A lipid metabolism and immune-related gene signature for predicting the prognosis of human gliomas

A lipid metabolism and immune-related gene signature for predicting the prognosis of human gliomas | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article A lipid metabolism and immune-related gene signature for predicting the prognosis of human gliomas Shanwen Liang, Xinin Zhang, Zhansheng Zhu, Yu Hong, Yangzhi Qi, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4408953/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Lipid metabolism reprogramming is critical in various biological processes and is considered a hallmark in cancers. The expression data of mRNA and corresponding follow-up information were obtained from TCGA used as a training set and the CGGA used as a validating set. Based on the expression of genes involved in lipid metabolism, 550 glioma samples of the training set were clustered by unsupervised classification method. Then, we construct a lipid metabolism-related risk signature based on the Lasso regression algorithm. The biological mechanism related to risk score was investigated by gene sets enrichment analysis (GSEA). 67 lipid metabolism- and immune-related genes were identified. Two robust groups were yielded by consensus clustering of the 550 samples. Subgroup2 correlated with a significantly better clinical outcome compared with Subgroup1. A 16-genes risk signature was constructed, and the overall survival of patients is dramatically better in the low-risk than the high-risk group. Consistently, the 16-gene signature showed pretty prognostically predicting ability by the receiver operating characteristic curve with areas under curve more than 0.8 in both TCGA and CGGA. Furthermore, the risk score was identified as an independent prognostic factor for glioma. Moreover, samples with a high-risk score were correlated with a higher level of immune infiltration and associated with a higher expression of immune checkpoints, which indicated an inhibitory tumor immune microenvironment. Our study demonstrated a new sight of lipid metabolism-related and immune-associated genes and constructed a 16-gene risk signature to predict prognosis and immunotherapy for glioma patients. Lipid metabolism immune profiling risk signature immunotherapy bioinformatics Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Background Gliomas are the most common malignant tumor among brain tumors (24.1% of all tumors). Moreover, glioblastoma accounts for 60% of gliomas[ 1 ]. Although standard treatment, including surgical resection, targeted radiation therapy, high-dose chemotherapy, has made tremendous development[ 2 ], the median survival for glioblastoma for all patients (regardless of treatment) remained eight months[ 1 ]. Immunotherapy and tumor-treating fields (TTFields)[ 3 , 4 ] have developed fast in recent years, but there was still no significant breakthrough. Also, more and more research were explored in metabolomic reprogramming, providing novel insights into malignant phenotypes of gliomas[ 5 , 6 ]. Metabolic reprogramming is recognized as an essential characteristic of cancer cells[ 7 ]. High rates of aerobic glycolysis and increased anabolism are prominent features of malignant tumor support[ 8 ]. Meanwhile, multiple metabolism-related pathways have been involved in the research of cancer. As one of the most significant metabolisms in cancer cells, lipid metabolism dysfunction has recently been more and more explored. As a highly diverse class of biological molecules, lipids exert various cells, including energy storage, cellular membrane structures, and metabolic signaling messengers[ 9 ]. Lipids were highly acquired in cancer cells, showing highly abnormal proliferation through lipid uptake enhancement, lipolysis, and fatty acid synthesis[ 10 ]. In tumor cells, abnormal lipid metabolism promotes malignant biological processes like migration, invasion, and angiogenesis[ 11 , 12 ]. Recently, various inhibitors that intervene in lipid synthesis have been investigated in preclinical studies and early clinical trials and showed anti-cancer effects [ 13 , 14 ]. However, mechanisms regulating lipid in cancer cells and the influence of lipid metabolism on the tumor microenvironment (TME) are incompletely understood. Thus, the development of cancer treatment by targeting altered lipid metabolism was minimal. In malignant gliomas, tumor tissues' lipid levels have been reported to be higher compared with normal tissues[ 15 ]. Revealing the molecular mechanism of altered lipid components involved in gliomas has become research hotspot. For example, extracellular lipid loading augments hypoxic paracrine signaling and stimulates the glioma malignancy process and immune cell infiltration[ 16 ]. Triglyceride- rich lipoproteins (TRLs) processing was facilitated, and a source of lipid nutrients for glioma cells was provided by GPIHBP1[ 17 ]. However, how lipid metabolism influence prognosis and the immune process in glioma progression need further systematic research. In this study, systematic and comprehensive research was conducted on the characteristics of the intersecting lipid metabolism- and immune-related genes in glioma. Firstly, the characteristics of clinical and molecular subtypes of gliomas samples could be well stratified by the intersecting genes. Furtherly, through the lasso regression algorithm and Cox regression analysis, a 16-gene risk signature was developed, and the association with the infiltration level and abundance of tumor-infiltrating immune cells (TIICs) was investigated. Finally, the signal pathways positively related to risk score were mainly involved in the immunosuppressive process identified by GSEA. Taken together, all the results might show a novel insight into the research of glioma malignancy and