Skip to main content
NCBI home page
Journal of Ovarian Research logoLink to Journal of Ovarian Research
. 2025 Dec 17;19:19. doi: 10.1186/s13048-025-01883-0

Integrative analysis and experiments to explore GAS1 as a prognostic target for ovarian cancer based on angiogenesis-related genes

Lingyun Zhai 1, Da Huang 2, Liya Lin 4, Ruqian Zheng 1, Qionghua He 5, Xiaoqing Zhu 1, Rui Liu 2, Jing Fei 1, Zhigang Zhang 1, Xiaoming Zhang 1,2,✉, Jianwei Zhou 1,3,✉
PMCID: PMC12822331  PMID: 41408334

Abstract

Background

Intratumoral angiogenesis is crucial for the proliferation and metastasis of ovarian cancer. This study aims to comprehensively analyze the impact of angiogenesis-related genes on clinical outcomes, the immune landscape, and immunotherapy response in ovarian cancer.

Results

Through integrated analysis of transcriptomic data from The Cancer Genome Atlas and Gene Expression Omnibus databases, we identified two novel angiogenesis subtypes that exhibited significant differences in clinicopathological features, prognostic outcomes, and tumor microenvironment characteristics. An angiogenesis-based risk score was developed, which effectively stratified patients into distinct risk groups. These groups demonstrated divergent clinical prognosis, immune cell infiltration patterns, expression levels of immune checkpoint genes, and sensitivity to chemotherapeutic agents. Furthermore, growth arrest-specific 1 was validated as a hub prognostic gene, showing abnormal expression in ovarian cancer tissues. Functional experiments confirmed that upregulation of growth arrest-specific 1 significantly suppressed the proliferative and migratory capacities of ovarian cancer cells.

Conclusions

Our findings underscore the integral relationship between angiogenesis and the immune microenvironment in ovarian cancer. The angiogenesis-based risk score provides a promising prognostic tool, and growth arrest-specific 1 is implicated as a potential diagnostic and prognostic biomarker for ovarian cancer patients.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13048-025-01883-0.

Keywords: Angiogenesis, Tumor microenvironment, Immunotherapy, Ovarian cancer, Growth arrest-specific 1

Introduction

Ovarian cancer (OV) is a highly aggressive malignancy. As approximately 75% of patients are diagnosed at an advanced stage with abdominal and pelvic metastases, the primary treatment involves cytoreductive surgery followed by platinum-based chemotherapy. However, recurrence is common, and the five-year survival rate remains persistently low at approximately30% [1, 2]. Although recent therapeutic advances— including combination regimens, targeted therapy, and hyperthermic intraperitoneal chemotherapy (HIPEC) have been achieved, the long-term survival benefits of current strategies are still unsatisfactory [3–6]. Therefore, novel and more effective therapies are urgently required to improve the overall survival (OS) of ovarian cancer patients.

Tumor immunotherapy, which mobilizes the patient's immune system against cancer, represents a powerful approach in oncology. The consistent association between the presence of intraepithelial tumor-infiltrating lymphocytes (TILs) and improved clinical outcomes suggests that epithelial ovarian cancer may be immunoresponsive. Studies have indicated that patients with high levels of CD8 + T-cell infiltration in tumors experience longer OS [7–10]. Additionally, clinical activity of programmed cell death-1/programmed death ligand-1 (PD-1/PD-L1) inhibitors has been reported in recurrent ovarian cancer [11–13]. However, compared to progress seen in other malignancies such as non-small cell lung cancer (NSCLC), metastatic melanoma, and renal cell carcinoma, only a small subset of OV patients achieves durable clinical benefits. This limitation may be partly attributed to the highly immunosuppressive tumor microenvironment (TME) characteristic of ovarian carcinomas. Research indicates that OV establishes a multilayered immunosuppressive network to evade immune attack [14–17]. Notably, angiogenesis within the TME is a key driver of this immunosuppression[14, 18]. For instance, vascular endothelial growth factor (VEGF) can disrupt dendritic cell (DC) maturation and function [19, 20], recruit regulatory T cells (Tregs) and myeloid-derived suppressor cells (MDSCs), and potentiate their immunosuppressive activities within the TME [21–23]. Antiangiogenic therapy has the potential to enhance immunotherapy efficacy by normalizing the aberrant tumor vasculature and facilitating the infiltration of immune effector cells [24–26]. However, most current studies have focused on evaluating the role of individual angiogenesis-related genes (ARGs) in OV progression and prognosis. Given the suboptimal efficacy of immunotherapy alone in OV and the established influence of angiogenic factors on disease progression, investigating the correlation between ARG expression and the tumor-immune microenvironment (TIME) could help identify distinct immunophenotypes and thus improve the prediction of immunotherapy responses.

Herein, we systematically evaluated the expression profiles of ARGs and their impact on prognosis, TME, and treatment outcomes. We then identifiedtwo distinct angiogenesis-based molecular subtypesand comprehensively analyzed their associated clinical features, prognostic implications, and immune infiltration patterns.

. Based on differentially expressed genes (DEGs) between these subtypes, patients were further classified into two gene-based subgroups.Finally, we constructed an ARG_Score that serves as a robust predictor of clinical outcomes and immunotherapy response.

Materials and methods

Data source and process

Clinical information and gene expression data for OV were acquired from two public databases, the Gene Expression Omnibus (GEO) and UCSC Xena databases. Two eligible OV cohorts, GSE63885 and TCGA-OV were selected for subsequent analyses. Batch effects were corrected using the “sva” and “limma” packages in R. After excluding patients with missing survival data, a final cohort of 480 OV patients was established for subsequent analysis (clinical details in Table S1). A unified set of 75 ARGs was curated from the Molecular Signatures Database (MSigDB) by merging the HALLMARK_ANGIOGENESIS and GOBP_BLOOD_VESSEL_REMODELING gene sets (Table S2).

