Integrin-binding sialoprotein as an extracellular matrix–associated independent prognostic biomarker in glioma Authors: Wei-wei Chen, Yu-xing Hao, Wen-qing Hong, Qi-you Yuan, Hao-dong Jin, Jin Peng, Ming Shan, Cheng-cong Li Published: 2026 ABSTRACT BackgroundIntegrin-binding sialoprotein (IBSP) has been implicated in bone metastasis in breast, lung, and prostate cancers. However, its role in glioma remains insufficiently characterized. The aim of this study was to assess the expression profile and prognostic significance of IBSP in patients with glioma.MethodsGene expression data associated with glioma prognosis were obtained from the Chinese Glioma Genome Atlas and Gene Expression Omnibus Series databases and analyzed in relation to clinical characteristics. The prognostic value of IBSP was assessed using univariate and multivariate Cox regression analyses. Immunohistochemical analysis was performed to validate these findings. Gene set enrichment analysis (GSEA) was conducted to identify molecular pathways associated with IBSP expression and its contribution to adverse prognosis. Additionally, gene set cancer analysis was used to explore potential therapeutic agents targeting IBSP.ResultsIBSP expression was elevated in glioma tissues and demonstrated a positive correlation with increasing World Health Organization grade. Multivariate Cox regression analysis identified IBSP as an independent prognostic factor across multiple datasets (HR = 1.531; 95% CI: 1.07–2.192; p = 0.020). GSEA indicated that IBSP is involved in apoptosis, coagulation, and complement pathways. Several genes, including CASP4, KYNU, MSR1, PLAUR, SIGLEC9, and SLC16A3, were positively correlated with IBSP expression and are implicated in glioma pathogenesis. Furthermore, alvocidib and dinaciclib were identified as potential therapeutic agents targeting IBSP.ConclusionsIBSP represents an independent prognostic biomarker associated with adverse outcomes in glioma. These findings provide insight into the molecular mechanisms underlying IBSP-associated glioma progression and highlight potential therapeutic targets that may contribute to improved clinical outcomes in patients with glioma. Introduction Glioma is a type of tumor arising from glial cells and represents the most common primary intracranial neoplasm [1]. Due to its locally aggressive behavior and limited responsiveness to current therapeutic strategies, glioma is classified as a highly malignant condition [1]. Along with surgical intervention, radiotherapy and chemotherapy have been incorporated into clinical management since 2005 [2]. Among these tumors, glioblastoma (GBM), which accounts for the highest incidence of malignant tumors of the central nervous system, is associated with a median survival of approximately 8 months [3]. The fifth edition of the World Health Organization Classification of Tumors of the Central Nervous System (WHO CNS5), published in 2021, highlights the critical role of molecular diagnostics in tumor classification, representing a significant advancement compared with the fourth edition released five years earlier. Although the diagnosis of glioma continues to rely on conventional histopathological approaches, including histomorphology and immunohistochemistry, molecular data are increasingly used as reliable biomarkers for tumor classification, grading, and prognostic assessment [4, 5]. Isocitrate dehydrogenase (IDH) is implicated in multiple tumor types, including but not limited to glioma. Variants in IDH are generally regarded as a favorable prognostic factor [6]. The latest WHO CNS5 recommends that IDH variant status be incorporated into the classification and grading of gliomas [7]. The combined deletion of chromosome arms 1p and 19q refers to the concurrent loss of the short arm of chromosome 1 and the long arm of chromosome 19, resulting in altered gene expression associated with increased chemosensitivity and improved prognosis [8]. These molecular alterations are well-established biomarkers that contribute to the therapeutic stratification and prognostic evaluation of patients with glioma. Integrin-binding sialoprotein (IBSP) is produced by bone cells and plays a key role in bone structure and remodeling [9]. IBSP belongs to the family of small integrin-binding ligand N-linked glycoproteins (SIBLING) [10]. As soluble proteins, members of the SIBLING family participate in various physiological processes, including cell proliferation, apoptosis, angiogenesis, wound healing, and regulation of extracellular matrix remodeling, and are frequently associated with inflammation, infectious diseases, and tumorigenesis [11, 12]. Members of the SIBLING family have also been reported as molecular biomarkers in multiple pathological conditions. In advanced breast cancer, IBSP has been demonstrated to promote bone metastasis in estrogen receptor–positive tumor cells, potentially through the recruitment of osteoclasts and the establishment of a microenvironment conducive to bone colonization [13]. In prostate cancer, IBSP has similarly been implicated in bone metastasis, with its expression reported to be positively regulated by HOXB13/HOXA11-AS axis [14]. IBSP has been identified as a significant biomarker associated with bone metastasis in several malignancies, including breast, prostate, lung, and