immunotherapy. Methods Data and samples collection We collected 698 glioma samples with whole-genome RNA-seq data and clinical information in the TCGA database ( http://cancergenome.nih.gov/ ), used as a training set. In addition, four hundred thirteen samples with complete follow-up information screened out of 693 samples were obtained from CGGA ( http://www.cgga.org.cn ) part B, which was used as a validation set. Patient samples The Institutional Ethics Committee approved this study of the Faculty of Medicine at Renmin Hospital of Wuhan University. Informed consent was obtained from all patients whose tissues were used. In total, 6 control samples from patients with cerebral hemorrhage and 24 lower-grade glioma samples (WHO grade 2–3), and 40 GBM samples were collected during May 2019 and April 2021. All patients were not treated with chemotherapy or radiotherapy before surgery. Screening of DEGs regarding the level of TIICs R package "estimate" was used to estimate the infiltrating level of TIICs in each sample. Kaplan–Meier curve was plotted by R package "survival" [ 18 ] (log-rank p < 0.05 was considered significant). All glioma samples were labeled with a high score or low score group by the median value of Immune Score, Stromal Score, and ESTIMATE Score. Differential expressed genes (DEGs) were identified regarding the high- vs. low-Immune and Stromal score groups by R Package "limma" [ 19 ] separately (False discovery rate (FDR) < 0.05 and the abs of fold change after transformation of log2 were larger than one considered significant. Acquisition of lipid metabolism-related genes Lipids metabolism-related gene set (REACTOME_METABOLISM_OF_LIPIDS) was obtained in Molecular Signatures Database ( http://www.broad.mit.edu/gsea/msigdb/ ). PPI network construction and Functional annotation of target genes Nodes with confidence of interactive relationship larger than 0.40 were obtained from the STRING database ( http://www.string-db.org/ ), then, the protein-protein interaction (PPI) network was reconstructed by Cytoscape software[ 20 ]. Finally, by R package "clusterProfiler" [ 21 ], then Kyoto Encyclopedia of Genes and Genomes(KEGG)pathways and Gene Ontology༈GO༉analysis were performed. (pathways with both p and q values were below 0.05 were considered significantly enriched). Consensus Clustering of Glioma Through the R package "Consensus Cluster Plus", k-means consensus clustering was carried out in698 glioma samples based on the expression profile of these 67 genes [ 22 ]. In addition, quantitative stability evidence determined the optimal number of the glioma clusters. Different Clinical Characteristics analysis among clusters The clinical and pathological characteristics data and Tumor immune cell infiltration levels corresponding to the glioma in TCGA were selected for correlation analysis with different molecular subtypes by Wilcoxon rank-sum or Kruskal–Wallis rank-sum test depending on the number of comparisons. The results were visualized as a heatmap by R package "pheatmap". Construction of risk signature Univariate COX regression analysis was performed by the "survival" package to select genes with prognostic value in glioma from 67 genes (p value < 0.05). The Least Absolute Shrinkage and Selection Operator (LASSO) regression algorithm[ 23 ] was used to formulate a risk signature (The penalty parameter λ was chosen based on a 10-fold cross-validation). Genes and their regression coefficients were obtained according to the minimum λ value. The formula was: Risk score = exprgene(1)×coefficientgene(1) + exprgene(2)×coefficientgene(2)+⋯+exprgene(n) ×coefficientgene(n). Prognosis analysis of risk signature Patients were separated into high- and low-risk groups in TCGA-glioma and CGGA cohorts by the median risk score. Then, evaluate the prognostic significance of the risk signature by Kaplan-Meier survival curves (performed through R package “survival”) in TCGA-glioma cohort and verified in CGGA cohort (log-rank test p-value < 0.05 was considered significant). Next, univariate and multivariate Cox regression analyses (performed through R packages “'survival” and “forestplot”) investigated independent prognostic factors, including the risk score in glioma. Subsequently, investigate the specificity and sensitivity of risk score in the prediction of 1-, 3- and 5-year overall survival (OS) by analyzing the receiver operating characteristic (ROC) curve [ 24 ]. Student's t or chi-squared tests tested differences in clinicopathological features and the level of TIICs of risk score. Gene Set Enrichment Analysis In the Molecular Signatures Database, Hallmark and C7 gene sets were downloaded, which were used as the target gene sets to investigate the gene sets associated with risk score in the whole transcriptome of all glioma samples in TCGA performed by the software GSEA-3.0. (NOM p < 0.05 and FDR q < 0.05 were considered significant). Immune infiltration Profile The abundance of TIICs was estimated using CIBERSORT[ 25 ] computational method in low and high-risk groups separately. Pearson correlation analysis and the Wilcoxon test investigate the correlation between TIICs and risk scores in glioma. In addition, for predicting the efficacy of immunotherapy, the correlation between risk score and the expression level of immune checkpoints, including PD-1 and PD-L1, was investigated by the Wilcoxon test. Results The research flow is shown in Fig. 1 . Identification of 67 immune- and lipid metabolism-related genes Excluding the cases with incomplete clinical information and overall survival time less than 3 months, 598 Cases were in TCGA and 657 cases in CGGA. Samples with higher immune infiltrating levels, including Stromal, Immune, and ESTIMATE score, had worse overall survival (P-value < 0.05, Figure S1 A-C). The results showed that it is reasonable