Translate this page into:
IL-32-mediated caspase-43 induction: Shaping macrophage differentiation & immunomodulation in glioblastoma multiforme
#Equal contribution
For correspondence: Bing-Qian Zhang, Chongqing Medical and Pharmaceutical College and Chongqing Engineering Research Center of Pharmaceutical Sciences, Chongqing, China 401331 e-mail: 10872@cqmpc.edu.cn
-
Received: ,
Accepted: ,
Abstract
Background & objectives
Glioblastoma multiforme (GBM) is the most aggressive form of brain tumour, characterised by rapid progression, high recurrence, and poor prognosis. The immunosuppressive tumour microenvironment (TME) of GBM poses a major barrier to effective therapy. Interleukin (IL)-32, a pro-inflammatory cytokine, has been implicated in cancer progression, but its specific role in GBM remains unclear. This study aimed to investigate the immunoregulatory functions of IL-32 in GBM, particularly its impact on monocyte differentiation and macrophage polarisation.
Methods
Transcriptomic data from TCGA-GBM and GEO-GSE156902 were analysed to identify differentially expressed genes (DEGs), with a focus on IL-32. Functional enrichment analyses (GO and Kyoto Encyclopaedia of Genes and Genomes- KEGG) and protein-protein interaction (PPI) network construction were conducted using R software and Cytoscape. IL-32 expression was validated by qPCR and Western blot (WB) in GBM cell lines. Single-cell RNA sequencing data were integrated to determine IL-32 expression across brain cell types. Additionally, correlations between IL-32 and RNA modification genes (m6A, m5C, m1A) were analysed.
Results
IL-32 was significantly upregulated in GBM tissues and particularly enriched in microglial cells. Functional studies revealed that IL-32 promotes caspase-43-mediated monocyte differentiation into macrophages. Moreover, IL-32 induced a phenotypic shift from M1 to M2 macrophages via NF-κB pathway activation. PPI analysis confirmed IL-32 as a hub gene involved in immune regulation. RNA modification analysis showed positive correlations between IL-32 and METTL3, and a negative correlation with TET2, indicating epigenetic modulation of IL-32-related immune functions.
Interpretation & conclusions
IL-32 plays a pivotal immunoregulatory role in the GBM microenvironment by driving macrophage differentiation and M2 polarisation, contributing to tumour immune evasion. These findings highlight IL-32 as a potential therapeutic target for modulating immune responses in GBM and underscore its relevance in the design of future immunotherapeutic strategies.
Keywords
Caspase-43
GBM
IL32
NF-κB
Glioblastoma, often referred to as glioblastoma multiforme (GBM), stands as one of the most devastating and challenging brain tumours encountered in clinical practice1. Characterised by its aggressive nature, rapid progression, and resistance to conventional treatments, GBM presents formidable hurdles to both patients and clinicians2. Despite decades of research and numerous therapeutic approaches, the prognosis for GBM patients remains bleak, underscoring the urgent need for a deeper understanding of the molecular intricacies driving this malignancy3. The history of glioblastoma research is marked by significant milestones, including the pioneering work of Jules Bernard Luys in the 19th century, who is credited with the initial description of these tumors4. Subsequent advancements in neuroimaging, such as the advent of magnetic resonance imaging (MRI) and positron emission tomography (PET), have revolutionised the detection and characterisation of GBM lesions5. However, despite these diagnostic innovations, GBM remains a formidable diagnostic challenge due to its heterogeneity and infiltrative growth pattern within the brain parenchyma6. Glioblastoma accounts for approximately 15 per cent of all primary brain tumours and represents the most common malignant glioma subtype7. This tumour primarily affects adults, with a median age of diagnosis around 64 yr, although it can occur at any age8. Sex differences in incidence have been noted, with a slightly higher predilection for males9. The clinical presentation of GBM is often insidious, with symptoms such as headache, cognitive changes, seizures, and focal neurological deficits arising gradually10. The aggressiveness of the disease is further underscored by its rapid progression, frequently leading to a diagnosis at advanced stages of the tumour11. Histopathologically, GBM is characterised by features including cellular pleomorphism, microvascular proliferation, and necrosis, often referred to as the hallmark features of this malignancy12. Additionally, GBM exhibits significant molecular heterogeneity, with the recognition of distinct subtypes based on gene expression profiles13. The identification of molecular markers, such as isocitrate dehydrogenase (IDH) mutations and O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation status, has added depth to our understanding of the disease, aiding in prognostication and therapeutic decision-making14. The current standard of care for GBM typically involves maximal safe surgical resection, followed by adjuvant radiation therapy and chemotherapy with temozolomide15. Despite these multimodal approaches, GBM almost invariably recurs, and long-term survival rates remain dismal16. The challenges in treating GBM are multifaceted, encompassing the intrinsic resistance of glioma cells to therapy, the intricate interaction between the tumour and its microenvironment, and the presence of the blood-brain barrier, which limits drug penetration into the tumour17. High-throughput technologies, such as next-generation sequencing (NGS) and single-cell RNA sequencing, have enabled comprehensive profiling of GBM genomes, transcriptomes, and epigenomes. These endeavours have led to the identification of novel therapeutic targets and promising biomarkers that hold the potential to revolutionise GBM treatment strategies18.