thyroid cancers [15]. IBSP has been implicated in traumatic brain injury (TBI). Upregulation of microRNA-92 A-3P following TBI has been demonstrated to reduce IBSP expression. IBSP is involved in bone growth through the phosphatidylinositol 3-kinase/threonine kinase 1 (PI3K/AKT) signaling pathway, and this regulatory mechanism may contribute to the accelerated healing observed in fractures associated with TBI compared with isolated fractures [16]. Although IBSP has been implicated in bone metastasis across multiple tumor types, its role in glioma remains insufficiently characterized. A recent bioinformatics study identified IBSP as a characteristic gene in GBM [17]. In this study, the biological function of IBSP and its prognostic significance in glioma were systematically explored. The findings indicate that IBSP plays a key role in glioma tumorigenesis. Materials and methods Patients and data collection Gene expression data from human glioma samples were obtained from the CGGA datasets (http://www.cgga.org.cn) and the GSE16011 dataset (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=gse16011). The CGGA datasets included CGGA Microarray, CGGA RNA-Seq-1, and CGGA RNA-Seq-2. Clinical and prognostic data from 1,497 patients with glioma were also retrieved from these datasets and are presented in Table 1. Patients with an overall survival (OS) of less than one month or with incomplete prognostic data were excluded from the analysis. Prior to clustering, single-cell data were pre-processed, including quality control, normalization, and feature selection. The Seurat toolkit was used as the primary analytical framework. Data processing steps were performed, including normalization, identification of highly variable genes, scaling, and principal component analysis, which enabled dimensionality reduction for subsequent clustering analyses. Table 1Multiple Cox regression analysis of IBSP and clinical parametersParametersHR95% CIP valueCGGA Microarray Age, years1.013(0.999–1.027)0.072 Gender, Male vs. Female1.002(0.724–1.387)0.992 Grade, III vs. II3.307(1.986–5.507)< 0.001* Grade, IV vs. II5.853(3.581–9.567)< 0.001* Radiotherapy, Yes vs. No0.498(0.327–0.756)0.001* Chemotherapy, Yes vs. No0.681(0.49–0.946)0.022* IDH, Mutant vs. Wild0.871(0.602–1.262)0.466 IBSP, High vs. Low1.531(1.07–2.192)0.020*CGGA RNA-Seq-1 Age, years1.013(1–1.027)0.054 Gender, Male vs. Female0.989(0.734–1.331)0.94 Grade, III vs. II3.59(2.244–5.745)< 0.001* Grade, IV vs. II6.05(3.693–9.911)< 0.001* Radiotherapy, Yes vs. No0.749(0.525–1.067)0.109 Chemotherapy, Yes vs. No0.713(0.512–0.992)0.045* IDH, Mutant vs. Wild1.038(0.728–1.479)0.838 1p19q codeletion, Yes vs. No0.326(0.189–0.562)< 0.001* IBSP, High vs. Low1.527(1.09–2.138)0.014*CGGA RNA-Seq-2 Age, years1.011(1.002–1.02)0.017* Gender, Male vs. Female1.09(0.864–1.376)0.467 Grade, III vs. II3.456(2.37–5.039)< 0.001* Grade, IV vs. II5.274(3.464–8.03)< 0.001* Radiotherapy, Yes vs. No0.747(0.545–1.022)0.068 Chemotherapy, Yes vs. No0.679(0.501–0.92)0.013* IDH, Mutant vs. Wild1.533(1.15–2.044)0.004* 1p19q codeletion, Yes vs. No2.195(1.498–3.217)< 0.001* IBSP, High vs. Low1.632(1.252–2.128)< 0.001*GSE11601 Age, years1.033(1.02–1.047)< 0.001* Gender, Male vs. Female1.002(0.727–1.38)0.991 Grade, III vs. II1.342(0.634–2.839)0.442 Grade, IV vs. II3.475(1.655–7.295)0.001* Chemotherapy, Yes vs. No0.714(0.443–1.15)0.166 IBSP, High vs. Low1.84(1.334–2.538)< 0.001**: P < 0.001 Table 1Multiple Cox regression analysis of IBSP and clinical parametersParametersHR95% CIP valueCGGA Microarray Age, years1.013(0.999–1.027)0.072 Gender, Male vs. Female1.002(0.724–1.387)0.992 Grade, III vs. II3.307(1.986–5.507)< 0.001* Grade, IV vs. II5.853(3.581–9.567)< 0.001* Radiotherapy, Yes vs. No0.498(0.327–0.756)0.001* Chemotherapy, Yes vs. No0.681(0.49–0.946)0.022* IDH, Mutant vs. Wild0.871(0.602–1.262)0.466 IBSP, High vs. Low1.531(1.07–2.192)0.020*CGGA RNA-Seq-1 Age, years1.013(1–1.027)0.054 Gender, Male vs. Female0.989(0.734–1.331)0.94 Grade, III vs. II3.59(2.244–5.745)< 0.001* Grade, IV vs. II6.05(3.693–9.911)< 0.001* Radiotherapy, Yes vs. No0.749(0.525–1.067)0.109 Chemotherapy, Yes vs. No0.713(0.512–0.992)0.045* IDH, Mutant vs. Wild1.038(0.728–1.479)0.838 1p19q codeletion, Yes vs. No0.326(0.189–0.562)< 0.001* IBSP, High vs. Low1.527(1.09–2.138)0.014*CGGA RNA-Seq-2 Age, years1.011(1.002–1.02)0.017* Gender, Male vs. Female1.09(0.864–1.376)0.467 Grade, III vs. II3.456(2.37–5.039)< 0.001* Grade, IV vs. II5.274(3.464–8.03)< 0.001* Radiotherapy, Yes vs. No0.747(0.545–1.022)0.068 Chemotherapy, Yes vs. No0.679(0.501–0.92)0.013* IDH, Mutant vs. Wild1.533(1.15–2.044)0.004* 1p19q codeletion, Yes vs. No2.195(1.498–3.217)< 0.001* IBSP, High vs. Low1.632(1.252–2.128)< 0.001*GSE11601 Age, years1.033(1.02–1.047)< 0.001* Gender, Male vs. Female1.002(0.727–1.38)0.991 Grade, III vs. II1.342(0.634–2.839)0.442 Grade, IV vs. II3.475(1.655–7.295)0.001* Chemotherapy, Yes vs. No0.714(0.443–1.15)0.166 IBSP, High vs. Low1.84(1.334–2.538)< 0.001**: P < 0.001 Multiple Cox regression analysis of IBSP and clinical parameters *: P < 0.001 Detection of biomarkers of gliomas In this study, clinical data and differential expression analyses of the GSE16011 and CGGA datasets were used to identify differentially expressed long non-coding RNAs and mRNAs associated with glioma. Univariate Cox regression