for the immune components in TME of glioma to indicate glioma patients' prognosis. A total of 4033 DEGs were obtained among high and low Immune Score groups, and 4629 DEGs were identified between different Stromal Score groups. The top 50 DEGs in immune score and stromal score goups were respectively shown in figure S1 D, E. Then, we obtained the 67 intersecting genes among lipid metabolism related genes, DEGs of immune score and DEGs of stromal score (Fig. 2 A). GO and KEGG analysis results confirmed that these genes are involved in lipid metabolism-related biological processes and pathways. PPI networks also showed that the 67 genes were closely related (Fig. 2 B-D). Consensus clustering of glioma samples Two robust clusters with clustering stability increasing between k = 2–9 were determined by k-means consensus clustering of the samples in TCGA-glioma (Fig. 3 A-C). We have also explored the stability of subtyping by heirarchical clustering and partitioning around medoids (figure S2 ), and found that K-means can achieve better grouping performance. PCA's (principal component analysis) results showed that different clusters have significantly different components when k = 2 (Fig. 3 D). Overall survival analysis showed that the Cluster1 subgroup predicted a better prognosis in glioma patients than the Cluster2 subgroup (P-value < 0.05, Fig. 3 E). Furthermore, we observed that striking differences were determined by consensus clustering in the clinical and molecular features of the two glioma subgroups (Fig. 3 F). In the TCGA database, Cluster2 was significantly correlated with higher Stromal Score, higher Immune Score, 1p19q non-codeletion, IDH wild type, older age group (> 45 years old), and higher WHO grade. Moreover, we have validated these results in CGGA cohorts (figure S3 ). These results showed that the clustering based on 67 could distinguish the clinical, pathological, and immune characteristics of glioma samples and predict the prognosis of patients with glioma. Construction of a 16-gene risk signature Fifty-two genes with prognostic significance were selected through univariate cox regression analysis (P-value < 0.05, Fig. 4 A). Then, 16 genes were generated by LASSO regression analysis, including PIK3R6, PLIN2, CYP2E1, PLA2G5, PLBD1, FABP5, PON1, HEXB, CYP17A1, MTMR7, ARSJ, HILPDA, PPARG, FABP7, ANGPTL4, and ACSL6 (Fig. 4 B-D). The risk score of each sample was then calculated using the coefficients and expression value of these 16 genes through the formula listed above. Then by the median value of risk score, all glioma samples are divided into two groups (low and high risk group). There was a significant difference between the two groups about the correlation between risk score and clinicopathological factors of glioma patients in the TCGA and CGGA cohorts, the clinicopathological characteristics and corresponding number of patients were shown in Table 1 . Table 1 Validation set TCGA RNA-seq cohort (n = 587) Training set CGGA RNA-seq cohort (n = 309) Features Low-risk Score(n = 297) High-risk Score(n = 290) P-value Low-risk Score(n = 206) High-risk Score(n = 207) P-value Age < 0.001 < 0.001 45 108 198 74 92 Gender 0.62 0.44 Female 131 115 96 86 Male 166 175 110 121 WHO Grade < 0.001 < 0.001 II 169 42 72 126 III 127 107 104 57 IV 1 141 30 24 IDH status < 0.001 < 0.001 WT 12 207 168 147 Mut 285 83 38 60 1p/19q status < 0.001 < 0.001 Codel 140 9 72 15 Noncodel 157 281 134 192 Difference analysis of risk score among clinicopathological subtypes Glioblastoma had a higher risk score than lower grade gliomas in the heatmap. Risk score was positively associated with immune and stromal score, 1p/19q non-codeletion, IDH wildtype, and > 45 age group in the training set (Fig. 4 E). Furthermore, we classified the patients into different groups according to the different clinical and pathological features of samples. As shown in Fig. 5 and Figure S4 , in > 45 age, higher grade, wildtype IDH, and non-codeletion 1p19q groups, glioma samples had higher risk scores (p < 0.001). However, there was no significant difference of risk score among different gender in both training and validation cohorts. In addition, in the TCGA cohort, patients in cluster 2 had higher risk scores (p < 0.001). In the CGGA-B cohort, samples with unmethylated MGMT promoter and accepting radiotherapy had a higher risk score (p 0.05). Prognostic value of risk signature constructed In TCGA, patients with higher risk scores had a significantly poor prognosis as shown in the Kaplan-Meier curve performed among all low and high grade gliomas (P-value < 0.05, Fig. 6 A-C), and the area under ROC curves for predicting overall survival were 0.838 (Fig. 6 D). Consistently, patients with higher risk scores were also correlated to worse clinical outcomes in gliomas in CGGA (Fig. 6 E-G). The AUC for predicting survival time in the CGGA dataset is 0.848 (Fig. 6 H). In addition, the risk score was demonstrated as an independent prognostic factor of prognosis in both TCGA and CGGA cohorts by Cox regression analysis (Table 2 ). These results suggested that the immune- and lipid metabolism-related risk signature could be a prognostic predictor for gliomas with relatively high sensitivity and specificity. Table 2 Univariate analysis Multivariate analysis Variables HR 95%CI P-value HR 95%CI P-value Grade 4.986 3.873–6.421 P < 0.001 1.993 1.456–2.729 P < 0.001 Gender 1.011 0.747–1.368 0.944 0.934 0.682–1.278 0.669 Age 4.863 3.391–6.975 P < 0.001 2.178 1.380–3.439 P < 0.001 IDH mutation status 0.090 0.063–0.129 P < 0.001 0.332 0.194–0.569 P < 0.001 1p19q codeletion status 0.217 0.128–0.370 P < 0.001 0.572 0.306–1.071 0.081 Risk Score 1.269 1.225–1.314 P < 0.001 1.109 1.051–1.171 P < 0.001 Correlation of Risk score with the Proportion of TIICs Twenty-one kinds of immune cell profiles in