One of the promising avenues of research in GBM involves the investigation of interleukin (IL)-32, a pro-inflammatory cytokine with diverse roles in inflammation, immunity, and cell survival. IL-32 may play a pivotal role in the pathogenesis of GBM, influencing tumour cell behaviour, immune responses, and the intricate crosstalk between tumour and stromal cells within the tumour microenvironment (TME)19,20. This research involved an in-depth analysis of the GEO-GBM dataset, encompassing multiple aspects of GBM. The primary objectives encompass identifying differential gene expression patterns, conducting functional enrichment analysis through GO-BP and KEGG pathways, and classifying samples based on immune cell infiltration.
Materials & Methods
Study Location
This study was conducted in Jinfeng Laboratory of Molecular Medicine, Chongqing, China, from June 2023 to December 2023. This study was reviewed and approved by the Ethics Committee of Chongqing Traditional Chinese Medicine Hospital. The protocol approved by the ethics committee was followed to carry out experimental research.
Sample collection and processing
Total RNA was extracted from the T98G glioblastoma cell line for both the control and IL-32 overexpression (IL-32-OE) groups. The T98G cell line was obtained from Guangzhou Baiwei Biotechnology Co., Ltd. The control group consisted of three samples collected from untreated T98G cells, while the IL-32-OE group comprised three samples of T98G cells transfected with an IL-32 expression plasmid. Total RNA from each group was rapidly isolated and frozen to preserve RNA quality. Before RNA extraction, the samples were homogenised and pooled to ensure representativeness4.
RNA extraction and quality control
Total RNA was isolated from the T98G glioblastoma cell line using TRIzol reagent (Invitrogen, USA), according to the manufacturer’s protocol (referenced in the revised manuscript). Approximately 1 × 10⁶ cells were used for RNA extraction per sample, and the average RNA yield was approximately 10 µg. The concentration and purity of RNA were determined using a NanoDrop spectrophotometer, and RNA integrity was further evaluated by agarose gel electrophoresis4.
RNA sequencing
For each of the control and IL-32 groups, a total of three murine tissue samples were used for RNA sequencing. The murine models were C57BL/6J male mice aged 8-10 wk, and the harvested tissue was glioblastoma-infiltrated brain tissue. The RNA sequencing library was prepared using the Novogene platform. RNA samples underwent poly-A mRNA purification using oligo(dT) beads. The purified mRNA was then fragmented into small pieces, followed by cDNA synthesis using random hexamers. The resulting cDNA fragments were end-repaired, A-tailed, and ligated with sequencing adapters. These libraries were amplified by PCR to enrich adapter-ligated fragments. Sequencing was conducted on the Novogene platform using paired-end mode, with sequencing depth and read length parameters in accordance with Novogene’s standard protocol. Raw data underwent strict quality control to eliminate adapter sequences, low-quality reads, and other artifacts, ensuring clean reads for downstream analyses8. The raw sequencing data underwent quality control and preprocessing using R packages such as ‘FastQC’ and ‘Trim Galore!’. This step aimed to remove low-quality reads and adapter sequences, resulting in clean reads suitable for subsequent analysis. Differential expression analysis was carried out using R packages like ‘DESeq2’ or ‘edgeR’12.
