improving the effectiveness of adoptive CAR T cell therapy [45]. In addition, we found that high CXCR6 expression was associated with an immunoreactive microenvironment, such as increased active or mature DC, and immunoreactive pathways, such as interferon-alpha and interferon-gamma.

Therefore, we hypothesised that CXCR6-based immunotherapy may be a promising approach for treating patients with GC. Our results indicate that the expression levels of CXCR6 are higher in immunotherapy responders than in non-responders, as demonstrated in the PRJEB25780 database and PD-1- and CTLA-4-treated murine GC models using the TISMO database. We found that CXCR6 + T-cell effector functions could be beneficial in patients with GC receiving immunotherapy.

Our study has limitations, including the lack of experimental validation of the mechanism by which CXCR6 enhances tumour immunity and the potential bias introduced from the limited sample size in the subset survival analyses.

Conclusion

This study is the first to demonstrate that increased expression of both CD8 + T cells and CXCR6 is associated with improved OS in patients with GC using combined analyses of multiple datasets. In addition, we suggest that a high level of CXCR6 expression characterises an immunologically active T cell-inflamed TME that may enhance the response to immunotherapy. Our findings may not only help predict response to immunotherapy but also provide insights into developing novel immunotherapeutic strategies for GC treatment.

Data availability

This study utilised a repository containing RNA sequencing data and clinical details of 420 patients with GC from The Cancer Genome Atlas (TCGA). An additional dataset (GSE265254, comprising 257 participants) was retrieved from the Gene Expression Omnibus (GEO) database. Furthermore, RNA expression data and clinical outcomes associated with anti-PD-1 treatment in the cohort PRJEB25780 were acquired from the Tumour Immune Dysfunction and Exclusion (TIDE) database.

Abbreviations

GC:

Gastric cancer

CB:

Immune checkpoint blockade

TME:

Tumour microenvironment

TILs:

Tumour-infiltrating lymphocytes

TMB:

Tumour mutational burden

TCGA:

The Cancer Genome Atlas

GEO:

Gene Expression Omnibus

TIDE:

Tumor Immune Dysfunction and Exclusion

DEGs:

Differentially expressed genes

TISCH:

Tumor Immune Single-cell Hub

TISMO:

Tumour Immune Syngeneic Mouse

DFS:

Disease-free survival

EBV:

Epstein–Barr virus

MSI:

Microsatellite instability

GS:

Genomically stable

CIN:

Chromosomally unstable

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(3):209–49.

  2. Park SH, Kang MJ, Yun EH, Jung KW. Epidemiology of Gastric Cancer in Korea: Trends in Incidence and Survival Based on Korea Central Cancer Registry Data (1999–2019). Journal of gastric cancer. 2022;22(3):160–8.

    Article  PubMed  PubMed Central  Google Scholar 

  3. Robert C. A decade of immune-checkpoint inhibitors in cancer therapy. Nat Commun. 2020;11(1):3801.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  4. Raskov H, Orhan A, Christensen JP, Gogenur I. Cytotoxic CD8(+) T cells in cancer and cancer immunotherapy. Br J Cancer. 2021;124(2):359–67.

    Article  PubMed  CAS  Google Scholar 

  5. Narita Y, Muro K. Updated Immunotherapy for Gastric Cancer. J Clin Med. 2023;12(7):2636.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  6. Kawazoe A, Shitara K, Boku N, Yoshikawa T, Terashima M. Current status of immunotherapy for advanced gastric cancer. Jpn J Clin Oncol. 2021;51(1):20–7.

    Article  PubMed  Google Scholar 

  7. Wiedemann A, Depoil D, Faroudi M, Valitutti S. Cytotoxic T lymphocytes kill multiple targets simultaneously via spatiotemporal uncoupling of lytic and stimulatory synapses. Proc Natl Acad Sci USA. 2006;103(29):10985–90.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  8. Holm JS, Funt SA, Borch A, Munk KK, Bjerregaard AM, Reading JL, et al. Neoantigen-specific CD8 T cell responses in the peripheral blood following PD-L1 blockade might predict therapy outcome in metastatic urothelial carcinoma. Nat Commun. 2022;13(1):1935.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  9. Dornmair K, Goebels N, Weltzien HU, Wekerle H, Hohlfeld R. T-cell-mediated autoimmunity: novel techniques to characterize autoreactive T-cell receptors. Am J Pathol. 2003;163(4):1215–26.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  10. Shevchenko I, Bazhin AV. Metabolic Checkpoints: Novel Avenues for Immunotherapy of Cancer. Front Immunol. 2018;9:1816.