low (n = 349) and high (n = 349) risk glioma samples were constructed(Fig. 7 A-D). Macrophages M2 (the median value of the component proportion was 33.89%), Monocytes (16.33%), and T cells CD4 memory resting (13.08%) were the prominent TIICs in the high-risk group. T cells regulatory (Tregs) were positively correlated with Macrophages M0 (rho = 0.47, p < 0.05). Meanwhile, Monocytes were significantly negatively correlated with macrophages M0 (r=-0.79, p < 0.05). In the low-risk score group, the Macrophages M2 (26.65%), Monocytes (26.45%), and T cells CD4 memory resting (11.13%) were the prominent TIICs. As the Fig. 7 E-I shown, the abundance of T cells regulatory (Tregs) (r = 0.25, p < 0.001) and Macrophages M2 (r = 0.32, p < 0.001) were significantly positively correlated with risk score. In contrast, the abundance of NK cells activated (r=-0.37, p < 0.001) and T cells CD4 memory resting (r=-0.16, p < 0.05) were significantly negatively correlated with risk score. We verified these results in the CGGA cohort (figure S5 ). These results further indicated that the immune activity in TME of glioma might be associated with the level of risk score. Risk score could predict the effect of immunotherapy As Fig. 8 A-F shown, GSEA analyses were carried out for investigating the immune signature gene sets related to the samples in TCGA-glioma, indicating that the gene sets "Genes up-regulated in myeloid-derived suppressor cells from colon tumors ", "Genes down-regulated in activated T lymphocytes " and "Genes up-regulated in CD4 T cells " were positively associated with the high-risk group, while the gene sets "Genes up-regulated in comparison of naive CD4 T cells versus CD4 and effector memory T cells ", "Genes up-regulated in comparison of unstimulated NK cells versus those stimulated with IL2" and "Genes down-regulated in comparison of wild type CD8 effector T cells " were negatively related to the high-risk group. According to the ranking of NES values from high to low, the top five genes sets positively enriched in high-risk score group and top five terms positively enriched in low-risk group were shown in Fig. 8 G. Then, the expression level of immune checkpoints, including PD-1 and PD-L1, was significantly higher in the high risk group (P-value < 0.05, Fig. 8 H, I), the expression of immune checkpoints is essential for immune escape and treatment with therapies. These results indicated that anti-PD-1 or PD-L1 immunotherapy might have a better therapeutic effect in glioma patients with a higher risk score. Discussion Metabolomic associated studies have provided new molecular mechanisms into the malignant process of brain tumors[ 5 , 6 ]. The lipid compositions of tissues are altered in malignant gliomas compared with normal brain tissues, particularly with the exclusive presence of cholesterol esters and significant elevation of unsaturated fatty acids and phosphatidylcholine in tumor tissues[ 25 , 26 ]. Previous studies have identified glucose-related and amino acid-related risk signatures for the biological process in glioma and the survival of glioma patients through bioinformatic profiling. The characteristics of novel lipid metabolism-related biomarkers need a systematical understanding of glioma. Our study first identified a lipid metabolism-related and immune-associated risk signature, which had instructional significance in the differentiation of malignancy degree of tumor and prognosis judgment. A risk score is used for meaningful signature construction as a standard approach [ 22 ]. Then, we constructed a 16-gene risk signature to connect lipid metabolism, immune microenvironment, and prognosis of glioma. The 16-gene signature could predict glioma's prognosis in different subgroups and distinguish the status of IDH mutation type and 1p/19q co-deletion among tumors. Furthermore, the signature was furtherly confirmed as an independent risk factor for poor prognosis of patients with glioma by performing a cox regression analysis. As a novel biomarker in glioma, risk score may be a better prognostic indictor compared to single vital clinical factor like age, WHO grade, which were correlated with the prognosis of patients[ 27 ]. Previous studies have shown that the seven factors including age, histopathology, IDH status, 1p/19q status, radiotherapy, chemotherapy and recurrence were identified as independent prognostic factors for high grade glioma patients[ 28 ]. In this study, we confirmed that the lipid metabolism related risk score could also be used as an independent risk factor in gliomas. Moreover, the area under the receiver operating characteristic curve values for overall survival probabilities was 0.838, which is close to the prediction efficiency of previous autophagy related prognosis models[ 29 ]. Accumulating shreds of evidence suggest that lipid metabolism regulates the tumor-associated immune cells. The activation program of M1 macrophages tends to utilize glycolysis. However, the M2 macrophages mainly resort to fat oxidation (FAO)[ 30 , 31 ]. T cells were also influenced by lipid metabolism, including differentiation, activation, and migration. Thus, lipid metabolism could also play an essential role in modulating T cell-mediated immunity. In this study, the risk score was correlated to the abundance of M2 macrophages and T cells (including CD4 and CD8), which indicated that the signature, an immune-associated biomarker, could distinguish between the anti-tumor promoting-tumor escape of characteristics in TIME. Moreover, the correlation of lipid-metabolism-related risk signature and distribution of TIICs furtherly provides the potential relationship between lipid metabolism and the immune microenvironments in gliomas. Samples with higher risk scores are often associated with higher levels of immune infiltration, which