Identification of ARGs-based molecular subtypes

Unsupervised consensus clustering was applied to the expression profiles of ARGs to define molecular subtypes of OV. The optimal number of clusters was determined by assessing the consensus matrix heatmaps and the cumulative distribution function (CDF) of consensus scores, with the robustness of the classification ensured using the “Consensus Cluster Plus” R package. Survival analysis was performed to compare OS across the identified subtypes and to evaluate their associations with clinicopathological characteristics. Principal component analysis (PCA) was employed for dimensionality reduction to visualize the distribution and separation of the subtypes. To investigate the differences in underlying biological processes, Gene Set Variation Analysis (GSVA) was conducted using the respective R package. The reference gene sets “c2.cp.kegg.v7.5.1.symbols.gmt”, “c2.cp.reactome.v7.5.1.symbols.gmt”, and “h.all.v7.5.1.symbols.gmt” were obtained from the MSigDB database for this enrichment analysis.

Correlation analysis between TME and molecular subtypes

To determine between-cluster immune infiltration landscape, single-sample gene set enrichment analysis (ssGSEA) was utilized in “GSVA” to anticipate the infiltration levels of 23 distinct kinds of immune cells [27].The Stromal Score, Immune Score, as well as ESTIMATE Score were established utilizing the “ESTIMATE” tool. These values represent the proportion of stromal and immunological components relative to their total proportion in the TME.

Construction of an angiogenesis -related prognostic ARG_score

Various scoring systems were established to assess the angiogenesis levels of individuals utilizing the PCA method. First, univariate Cox regression analysis was performed to detect OS-associated DEGs. Second, the predictive DEGs (p < 0.01) were subjected to PCA to generate the angiogenesis-associated gene signature, after which principal components 1 and 2 were obtained as signature scores. Finally, PCA was performed to determine ARG score using the following formula: ARG_score = PC1 + PC2. Then, correlation analyses were done to establish the association of the ARG_score with TME cell infiltration.

Mutation and drug susceptibility analysis

The Gene Set Cancer Analysis (GSCA) database (http://bioinfo.life.hust.edu.cn/GSCA) is a cross-type tumor analysis database integrating single- and multi-gene, immune infiltration, and mutation as well as drug sensitivity analyses. We used this online database to determine the association of ARGs levels with somatic mutations as well as infiltration of immune cells.

Cell culture and transient transfection

The OV cell lines (A2780, SKOV3, OVCAR-3 and HEY) were purchased from Shanghai Cell Bank, Chinese Academy of Sciences (Shanghai, China). A2780, SKOV3 and OVCAR-3 were cultured in RPMI-1640, with 10% FBS and 1% P/S. HEY cells were maintained in DMEM, with 10% FBS and 1% P/S. All cell lines were cultured at 37 °C in a humidified incubator containing 5% CO2. The overexpressed plasmid of GAS1 and GAS1 small interfering RNA (siRNA) (Genechem, Shanghai, China) were transfected into OV cells by using Lipofectamine™ 3000 (Invitrogen, Themo Fisher Scientific, USA) according to the manufacturer’s instruction.

RNA extraction, RT-PCR, and RT-qPCR

Total RNA was retrieved from patients with OV utilizing Trizol reagent (Macherey-Nagel, Germany), and reverse transcription was done utilizing PrimeScript™ RT reagent Kit with gDNA Eraser (Takara, Japan). RT-qPCR was done utilizing TB Green® Premix Ex Taq™ II (Takara, Japan) on 7500FAST system (ABI, USA) following the supplier’s guidelines. The relative quantification of mRNA expression was expressed using 2−ΔΔCT. Table S3 displays the primers for genes.

Wound‑healing assay

A2780, OVCAR-3 and HEY cell lines were seeded into 6-well plates, transfected with the indicated plasmids and siRNAs. After 24 h, cells were digested using trypsin and seeded into 12-well plates for 24 h, to reach 90–100% confluency. A line wound was scratched by a 10 µl tip and the detached cells were removed by washing with PBS. Subsequently, the cells were cultured in corresponding serum-free culture medium. The wounds were observed at 0 h, 24 h.

Invasion and migration assay

Transfected cells were seeded into the upper chambers with serum-free medium, which were coated with or without VitroGel Hydrogel Matrix (The Well Bioscience, USA) for the transwell assay. Specifically, the OV cells were cultured in the upper chambers of 24-well culture plates with 5 × 104 cells and 200 µl serum-free culture medium. The lower chambers were incubated with 500ul complete medium. After incubation at 37 °C for 48 h, cells on the lower membrane were stained with 0.1% crystal violet, and the migrated cells were counted under a light microscope.

CCK8 assay

Transfected cells were seeded into 96-well plates with 2500 cells per well. At days 1, 2, 3 and 4, 10 µl CCK8 solution (Dojindo, China) was added into each well and incubated at 37 °C for 2 h. Finally, the absorbance at 450 nm was detected to determine the proliferative ability of cancer cells.

Cell apoptosis experiments

Annexin V-FITC/PI apoptosis assay kit (Absin, Shanghai) was used to perform apoptosis assay according to the manufacturer’s instructions. Briefly, cells were harvested at 48 h transfection, and were suspended with binding buffer, then, cells were treated with Annexin V-FITC and PI. The detection was performed using flow cytometry (BD Biosciences, USA).

Western blotting

OV cells were harvested after 72 h transfection and lysed in RIPA lysis buffer (Beyotime, China). Then protein lysates were boiled at 95˚C with 5×loading buffer for 10 min. The sample were separated using 4%–20% FuturePAGE Gels (ACE Biotechnology, China) and transfer to the PVDF membranes (Millipore, USA). The PVDF membranes were blocked with 5% non-fat dry milk in TBST for 2 h. The anti-GAS1 (Proteintech, China, 1:1000 dilution), anti-β-actin(Servicebio, China, 1:10000 dilution) were used as primary antibody. The anti-rabbit IgG HRP (CST, USA, 1:10000 dilution) as secondary antibody.

Statistical analyses

Statistical analyses were carried out in R (v4.1.1), ImageJ, and GraphPad Prism 9. Group comparisons were made using an unpaired two-sided t-test (two groups) or one-way ANOVA (multiple groups). For multiple testing in survival and pathway analyses, the Benjamini-Hochberg FDR method was applied, with significance set at FDR < 0.05. For all other tests, a two-sided P < 0.05 indicated significance.

