Introduction

Glioma, as the most aggressive malignant tumor of the central nervous system, accounts for approximately 80% of adult primary brain tumors1. Its clinical heterogeneity and treatment resistance have always been important challenges in the field of neuro-oncology2,3,4. Although the current standard treatment regimens combine maximum surgical resection, temozolomide chemotherapy and radiotherapy5,6,7, the high recurrence rate and drug resistance of glioma still exist, and the prognosis of patients is poor8,9,10. Studies have shown that the heterogeneity of the tumor microenvironment, the limitation of the blood-brain barrier, and abnormal epigenetic regulation jointly constitute the three core obstacles in the treatment of glioma11, resulting in a five-year survival rate of less than 5% for patients12,13. In addition to conventional clinical parameters, molecular markers such as isocitrate dehydrogenase (IDH) mutation and 1p/19q codeletion have greatly improved the classification and risk stratification of gliomas and have even led to the development of targeted therapies, including IDH inhibitors, in lower-grade gliomas14. Recent reviews have further emphasized that the complex tumor microenvironment (TME) of glioblastoma profoundly shapes treatment resistance and represents a critical target for emerging strategies, including gene therapy and combination regimens15. Moreover, bioinformatics-driven analyses have identified novel biomarkers such as AEBP1, which is overexpressed in glioblastoma and associated with poor survival, further underscoring the importance of discovering new molecular targets in glioma16.

Recent research has highlighted the potential regulatory role of the Transcription Termination Factors (TTFs) family in tumorigenesis and disease progression. For example, TTF1 (RNA Polymerase I - Specific Transcription Termination Factor 1) drives ribosomal biosynthesis by regulating ribosomal RNA (rRNA) transcription termination, thereby promoting the abnormal proliferation of hepatocellular carcinoma. Clinical data analysis has revealed that high TTF1 expression is significantly associated with a shortened overall survival in liver cancer patients (hazard ratio HR = 1.89, p < 0.001)17. Similarly, TTF1 is markedly overexpressed in tumors such as thyroid cancer and lung cancer, with its expression levels correlating with adverse prognoses18,19. However, TTF2 is another member of this family, an ATP-dependent DNA translocase belonging to the SWI2/SNF2 superfamily, and is related to mitosis. The high expression of TTF2 may promote cell proliferation20,21, and it has received relatively less attention in the field of oncology. The expression pattern and functional mechanism of TTF2 in glioma remain unclear.

This study sets out to investigate the potential role of TTF2 in glioma and its association with patient prognosis. Using the TCGA database and the CGGA dataset, we aim to evaluate the expression levels of TTF2 and its prognostic value in glioma. Through univariate and multivariate Cox analyses, we strive to identify independent prognostic factors and construct clinical prediction models. Additionally, bioinformatics analyses, including GO/KEGG/GSEA enrichment analyses, will be employed to elucidate the biological functions of TTF2 in glioma pathogenesis. Finally, PCR validation will be conducted to assess TTF2 mRNA expression in glioma. Our goal is to clarify the role of TTF2 in glioma progression and to explore its potential as a novel biomarker for early diagnosis and a promising therapeutic target for glioma treatment.

Materials and methods

Data sets

Patient clinical annotations and gene expression data were used in this study from public databases. TCGA gliomas data set (https://www.cancer.gov/about-nci/organization/ccg/research/structural-genomics/tcga), including genetic and phenotypic data, WHO Classification, IDH mutation status, 1P/19q codeletion were missing from Ceccarelli, from the UCSC XENA (https://xenabrowser.net/datapages/) by the Toil process unified handling TCGA GTEx and TPM RNAseq data format. Glioma from TCGA and corresponding normal tissue data from GTEx were extracted. Glioma genome from China (CGGA, http://www.cgga.org.cn/) from the other patients with gliomas, the download mRNA sequencing data (RSEM) and clinical data22,23,24.Five normal tissue samples and ten glioma samples were obtained from the Department of Neurosurgery of Wuxi Clinical College of Anhui Medical University and the 904th Hospital of the Joint Logistics Support Force of the People’s Liberation Army (located in Wuxi, China).