contains mainly inhibitory immune components, such as M2 macrophages. It may be one of the reasons for the poor prognosis of patients in the high-risk group. Immune checkpoint inhibitor therapy is emerging as a valid treatment for diverse solid tumors, including anti-PD-1/PD-L1. However, the therapeutic efficacy in glioma largely depends on the activity of tumor-killing immune cells and the expression level of an immune checkpoint, and optimal outcomes have been demonstrated by various preclinical studies[ 32 ]. Different genomic subtypes or molecular profiles are the main challenges in response to PD-1/PD-L1 checkpoint blockades[ 33 ]. Therefore, we detected the expression of immune checkpoints in both low-risk and high-risk groups, respectively. Compared with the high-risk group, the expression level of PD-L1 and PD-1 were significantly decreased in the low-risk group. These results indicated a better efficacy and more sensitivity of immune checkpoint inhibitors (ICIs) therapy for patients with higher risk scores in glioma. However, resistance to ICIs therapy was attributed to macrophage polarization[ 34 ]. Due to the high proportion of macrophages in the high-risk group, inducing macrophages to differentiate into M1 tumor-killing types may provide us with new ideas. In addition, with the help of mediators and cell adhesion molecules, it is necessary to promote the homing of peripheral CD4 and CD8 T cells across the blood-brain barrier to improve the level of tumor immune cells infiltration, ultimately improve the therapeutic effect of ICIs. Consequently, we have identified and validated a novel and reliable risk signature, which could accurately predict the effect of ICIs treatment and provide appropriate treatment strategies according to the risk score of different populations with glioma, considering the level of two vital immune checkpoints and the characteristics of immune microenvironment. However, certain limitations should be noted in the present work. Firstly, the present study conducted independent external validation, yet all the changes across glioma cases derived from diverse geographical regions might not be covered, since all information and tissues were retrospectively obtained from public databases. Secondly, there are different microenvironment features among various tumor sites, like the invasion margin or the tumor core. Thirdly, the expression of lipid metabolism-related markers could be different among immune cells and tumor cells, making it difficult to assess the lipid-metabolism and immune status among diverse cells in gliomas. As a result, our results should be further validated in the well-designed, multicenter, prospective studies (including conducting next-generation sequencing for glioma tissues), and the expression pattern of lipid metabolism-related genes may also need to be investigated in the single-cell RNA sequence data. Conclusion In conclusion, we identified that the lipid metabolism-related gene set could distinguish gliomas' clinical and molecular features. A 16-genes risk signature developed related to immunity and lipid metabolism was confirmed as a novel biomarker that could predict the prognosis and the effect of immunotherapy for patients with gliomas. Abbreviations LGG lower-grade glioma WHO World Health Organization TIME tumor immune microenvironment TME tumor microenvironment TCGA The Cancer Genome Atlas CGGA Chinese Glioma Genome Atlas OS overall survival TIICs tumor immune infiltrating cells ICIs Immune checkpoints inhibitors GO Gene Ontology KEGG Kyoto Encyclopedia of Genes and Genomes PPI protein-protein interaction Fig figure Tab table. Declarations Consent for publication Not applicable. Availability of data and materials Publicly available datasets were analyzed in this study. This data can be found below: TCGA, https://www.cancer.gov/; CGGA, http://www.cgga.org.cn/; STRING, https://string-db.org/cgi/input.pl Competing interest The authors declare that they have no conflict of interest. Funding This work was supported by the National Natural Science Foundation of China (No.82072764). Authors' contributions QX Chen and LG Ye contributed to the conception and design of this study. SW Liang and XY Zhang contributed to the analysis and interpretation of data. All authors read and approved the final manuscript. Acknowledgments We gratefully acknowledge The Cancer Genome Atlas pilot project, which made the genomic data and clinical data of glioma available. References Ostrom QT, Patil N, Cioffi G, Waite K, Kruchko C, Barnholtz-Sloan JS. CBTRUS Statistical Report: Primary Brain and Other Central Nervous System Tumors Diagnosed in the United States in 2013–2017. Neuro Oncol. 2020;22:iv1–96. Aldape K, Brindle KM, Chesler L, Chopra R, Gajjar A, Gilbert MR, et al. Challenges to curing primary brain tumours. Nat Rev Clin Oncol. 2019;16:509–20. Stupp R, Taillibert S, Kanner A, Read W, Steinberg D, Lhermitte B, et al. Effect of Tumor-Treating Fields Plus Maintenance Temozolomide vs Maintenance Temozolomide Alone on Survival in Patients With Glioblastoma: A Randomized Clinical Trial. JAMA. 2017;318:2306–16. Sampson JH, Gunn MD, Fecci PE, Ashley DM. Brain immunology and immunotherapy in brain tumours. Nat Rev Cancer. 2020;20:12–25. Ahmed KA, Chinnaiyan P. Applying metabolomics to understand the aggressive phenotype and identify novel therapeutic targets in glioblastoma. Metabolites. 2014;4:740–50. Griffin JL, Kauppinen RA. A metabolomics perspective of human brain tumours. FEBS J. 2007;274:1132–9. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011;144:646–74. Pavlova NN, Thompson CB. The Emerging Hallmarks of Cancer Metabolism. Cell Metab. 2016;23:27–47. van Meer G, Voelker DR, Feigenson GW. Membrane lipids: where they are and how they behave. Nat Rev Mol Cell Biol. 2008;9:112–24. Beloribi-Djefaflia S, Vasseur S, Guillaumond F. Lipid metabolic reprogramming in cancer cells. Oncogenesis. 