Results

Survival analysis of angiogenesis-related genes in OV

Figure S1 shows the present research analyses. We collected 75 ARGs from the database. We first used these genes for survival analysis. A total of 480 patients were enrolled in the study, including 379 patients from the TCGA database and 101 patients from GEO database. To remove batch effects among different datasets, the R “sva” ComBat algorithm was performed, and the results without and with adjustment visualized by PCA plots (Fig. 1A). The present results confirmed that the quality of the sample data set used in this study was reliable. Subsequently, survival analysis of all ARGs was performed. The deregulation of most ARGs was significantly associated with overall survival. Specifically, high expression of TNFRSF21, STC1, MSX1, JAG1, HRG, EXT1, CEACAM1, CCR2, and ACE was correlated with poorer prognosis, whereas high expression of THBD, TGM2, PGLYRP1, CXCL6, AXL, BGN, BMPR2, AGTR2, LPL, POSTN, SLCO2A1, PF4, FGF10, FSTL1, NOS3, FLNA, LUM, FGF8, EPAS1, ACVRL1, COL5A2, VAV2, COL3A1, and CCND2 predicted a more favorable outcome (Fig. 1B).

Fig. 1.

Fig. 1

Open in a new tab

Prognostic analysis of ARGs in OV. (A) PCA plot before and after batch effect removal. (B) Kaplan-Meier survival curves of ARGs. ARGs, angiogenesis-related genes; PCA, Principal component analysis

Angiogenesis subtypes in OV

To establish the expression patterns of ARGs, the landscape of ARGs associations, regulator connections, and their predictive importance for OV patients were determined in an angiogenesis network (Fig. 2A). We then used a consensus clustering algorithm to categorize patients with qualitatively different angiogenesis modification patterns (Figure S2). It was found that k = 2 was optimal for categorization of the entire cohort into subtypes A (n = 318) and B (n = 162) (Fig. 2B). PCA further confirmed a clear separation between the two clusters (Fig. 2C). With regards to OS time, relative to subtype B, patients with subtype A exhibited longer OS outcomes (Fig. 2D). Furthermore, heatmap showed that the intrinsic differences of genomic expression and clinicopathological variables between the two subtypes (Fig. 2E). Among the two main angiogenesis subtypes, there were marked differences in the expression of ARGs between subtypes, and most ARGs were upregulated in cluster B (Fig. 2F). Thus, findings from consensus clustering analysis were related to OV development and survival time.

Fig. 2.

Fig. 2

Open in a new tab

Identification of angiogenesis subtypes in OV. (A) Associations among ARGs in OV (shown by lines connecting ARGs). Blue and pink lines denote negative and positive associations, respectively. (B) Consensus matrix heatmap defining two clusters (k = 2) and their association area. (C) PCA analysis revealed marked variations in transcriptomes between the clusters. (D) Kaplan–Meier curves for OS between the clusters. (E) Unsupervised clustering of ARGs in the meta-cohort divided the patients into two distinct clusters. (F) Differential expression analysis of ARGs between the clusters. ARGs, angiogenesis-related genes; PCA, principal component analysis; OS, Overall survival; *p < 0.05, **p < 0.01, ***p < 0.001; ns, no significance

Characteristics of the TME in distinct subtypes

To elucidate differences in survival among clusters, we first performed KEGG hallmark and REACTOME pathway enrichment analyses using GSVA to establish biological and functional differences between the subtypes. Enrichment pathway analyses for every sample showed that the two subtypes had precise expression profiles. Cluster B was enriched in metastasis-associated pathways (focal adhesion, ECM receptor interaction, actin cytoskeleton regulation, epithelial-to-mesenchymal transition, extracellular matrix organization, angiogenesis, as well as integrin cell surface interaction pathways) and immune-related pathways (cytokine receptor interactions, natural killer cell-mediated cytotoxicity, chemokine signaling pathways, T and B cell receptor signaling pathways, NOD- and Toll-like receptor signaling pathways) (Fig. 3A-C). To identify the relationship among the subtypes and TME of OV, we use the ESTIMATE tool to evaluate the TME score (stromal, immune, and estimate scores) of the two subtypes. The findings demonstrated clear variations in TME scores between the two categories, with patients in cluster B having higher TME scores in to assess the quantity of stromal and immunological elements in TME (Fig. 3D). Subsequently, we evaluated the associations between subtypes and the same and 23 kinds of immune cell infiltration emplying the ssGSEA algorithm. We found that a considerable variation in infiltration between subtype A and subtype B was markedly rich in immune cells, including B cells, NK cells, mast cells, CD4 T + cells, CD8 T + cells, regulatory T cells, dendritic cells, MDSCs, macrophages, and neutrophil (Fig. 3E).

Fig. 3.

Fig. 3

Open in a new tab

Correlations of tumor immune microenvironments and two OV subtypes. (A-C) GSVA enrichment analysis showing enriched biological pathways between the clusters. These biological processes were visualized using a heatmap; red and blue denote activated and inhibited pathways, respectively. (D) Association between TME scores and the two clusters. (E) Relative abundances of 23 types of immune cells between the clusters. TME, tumor microenvironment; GSVA, gene set variation analysis

Gene subtypes based on DEGs in OV

To establish the probable biological behavior for every angiogenesis pattern, “limma” in R was used for identification of DEGs among the two clusters, and 175 genes were identified (Fig. 4A and Table S4). Next, the above DEGs were performed using function enrichment analysis. These DEGs were markedly enriched in metastasis-associated biological processes, involving extracellular matrix, extracellular structure organization, and extracellular matrix organization. KEGG analysis also showed enriched metastasis-related pathways (Fig. 4B-C), suggesting that angiogenesis is crucial in tumor metastasis modulation. Following that, 11 genes were selected out using univariate Cox analysis to determine their predictive significance (p < 0.01) besides being employed in further analysis (Fig. 4D). To investigate specific adjustment mechanisms, we used a consensus clustering approach to assign patients into two genomic subtypes based on predictive genes (Figure S3). Patients in gene cluster A exhibited better OS time than those in gene cluster (Fig. 4E). In addition, marked variations existed between the gene subtypes regarding ARGs expression, FIGO stage, grade, survival time, and status (Fig. 4F).

Fig. 4.