    Article  PubMed  PubMed Central  Google Scholar 

  11. Mortezaee K, Majidpoor J. Mechanisms of CD8(+) T cell exclusion and dysfunction in cancer resistance to anti-PD-(L)1. Biomed Pharmacother. 2023;163:114824.

    Article  PubMed  Google Scholar 

  12. Aran D, Hu Z, Butte AJ. xCell: digitally portraying the tissue cellular heterogeneity landscape. Genome Biol. 2017;18(1):220.

    Article  PubMed  PubMed Central  Google Scholar 

  13. 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(5):453–7.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  14. Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015;43(7):e47.

    Article  PubMed  PubMed Central  Google Scholar 

  15. Zeng Z, Wong CJ, Yang L, Ouardaoui N, Li D, Zhang W, et al. TISMO: syngeneic mouse tumour database to model tumour immunity and immunotherapy response. Nucleic Acids Res. 2022;50(D1):D1391–7.

    Article  PubMed  CAS  Google Scholar 

  16. Wu T, Hu E, Xu S, Chen M, Guo P, Dai Z, et al. clusterProfiler 40: A universal enrichment tool for interpreting omics data. Innovation. 2021;2(3):100141.

    PubMed  PubMed Central  CAS  Google Scholar 

  17. Langfelder P, Horvath S. WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics. 2008;9:559.

    Article  PubMed  PubMed Central  Google Scholar 

  18. Gartrell RD, Marks DK, Hart TD, Li G, Davari DR, Wu A, et al. Quantitative Analysis of Immune Infiltrates in Primary Melanoma. Cancer Immunol Res. 2018;6(4):481–93.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  19. Galon J, Costes A, Sanchez-Cabo F, Kirilovsky A, Mlecnik B, Lagorce-Pages C, et al. Type, density, and location of immune cells within human colorectal tumours predict clinical outcome. Science. 2006;313(5795):1960–4.

    Article  PubMed  CAS  Google Scholar 

  20. Yao W, He JC, Yang Y, Wang JM, Qian YW, Yang T, et al. The Prognostic Value of Tumour-infiltrating Lymphocytes in Hepatocellular Carcinoma: a Systematic Review and Meta-analysis. Sci Rep. 2017;7(1):7525.

    Article  PubMed  PubMed Central  Google Scholar 

  21. Oshi M, Asaoka M, Tokumaru Y, Yan L, Matsuyama R, Ishikawa T, et al. CD8 T Cell Score as a Prognostic Biomarker for Triple Negative Breast Cancer. International journal of molecular sciences. 2020;21(18):6968.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  22. Uso M, Jantus-Lewintre E, Bremnes RM, Calabuig S, Blasco A, Pastor E, et al. Analysis of the immune microenvironment in resected non-small cell lung cancer: the prognostic value of different T lymphocyte markers. Oncotarget. 2016;7(33):52849–61.

    Article  PubMed  PubMed Central  Google Scholar 

  23. Blessin NC, Li W, Mandelkow T, Jansen HL, Yang C, Raedler JB, et al. Prognostic role of proliferating CD8(+) cytotoxic Tcells in human cancers. Cell Oncol. 2021;44(4):793–803.

    Article  CAS  Google Scholar 

  24. Tian C, Jing H, Wang C, Wang W, Cui Y, Chen J, et al. Prognostic role of tumour-infiltrating lymphocytes assessed by H&E-stained section in gastric cancer: a systematic review and meta-analysis. BMJ Open. 2021;11(1):e044163.