analysis was conducted using R packages, including survival, glmnet, survminer, and survivalROC. Genes associated with OS in each dataset were selected based on the criteria (HR > 2, p < 0.001). The UpSetR package in R was used to visualize the intersections among gene sets. Gene set enrichment analysis Gene set enrichment analysis (GSEA) was conducted using the R package clusterProfiler (version 4.0) to perform Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses, enabling the comparison of biological themes across gene sets. GO analysis was used to categorize genes into biological processes, molecular functions, and cellular components within a directed acyclic graph framework, while KEGG analysis was used to annotate genes to signaling pathways. The cnetplot function was used to visualize gene–pathway interaction networks. KEGG and GO datasets were obtained from the GSEA database (https://www.gsea-msigdb.org/gsea/register.jsp). Enrichment analyses for GO, KEGG, and HALLMARK gene sets were performed using the R packages clusterProfiler and msigdbr. p < 0.05 was considered statistically significant for pathway enrichment. Derivation and validation of a prognostic model Patients with glioma were stratified into high and low IBSP expression groups based on IBSP expression levels. Kaplan–Meier survival analysis and the log-rank test were used to evaluate differences in survival outcomes between groups. IBSP transcriptional expression in glioma tissues and adjacent normal brain tissues was analyzed, and differences were assessed using Student’s t-test. Cox regression analysis was performed to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for OS. The predictive performance of IBSP for OS was evaluated using receiver operating characteristic (ROC) analysis with the R package pROC [14]. Meta-analysis of HRs across datasets was conducted using the metafor package to assess the overall prognostic value of IBSP. Associations between IBSP expression and clinical characteristics in patients with glioma were analyzed using the Wilcoxon or Kruskal–Wallis tests. Differences among multiple groups were further assessed using the Kruskal–Wallis test and one-way analysis of variance. Immunohistochemistry and evaluation In 2017, tumor tissue samples were obtained from the Department of Neurosurgery at the First Affiliated Hospital of Anhui Medical University and histologically classified as grade II, III, or IV glioma. The study protocol was approved by the Human Ethics Committee of the First Affiliated Hospital of Anhui Medical University, and written informed consent was obtained from all patients. Tissue samples were fixed in 4% paraformaldehyde, followed by dehydration, paraffin embedding, and sectioning. Immunohistochemical staining was performed according to the protocol provided by the manufacturer using anti-IBSP antibodies (BOSTER BA2329, 1:100 dilution). IHC staining results were evaluated using ImageScope software (Leica). For each section, three representative high-power fields (400× magnification) were captured and analyzed. Staining intensity was categorized as follows: dark brown nuclei were considered strongly positive, brown nuclei were considered moderately positive, light yellow nuclei were considered weakly positive, and blue nuclei were considered negative. For each tissue section, quantitative analysis included the measurement of pixel area corresponding to strong, moderate, weak, and negative staining, as well as the proportion of positive cells and the calculation of the H-score [18]. Potential drugs for patients with glioma Gene Set Cancer Analysis (GSCA) is an integrated platform for genomic, pharmacogenomic, and immunogenomic analyses that facilitates the identification of candidate therapeutic agents in cancer research. In this study, the GSCA platform (http://bioinfo.life.hust.edu.cn/GSCA) was used to identify potential therapeutic agents targeting IBSP in glioma. The GSCA database was employed to screen candidate drugs by analyzing the correlation between mRNA expression levels and drug half-maximal inhibitory concentration (IC50) values. Drug sensitivity data were derived from the Genomics of Drug Sensitivity in Cancer and the Cancer Therapeutics Response Portal (CTRP) databases. Associations between gene expression and drug sensitivity were visualized using bubble plots. In these plots, color gradients represented the correlation between mRNA expression and IC50 values, while bubble size was positively associated with the false discovery rate (FDR) significance. Data points outlined in black indicated FDR values < 0.05. Statistical analysis All statistical analyses were conducted using R software (version 4.1.0). The Pearson correlation coefficient was calculated to identify genes co-expressed with IBSP. p ≤ 0.05 was considered statistically significant. Results General data and clinical characteristics of patients with gliomas A total of 1,248 samples were obtained from the CGGA datasets, including 309 from CGGA RNA-seq-1, 656 from CGGA RNA-seq-2, and 283 from CGGA microarray datasets, along with 249 samples from the GSE16011 dataset. These datasets included gene expression profiles and corresponding clinical information for patients with glioma. The clinical variables comprised of age, sex, histological classification, and tumor