Fig. 4

Open in a new tab

DEGs-based gene subtypes. (A) Volcano map of DEGs. (B-C) GO and KEGG analyses of DEGs between the two angiogenesis subtypes. (D) Forest map showing findings from univariate Cox regression analysis for 11 DEGs. (E) Kaplan–Meier curves for OS of two gene clusters. (F) Association between clinic-pathologic characteristics and the two gene clusters. DEGs, differentially expressed genes

Generation and immunological features of ARG_score

The ARG_score was developed using cluster-related DEGs. Of these 175 DEGs, we found that 36 genes showed prognostic significance in relation to OV patients. As previously described, eleven key genes (GAS1, KIF26B, COL16A1, CCDC80, TIMP3, FZD1, TGFBI, VSIG4, EMILIN1, NUAK1 and MFAP4) among the angiogenesis cluster related prognostic genes were obtained to establish a scoring system using the PCA algorithm. Patients were categorized into two risk score groups, and two gene subtypes were visualized using the alluvial diagram (Fig. 5A). Patients with low scores exhibited markedly better OS outcomes, relative to high score patients (Fig. 5B). Next, we examined the association of the risk score with angiogenesis clusters A-B and geneclusters A-B. Angiogenesis cluster B (Fig. 5C) and gene cluster B (Fig. 5D) exhibited high ARG_scores, whereas gene cluster B (Fig. 4F) and angiogenesis cluster B (Fig. 2D) exhibited poor prognostic outcomes. Interestingly, to identify the relationship of immune cell infiltration with score, we established that most immune cells were positively related to ARG_score, particularly regulatory T cells, macrophages, MDSC, follicular helper T cells and helper T cell type 1(Fig. 5E). Based on the above results, we assessed the relationship of selected gene expression and immune cell infiltration. As shown in Fig. 5F, almost all genes had positive correlations with monocytes, Th2 cells, induced regulatory T (Treg) cells, and macrophages, while negatively with B cells, neutrophils and Th17 cells. These findings also validate our analysis results and low-risk scores could be strongly correlated with immune activation-associated characteristics, while high-risk scores could be correlated with immune suppression-related characteristics.

Fig. 5.

Fig. 5

Open in a new tab

Construction of risk score and its relationship with immune cells. (A) Alluvial diagram of subtype distributions in groups with different ARG_scores and survival outcomes. (B) Kaplan-Meier survival analysis of high and low risk score group in the meta-cohort. (C) Differences in ARG_score between the two angiogenesis clusters. (D) Differences in ARG_score between the two geneclusters. (E) Correlation between ARG_score and immune cell infiltration. (F) The relationship between the expression of eleven differential genes and immune cell infiltration

Associations between ARG_scores and immunotherapy

Immunotherapy, particularly immune checkpoint blockade (ICB) therapy, is an efficient option for solid cancer treatment. By investigating the link of ARG_score to key immune checkpoint genes (TGFB1, PDCD1, LAG3, CTLA4, TIGIT, and CD274) in OV, we observed that the expressions of these 6 immune checkpoint genes were markedly high in the high ARG_score group, relative to low ARG_score group(Fig. 6A).This suggested that high ARG_score patients might be more appropriate for ICB therapy in TCGA-OV. However, the immunosuppression microenvironment also occurred due to high expression of immune checkpoint genes. This also explains why high ARG_score patients had a relative worse prognostic outcome. To evaluate if the ARG_score can predict patient response to ICB therapy, we assessed the prognostic significance of ARG_score for patients administered with ICB therapy in urothelial carcinoma and advanced clear cell renal cell carcinoma (ccRCC) immunotherapy cohorts by assigning them into ARG_score-low and–high groups. High ARG_score patients had a significantly long overall survival time, relative to those with lower ARG_score in urothelial carcinoma of IMvigor210 cohort (Fig. 6B, C) and the ccRCC cohort (Fig. 6D, E).

Fig. 6.

Fig. 6

Open in a new tab

Evaluation of immune checkpoint genes and immunotherapy response between the high- and low- risk groups. (A) Differences in expression of immune checkpoints between high- and low- risk groups. (B) The proportion of patients with or without response to PD-L1 blockade therapy in the ARG_score-high and ARG_score-low groups in the IMvigor210 cohort. (C) Survival analysis for patients with high and low ARG_score in the IMvigor210 cohort. (D) The proportion of patients with or without response to PD-1 blockade therapy therapy in the ARG_score-high and ARG_score-low groups in the clear cell renal cell carcinoma cohort. (E) Survival analysis for patients with high and low ARG_score in the clear cell renal cell carcinoma cohort. PR, partial response; CR, complete response; PD, progressive disease; SD, stable disease

GAS1 inhibited the migration and invasion of OV cells in vitro

Based on the eleven key genes identified in the previous analysis, we subsequently investigated their expression profiles using PCR. Notably, the expression of these genes varied between normal ovarian cells and four ovarian cancer cell lines (Figure S4). Among them, GAS1 exhibited the most significant differential expression (Fig. 7A). Considering gene GAS1 have not been functionally characterized in OV, we conducted experiments in tumor cells. We first constructed the GAS1 knockdown in OVCAR3 cells (high endogenous GAS1 expression) and GAS1 overexpression in A2780 and HEY cells (low endogenous GAS1 expression). RT-qPCR and Western blotting showed that GAS1 could be effectively silenced (Fig. 7B-C) or overexpressed (Fig. 8A-C). As indicated in the results, the CCK8 and wounding healing assay showed that overexpression of GAS1 significantly inhibited SKOV3 and HEY cell proliferation and migration (Fig. 8D-F), whereas GAS1 depletion significantly increased the proliferation and migration ability in OVCAR3 cells (Fig. 7D-E). In the meantime, the transwell assay showed that GAS1 overexpression significantly inhibited SKOV3 and HEY cells invasion and migration (Fig. 8G) but that siRNA-mediated GAS1 knockdown promoted the invasion and migration of OVCAR3 cells (Fig. 7F). Moreover, the apoptotic assay also demonstrated that overexpressed GAS1 induced SKOV3 and HEY cells apoptosis (Fig. 8H) but decreased apoptosis when GAS1 was silenced in OVCAR3 cells (Fig. 7G). These results suggest that GAS1 may play a vital anticancer role in the progress of OV.

Fig. 7.