Analysis of the TTF2 expression level between cancer tissue and corresponding normal tissue

Gene expression profiles were obtained from the UCSC Xena platform (https://xenabrowser.net/), which integrates TCGA and GTEx datasets containing both tumor and matched normal samples. RNA-Seq data (TPM) from the UCSC Xena project were reprocessed using a standardized analytical pipeline to minimize disparities between tumor and normal datasets. For external validation, additional glioma and normal brain tissue data from the CGGA database were analyzed. The differential expression of TTF2 between groups was evaluated using the R package ggplot2.

Survival analysis of TTF2 in glioma

Kaplan–Meier survival estimation and Cox proportional hazards regression were applied to determine the prognostic significance of TTF2 expression in glioma. All analyses were conducted in R, primarily using the survival package for modeling and the survminer package for visualization. Data from both TCGA and CGGA cohorts were incorporated into the analysis.

Analysis of differentially expressed genes (DEGs) between the high and low TTF2 expression groups in patients with Gliomas

Differential expression analysis was conducted to compare glioma samples with high versus low TTF2 expression levels. Gene expression data (HTSeq-FPKM) were analyzed using the limma package in R, applying an unpaired Student’s t-test to detect expression differences between groups. Genes meeting the criteria of |log₂Fold Change| > 2 and adjusted P < 0.05 were regarded as differentially expressed and retained for subsequent analyses.

Gene ontology (GO) and kyoto encyclopedia of genes and genomes (KEGG) enrichment analysis

Functional enrichment analyses were conducted using the clusterProfiler package in R, encompassing Gene Ontology (GO) categories—biological process (BP), cellular component (CC), and molecular function (MF)—as well as Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways21,22,23. GO terms or KEGG pathways with an adjusted P value below 0.05 were considered significantly enriched.

Gene set enrichment analysis (GSEA)

Gene Set Enrichment Analysis (GSEA) was conducted to identify pathways and biological functions significantly associated with TTF2 expression levels. The analysis was implemented using the clusterProfiler package in R, comparing high and low TTF2 expression groups. Each gene set was permuted 1,000 times to assess enrichment significance, with the mRNA expression level of TTF2 serving as the phenotypic label. The h.all.v7.0.symbols.gmt file from the MSigDB Hallmark collection was used as the reference gene set. Pathways meeting the thresholds of P < 0.05, false discovery rate (FDR) < 0.25, and |normalized enrichment score |(NES)| > 1 were considered significantly enriched.

Analysis of the connection of TTF2 expression level and immune infiltrates

The relative infiltration levels of various immune cell types were estimated using the single-sample Gene Set Enrichment Analysis (ssGSEA) algorithm implemented in the GSVA package in R. Immune cell scores were calculated based on the expression of predefined marker genes from published immune cell signature lists. To assess the association between immune infiltration and TT.

Prognostic model generation and prediction

Univariate and multivariate Cox proportional hazards regression analyses were conducted to assess factors associated with overall survival, with P < 0.05 regarded as statistically significant. Clinical variables, including WHO grade, 1p/19q codeletion, IDH mutation status, and age, were incorporated into a nomogram model to predict the probabilities of 1-, 3-, and 5-year overall survival (OS).

RNA isolation and qRT-PCR

Total RNA was isolated from glioma tissue specimens and matched normal brain samples from a cohort of 15 patients, including 5 normal brain tissue samples and 10 glioma tissue samples. The clinical characteristics of these patients, including age, gender, and glioma subtype (if applicable), are detailed in Supplementary Table. cDNA synthesis and quantitative real-time PCR were performed as described previously. Total RNA was isolated from glioma tissues and normal brain samples using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) in accordance with the manufacturer’s protocol. Complementary DNA (cDNA) was synthesized from 1 µg of total RNA with the PrimeScript™ RT Reagent Kit (Takara, Japan). Quantitative real-time PCR was then performed using the Eraser™ qPCR Kit (Takara, Dalian, China) to determine TTF2 mRNA expression levels. The primer sequences were as follows: TTF2-F 5’-GCCAGTGTTGCTGTCATCTT-3’ and TTF2-R 5’-GCTCTGAGTCACGGAGTTCT-3’; GAPDH-F 5’-GGTGTGAACCATGAGAAGTATGA-3’ and GAPDH-R 5’-GAGTCCTTCCACGATACCAAAG-3’. GAPDH served as the internal reference for normalization of target gene expression across samples.