Differentially expressed genes (DEGs) analysis methodology
The DEGs analysis entailed rigorous statistical comparison utilising tools such as DESeq2 or edgeR to pinpoint genes exhibiting significant expression differences across various experimental conditions or sample groups. This process accounted for data variability and applied stringent multiple testing corrections. Subsequently, the analysis extended to functional enrichment assessment, encompassing GO and KEGG pathway analysis, which facilitated the categorisation of DEGs based on their biological roles. To delve deeper into the molecular context, a Protein-Protein Interaction (PPI) network analysis was performed. This analysis aimed to elucidate interactions among DEGs and identify central genes, referred to as hub genes, using specialised network analysis algorithms. Hub genes, recognised for their critical roles within biological networks, were then identified. In parallel, a survival analysis was conducted to evaluate the prognostic significance of these hub genes. This analysis explored the association between the expression levels of hub genes and clinical outcomes, shedding light on their potential as prognostic indicators. To further explore the functional and localisation aspects of hub genes, the Protein Atlas website as described previously12.
Studying IL-32 expression through cell-based experiments
In the investigation of IL-32 expression through cell-based experiments, the T98G glioblastoma cell line was cultured under standard conditions and divided into experimental and control groups. The IL-32-overexpression (IL-32-OE) group was generated by transfecting cells with an IL-32-overexpression plasmid using Lipofectamine 3000 (Thermo Fisher Scientific, USA), while the control group received an empty plasmid vector. Each transfection was performed using 2 × 10⁶ cells per well. Cells were maintained in RPMI-1640 medium supplemented with 10 per cent foetal bovine serum and one per cent penicillin-streptomycin, and incubated at 37°C with 5 per cent CO₂ for 48 hours post-transfection. After the designated incubation period, cell samples were collected for protein extraction16.WB analysis was performed following standard protocols, including SDS-PAGE electrophoresis, protein transfer to membranes, blocking with 5 per cent BSA, and incubation with IL-32-specific primary antibodies followed by HRP-conjugated secondary antibodies. Protein bands were visualised using chemiluminescence detection (ECL). This experimental setup is consistent with previously published protocols.
Analysis of the relationship between IL-32 expression levels and immune infiltration
The study focuses on investigating the correlation between IL-32 expression and immune infiltration levels, particularly tumour-infiltrating lymphocytes (TILs), which have shown promise in improving prognosis and therapeutic efficacy across different cancer types. IL-32 expression and immune infiltration levels were studied in various types of tumours. To explore the relationship between IL-32 expression and immune infiltration levels, we harnessed the resources of TCGA tumours, including B-cells, CD4+ T-cells, CD8+ T-cells, macrophages, neutrophils, and dendritic cells. Specifically, we utilised the immune-gene module of TIMER2 (Tumour Immune Estimation Resource, version 2, http://timer.cistrome.org/ ) for this analysis. The relative proportions of B-cells, CD4+ T-cells, and macrophages were calculated using the EPIC (Estimation of Proportions of Immune and Cancer cells) method across multiple tumour types, while the relative proportion of CD8+ T-cells was determined using the QUANTISEQ method. Additionally, the relative proportions of neutrophils and dendritic cells were computed using the MCPCOUNTER method. These analyses provide valuable insights into the intricate interplay between IL-32 expression and immune infiltration, shedding light on potential therapeutic implications across a spectrum of cancer types17.
IL-32 and RNA modification gene analysis (m6A, m5C, m1A)
An analysis was conducted utilising the UCSC database ( https://xenabrowser.net/ ) on the pan-cancer dataset TCGA TARGET GTEx (PANCAN, N=19131, G=60499), which underwent uniform standardisation. The dataset included expression data for ENSG00000008517 (IL-32) and 44 genes associated with three RNA modifications: m6A (21 genes), m5C (13 genes), and m1A (10 genes). The analysis focused on samples categorised as Primary Solid Tumour, Primary Tumour, Primary Blood Derived Cancer - Bone Marrow, and Primary Blood Derived Cancer - Peripheral Blood, with all normal samples excluded. A log2(x+0.001) transformation was applied to each expression value. Subsequently, Pearson correlations were computed between ENSG00000008517 (IL-32) and genes representing these RNA modifications. This analysis provides insights into potential interactions between IL-32 and genes involved in RNA modifications across diverse cancer types11.
Analysis of GEO-GBM data for differential expression, enrichment analysis (GO-BP-KEGG), and immune infiltration classification
Beginning with data acquisition and preprocessing, the GEO-GBM dataset was gathered from the GEO database and subjected to rigorous quality control measures, including noise reduction, batch effect correction, and normalisation. Subsequently, the samples were categorised into high-expression and low-expression groups in accordance with the experimental design. Moving forward, enrichment analysis was performed using specialised bioinformatics tools, such as GSEA, DAVID, or Metascape.