2016;5:e189. Fisher KE, Pop A, Koh W, Anthis NJ, Saunders WB, Davis GE. Tumor cell invasion of collagen matrices requires coordinate lipid agonist-induced G-protein and membrane-type matrix metalloproteinase-1-dependent signaling. Mol Cancer. 2006;5:69. Luo X, Cheng C, Tan Z, Li N, Tang M, Yang L, et al. Emerging roles of lipid metabolism in cancer metastasis. Mol Cancer. 2017;16:76. Geng F, Guo D. Lipid droplets, potential biomarker and metabolic target in glioblastoma. Intern Med Rev (Wash D C). 2017;3. Cheng C, Ru P, Geng F, Liu J, Yoo JY, Wu X, et al. Glucose-Mediated N-glycosylation of SCAP Is Essential for SREBP-1 Activation and Tumor Growth. Cancer Cell. 2015;28:569–81. Ward PS, Thompson CB. Metabolic reprogramming: a cancer hallmark even warburg did not anticipate. Cancer Cell. 2012;21:297–308. Offer S, Menard JA, Pérez JE, de Oliveira KG, Chandran VI, Johansson MC, et al. Extracellular lipid loading augments hypoxic paracrine signaling and promotes glioma angiogenesis and macrophage infiltration. J Exp Clin Cancer Res. 2019;38:241. Hu X, Matsumoto K, Jung RS, Weston TA, Heizer PJ, He C et al. GPIHBP1 expression in gliomas promotes utilization of lipoprotein-derived nutrients. Elife. 2019;8. Upton A, Arvanitis TN. Using evolutional properties of gene networks in understanding survival prognosis of glioblastoma. IEEE J Biomed Health Inf. 2014;18:810–6. Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015;43:e47. Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, et al. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 2003;13:2498–504. Yu G, Wang L-G, Han Y, He Q-Y. clusterProfiler: an R package for comparing biological themes among gene clusters. OMICS. 2012;16:284–7. Liu Y-Q, Chai R-C, Wang Y-Z, Wang Z, Liu X, Wu F, et al. Amino acid metabolism-related gene expression-based risk signature can better predict overall survival for glioma. Cancer Sci. 2019;110:321–33. Gao J, Kwan PW, Shi D. Sparse kernel learning with LASSO and Bayesian inference algorithm. Neural Netw. 2010;23:257–64. Heagerty PJ, Lumley T, Pepe MS. Time-dependent ROC curves for censored survival data and a diagnostic marker. Biometrics. 2000;56:337–44. Srivastava NK, Pradhan S, Gowda GAN, Kumar R. In vitro, high-resolution 1H and 31P NMR based analysis of the lipid components in the tissue, serum, and CSF of the patients with primary brain tumors: one possible diagnostic view. NMR Biomed. 2010;23:113–22. Tosi MR, Tugnoli V. Cholesteryl esters in malignancy. Clin Chim Acta. 2005;359:27–45. Hughey JJ, Butte AJ. Robust meta-analysis of gene expression using the elastic net. Nucleic Acids Res. 2015;43:e79. Qu S, Qiu O, Hu Z. The prognostic factors and nomogram for patients with high-grade gliomas. Fundamental Research [Internet]. 2021;1:824–8. https://www.sciencedirect.com/science/article/pii/S2667325821001345 . Qu S, Liu S, Qiu W, Liu J, Wang H. Screening of autophagy genes as prognostic indicators for glioma patients. Am J Transl Res [Internet]. e-Century Publishing Corporation; 2020;12:5320–31. https://pubmed.ncbi.nlm.nih.gov/33042422 . Huang SC-C, Everts B, Ivanova Y, O’Sullivan D, Nascimento M, Smith AM, et al. Cell-intrinsic lysosomal lipolysis is essential for alternative activation of macrophages. Nat Immunol. 2014;15:846–55. Pearce EL, Pearce EJ. Metabolic pathways in immune cell activation and quiescence. Immunity. 2013;38:633–43. Cloughesy TF, Mochizuki AY, Orpilla JR, Hugo W, Lee AH, Davidson TB, et al. Neoadjuvant anti-PD-1 immunotherapy promotes a survival benefit with intratumoral and systemic immune responses in recurrent glioblastoma. Nat Med. 2019;25:477–86. Berghoff AS, Kiesel B, Widhalm G, Rajky O, Ricken G, Wöhrer A, et al. Programmed death ligand 1 expression and tumor-infiltrating lymphocytes in glioblastoma. Neuro Oncol. 2015;17:1064–75. Soto-Pantoja DR, Wilson AS, Clear KY, Westwood B, Triozzi PL, Cook KL. Unfolded protein response signaling impacts macrophage polarity to modulate breast cancer cell clearance and melanoma immune checkpoint therapy responsiveness. Oncotarget [Internet]. Impact Journals LLC; 2017;8:80545–59. https://pubmed.ncbi.nlm.nih.gov/29113324 . Additional Declarations No competing interests reported. Supplementary Files SupplementaryMaterial.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4408953","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":306375256,"identity":"bec24a0e-c89c-48f5-b343-efaee22244af","order_by":0,"name":"Shanwen Liang","email":"","orcid":"","institution":"Renmin Hospital of Wuhan University","correspondingAuthor":false,"prefix":"","firstName":"Shanwen","middleName":"","lastName":"Liang","suffix":""},{"id":306375257,"identity":"f68a25f9-d365-4771-9901-a2ddbdb77f85","order_by":1,"name":"Xinin Zhang","email":"","orcid":"","institution":"Renmin Hospital of Wuhan University","correspondingAuthor":false,"prefix":"","firstName":"Xinin","middleName":"","lastName":"Zhang","suffix":""},{"id":306375258,"identity":"b01901b7-a73b-4f59-92d7-0f18ad0c753c","order_by":2,"name":"Zhansheng Zhu","email":"","orcid":"","institution":"Renmin Hospital of Wuhan University","correspondingAuthor":false,"prefix":"","firstName":"Zhansheng","middleName":"","lastName":"Zhu","suffix":""},{"id":306375259,"identity":"753ec728-e57c-42e4-8471-7d71e46f6e65","order_by":3,"name":"Yu Hong","email":"","orcid":"","institution":"Renmin Hospital of Wuhan University","correspondingAuthor":false,"prefix":"","firstName":"Yu","middleName":"","lastName":"Hong","suffix":""},{"id":306375263,"identity":"e1346c87-9e10-4d66-a175-b5af5a604d0e","order_by":4,"name":"Yangzhi Qi","email":"","orcid":"","institution":"Renmin Hospital of Wuhan University","correspondingAuthor":false,"prefix":"","firstName":"Yangzhi","middleName":"","lastName":"Qi","suffix":""},{"id":306375269,"identity":"99604a05-e4e9-484a-a8b0-53510a10042e","order_by":5,"name":"Liguo