grade. Data on postoperative treatment modalities, including radiotherapy and chemotherapy, were collected. Molecular characteristics of the tumors were also included in the analysis. Results of glioma prognosis-related gene screens and enrichment of IBSP expressions in relevant signaling pathways As presented in Fig. 1, genes associated with glioma prognosis were analyzed using multiple enrichment approaches. The principal findings are summarized as follows: (1) The results demonstrate the reliability of the prognostic model, which demonstrated consistent performance across multiple datasets and high reproducibility in the external validation cohort. (2) A total of 298 prognosis-associated genes were identified as being statistically significant across different datasets. (3) During the identification of tumor-associated factors, risk factors associated with poor prognosis were clearly distinguished from protective factors associated with favorable outcomes. Fig. 1Identification of prognosis-associated genes in glioma. A–D Volcano plots depicting differential gene expression in the CGGA and GSE16011 datasets. Blue dots indicate significantly upregulated genes, yellow dots indicate significantly downregulated genes, and gray dots represent non-differentially expressed genes. E UpSet plot demonstrating the intersection of prognosis-associated genes across the four datasets. F Circos plot depicting the distribution of HRs of prognostic genes across different cohorts Identification of prognosis-associated genes in glioma. A–D Volcano plots depicting differential gene expression in the CGGA and GSE16011 datasets. Blue dots indicate significantly upregulated genes, yellow dots indicate significantly downregulated genes, and gray dots represent non-differentially expressed genes. E UpSet plot demonstrating the intersection of prognosis-associated genes across the four datasets. F Circos plot depicting the distribution of HRs of prognostic genes across different cohorts To further elucidate the biological role of IBSP in the malignant progression of glioma, gene enrichment analysis was performed. The results indicated that IBSP is involved in the epithelial–mesenchymal transition (EMT) process within the HALLMARK gene set (p < 0.004). EMT is a fundamental cellular program involved in embryogenesis, wound repair, and tumor progression, and has been strongly associated with glioma malignancy [19, 20]. Prior studies have reported that upregulation of IBSP is associated with poor prognosis in patients with esophageal squamous cell carcinoma, indicating that its role in tumor progression may be partially mediated through EMT-related mechanisms [21]. GO and KEGG enrichment analyses have demonstrated that IBSP is involved in extracellular matrix organization, PI3K–Akt signaling, stromal adhesion, and extracellular matrix interactions. Given that the prognostic significance of IBSP in glioma has not been fully established, this study focused on IBSP to further explore its functional role and clinical relevance in glioma. The IBSP expression has predictive value as a clinical feature of glioma To evaluate the association between IBSP expression and clinical as well as molecular characteristics, data from the CGGA microarray, CGGA RNA-seq, and GSE16011 datasets were analyzed. As presented in Fig. 2A, D, G, and J, IBSP expression was significantly associated with WHO grade (p < 0.05), with mean expression levels increasing progressively with higher pathological grades. This trend was particularly evident when comparing high-grade and low-grade gliomas. Molecular biomarkers, including 1p/19q codeletion and IDH variant status, are widely used in clinical practice for prognostic assessment. IBSP expression was significantly higher in the IDH1 wild-type group compared with the IDH1 variant group (p < 0.05; Fig. 2B, E, H). Similarly, IBSP expression was elevated in the 1p/19q non-codeletion group, which is typically associated with a less favorable prognosis, compared to the 1p/19q codeletion group (p < 0.05; Fig. 2C, F, I). These findings indicate that increased IBSP expression is associated with adverse prognostic features in glioma. Analysis of the GSE16011 dataset further demonstrated that elevated IBSP expression was associated with lower Karnofsky Performance Score (KPS) in patients with glioma. Among histological subtypes, GBM, the most aggressive astrocytic tumor, exhibited the highest levels of IBSP expression (Fig. 2K). This pattern was consistently observed across all four datasets (Fig. 3). Consistent with immunohistochemical findings, IBSP expression was significantly higher in high-grade gliomas compared with low-grade gliomas, with the highest expression observed in GBM (Fig. 4). Overall, IBSP expression increased progressively with glioma malignancy, indicating its potential value as a prognostic biomarker associated with survival outcomes in patients with glioma. Fig. 2IBSP expression level and clinical characteristics in glioma across the CGGA RNA-seq, CGGA microarray, and GSE16011 cohorts. Association between IBSP expression and glioma grade: A CGGA microarray cohort; D CGGA RNA-seq-1 cohort; G CGGA RNA-seq-2 cohort; J GSE16011 cohort. Association between IBSP expression and IDH variant status: B CGGA microarray cohort; E CGGA RNA-seq-1 cohort; H CGGA