    Article  PubMed  PubMed Central  Google Scholar 

  25. Deng M, Li SH, Fu X, Yan XP, Chen J, Qiu YD, et al. Relationship between PD-L1 expression, CD8+ T-cell infiltration and prognosis in intrahepatic cholangiocarcinoma patients. Cancer Cell Int. 2021;21(1):371.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  26. Zhang Y, Li L, Zheng W, Zhang L, Yao N. CD8(+) T-cell exhaustion in the tumour microenvironment of head and neck squamous cell carcinoma determines poor prognosis. Annals of translational medicine. 2022;10(6):273.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  27. Kim M, Chung YR, Kim HJ, Woo JW, Ahn S, Park SY. Immune microenvironment in ductal carcinoma in situ: a comparison with invasive carcinoma of the breast. Breast cancer research : BCR. 2020;22(1):32.

    Article  PubMed  PubMed Central  Google Scholar 

  28. Di Pilato M, Kfuri-Rubens R, Pruessmann JN, Ozga AJ, Messemaker M, Cadilha BL, et al. CXCR6 positions cytotoxic T cells to receive critical survival signals in the tumour microenvironment. Cell. 2021;184(17):4512-30 e22.

    Article  PubMed  PubMed Central  Google Scholar 

  29. Stegmann KA, Robertson F, Hansi N, Gill U, Pallant C, Christophides T, et al. CXCR6 marks a novel subset of T-bet(lo)Eomes(hi) natural killer cells residing in human liver. Sci Rep. 2016;6:26157.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  30. Mabrouk N, Tran T, Sam I, Pourmir I, Gruel N, Granier C, et al. CXCR6 expressing T cells: Functions and role in the control of tumours. Front Immunol. 2022;13:1022136.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  31. Deng L, Chen N, Li Y, Zheng H, Lei Q. CXCR6/CXCL16 functions as a regulator in metastasis and progression of cancer. Biochem Biophys Acta. 2010;1806(1):42–9.

    PubMed  CAS  Google Scholar 

  32. Gao Q, Zhao YJ, Wang XY, Qiu SJ, Shi YH, Sun J, et al. CXCR6 upregulation contributes to a proinflammatory tumour microenvironment that drives metastasis and poor patient outcomes in hepatocellular carcinoma. Can Res. 2012;72(14):3546–56.

    Article  CAS  Google Scholar 

  33. Chang Y, Zhou L, Xu L, Fu Q, Yang Y, Lin Z, et al. High expression of CXC chemokine receptor 6 associates with poor prognosis in patients with clear cell renal cell carcinoma. Urologic oncology. 2017;35(12):675-e17- e24.

    Article  Google Scholar 

  34. Jin JJ, Dai FX, Long ZW, Cai H, Liu XW, Zhou Y, et al. CXCR6 predicts poor prognosis in gastric cancer and promotes tumour metastasis through epithelial-mesenchymal transition. Oncol Rep. 2017;37(6):3279–86.

    Article  PubMed  Google Scholar 

  35. Li Y, Fu LX, Zhu WL, Shi H, Chen LJ, Ye B. Blockade of CXCR6 reduces invasive potential of gastric cancer cells through inhibition of AKT signaling. Int J Immunopathol Pharmacol. 2015;28(2):194–200.

    Article  PubMed  CAS  Google Scholar 

  36. Yu C, Zhang Y. Characterization of the prognostic values of CXCR family in gastric cancer. Cytokine. 2019;123:154785.

    Article  PubMed  CAS  Google Scholar 

  37. Janjigian YY, Kawazoe A, Yanez P, Li N, Lonardi S, Kolesnik O, et al. The KEYNOTE-811 trial of dual PD-1 and HER2 blockade in HER2-positive gastric cancer. Nature. 2021;600(7890):727–30.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  38. Brahmer JR, Drake CG, Wollner I, Powderly JD, Picus J, Sharfman WH, et al. Phase I study of single-agent anti-programmed death-1 (MDX-1106) in refractory solid tumours: safety, clinical activity, pharmacodynamics, and immunologic correlates. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2010;28(19):3167–75.