Statistical analysis

Associations between clinicopathological characteristics and TTF2 expression were evaluated using the Kruskal–Wallis test, Wilcoxon rank-sum test, and Chi-square test. Kaplan–Meier survival curves were constructed to visualize overall survival, and differences between groups were assessed by the log-rank test. Univariate and multivariate Cox proportional hazards regression analyses were applied to estimate mortality risk. Statistical significance was defined as a two-tailed P value < 0.05. All analyses were performed in R v4.2.1. Package versions: survival (v3.5–7), survminer (v0.4.9), limma (v3.52.2), clusterProfiler (v4.4.4), GSVA (v1.46.0). GSEA was conducted with 1,000 permutations using hallmark gene sets from MSigDB v7.0. Significant pathways were defined by an FDR < 0.25, a normalized enrichment score (NES) > 1, and a P value < 0.05. For GO and KEGG enrichment, the adjusted P value threshold was set to < 0.05 to identify significantly enriched biological processes, molecular functions, and cellular components.

Results

TTF2 overexpression in glioma and its link to poor prognosis

Compared to normal tissue, TTF2 for almost all tumor types in the TCGA database mRNA expression was significantly overexpressed in invasive glioma (GBM and LGG), breast carcinoma (BRCA), colonic adenocarcinoma (COAD), cholangiocarcinoma (CHOL), esophageal carcinoma (ESCA), head and neck squamous cell carcinoma (HNSC), and renal Chromophobe (KICH), hepatocellular carcinoma (LIHC), Lung adenocarcinoma (LUAD), Lung squamous cell carcinoma (LUSC), pheochromocytoma, paraganglioma (PCPG), Rectal adenocarcinoma (READ), Gastric adenocarcinoma (STAD), Thymic carcinoma, prostate carcinoma (PRAD)(Fig. 1A). To further investigate the expression pattern of TTF2 specifically in glioma, UCSC Xena and CGGA datasets were analyzed. Both datasets consistently showed that TTF2 expression in glioma tissues was significantly higher than in normal brain tissues (Fig. 1B–C).Kaplan-meier survival analysis showed that the high expression of TTF2 was significantly associated with a poor prognosis (Fig. 1D). The prognosis of glioma patients was analyzed using the expression level of TTF2mRNA, and the AUC value was 0.7–0.8 (Fig. 1E).

Fig. 1
figure 1

TTF2 mRNA in Glioma and other types of human cancers from TCGA and CGGA data: (A) The expression level of TTF2 in different tumor types in the TCGA database. (B) UCSC Xena analyzed the expression levels of TTF2 in Glioma and normal tissues. The expression of TTF2 in glioma was significantly higher than the normal level. (C) The CCGA database shows the expression levels of TTF2 in Glioma and normal tissues. (D)TCGA data analysis showed that high expression of TTF2 was associated with poor prognosis. (E) Relationship between the expression level of TTF2 and the AUC curve.

Full size image

High TTF2 expression linked to poor prognosis and a predictive model for glioma survival