RNA-seq-2 cohort. Association between IBSP expression and 1p/19q codeletion status: C CGGA microarray cohort; F CGGA RNA-seq-1 cohort; I CGGA RNA-seq-2 cohort. K Association between IBSP expression and KPS in the GSE16011 cohort IBSP expression level and clinical characteristics in glioma across the CGGA RNA-seq, CGGA microarray, and GSE16011 cohorts. Association between IBSP expression and glioma grade: A CGGA microarray cohort; D CGGA RNA-seq-1 cohort; G CGGA RNA-seq-2 cohort; J GSE16011 cohort. Association between IBSP expression and IDH variant status: B CGGA microarray cohort; E CGGA RNA-seq-1 cohort; H CGGA RNA-seq-2 cohort. Association between IBSP expression and 1p/19q codeletion status: C CGGA microarray cohort; F CGGA RNA-seq-1 cohort; I CGGA RNA-seq-2 cohort. K Association between IBSP expression and KPS in the GSE16011 cohort Fig. 3Expression level of IBSP in gliomas of various histological subtypes. A CGGA microarray cohort; B CGGA RNA-seq-1 cohort; C CGGA RNA-seq-2 cohort; D GSE16011 cohort Expression level of IBSP in gliomas of various histological subtypes. A CGGA microarray cohort; B CGGA RNA-seq-1 cohort; C CGGA RNA-seq-2 cohort; D GSE16011 cohort Fig. 4IBSP immunohistochemical staining in glioma tissues of different grades. A Representative IBSP staining in WHO grade II glioma specimens; B weakly positive IBSP staining in WHO grade III glioma specimens; C strongly positive IBSP staining in WHO grade IV glioma specimens; D quantitative analysis of IBSP expression across WHO grades. *p < 0.05, ***p < 0.001 IBSP immunohistochemical staining in glioma tissues of different grades. A Representative IBSP staining in WHO grade II glioma specimens; B weakly positive IBSP staining in WHO grade III glioma specimens; C strongly positive IBSP staining in WHO grade IV glioma specimens; D quantitative analysis of IBSP expression across WHO grades. *p < 0.05, ***p < 0.001 A high IBSP level is an independent predictive marker of OS in patients with malignant gliomas To evaluate the prognostic significance of IBSP, Kaplan–Meier survival analysis was performed using data from the CGGA and GSE16011 datasets. Increased IBSP expression was significantly associated with reduced OS across all glioma grades in both datasets (Fig. 5A–C; p < 0.001 for all cohorts). ROC curve analysis was subsequently conducted to assess the predictive performance of IBSP. The area under the curve values for 1-year and 3-year survival exceeded 0.7 in both the CGGA RNA-seq-1 and GSE16011 datasets (Fig. 5B, D), indicating that IBSP demonstrates acceptable prognostic accuracy for short- and intermediate-term survival outcomes in patients with glioma. To account for potential confounding clinical variables, univariate and multivariate Cox regression analyses were conducted. Variables including IBSP expression, WHO grade, and molecular characteristics were assessed as predictors of prognosis (Table 1). Multivariate analysis across all datasets confirmed that IBSP is an independent prognostic factor for glioma (Table 1). A meta-analysis incorporating data from multiple datasets was conducted to validate the prognostic significance of IBSP. As presented in Fig. 5E, the pooled HR was 2.92 (95% CI: 2.57–3.33), with low heterogeneity (I² = 5%, p < 0.01), supporting the robustness of the findings. Overall, elevated IBSP expression is significantly associated with poorer survival outcomes, indicating its potential use as an independent prognostic biomarker in patients with glioma. Fig. 5The correlation between IBSP expression and OS in glioma. A Kaplan–Meier survival analysis based on CGGA RNA-seq data; B Kaplan–Meier survival analysis based on CGGA microarray data; C Kaplan–Meier survival analysis based on GSE16011 data; D ROC curve analysis evaluating the predictive performance of IBSP; E meta-analysis demonstrating the prognostic significance of IBSP across four datasets The correlation between IBSP expression and OS in glioma. A Kaplan–Meier survival analysis based on CGGA RNA-seq data; B Kaplan–Meier survival analysis based on CGGA microarray data; C Kaplan–Meier survival analysis based on GSE16011 data; D ROC curve analysis evaluating the predictive performance of IBSP; E meta-analysis demonstrating the prognostic significance of IBSP across four datasets The regulation of multiple cellular pathways and co-expressed genes related to glioma From the preceding findings, the biological role of IBSP in glioma was further explored using GSEA. Analysis of the CGGA and GSE16011 datasets demonstrated that high IBSP expression was significantly enriched in pathways related to allograft rejection, apoptosis, EMT, and coagulation (Fig. 6A–D; all FDR < 0.05). These pathways are primarily associated with tumor immunity, apoptotic processes, and hypoxia-related responses. Elevated IBSP expression was significantly correlated with increased immune response signatures (p < 0.001). Fig. 6GSEA results for IBSP-associated signaling pathways in glioma. A CGGA microarray cohort; B CGGA RNA-seq-1 cohort; C CGGA RNA-seq-2 cohort; D GSE16011 cohort GSEA results for IBSP-associated signaling pathways in glioma. A CGGA microarray cohort; B CGGA RNA-seq-1 cohort; C CGGA RNA-seq-2 cohort; D GSE16011 cohort To identify potential downstream effectors of IBSP, co-expression analysis was conducted. Six glioma-associated