Fig. 7

Open in a new tab

GAS1 promotes OV cell proliferation, migration and inhibits cell apoptosis. (A) The mRNA expression level of GAS1 in cells detected by RT-qPCR. (B-C)The expression level of GAS1 detected by RT-qPCR and Western blot in OVCAR3 cells transfected with siRNA. (D) Cell proliferation of OVCAR3 cells transfected with control or si-GAS1 was measured by CCK8. (E) Cell migration of OVCAR3 cells transfected with control or si-GAS1 was measured by wound healing assay. (F) Cell migration and invasion of OVCAR3 cells transfected with control or si-GAS1 was measured by transwell assay. (G) Cell apoptosis of OVCAR3 cells transfected with control or si-GAS1 was measured by flow cytometry. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001

Fig. 8.

Fig. 8

Open in a new tab

GAS1 inhibits OV cell proliferation, migration and promotes cell apoptosis. (A-C) The expression level of GAS1 detected by RT-qPCR and Western blot in A2780 and HEY cells transfected with GAS1 overexpression plasmid. (D) Cell proliferation of A2780 and HEY cells transfected with control or GAS1 overexpression plasmid was measured by CCK8. (E-F) Cell migration of A2780 and HEY cells transfected with control or GAS1 overexpression plasmid was measured by wound healing assay. (G) Cell migration and invasion of A2780 and HEY cells transfected with control or GAS1 overexpression plasmid was measured by transwell assay. (H) Cell apoptosis of A2780 and HEY cells transfected with control or GAS1 overexpression plasmid was measured by flow cytometry. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001

Discussion

Angiogenesis is an essential hallmark of OV and is involved in tumor immune escape. In the TME, tumor cells secrete proangiogenic factors which are involved in the creation of an abnormal vascular network (tortuous, disorganized and excessively leaky), resulting in irregular blood perfusion and oxygenation. This dysregulated morphology leads to a hypoxic and acidic microenvironment, thereby hindering immune cell infiltration into the tumor and impairing its function.[28–30]. Besides, the crosstalk between pro-angiogenic factors and the immune system could directly suppress the immune effector cells and antigen-presenting cells or by augmenting the effects of immune-suppressive cells, which could in turn also facilitate angiogenesis, establishing a vicious cycle of dysregulated immune activations [31–34]. For this reason, most anti-angiogenic agents employed as monotherapy in the clinic exhibited only limit efficacy [35–37]. However, combining anti-angiogenic treatment with immunotherapy in recent years was demonstrated the potential clinical benefits in various of cancers [38–40].In ovarian carcinoma, combination therapy with bevacizumab and immunotherapy has shown a signal for enhanced objective response rates compared to monotherapy [41, 42]. As many current investigations have primarily focused on a single angiogenesis-related gene, especially VEGF/VEGFR, there is a need for comprehensive analyses of the roles of angiogenesis in mediating TME immune cell infiltrations.

To advance the in-depth study of the correlation between angiogenesis and the intrinsic TME in ovarian carcinoma, it is necessary to elucidate how different angiogenic patterns influence immune infiltration. This will help us comprehend the underlying mechanisms of the immune response and improve the efficacy of existing immunotherapies.

In this study, we found that ARGs were correlated with clinical outcomes in OV patients and higher levels of many of these genes were related to inferior prognosis. Based on the expression patterns of these genes, we classified OV patients into two distinct angiogenesis subtypes (Cluster A and Cluster B). These subtypes exhibited different clinicopathological characteristics and TME immune landscapes. Patients in Cluster B showed more advanced pathological features and significantly worse overall survival compared with those in Cluster A. Differences in TME characteristics between the two subtypes were also marked, with cluster B being associated with high immune cell infiltration, including B cells, CD4 + T cells, dendritic cells, CD8 + T cells, eosinophils, MDSCs, macrophages, mast cells, neutrophils and T helper cells. To determine the sources of these variations, GSVA enrichment analysis was performed.

The results revealed that Cluster B was markedly enriched in immune- and tumor-related biological pathways, which is consistent with previously reported findings.Angiogenic factors could directly interact with these signaling pathways to promote tumor progression and may serve as immune modulators, thereby augmenting angiogenesis and immune suppression. [43, 44]. Furthermore, we established two gene clusters based on the DEGs between the two angiogenesis subtypes (termed Gene Clusters A and B). The two gene clusters also exhibited distinct clinical features. Given individual heterogeneity of angiogenesis patterns, we developed a risk score based on 11 DEGs (GAS1, KIF26B, COL16A1, CCDC80, TIMP3, FZD1, TGFBI, VSIG4, EMILIN1, NUAK1 and MFAP4) that can evaluate risk of OV patients individually. Stratifying patients into high- and low-risk score groups resulted in significant variations in clinicopathological characteristics, prognostic outcomes, immune cell infiltration, immune checkpoints, immunotherapeutic responses and drug susceptibility. As predicted, cluster B and gene cluster B with worse clinical outcomes were linked to a greater ARG_score. Consistently, compared with patients with low ARG_score, high ARG_score patients had worse OS outcomes. In addition, we performed an comprehensive analysis of the relationship between TME cell infiltration and risk scores. And we found the significant differences in TME features and relative abundances of 23 TIICs between the molecular subtypes with different ARG_score. The angiogenesis score showed a positive association with immunosuppressive cells, involving Tregs, MDSCs, and macrophages. Furthermore, the individual expression of these 11 ARGs were closely associated with immune cell infiltrations. OV is an immune cell infiltrating tumor that is characterized with an immune-suppressive TME [45–48]. The high ARG_score group had elevated immune checkpoint gene levels (TGFB1, PDCD1, LAG3, TIGIT, CD274 and CTLA4), which act as pivotal regulators of immune escape in cancer by inhibiting effector immune cells activity, facilitating cells anergy and exhaustion [49]. All of these indicate poor prognostic outcomes for patients with high-risk ARGs that may be induced by up‐regulation of immune checkpoints and formation of immunosuppressive TME.