The total sample size was 698 cases, among which 401 were male and 297 were female. According to the median expression level of TTF2 in low-grade gliomas, the total samples were divided into the low-expression group and the high-expression group. The detailed clinicopathological features are shown in Table 1. Univariate and multivariate Cox analyses of clinical data showed that: WHO classification (HR = 9.538, 95%CI:7.243–12.560, P < 0.001), 1p/19q coding (HR = 0.225, 95%CI:0.147–0.346), IDH status (HR = 0.116, 95%CI: 0.089–0.151, P < 0.001), Age(HR = 4.696, 95%CI: 3.620–6.093, P < 0.001), TTF2 (HR = 4.645, 95%CI:3.494–6.177, P < 0.001), multivariate analysis showed: WHO classification (HR = 2.571, 95%CI:1.795–3.682, P < 0.001), IDH status (HR = 0.266, 95%CI: 0.179–0.396, P < 0.001), Age(HR = 1.489, 95%CI: 1.090–2.035, P < 0.001) and TTF2 (HR = 1.608, 95%CI:1.113–2.323, P = 0.011) were independent prognostic factors(Table 2) (Fig. 2A-B). We verified this result by fitting TTF2 mRNA expression and other clinicopathological parameters, and established an OS prediction model in TCGA data, including TTF2 and other independent prognostic factors, such as WHO grade, IDH mutation status and age (Fig. 2C). The higher the point on the chart is, the worse the indicative factor is. The performance of the model diagram is evaluated using the calibration curve (Fig. 2D). The accuracy of the model was further verified using the CGGA database (Supplementary Figure S1). Interestingly, using TCGA data analysis, we found that the expression of TTF2 was significantly increased in high-grade gliomas (Fig. 3A). In the analyses of PD (Progressive disease), SD (StableDisease), PR (PartialResponse), and CR (CompleteResponse), it was found that the expression of TTF2 was inversely proportional to the treatment correlation (Fig. 3B). The relationship between progression-free survival (PFI), disease-specific survival (DSS) and the expression of TTF2 also proves this point (Fig. 3C-D).

Table 1 Basic clinical baseline table of glioma.
Full size table
Table 2 Univariate and multivariate cox regression analysis.
Full size table
Fig. 2
figure 2

Univariate and multivariate cox regression analysis of glioma and construction of TTF2 prognostic model: (A-B) Univariate and multivariate Cox regression analysis was conducted using the TCGA database to explore the independent risk factors in glioma. WHO grade, IDH status, Age, and TTF2 were the independent risk factors related to prognosis. (C) nomogram integrating TTF2 and other prognostic factors from TCGA data. (D) Calibration curves of the prediction models for 1/3/5 years.

Full size image
Fig. 3
figure 3

High TTF2 expression is inversely proportional to the therapeutic effect: (A) The expression of TTF2 significantly increases in high-grade gliomas. (B) In PD (progressive disease), SD (stable disease), PR (partial response), and CR (complete response), the expression of TTF2 is inversely proportional to the therapeutic correlation. (C-D) The relationship between progression-free survival (PFI), disease-specific survival (DSS) and the expression of TTF2 also proves this point.

Full size image

Functional enrichment analysis of samples with high and low TTF2 expression

To explore the potential mechanism by which TTF2 promotes tumor progression, we analyzed samples with high and low expression of TTF2 and subsequently presented genes co-expressed with TTF2, including up-regulated genes and down-regulated genes (Fig. 4A-B). The correlations of co-expressed genes were demonstrated, with red representing positive correlations and blue representing negative correlations (Fig. 4C). Subsequently, GO enrichment analysis was used to predict the co-expression function of glioma patients. The Top go bioenrichment program (BP), molecular function (MF), and cellular component (CC) groups, including immunoglobulin complexes, signal receptor activation, receptory-ligand activity, cell recognition, etc. (Fig. 4D), KEGG analysis revealed that TTF2 may be involved in a variety of pathways including cell adhesion, the PI3K-AKT signaling pathway, the AGE-RAGE signaling pathway, etc. (Fig. 4E) (Table 3), and the key pathways related to TTF2 were determined through GSEA analysis. GSEA analysis revealed that the data set satisfied FDR < 0.25, P < 0.05. Enrichment analysis and GSEA analysis showed that the expression of TTF2 was related to the production of immunoglobulins, adaptive immune responses, immune regulation, and the transmission of immune cell signaling factors, etc. (Fig. 4F-I).

Fig. 4
figure 4

Functional enrichment of TTF2 in Glioma: (A-B) The samples were divided into high-expression and low-expression groups using the median of TTF2 to explore the co-expressed genes of TTF2. The heat maps respectively showed the top 20 genes positively correlated with TTF2 expression and the top 20 genes negatively correlated with TTF2 expression. (C) The correlation matrix diagram shows the genes associated with TTF2. (D) GO enrichment analysis diagram. (E) KEGG analysis revealed that TTF2 might be involved in multiple pathways such as cell adhesion, the PI3K-AKT signaling pathway, and the AGE-RAGE signaling pathway. (F-I)GSEA analysis showed a correlation with immune responses, such as the production of immunoglobulins, adaptive immune responses, immune regulation, and the transmission of immune cell signaling factors.