genes were identified as significantly positively correlated with IBSP expression: CASP4 (r = 0.42, p < 0.001), KYNU (r = 0.38, p < 0.001), MSR1 (r = 0.35, p < 0.001), PLAUR (r = 0.41, p < 0.001), SIGLEC9 (r = 0.36, p < 0.001), and SLC16A3 (r = 0.39, p < 0.001) (Table 2). These associations were consistently observed across both the CGGA and GSE16011 datasets. Table 2Top correlated genes to IBSPGene symbolCGGA MicroarrayCGGA RNA-Seq-1CGGA RNA-Seq-2GSE16011CorrelationP valueCorrelationP valueCorrelationP valueCorrelationP valueCASP40.682< 0.0010.73< 0.0010.602< 0.0010.608< 0.001KYNU0.69< 0.0010.713< 0.0010.639< 0.0010.605< 0.001MSR10.711< 0.0010.656< 0.0010.651< 0.0010.664< 0.001PLAUR0.696< 0.0010.765< 0.0010.644< 0.0010.638< 0.001SIGLEC90.601< 0.0010.684< 0.0010.617< 0.0010.681< 0.001SLC16A30.675< 0.0010.774< 0.0010.638< 0.0010.600< 0.001 Table 2Top correlated genes to IBSPGene symbolCGGA MicroarrayCGGA RNA-Seq-1CGGA RNA-Seq-2GSE16011CorrelationP valueCorrelationP valueCorrelationP valueCorrelationP valueCASP40.682< 0.0010.73< 0.0010.602< 0.0010.608< 0.001KYNU0.69< 0.0010.713< 0.0010.639< 0.0010.605< 0.001MSR10.711< 0.0010.656< 0.0010.651< 0.0010.664< 0.001PLAUR0.696< 0.0010.765< 0.0010.644< 0.0010.638< 0.001SIGLEC90.601< 0.0010.684< 0.0010.617< 0.0010.681< 0.001SLC16A30.675< 0.0010.774< 0.0010.638< 0.0010.600< 0.001 Top correlated genes to IBSP Predictive drugs in gliomas targeting IBSP variants From these findings, IBSP plays a key role in glioma malignancy. Therefore, therapeutic agents targeting IBSP-associated molecular mechanisms may have potential to inhibit tumor progression. Pharmacogenomic analysis was conducted using data from the CTRP to identify candidate small-molecule compounds associated with IBSP expression. A total of fourteen potential therapeutic agents were identified (all r > 0.08, p < 0.05) (Table 3). Among these, alvocidib and dinaciclib demonstrated the strongest correlations with IBSP expression (r > 0.1, p < 0.005). These findings indicate that IBSP expression is associated with differential sensitivity to multiple therapeutic agents, highlighting its potential relevance as a target for pharmacological intervention in patients with glioma. Table 3Prediction of potential drugs to IBSP via CTRP datasetSymbolDrugCorrelationFdrIBSPalvocidib-0.171570.002496IBSPdinaciclib-0.155930.005291IBSPtipifarnib-P1-0.096380.014683IBSPBMS-345,541-0.093430.017653IBSPN9-isopropylolomoucine-0.093150.022463IBSPsirolimus-0.092440.019467IBSPlinifanib-0.08980.021497IBSPmomelotinib-0.088070.031504IBSPMK-2206-0.087910.040589IBSPBRD6340-0.087730.027927IBSPCR-1-31B-0.084260.025991IBSPpazopanib-0.083260.038391IBSPMST-312-0.082970.033848IBSPKo-143-0.080050.04832 Table 3Prediction of potential drugs to IBSP via CTRP datasetSymbolDrugCorrelationFdrIBSPalvocidib-0.171570.002496IBSPdinaciclib-0.155930.005291IBSPtipifarnib-P1-0.096380.014683IBSPBMS-345,541-0.093430.017653IBSPN9-isopropylolomoucine-0.093150.022463IBSPsirolimus-0.092440.019467IBSPlinifanib-0.08980.021497IBSPmomelotinib-0.088070.031504IBSPMK-2206-0.087910.040589IBSPBRD6340-0.087730.027927IBSPCR-1-31B-0.084260.025991IBSPpazopanib-0.083260.038391IBSPMST-312-0.082970.033848IBSPKo-143-0.080050.04832 Prediction of potential drugs to IBSP via CTRP dataset Discussion This study provides a comprehensive analysis that identifies IBSP as an independent prognostic biomarker in glioma, demonstrating significant associations with WHO grade, IDH status, and OS across multiple large-scale cohorts. IBSP was enriched in apoptosis- and coagulation-related pathways and positively correlated with six glioma-associated genes, supporting its potential relevance in molecular stratification and targeted therapeutic approaches for high-grade glioma. Glioma is a malignant tumor arising within brain tissue and remains one of the most challenging neoplasms to treat due to its rapid progression and high recurrence rate [22]. Advances in high-throughput sequencing technologies have substantially improved the understanding of glioma-associated molecular markers, leading to the identification of numerous biomarkers that are increasingly incorporated into clinical decision-making. The DNA-alkylating agent temozolomide, introduced in 2005, improve survival outcomes in patients with glioma by inducing DNA damage in tumor cells [2]. However, therapeutic resistance may occur due to increased expression of DNA repair proteins, which can attenuate temozolomide-induced DNA damage [23]. One key mechanism involves epigenetic silencing of the O6-methylguanine-DNA methyltransferase (MGMT) gene through promoter methylation, which impairs DNA repair capacity in glioma cells [24]. Consequently, MGMT promoter methylation status serves as an important prognostic and predictive biomarker for response to temozolomide in GBM [25]. Overexpression of vascular endothelial growth factor (VEGF) has been observed in recurrent high-grade gliomas [26]. Bevacizumab, a monoclonal antibody targeting VEGF-A, has been used as a therapeutic option for patients with recurrent high-grade glioma [27]. Gene expression data for glioma were obtained from the CGGA and GSE public databases and analyzed to evaluate the prognostic significance of IBSP. The findings indicate that IBSP serves as a biomarker associated with poor prognosis in glioma. Further analysis