GAS1 is a glycosylphosphatidylinositol (GPI)-anchored protein that is widely expressed in both epithelial cells and fibroblasts. The available studies demonstrate that GAS1 could induce growth arrest and apoptosis in cells [49, 50].In colorectal cancer (CRC) cells, GAS1 inhibited the Warburg effect and epithelial–mesenchymal transition (EMT) via suppression of the mTOR signaling pathway and activation of the AMPK signaling pathway [51]. The latest research suggests that Bifidobacterium adolescentis (B.a) suppress tumorigenesis by facilitating a subset of CD143 + cancer-associated fibroblasts, which activate Wnt/β-catenin signaling pathways, and the activation of Wnt/β-catenin promotes GAS1 expression [52]. In addition, the enhancement of human brain tumor initiating cell growth inhibition mediated by microglia can be achieved through the upregulation of GAS1 expression in microglia [53]. Therefore, GAS1 have been acknowledged as a negative regulator in tumorigenesis. This study further elucidated the involvement of GAS1 in ovarian cancer cell lines through the establishment of a model encompassing both GAS1 overexpression and inhibition. The cell experiments indicated that overexpressing of GAS1 obviously inhibited the proliferation and migration of OV cells. In contrast, the suppression of GAS1 expression has been observed to facilitate the proliferation, invasion, and metastasis of tumor cells. These findings suggest that targeting GAS1 regulation could be a viable therapeutic approach for individuals diagnosed with ovarian cancer.

However, our study also has several limitations. First, all samples were retrospectively obtained from public databases, and thus inherent selection bias may have affected the robustness of the results. Second, although we observed a potential immunologic association of the ARG score, the validation using the IMvigor210 immunotherapy cohort should be regarded as exploratory, since ovarian cancer differs from urothelial carcinoma in histologic origin and immune microenvironment. Third, the function of GAS1 was only validated in vitro, necessitating further investigation into its underlying mechanism and in vivo effects. In addition, rescue experiments for GAS1 were not performed in this study, which may limit direct mechanistic validation. Future studies should include ovarian cancer–specific immunotherapy datasets or patient-derived data to further confirm the clinical applicability of the ARG score, and to elucidate differences between risk groups in the tumor microenvironment, immune cell infiltration, and intercellular communication.

Conclusion

In this study, we divided OV patients into two subtypes of angiogenesis, which have different prognoses, clinical and pathological features, functions, time patterns, and immunotherapy responses. Based on these subtypes, we have identified a prognostic gene, GAS1, which plays an important role in ovarian cancer cells, affecting the proliferation, invasion, and migration of this type of cancer. GAS1 may be a crucial target for the treatment of ovarian cancer.

Supplementary Information

Supplementary Material 1. (844.7KB, docx)
Supplementary Material 2. (886.3KB, docx)
Supplementary Material 3. (135KB, xls)
Supplementary Material 4. (355.5KB, xls)
Supplementary Material 5. (775.7KB, zip)

Acknowledgements

We would like to thank the researchers and study participants for their contributions.

Authors’ contributions

Lingyun Zhai and Da Huang participated in the study design and wrote the manuscript. Ruqian Zheng and Qionghua He collected the data. Xiaoqing Zhu, Liya Lin and Rui Liu performed functional experiments. Jing Fei and Zhigang Zhang conducted data analysis. Xiaoming Zhang and Jianwei Zhou helped the revision. All authors contributed to the article and approved the submitted version.

Funding

The study was supported by the National Natural Science Foundation of China (Grant number 32471290 and Grant number 82203620). The Natural Science Foundation of Zhejiang Province (grant number QN25H160093) and the Zhejiang Provincial Leading Project for Leading Geese Plan（grant number 2025C02114).

Data availability

The data used in this study were obtained from public repositories, specifically The Cancer Genome Atlas (TCGA) and the Gene Expression Omnibus (GEO) databases. No new data were created or analyzed during this study.

Declarations

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s Note

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

Contributor Information

Xiaoming Zhang, Email: zxm@zju.edu.cn.

Jianwei Zhou, Email: jianwei-zhou@zju.edu.cn.