Full size image
Table 3 GO and KEGG enrichment analysis.
Full size table

The expression of TTF2 in glioma is related to the level of immune infiltration

Considering that both KEGG and GSEA enrichment analyses found that TTF2 might be involved in the tumor immune response, we further used ssGSEA to analyze the relationship between TTF2 mRNA expression and the infiltration level of immune cells. The correlation between immune cell infiltration and TTF2 mRNA expression (Fig. 5A). The results showed that the expression level of TTF2 mRNA was higher than that of Th2 Cells (R = 0.595, P < 0.001) and macrophages (R = 0.509, P < 0.001;) (Fig. 5B-C), neutrophils (R = 0.422, P < 0.001; (Fig. 5D) shows a positive correlation. Additionally, ssGSEA also indicates that the expression of TTF2 is negatively correlated with pDC (R=−0.455, P < 0.001) (Fig. 5E).

Fig. 5
figure 5

TTF2 immune correlation analysis: (A) ssGSEA was used to analyze the relationship between TTF2 mRNA expression and the infiltration level of immune cells. (B) The expression of TTF2 was positively correlated with that of Th2 cells, p < 0.001. (C) The expression of TTF2 was positively correlated with that of macrophages. (D) The expression of TTF2 was positively correlated with that of neutrophils, p < 0.001. (E) The expression of TTF2 was negatively correlated with that of pDC cells, p < 0.001.

Full size image

To verify the clinical characteristics and expression levels of TTF2 in glioma

The results showed that after dividing the samples into two groups with high and low expression of TTF2, the expression of TTF2 was inversely proportional to 1/19q (Fig. 6A). The expression of TTF2 in mutant IDH was also lower than that in wild-type IDH (Fig. 6B). However, in terms of gender (Fig. 6C), there was no significant difference in the expression of TTF2. In those over 60 years old (Fig. 6D), the expression of TTF2 was significantly increased. The expression of TTF2 was verified using brain tissue and glioma, and it was found that the expression in tumor tissue was significantly higher than that in normal tissue (P < 0.001;) (Fig. 6E).

Fig. 6
figure 6

Correlation analysis of TTF2 expression with clinical characteristics and verification of tissue specimens: (A) In 1p/19q codeletion, the expression of TTF2 was higher in the Non-codel group. (B) In IDH status, TTF2 is expressed higher in WT. (C) In terms of gender, there was no significant difference in the expression of TTF2. \ n (D) In terms of age, the expression of TTF2 was significantly increased in those over 60 years old. (E) The expression of TTF2 was verified using glioma and normal tissue samples. The expression of TTF2 was significantly increased in glioma tumor samples.

Full size image

Discussion

In this study, we comprehensively characterized the expression pattern, clinical relevance, and potential biological functions of TTF2 in glioma by integrating large-scale transcriptomic data from TCGA and CGGA with qRT-PCR validation in clinical samples. We demonstrated that TTF2 is markedly overexpressed in glioma tissues compared with normal brain tissues, and elevated TTF2 mRNA expression is consistently associated with unfavorable clinical outcomes. Importantly, TTF2 remained an independent prognostic factor for overall survival (OS) after adjustment for established clinicopathological variables, suggesting that it may play a functional rather than incidental role in glioma progression.

TTF2 expression was closely linked to several key molecular and clinical features, including advanced WHO grade, IDH wild-type status, non–1p/19q codeletion, and older age—all recognized markers of aggressive glioma biology. Incorporating TTF2 into a prognostic model alongside WHO grade, IDH status, and age significantly improved risk stratification, and the resulting nomogram displayed good calibration for predicting 1-, 3-, and 5-year OS. Analyses of multiple clinical endpoints further revealed that TTF2 expression was higher in patients experiencing progression-free interval (PFI) and disease-specific survival (DSS) events. Although statistical significance varied across OS, PFI, and DSS, the overall pattern consistently indicated that elevated TTF2 expression is associated with faster disease progression and increased disease-specific mortality. This result is consistent with previous studies, which also found that the TTF2 protein is highly expressed in various types of cancer, including papillary thyroid carcinoma, colorectal adenocarcinoma and breast cancer, etc25,26,27,28,29.