of the relationship between IBSP expression and clinical characteristics demonstrated that IBSP expression was positively correlated with WHO grade, with the highest expression observed in GBM. Additional clinical parameters, including IDH variant status, 1p/19q codeletion status, and KPS, were also significantly associated with IBSP expression. Univariate and multivariate survival analyses further confirmed that IBSP is an independent prognostic factor for OS in patients with glioma. Collectively, these results indicate that elevated IBSP expressions are associated with adverse clinical outcomes; however, the underlying molecular mechanisms require further investigation. GSEA was performed using CGGA and GSE datasets to identify pathways associated with IBSP expressions. High IBSP expression was significantly enriched in apoptosis, coagulation, and complement pathways (all FDR < 0.05). Although apoptosis is a fundamental physiological process, it is frequently dysregulated in cancer cells, thereby contributing to chemoresistance [28–30]. Glioma is associated with an increased risk of coagulopathies, and intravascular thrombosis has been proposed as a histopathological feature that may help in distinguishing tumor malignancy grade [31, 32]. Inflammatory responses triggered by endothelial injury activate the coagulation cascade, and coagulation processes, similar to inflammation, play a key role in tumor progression and are considered important biological markers of malignancy. The complement system contributes to the malignant progression of glioma, with its activity correlated with established clinical prognostic factors, including WHO grade and IDH status [33]. As a high-throughput analytical approach, GSEA provides reliable identification of biologically relevant pathways in large-scale datasets. The results of GSEA indicate that IBSP may influence glioma progression through these interconnected biological pathways. In this study, IBSP expression was significantly elevated in GBM compared with lower-grade gliomas (p < 0.001), and increased IBSP expression was independently associated with reduced OS (HR = 1.531, 95% CI: 1.07–2.192, p = 0.020). Multivariate Cox regression analysis confirmed IBSP as an independent prognostic factor, irrespective of IDH variant status and 1p/19q codeletion. Furthermore, GSEA demonstrated that high IBSP expression was significantly enriched in apoptosis and coagulation pathways (both FDR < 0.05), with positive correlations observed between IBSP and CASP4 (r = 0.42, p < 0.001) and PLAUR (r = 0.41, p < 0.001). Collectively, these findings support the role of IBSP as a reliable prognostic biomarker in glioma. Co-expression analysis identified several genes potentially associated with glioma pathogenesis, many of which have been previously reported to be positively correlated with IBSP expression. CASP4 has been implicated in apoptotic signaling pathways in tumor cells and is highly expressed in high-risk glioma subgroups [34, 35]. KYNU, a 465-amino acid protein, is involved in tryptophan metabolism and contributes to regulation of the tumor immune microenvironment, thereby influencing cancer prognosis [36]. Furthermore, KYNU has been associated with the PI3K/AKT signaling pathway, which plays a key role in tumor progression. MSR1 has been demonstrated to regulate macrophage-mediated inflammatory responses, partly through modulation of CD44 expression, and has been explored as a potential therapeutic target in cancer [37, 38]. It is also closely associated with the tumor microenvironment, is significantly overexpressed in low-grade glioma compared with normal brain tissue, and has prognostic value in these tumors [39]. PLAUR is involved in tumorigenesis and immunomodulation and has been demonstrated to promote malignant progression, with its overexpression associated with poor prognosis in patients with glioma [40]. In immunotherapy, modulation of the tumor microenvironment from an anti-inflammatory to a pro-inflammatory state is considered critical, and SIGLEC9 may contribute to regulation of the immunosuppressive microenvironment in glioma [41]. The transmembrane protein MCT4 (SLC16A3) is involved in lactate transport in metabolically active tissues [42]. Overexpression of MCT4 has been associated with increased tumor cell activity, and SLC16A3 has been identified as a key gene associated with glioma malignancy [43]. The potential regulatory relationships between IBSP and its co-expressed genes were further explored. CASP4, a key mediator of pyroptosis, may be transcriptionally regulated through IBSP-associated NF-κB signaling, as both genes are enriched in inflammatory pathways. The involvement of KYNU in tryptophan metabolism indicates that IBSP may influence the tumor immune microenvironment via the kynurenine–AhR axis. The observed correlation between IBSP and MSR1 (r = 0.35) indicates a possible role in the co-regulation of tumor-associated macrophage infiltration. These proposed mechanisms require further experimental validation, including chromatin immunoprecipitation sequencing and promoter luciferase reporter assays, to determine whether these regulatory interactions are direct or indirect. Given the involvement of IBSP in glioma progression, the identification of therapeutic agents targeting this biomarker is