References

  • 1.Murphy B, Miyamoto T, Manning BS, Mirji G, Ugolini A, Kannan T, et al. Myeloid activation clears Ascites and reveals IL27-dependent regression of metastatic ovarian cancer. J Exp Med. 2024;221(12):e20231967. [DOI] [PMC free article] [PubMed]
  • 2.Kahn R, Filippova O, Gordhandas S, An A, Straubhar AM, Zivanovic O, et al. Ten-year conditional probability of survival for patients with ovarian cancer: A new metric tailored to Long-term survivors. Gynecol Oncol. 2023;169:85–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Min Y, Park HB, Baek KH, Hwang S. Cellular functions of deubiquitinating enzymes in ovarian adenocarcinoma. Genes (Basel). 2023;14(4):886. [DOI] [PMC free article] [PubMed]
  • 4.Nguyen QM, Dupre PF, Haute T, Montier T, d’Arbonneau F. Suicide gene strategies applied in ovarian cancer studies. Cancer Gene Ther. 2023;30(6):812–21. [DOI] [PubMed] [Google Scholar]
  • 5.D’Indinosante M, Guidi G, Giannarelli D, Diella C, Rosati A, Oliva R, et al. Laparoscopic tumor load as an independent prognostic marker in advanced ovarian cancer: a 3-year cohort study. Int J Gynecol Cancer. 2025;35(8):101965. [DOI] [PubMed] [Google Scholar]
  • 6.Ray-Coquard I, Leary A, Pignata S, Cropet C, Gonzalez-Martin A, Marth C, et al. Olaparib plus bevacizumab first-line maintenance in ovarian cancer: final overall survival results from the PAOLA-1/ENGOT-ov25 trial. Ann Oncol. 2023;34(8):681–92. [DOI] [PubMed] [Google Scholar]
  • 7.Kasikova L, Rakova J, Hensler M, Lanickova T, Tomankova J, Pasulka J, et al. Tertiary lymphoid structures and B cells determine clinically relevant T cell phenotypes in ovarian cancer. Nat Commun. 2024;15(1):2528. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Bareche Y, Pommey S, Carneiro M, Buisseret L, Cousineau I, Thebault P et al. High-dimensional analysis of the adenosine pathway in high-grade serous ovarian cancer. J Immunother Cancer. 2021;9(3). [DOI] [PMC free article] [PubMed]
  • 9.Olalekan S, Xie B, Back R, Eckart H, Basu A. Characterizing the tumor microenvironment of metastatic ovarian cancer by single-cell transcriptomics. Cell Rep. 2021;35(8):109165. [DOI] [PubMed] [Google Scholar]
  • 10.Guo N, Yang A, Farooq FB, Kalaria S, Moss E, DeVorkin L, et al. CD8 + T cell infiltration is associated with improved survival and negatively correlates with hypoxia in clear cell ovarian cancer. Sci Rep. 2023;13(1):6530. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Zsiros E, Lynam S, Attwood KM, Wang C, Chilakapati S, Gomez EC, et al. Efficacy and safety of pembrolizumab in combination with bevacizumab and oral metronomic cyclophosphamide in the treatment of recurrent ovarian cancer: A phase 2 nonrandomized clinical trial. JAMA Oncol. 2021;7(1):78–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Hinchcliff E, Hong D, Le H, Chisholm G, Iyer R, Naing A, et al. Characteristics and outcomes of patients with recurrent ovarian cancer undergoing early phase immune checkpoint inhibitor clinical trials. Gynecol Oncol. 2018;151(3):407–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Ledermann JA, Shapira-Frommer R, Santin AD, Lisyanskaya AS, Pignata S, Vergote I, et al. Molecular determinants of clinical outcomes of pembrolizumab in recurrent ovarian cancer: exploratory analysis of KEYNOTE-100. Gynecol Oncol. 2023;178:119–29. [DOI] [PubMed] [Google Scholar]
  • 14.Zhang AW, McPherson A, Milne K, Kroeger DR, Hamilton PT, Miranda A, et al. Interfaces of malignant and Immunologic clonal dynamics in ovarian cancer. Cell. 2018;173(7):1755–e6922. [DOI] [PubMed] [Google Scholar]
  • 15.Zeng S, Liu D, Yu Y, Zou L, Jin X, Liu B, et al. Efficacy and safety of PD-1/PD-L1 inhibitors in the treatment of recurrent and refractory ovarian cancer: A systematic review and a meta-analysis. Front Pharmacol. 2023;14:1111061. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Kader T, Lin JR, Hug CB, Coy S, Chen YA, de Bruijn I, et al. Multimodal Spatial profiling reveals immune suppression and microenvironment remodeling in fallopian tube precursors to High-Grade serous ovarian carcinoma. Cancer Discov. 2025;15(6):1180–202. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Matsushita H, Hasegawa K, Oda K, Yamamoto S, Asada K, Karasaki T, et al. Neoantigen load and HLA-class I expression identify a subgroup of tumors with a T-cell-inflamed phenotype and favorable prognosis in homologous recombination-proficient high-grade serous ovarian carcinoma. J Immunother Cancer. 2020;8(1):e000375. [DOI] [PMC free article] [PubMed]
  • 18.Tada Y, Togashi Y, Kotani D, Kuwata T, Sato E, Kawazoe A, et al. Targeting VEGFR2 with ramucirumab strongly impacts effector/ activated regulatory T cells and CD8(+) T cells in the tumor microenvironment. J Immunother Cancer. 2018;6(1):106. [DOI] [PMC free article] [PubMed]
  • 19.Ibanez-Vea M, Zuazo M, Gato M, Arasanz H, Fernandez-Hinojal G, Escors D, et al. Myeloid-Derived suppressor cells in the tumor microenvironment: current knowledge and future perspectives. Arch Immunol Ther Exp (Warsz). 2018;66(2):113–23. [DOI] [PubMed] [Google Scholar]
  • 20.Bohn T, Rapp S, Luther N, Klein M, Bruehl TJ, Kojima N, et al. Tumor immunoevasion via acidosis-dependent induction of regulatory tumor-associated macrophages. Nat Immunol. 2018;19(12):1319–29. [DOI] [PubMed] [Google Scholar]
  • 21.Cao K, Zhang G, Zhang X, Yang M, Wang Y, He M, et al. Stromal infiltrating mast cells identify immunoevasive subtype high-grade serous ovarian cancer with poor prognosis and inferior immunotherapeutic response. Oncoimmunology. 2021;10(1):1969075. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Qiao S, Hou Y, Rong Q, Han B, Liu P. Tregs are involved in VEGFA/ VASH1-related angiogenesis pathway in ovarian cancer. Transl Oncol. 2023;32:101665. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Lee WS, Yang H, Chon HJ, Kim C. Combination of anti-angiogenic therapy and immune checkpoint Blockade normalizes vascular-immune crosstalk to potentiate cancer immunity. Exp Mol Med. 2020;52(9):1475–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Khan KA, Kerbel RS. Improving immunotherapy outcomes with anti-angiogenic treatments and vice versa. Nat Rev Clin Oncol. 2018;15(5):310–24. [DOI] [PubMed] [Google Scholar]
  • 25.Chryplewicz A, Scotton J, Tichet M, Zomer A, Shchors K, Joyce JA, et al. Cancer cell autophagy, reprogrammed macrophages, and remodeled vasculature in glioblastoma triggers tumor immunity. Cancer Cell. 2022;40(10):1111–27. e9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Dong X, Ren J, Amoozgar Z, Lee S, Datta M, Roberge S, et al. Anti-VEGF therapy improves EGFR-vIII-CAR-T cell delivery and efficacy in syngeneic glioblastoma models in mice. J Immunother Cancer. 2023;11(3):e005583. [DOI] [PMC free article] [PubMed]