Given the enrichment of immune-related pathways, we further explored the association between TTF2 expression and immune cell infiltration using ssGSEA. High TTF2 expression was positively correlated with Th2 cells, macrophages, and neutrophils, and negatively correlated with plasmacytoid dendritic cells (pDCs). A Th2-polarized immune response, together with increased tumor-associated macrophages and neutrophils, is generally associated with an immunosuppressive microenvironment and poorer prognosis in many cancers. The observed immune infiltration pattern in TTF2-high gliomas is therefore consistent with their more aggressive clinical behavior. These data raise the possibility that TTF2 may facilitate immune evasion, at least in part, by promoting a Th2-dominant, immunosuppressive milieu. Although the present findings are based on computational inference, they provide a rationale for further investigation of TTF2 as a potential immunomodulatory target in glioma.

From a biological standpoint, TTF2 is the core regulatory factor of the transcription termination process mediated by RNA polymerase II. Its domain promotes chromatin unwinding and the dissociation of RNA-DNA heterozygotes by hydrolyzing ATP30. Some studies have shown that the protein level of the transcription termination-related factor TTF2 is regulated by APC/C-mediated ubiquitin-proteasome in a cyclic-dependent manner30. Knockout of TTF2 activates the spindle assembly checkpoint (SAC), leading to chromosomal separation errors and cytoplasmic division failure. This may be a possible mechanism of TTF2 in tumors21. Our enrichment analysis and GSEA found that TTF2 may be involved in some immune pathways. ssGSEA also showed that TTF2 was positively correlated with the infiltration of Th2 cells, macrophages and neutrophils, but negatively correlated with the infiltration of pDC cells. It has been found in previous studies that Th2 cell infiltration is associated with Th2 cell immunosuppression and poor survival in various tumors. We hypothesize that the TTF2-dependent transcriptional program may regulate the expression of cytokines such as IL-4 and IL-13, as well as the expression of chemokines including CCL17 and CCL22, which have been proven to promote Th2 configuration and migration. Secondly, by regulating the transcriptional termination of genes related to immunity, TTF2 may affect antigen presentation and co-stimulatory signal transduction in glioma cells or tumor-associated myeloid cells, thereby indirectly skewing the T cell response towards Th2 phenotype. In this study, we found a significant increase in Th2 cells, suggesting that TTF2 may be involved in glioma-mediated immune escape. A similar situation also occurs on the tumor-associated antigen EpCAM, which promotes Th2 cell-mediated immune escape31,32,33. Therefore, TTF2 has great value as a possible immunotherapy target in the future.

This study has several limitations. First, our analyses were based on retrospective public cohorts from TCGA and CGGA. Despite rigorous data processing, treatment-related variables such as radiotherapy, chemotherapy, and temozolomide administration are known to strongly influence glioma outcomes, these data are largely missing or poorly annotated in TCGA and CGGA. Including such incomplete information in multivariable Cox models would introduce considerable bias; The prognostic model we constructed was subjected to internal validation, but was not tested in an independent external clinical cohort. This might limit its general applicability. We will continue to collect data in future studies and conduct validation in a refined independent cohort to confirm and refine our findings. Additionally, the immune infiltration map was inferred through a computational deconvolution method and has not yet been validated through functional experiments (such as TTF2 knockdown or overexpression followed by co-culture with immune cells or validation in animal models). We will further investigate this in future studies.

In summary, our work identifies TTF2 as a gene markedly overexpressed in glioma and strongly associated with adverse clinical features and poor outcomes. High TTF2 expression correlates with shorter OS and is enriched in patients with PFI and DSS events, supporting its role as a marker of aggressive disease. Functional enrichment and immune infiltration analyses suggest that TTF2 may promote tumor progression through coordinated regulation of oncogenic signaling pathways and the immune microenvironment. Although further mechanistic and experimental studies are required, TTF2 holds promise as a diagnostic and prognostic biomarker and as a potential immunotherapeutic target in glioma.

Conclusion

TTF2 mRNA is overexpressed in glioma, and high TTF2 mRNA expression is OS-related. TTF2 is a potential biomarker for the diagnosis and prognosis of glioma and may be a potential target for immunotherapy.