Ye","email":"","orcid":"","institution":"Renmin Hospital of Wuhan University","correspondingAuthor":false,"prefix":"","firstName":"Liguo","middleName":"","lastName":"Ye","suffix":""},{"id":306375270,"identity":"0d51b932-c689-4ec6-86b5-62e6a9b89f82","order_by":6,"name":"Qianxue Chen","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA20lEQVRIiWNgGAWjYDACCTBpI8fP3sAGYjE2EKklzViy5wBpWg4lGsxIIFIL/+zmYw/eth1IMJB8Y/aYh8FGdsMB5mcP8Fpy51i64Zwzd/LMpXPMjXmALtxwgM3cAJ8WA4kcM2meimfFlrNzt0nzMBxO3HCAh00Cv5b8b9I8BkCVN8+CtPwnRksOG9AWoJYbvCAtBwhrkbiRZiY55wwokPO/Sc4xSDaeeZjNDK8W/hnJzyTetoGi8liaxJsKO9m+483P8GoBAx6EO4GYmaB6FC2jYBSMglEwCrAAAOa+Rr5tnnRdAAAAAElFTkSuQmCC","orcid":"","institution":"Renmin Hospital of Wuhan University","correspondingAuthor":true,"prefix":"","firstName":"Qianxue","middleName":"","lastName":"Chen","suffix":""}],"badges":[],"createdAt":"2024-05-12 15:08:38","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4408953/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4408953/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":57517940,"identity":"77e50f3f-0bd9-4d0c-9b88-64fd213bb0d8","added_by":"auto","created_at":"2024-05-31 20:25:57","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":596861,"visible":true,"origin":"","legend":"\u003cp\u003eThe flowchart of this reseach.\u003c/p\u003e","description":"","filename":"Figure1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4408953/v1/080deb0c155d0e55d894ca3b.jpg"},{"id":57517942,"identity":"6d54bdad-a8b6-4e6f-9185-4406920d8494","added_by":"auto","created_at":"2024-05-31 20:25:57","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1479848,"visible":true,"origin":"","legend":"\u003cp\u003eIdentification of genes among lipids metabolism and immune-related genes. (A) The Venn diagram of genes. The pink circle represents genes related to lipids metabolism. The yellow circle represents the differentially expressed genes among low and high immune score groups, and the blue circle represents the differentially expressed genes among low and high stromal score groups. (B) GO enrichment analysis for 67 DEGs, terms with p- and q-value \u0026lt; 0.05 were considered enriched significantly. (C) Protein-protein interact network of 67 genes. (D) In the bubble plot of KEGG enrichment analysis, the size of the spots represents the gene number. The darkness of color represents the level of adjusted p-value.\u003c/p\u003e","description":"","filename":"Figure2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4408953/v1/99e02d9c832a3584d45ebc78.jpg"},{"id":57517943,"identity":"3fb3f6f4-2cbd-4840-9ba8-8a1c9cc877dd","added_by":"auto","created_at":"2024-05-31 20:25:57","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":988311,"visible":true,"origin":"","legend":"\u003cp\u003eThe 67 genes could classify the clinical and molecular features of gliomas. (A, B) Consensus clustering matrix of TCGA samples for k = 2 and k = 3. (C) Consensus clustering CDF for k = 2 to k = 9. (D) Principal Component Analysis (PCA) in glioma when K=2. (E) Survival analysis of Cluster 1 and Cluster 2 subgroups in training datasets. (F) Heatmap for clinicopathological features of the two clusters defined by the 67 genes. CDF, cumulative distribution function; Codel, codeletion; IDH, isocitrate dehydrogenase; Noncodel, non-codeletion; OS, overall survival.\u003c/p\u003e","description":"","filename":"Figure3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4408953/v1/6f8f91ac67d7edafaf2dc763.jpg"},{"id":57518422,"identity":"6b606318-5ece-4d3e-ad1b-0f6099d540c1","added_by":"auto","created_at":"2024-05-31 20:33:57","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1349494,"visible":true,"origin":"","legend":"\u003cp\u003eIdentification of 16-gene risk signature by LASSO regression analysis in TCGA datasets. (A) The forest plot of the univariate COX analysis for 67 genes showed that only the genes with p value\u0026lt; 0.001 were shown. (B) Partial likelihood deviance as a function of regularization parameter λ in the training dataset. Each red point marks a λ value along regularization paths, and gray error bars represent confidence intervals for the cross-validated error rate. The left vertical dotted line marks the minimum error, whereas the right vertical dotted line marks the most significant λ value, the error of which is within 1 standard deviation of the minimum. The horizontal row of numbers above the plot marks the gene number in each condition upon shrinkage and selection based on linear regression. (C) The craft plot for Partial likelihood deviance in LASSO. (D) Results of 16 genes selected and their regression coefficients by LASSO are shown. (E) The heat map shows the association of risk score and clinicopathological features based on the 16-gene risk signature. LASSO, Least Absolute Shrinkage and Selection Operator.\u003c/p\u003e","description":"","filename":"Figure4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4408953/v1/510b631a89d0d13b998499a1.jpg"},{"id":57517949,"identity":"e8fb163d-4b98-40e1-92bb-8a0d708fb2d6","added_by":"auto","created_at":"2024-05-31 20:25:57","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":694579,"visible":true,"origin":"","legend":"\u003cp\u003eAssociations between the risk signature and clinical features in TCGA and CGGA-B cohorts. (A) Distribution of the risk score in patients stratified by age group, (B)WHO grade, (C) IDH status, and (D) 1p/19q co-deletion status in TCGA. (E) Distribution of the risk score in patients stratified by age group, (F) WHO grade, (G) IDH status, and (H) by 1p/19q status in CGGA-B. * P \u0026lt; 0.05; ** P \u0026lt; 0.01; *** P \u0026lt; 0.001, WHO, World Health Organization.