  • 27.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(1):248–62. [DOI] [PubMed] [Google Scholar]
  • 28.Fukurnura D, Kloepper J, Amoozgar Z, Duda DG, Jain RK. Enhancing cancer immunotherapy using antiangiogenics: opportunities and challenges. Nat Rev Clin Oncol. 2018;15(5):325–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Bao W, Li Z. Efficacy and safety of neoadjuvant chemotherapy containing anti-angiogenic drugs, immunotherapy, or PARP inhibitors for ovarian cancer. Crit Rev Oncol Hematol. 2024;194:104238. [DOI] [PubMed] [Google Scholar]
  • 30.Miao D, Matulonis UA, Porter RL. Ovarian cancer therapy. Cold Spring Harb Perspect Med; 2025. [DOI] [PMC free article] [PubMed]
  • 31.Rahma OE, Hodi FS. The intersection between tumor angiogenesis and immune suppression. Clin Cancer Res. 2019;25(18):5449–57. [DOI] [PubMed] [Google Scholar]
  • 32.Missiaen R, Mazzone M, Bergers G. The reciprocal function and regulation of tumor vessels and immune cells offers new therapeutic opportunities in cancer. Semin Cancer Biol. 2018;52:107–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Zhang L, Xu J, Zhou S, Yao F, Zhang R, You W, et al. Endothelial DGKG promotes tumor angiogenesis and immune evasion in hepatocellular carcinoma. J Hepatol. 2024;80(1):82–98. [DOI] [PubMed] [Google Scholar]
  • 34.Zheng W, Qian C, Tang Y, Yang C, Zhou Y, Shen P, et al. Manipulation of the crosstalk between tumor angiogenesis and immunosuppression in the tumor microenvironment: insight into the combination therapy of anti-angiogenesis and immune checkpoint Blockade. Front Immunol. 2022;13:1035323. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Ferrara N, Adamis AP. Ten years of anti-vascular endothelial growth factor therapy. Nat Rev Drug Discov. 2016;15(6):385–403. [DOI] [PubMed] [Google Scholar]
  • 36.Gaitskell K, Rogozinska E, Platt S, Chen Y, Abd El Aziz M, Tattersall A, et al. Angiogenesis inhibitors for the treatment of epithelial ovarian cancer. Cochrane Database Syst Rev. 2023;4(4):CD007930. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Wang CW, Lee YC, Lin YJ, Chang CC, Sai AK, Wang CH, et al. Ensemble biomarkers for guiding anti-angiogenesis therapy for ovarian cancer using deep learning. Clin Transl Med. 2023;13(1):e1162. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Socinski MA, Jotte RM, Cappuzzo F, Orlandi F, Stroyakovskiy D, Nogami N, et al. Atezolizumab for First-Line treatment of metastatic nonsquamous NSCLC. N Engl J Med. 2018;378(24):2288–301. [DOI] [PubMed] [Google Scholar]
  • 39.Rini BI, Plimack ER, Stus V, Gafanov R, Hawkins R, Nosov D, et al. Pembrolizumab plus axitinib versus Sunitinib for advanced Renal-Cell carcinoma. N Engl J Med. 2019;380(12):1116–27. [DOI] [PubMed] [Google Scholar]
  • 40.Finn RS, Qin S, Ikeda M, Galle PR, Ducreux M, Kim TY, et al. Atezolizumab plus bevacizumab in unresectable hepatocellular carcinoma. N Engl J Med. 2020;382(20):1894–905. [DOI] [PubMed] [Google Scholar]
  • 41.Liu JF, Herold C, Gray KP, Penson RT, Horowitz N, Konstantinopoulos PA, et al. Assessment of combined nivolumab and bevacizumab in relapsed ovarian cancer: A phase 2 clinical trial. JAMA Oncol. 2019;5(12):1731–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Kurtz JE, Pujade-Lauraine E, Oaknin A, Belin L, Leitner K, Cibula D, et al. Atezolizumab combined with bevacizumab and Platinum-Based therapy for Platinum-Sensitive ovarian cancer: Placebo-Controlled randomized phase III ATALANTE/ENGOT-ov29 trial. J Clin Oncol. 2023;41(30):4768–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Albini A, Bruno A, Noonan DM, Mortara L. Contribution to tumor angiogenesis from innate immune cells within the tumor microenvironment: implications for immunotherapy. Front Immunol. 2018;9:527. [DOI] [PMC free article] [PubMed]
  • 44.Chen X, Wang K, Jiang S, Sun H, Che X, Zhang M et al. eEF2K promotes PD-L1 stabilization through inactivating GSK3beta in melanoma. J Immunother Cancer. 2022;10(3). [DOI] [PMC free article] [PubMed]
  • 45.Jiang YT, Wang CD, Zhou ST. Targeting tumor microenvironment in ovarian cancer: premise and promise. Bba-Rev Cancer. 2020;1873(2):188361. [DOI] [PubMed]
  • 46.Jimenez-Sanchez A, Memon D, Pourpe S, Veeraraghavan H, Li YY, Vargas HA, et al. Heterogeneous Tumor-Immune microenvironments among differentially growing metastases in an ovarian cancer patient. Cell. 2017;170(5):927–. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Mollaoglu G, Tepper A, Falcomata C, Potak HT, Pia L, Amabile A, et al. Ovarian cancer-derived IL-4 promotes immunotherapy resistance. Cell. 2024;187(26):7492–e51022. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Zhao J, Wang Z, Tian Y, Ning J, Ye H. T cell exhaustion and senescence for ovarian cancer immunotherapy. Semin Cancer Biol. 2024;104–105:1–15. [DOI] [PubMed] [Google Scholar]
  • 49.Zhang Y, Zheng J. Functions of immune checkpoint molecules beyond immune evasion. Adv Exp Med Biol. 2020;1248:201–26. [DOI] [PubMed] [Google Scholar]
  • 50.Tao J, Xue C, Wang X, Chen H, Liu Q, Jiang C, et al. GAS1 promotes ferroptosis of liver cells in Acetaminophen-Induced acute liver failure. Int J Med Sci. 2023;20(12):1616–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.King CM, Ding W, Eshelman MA, Yochum GS. TCF7L1 regulates colorectal cancer cell migration by repressing GAS1 expression. Sci Rep. 2024;14(1):12477. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Chen S, Fan L, Lin Y, Qi Y, Xu C, Ge Q, et al. Bifidobacterium adolescentis orchestrates CD143(+) cancer-associated fibroblasts to suppress colorectal tumorigenesis by Wnt signaling-regulated GAS1. Cancer Commun (Lond). 2023;43(9):1027–47. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Sarkar S, Poon CC, Mirzaei R, Rawji KS, Hader W, Bose P, et al. Microglia induces Gas1 expression in human brain tumor-initiating cells to reduce tumorigenecity. Sci Rep. 2018;8(1):15286. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Material 1. (844.7KB, docx)
Supplementary Material 2. (886.3KB, docx)
Supplementary Material 3. (135KB, xls)
Supplementary Material 4. (355.5KB, xls)
Supplementary Material 5. (775.7KB, zip)

Data Availability Statement

The data used in this study were obtained from public repositories, specifically The Cancer Genome Atlas (TCGA) and the Gene Expression Omnibus (GEO) databases. No new data were created or analyzed during this study.

Articles from Journal of Ovarian Research are provided here courtesy of BMC

RESOURCES