of considerable clinical relevance. In this study, pharmacogenomic analysis using the CTRP dataset identified several small-molecule compounds with potential therapeutic applicability in glioma [44]. Cyclin-dependent kinase inhibitors have demonstrated efficacy in preclinical models. For example, in an in vitro study of medulloblastoma, dinaciclib exhibited greater cytotoxic activity compared with the CDK4/6-selective inhibitor palbociclib, reflecting its broader inhibitory profile [45]. In pediatric diffuse glioma, the farnesyltransferase inhibitor tipifarnib enhances radiosensitivity and improves clinical outcomes. Consistent with these findings, CTRP-based analysis in this study demonstrated that elevated IBSP expression was positively correlated with increased sensitivity to CDK inhibitors, including alvocidib (r = 0.12, p < 0.005) and dinaciclib (r = 0.11, p < 0.005). These results indicate that IBSP may serve as a predictive biomarker for response to CDK inhibitor–based therapies in patients with glioma [46]. BMS-345,541 functions as an inhibitor of the IκB kinase complex and has been demonstrated to modulate interferon-induced gene expression and antiviral activity. It has also demonstrated the capacity to enhance the efficacy of oncolytic viruses in the elimination of glioma cells [47]. Collectively, these findings support the potential clinical relevance of IBSP as a biomarker for guiding therapeutic strategies in glioma. Beyond its prognostic significance, enrichment analysis in this study indicates that IBSP is significantly associated with EMT and allograft rejection pathways, indicating a role in the modulation of the glioma microenvironment, potentially contributing to immune evasion and tumor invasion. The observed positive correlation between IBSP and SIGLEC9 (r = 0.36, p < 0.001), an immunoregulatory checkpoint molecule, further supports its potential involvement in the establishment of an immunosuppressive microenvironment. The association between IBSP and PLAUR (r = 0.41, p < 0.001) indicates a role in extracellular matrix remodeling and tumor cell dissemination. Collectively, these findings indicate that IBSP may function not only as a prognostic biomarker but also as a regulator of tumor–microenvironment interactions in glioma. Further experimental studies are required to elucidate its mechanistic role in immune modulation and tumor progression. Consistent with findings in primary central nervous system lymphoma [48], where gene expression profiles associated with cancer morphology and the tumor microenvironment demonstrate robust prognostic value, the present study suggests that IBSP may similarly function as a microenvironment-related prognostic indicator in glioma. Furthermore, given that epithelial–mesenchymal transition and glioma stem cell characteristics have been identified as promising prognostic markers in glioblastoma [49], the significant enrichment of IBSP in EMT-related pathways observed herein implies that IBSP may contribute to glioma progression through mechanisms involving mesenchymal transformation and stemness maintenance. Collectively, these findings support the broader concept that stromal remodeling and microenvironmental factors are integral to survival prediction across diverse central nervous system malignancies, and highlight the potential of IBSP as a biomarker that bridges tumor grade, microenvironmental dynamics, and stem cell biology in glioma. However, several inherent limitations of this study should be acknowledged. First, as a retrospective analysis based on publicly available datasets, selection bias is unavoidable. The CGGA and GSE16011 cohorts predominantly represent Asian and Western populations, respectively, which may limit the generalizability of the findings to other ethnic groups. Second, incomplete clinical variables including detailed treatment regimens, MGMT promoter methylation status, and recurrence patterns may have introduced confounding effects in the prognostic analysis, despite adjustment using multivariate models. Third, although IBSP-associated pathways were identified through GSEA, these findings are based on bioinformatic predictions and require experimental validation. The absence of functional assays, such as IBSP knockdown or overexpression studies, limits the ability to establish a causal relationship between IBSP expression and glioma progression. Fourth, the drug sensitivity analysis was derived using in silico correlations rather than experimental models, such as patient-derived xenografts or clinical trial data, which may limit its translational relevance. Future prospective studies incorporating comprehensive clinical annotation and mechanistic investigations are warranted to validate IBSP as a therapeutic target in glioma. Conclusion IBSP is implicated in the malignant progression of glioma and its elevated expression is associated with poor prognosis. IBSP demonstrates independent prognostic value in patients with glioma and is significantly upregulated in high-grade tumors. IBSP provides predictive value beyond established clinical and molecular prognostic factors. These findings support the potential of IBSP as a therapeutic target for improving clinical outcomes in glioma and provide preliminary evidence for candidate agents targeting IBSP-associated pathways.