    Article  PubMed  CAS  Google Scholar 

  39. Dempke WCM, Fenchel K, Dale SP. Programmed cell death ligand-1 (PD-L1) as a biomarker for non-small cell lung cancer (NSCLC) treatment-are we barking up the wrong tree? Translational lung cancer research. 2018;7(Suppl 3):S275–9.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  40. Choucair K, Morand S, Stanbery L, Edelman G, Dworkin L, Nemunaitis J. TMB: a promising immune-response biomarker, and potential spearhead in advancing targeted therapy trials. Cancer Gene Ther. 2020;27(12):841–53.

    Article  PubMed  CAS  Google Scholar 

  41. Jardim DL, Goodman A, de Melo GD, Kurzrock R. The Challenges of Tumour Mutational Burden as an Immunotherapy Biomarker. Cancer Cell. 2021;39(2):154–73.

    Article  PubMed  CAS  Google Scholar 

  42. Tumeh PC, Harview CL, Yearley JH, Shintaku IP, Taylor EJ, Robert L, et al. PD-1 blockade induces responses by inhibiting adaptive immune resistance. Nature. 2014;515(7528):568–71.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  43. He J, Xiong X, Yang H, Li D, Liu X, Li S, et al. Defined tumour antigen-specific T cells potentiate personalized TCR-T cell therapy and prediction of immunotherapy response. Cell Res. 2022;32(6):530–42.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  44. Li T, Pan J, Chen H, Fang Y, Sun Y. CXCR6-based immunotherapy in autoimmune, cancer and inflammatory infliction. Acta pharmaceutica Sinica B. 2022;12(8):3255–62.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  45. Lesch S, Blumenberg V, Stoiber S, Gottschlich A, Ogonek J, Cadilha BL, et al. T cells armed with C-X-C chemokine receptor type 6 enhance adoptive cell therapy for pancreatic tumours. Nature biomedical engineering. 2021;5(11):1246–60.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

Download references

Funding

This study was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF), funded by the Ministry of Education (2022R1C1C1003307).

Author information

Authors and Affiliations

  1. Department of Pathology, Dong-A University College of Medicine, Busan, 49201, Republic of Korea

    Song-Hee Han, Mi Ha Ju & Min Gyoung Pak

Authors
  1. Song-Hee Han
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Mi Ha Ju
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Min Gyoung Pak
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

Song-hee Han contributed to the conception and design of the study and outlined the research objectives and methodology. Mi Ha Ju collected and assembled the necessary data. Song-hee Han and Min Gyoung Pak analyzed the data and interpreted the results. All the authors reviewed the manuscript.

Corresponding author

Correspondence to Song-Hee Han.

Ethics declarations

Ethics approval and consent to participate

This study was a retrospective analysis conducted using paraffin-embedded tissue blocks that were originally prepared for diagnostic purposes. This study was approved by the Institutional Review Board of the Dong-A University Hospital (approval no. DAUHIRB-22–007). As the research utilized previously collected samples and did not have a direct impact on the patients, the requirement for written informed consent was waived by the Institutional Review Board of Dong-A University Hospital (approval no. DAUHIRB-22–007). This study was conducted in accordance with the declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

12876_2025_3735_MOESM1_ESM.pdf

Additional file 1: Supplementary Fig. 1. WGCNA Module Identification in GC. (a) Sample clustering and heatmap based on CD8 infiltration, group, and clinical stage. (b) Heatmap correlating module feature genes with clinical traits, displaying correlation coefficients, and p-values. (c, d) Scatter plots of gene significance (GS) vs. module membership in yellow (c) and grey (d) modules. Supplementary Fig. 2. Cox Proportional Hazard Analysis of CXCR6 in GC. Analysis of the impact of CXCR6 expression on GC prognosis.

Additional file 2.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Han, SH., Ju, M.H. & Pak, M.G. Prognostic and therapeutic potential of CXCR6 expression on CD8 + T cells in gastric cancer: a retrospective cohort study. BMC Gastroenterol 25, 139 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12876-025-03735-z

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12876-025-03735-z

Keywords

BMC Gastroenterology

ISSN: 1471-230X

Contact us