Database prediction
Data Preprocessing involved the initial preparation of raw sequencing reads from both the TCGA-GBM and GEO-GSE156902 datasets. This preprocessing included quality control procedures to eliminate low-quality reads and adapter sequences, resulting in the acquisition of clean reads suitable for further analysis. Differential Expression Analysis was performed using R packages such as ‘DESeq2’ or ‘edgeR’. Comparisons were conducted between the Control and IL-32 groups to identify genes exhibiting significant differential expression2. The primary focus was on identifying gene expression changes associated with IL-32 in GBM. Volcano Plot Visualisation was employed as a means of visualising the results of the differential expression analysis. These plots utilised log2 fold change values and adjusted P values to represent the significance of gene expression changes, providing an intuitive overview of differentially expressed genes. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) Pathway Analysis involved the enrichment analysis of GO terms and KEGG pathways associated with the differentially expressed genes3. This analysis aimed to elucidate the biological processes and pathways linked to IL-32 in the context of GBM. Identification of Top Genes was conducted using the Cytoscape software in conjunction with the HUBBA plugin. The EcCentricity algorithm within HUBBA was utilised to select the top 10 genes for further investigation, with a focus on their potential significance in the context of IL-32 and GBM. This methodology outlines the sequential steps involved in data preprocessing, differential gene expression analysis, visualisation, functional enrichment analysis, and the selection of top genes for further analysis.
Results
Database prediction
The analysis involved data collection from public databases, revealing 12,098 overlapping genes related to inflammation and tumour biology. The top 10 hub genes, including IL-32, were identified using network analysis. Pathway analysis unveiled associations with critical pathways such as NF-ĸB and TNF signalling. GO analysis highlighted their functions, including roles in apoptosis and immune responses. Notably, IL-32 showed differential expression in glioblastoma. These findings provide valuable insights into the roles of these genes in cancer biology (Supplementary Fig. 1).
RNA sequencing data analysis
In the RNA sequencing analysis, 259 genes with significant differential expression (absolute log-fold change >1.5) were identified between the IL32 and control (con) groups. These genes are known to be involved in key pathways, including the PI3K-Akt signalling pathway, Th1 and Th2 cell differentiation, vascular smooth muscle contraction, viral carcinogenesis, GABAergic synapse, PD-L1 expression, salivary secretion, TGF-beta signalling, inflammatory mediator regulation of TRP channels, and non-homologous end-joining. Notably, the top 10 genes in terms of differential fold change included IL32, RPL7L1P3, UTS2B, DRICH1, AC034236.3, AL138976.2, AC245595.1, LINC00499, AC048344.4, and PRSS22.To further confirm IL-32 gene expression within the tissues, immunohistochemistry staining was performed, utilising TESTIS and haematoxylin eosin (HE) staining techniques. The results unequivocally demonstrated differential IL-32 gene expression within the tissues, confirming its varying levels of expression (Fig.1A-B; Supplementary Fig. 2 and 3).
- RNA sequencing data analysis.
IL-32 expression through cell-based experiments
WB and Quantitative Polymerase Chain Reaction (qPCR) analyses indicated significant differences in IL-32 expression levels (Fig. 2A-B).
- IL-32 expression through cell-based experiments. Model: IL-32-overexpressed groups (T98G cell lines were used. The T98G cell lines transfected with an IL-32 expression plasmid).
Analysis of the relationship between IL-32 expression levels and immune infiltration
Based on IL-32 gene expression, calculations were performed for stromal (r=0.49), immune (r=0.48), and estimate scores (r=0.51) within GBM tumours. These results indicate a significant correlation between IL-32 gene expression and immune infiltration. Furthermore, IL-32 gene expression was used to reassess the infiltration scores of various immune cell types within GBM tumours. In the immune cell analysis using TIMER, B-cells and T-cells CD8 showed a negative correlation, while the remaining cell types exhibited positive correlations. In the immune cell analysis using EPIC, B-cells, cancer-associated fibroblasts (CAFs), and other cells displayed negative correlations, while the other cell types showed positive correlations with IL-32 gene expression (Supplementary Fig. 4).
IL-32 and RNA modification gene analysis (m6A, m5C, m1A)
The results revealed that TRMT6, TRMT10C, and ELAVL1 exhibited negative correlations, while the remaining genes showed positive correlations (Supplementary Fig. 5). Notably, METTL3 and TET2 displayed the most significant positive correlations in this analysis. Furthermore, single-cell data analysis showed that IL-32 was highly expressed in microglial cells in brain single-cell sequencing data.
Analysis of GEO-GBM data for differential expression, enrichment analysis (GO-BP-KEGG), and immune infiltration classification
The results revealed varying patterns of immune cell infiltration across three distinct groups (group 1, group 2, and group 3). Notable differences were observed in the infiltration scores of B-cells, CD4+ T-cells, CD8+ T-cells, NK cells, monocytes, neutrophils, and Th2 cells among the groups. While groups 1 and 2 exhibit similar infiltration profiles, group 3 stands out with a unique immune cell composition, characterised by higher levels of CD4+ T-cells and CD8+ T-cells (Supplementary Fig. 6). Statistically significant differences in infiltration levels were observed for neutrophils and Th2 cells between the groups, as indicated by the provided P values.
IL-32 and functional interconnection with PI3K-Akt and NF-κB signalling pathways
To further investigate the molecular mechanisms by which IL-32 modulates immune responses and tumour behaviour in GBM, we analysed its association with the PI3K-Akt and NF-κB signalling pathways. Enrichment analysis of IL-32-related differentially expressed genes (DEGs) revealed significant involvement of these pathways. In particular, genes associated with the PI3K-Akt axis, including AKT1, PIK3R1, and PTEN, showed co-expression with IL-32, suggesting a potential regulatory influence. Similarly, components of the NF-κB pathway, such as NFKB1 and RELA, were positively correlated with IL-32 expression levels. These results support the hypothesis that IL-32 may exert its immunomodulatory effects in GBM by activating the NF-κB pathway to drive M2 macrophage polarisation, while concurrently engaging PI3K-Akt signalling to promote tumour cell survival and proliferation. These insights provide mechanistic evidence for IL-32’s role as a central regulator in GBM signalling networks.
Discussion
A comprehensive analysis of the multifaceted role of IL-32 in the context of GBM can provide critical insights. Integrative exploration beginning with database predictions identifies IL-32 as a central hub gene associated with inflammation and tumour biology pathways. Subsequent pathway analyses further reinforce the involvement of IL-32 in pivotal signalling cascades, including NF-κB and TNF signalling. Transitioning to transcriptional characterisation, RNA sequencing uncovers 259 genes exhibiting significant differential expression patterns associated with IL-32. These IL32-connected genes participate in numerous essential pathways such as PI3K-Akt signalling, Th1/Th2 cell differentiation, inflammation, and TRP channel modulation. Notably, our analysis highlighted the potential functional interconnection between IL-32 and the PI3K-Akt as well as NFĸB pathways. IL-32 expression showed positive correlation with key components of the PI3K-Akt pathway (e.g., AKT1, PIK3R1), suggesting its involvement in tumour cell survival, proliferation, and immune escape. Concurrently, IL-32 appears to activate the NFĸB pathway, as indicated by the upregulation of NFKB1 and RELA, which may drive the polarisation of macrophages from the M1 to the M2 phenotype, thereby facilitating an immunosuppressive TME. These findings point to IL-32 as a central upstream regulator capable of integrating immune signalling and oncogenic growth pathways within GBM. Immunohistochemical validation further confirms the differential expression of IL-32 within GBM tissues, consolidating its functional role in shaping the TME. Additional cell-based experiments, including Western blotting and qPCR, provide further evidence of IL-32 overexpression and its downstream signalling effects. Quantitative analyses underscore its regulatory capacity over immune cell infiltration, as revealed by strong correlations between IL-32 expression and various immune scores. Parallel investigations using TIMER and EPIC further confirm complex interactions between IL-32 and diverse immune cell subsets within the tumour milieu. Moreover, exploration of the association of IL-32 with RNA modification genes unveiled notable interactions, especially with METTL3 and TET2, indicating a possible role in epitranscriptomic regulation. Single-cell transcriptomic data reveal the preferential expression of IL-32 in microglial cells, suggesting involvement in critical processes such as phagocytosis, motility, inflammation, and immune signalling within the central nervous system.
In totality, these integrated findings establish IL-32 as a pivotal coordinator in GBM, orchestrating immune dynamics, oncogenic signalling (including PI3K-Akt and NF-κB), and RNA regulatory mechanisms. These insights provide a compelling rationale for targeting IL-32 as a novel immunotherapeutic entry point to reprogram the GBM immune microenvironment and improve anti-tumour immune responses.
Overall, IL-32 emerged as a central mediator within the glioblastoma microenvironment, coordinating diverse signalling pathways, immune dynamics, and RNA regulation. Additionally, it is suggested that IL-32 induces monocyte-to-macrophage differentiation via caspase-3 activation, enhancing systemic immunity. However, IL-32 also facilitates dendritic cell development, subsequently enabling immune evasion. Thus, IL-32 directly impacts specific immunity by promoting both proinflammatory mononuclear phagocyte differentiation and immunosuppressive polarisation. Hence, targeting the multifaceted roles of IL-32 may disrupt its immunomodulatory functions and offer novel immunotherapeutic approaches against glioblastoma and related cancers. Further research should, however, elaborate on the intricate molecular mechanisms governing the influence of IL-32 on immune phenotype and antitumour immunity.
Financial support & sponsorship
The study received funding support by the Natural Science Foundation of Chongqing (General Program; Grant no. cstc2020jcyj-msxmX0116 and cstc2020jcyj-msxmX0117). The study also received funding support by the Science and Technology Foundation of Chongqing Municipal Commission of Education through the Young Teachers Program (Grant no. KJQN202102802), as well as through the Key Program (Grant no. KJZD-K202202804) both awarded to corresponding author (BQZ). The study also received support from the Open Fund of the Ministry of Education of China Key Laboratory of Tumor Immunopathology (Grant no.2018jsz106) awarded to corresponding author (BQZ).
Conflicts of Interest
None.
Use of Artificial Intelligence (AI)-Assisted Technology for manuscript preparation
The authors confirm that there was no use of AI-assisted technology for assisting in the writing of the manuscript and no images were manipulated using AI.
References
- Glioblastoma and other primary brain malignancies in adults: A review. JAMA. 2023;329:574-87.
- [Google Scholar]
- Glioblastoma hijacks neuronal mechanisms for brain invasion. Cell. 2022;185:2899-917.
- [Google Scholar]
- Single-cell RNA sequencing and spatial transcriptomics reveal cancer-associated fibroblasts in glioblastoma with protumoral effects. J Clin Invest. 2023;133:e147087.
- [Google Scholar]
- A patient-derived glioblastoma organoid model and biobank recapitulates inter- and intra-tumoral heterogeneity. Cell. 2020;180:188-204.
- [Google Scholar]
- Updated diagnostic criteria and nomenclature for neurofibromatosis type 2 and schwannomatosis: An international consensus recommendation. Genet Med. 2022;24:1967-7.
- [Google Scholar]
- Molecular heterogeneity and immunosuppressive microenvironment in glioblastoma. Front Immunol. 2020;11:1402.
- [Google Scholar]
- Epidemiology of glioblastoma multiforme-literature review. Cancers (Basel). 2022;14:2412.
- [Google Scholar]
- Predictors of survival in elderly patients undergoing surgery for glioblastoma. Neurooncol Adv. 2021;3:vdab083.
- [Google Scholar]
- YAP promotes autophagy and progression of gliomas via upregulating HMGB1. J Exp Clin Cancer Res. 2021;40:99.
- [Google Scholar]
- Symptom burden in glioblastoma: A prospective pilot study from diagnosis to first progression. Oncology. 2023;101:145-52.
- [Google Scholar]
- Modeling patient-derived glioblastoma with cerebral organoids. Cell Reports. 2019;26:3203-3211.e5.
- [Google Scholar]
- The KDR (VEGFR-2) genetic polymorphism Q472H and c-KIT polymorphism M541L are associated with more aggressive behaviour in astrocytic gliomas. Cancer Genomics Proteomics. 2020;17:715-27.
- [Google Scholar]
- Molecular heterogeneity and immunosuppressive microenvironment in glioblastoma. Front Immunol. 2020;11:1402.
- [Google Scholar]
- Higher cytolytic score correlates with an immunosuppressive tumor microenvironment and reduced survival in glioblastoma. Scientific Reports. 2020;10:17580.
- [Google Scholar]
- Brain co-delivery of temozolomide and cisplatin for combinatorial glioblastoma chemotherapy. Adv Mater. 2022;34:e2203958.
- [Google Scholar]
- Fatostatin induces ferroptosis through inhibition of the AKT/mTORC1/GPX4 signaling pathway in glioblastoma. Cell Death Dis. 2023;14:211.
- [Google Scholar]
- Development of small-molecule tropomyosin receptor kinase (TRK) inhibitors for fusion cancers. Acta Pharmaceutica Sinica B. 2020;11:355-372.
- [Google Scholar]
- A paradoxical effect of interleukin-32 isoforms on cancer. Front Immunol. 2022;13:837590.
- [Google Scholar]
- IL-32 and its paradoxical role in neoplasia. Crit Rev Oncol Hematol. 2023;186:104011.
- [Google Scholar]