\u003c/p\u003e","description":"","filename":"Figure5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4408953/v1/d21ec565f257db026b8c22cc.jpg"},{"id":57517946,"identity":"5ff59f6f-9749-4699-aaf6-e156a4f7d192","added_by":"auto","created_at":"2024-05-31 20:25:57","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1855153,"visible":true,"origin":"","legend":"\u003cp\u003ePrognostic significance of the risk signature in different subtypes. (A) Prognosis efficiency of the risk signature in all grades, (B) lower grade (WHO II), and (C) in higher grade (WHOIII, IV) in TCGA. (D) ROC curves indicate the sensitivity and specificity of predicting 1-, 3- and 5-year survival with the risk signature in the TCGA datasets. (E) Prognosis efficiency of the risk signature in all grades, (F) in a lower grade (WHO II), and (G) in higher grade (WHOIII, IV) in CGGA. (H) ROC curves indicate the sensitivity and specificity of predicting 1-, 3- and 5-year survival with the risk signature in the CGGA datasets.\u003c/p\u003e","description":"","filename":"Figure6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4408953/v1/0ffad24fe111d14422e73350.jpg"},{"id":57517948,"identity":"8ebb0901-2b14-4340-bf8c-436a80ebb139","added_by":"auto","created_at":"2024-05-31 20:25:57","extension":"jpg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":783345,"visible":true,"origin":"","legend":"\u003cp\u003eTIICs profiling. (A) Barplot showed the proportion of 21 kinds of TIICs in low-risk glioma samples. The column names of the plot were sample ID. (B) Heatmap showed the correlation between 21 kinds of TIICs in the low-risk group. Numeric in each tiny box indicated the p-value of correlation between two kinds of cells. The shade of each tiny color box represented a corresponding correlation value between two cells. (C) The proportion of 21 kinds of TIICs in high-risk glioma samples. (D) The correlation analysis between 21 kinds of TIICs in the high-risk group. (E) The Violin plot showed the ratio differentiation of 21 immune cells between glioma samples with low or high risk scores. (F-I) The scatter plot showed the correlation of TIICs proportion with the risk score (p \u0026lt; 0.05). The blue line in each plot was a fitted linear model indicating the proportion tropism of the immune cell along with risk score, and Pearson coefficient was used for the correlation test.\u003c/p\u003e","description":"","filename":"Figure7.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4408953/v1/cbaf096769c522ea5d5b140a.jpg"},{"id":57517947,"identity":"1096d105-dacf-4318-a4a6-84f704b5ffdc","added_by":"auto","created_at":"2024-05-31 20:25:57","extension":"jpg","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":1291417,"visible":true,"origin":"","legend":"\u003cp\u003eGene set enrichment analysis (GSEA) of lipid metabolism related-risk signature. GSEA results showing gene sets in (A, B, and C) were negatively correlated with the riskScore. In (D, E, and F) were positively correlated with the risk Score. (G) summarizes the above top ten gene sets. (H) The expression level of PD-1 in low and high-risk glioma samples in TCGA. (I) The expression level of PD-L1 in low and high-risk glioma samples in TCGA. Wilcoxon rank test was performed, *P \u0026lt; 0.05, **P \u0026lt; 0.01, ***P \u0026lt; 0.001.\u003c/p\u003e","description":"","filename":"Figure8.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4408953/v1/29d18c64432ceaf17502eb0e.jpg"},{"id":60556179,"identity":"1e1c341a-31aa-4028-a256-e551181c8eea","added_by":"auto","created_at":"2024-07-18 06:32:05","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":9661206,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4408953/v1/0aa7c7da-f35e-4774-aa4c-c9daa8193c23.pdf"},{"id":57517944,"identity":"61f82d93-b79f-4a04-b076-a9b150fff228","added_by":"auto","created_at":"2024-05-31 20:25:57","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":2486866,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterial.docx","url":"https://assets-eu.researchsquare.com/files/rs-4408953/v1/4bf0246d16e467f2f69d7c82.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"\u003cp\u003eA lipid metabolism and immune-related gene signature for predicting the prognosis of human gliomas \u003c/p\u003e","fulltext":[{"header":"Background","content":"\u003cp\u003eGliomas are the most common malignant tumor among brain tumors (24.1% of all tumors). Moreover, glioblastoma accounts for 60% of gliomas[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. Although standard treatment, including surgical resection, targeted radiation therapy, high-dose chemotherapy, has made tremendous development[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e], the median survival for glioblastoma for all patients (regardless of treatment) remained eight months[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. Immunotherapy and tumor-treating fields (TTFields)[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e] have developed fast in recent years, but there was still no significant breakthrough. Also, more and more research were explored in metabolomic reprogramming, providing novel insights into malignant phenotypes of gliomas[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eMetabolic reprogramming is recognized as an essential characteristic of cancer cells[\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. High rates of aerobic glycolysis and increased anabolism are prominent features of malignant tumor support[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Meanwhile, multiple metabolism-related pathways have been involved in the research of cancer. As one of the most significant metabolisms in cancer cells, lipid metabolism dysfunction has recently been more and more explored. As a highly diverse class of biological molecules, lipids exert various cells, including energy storage, cellular membrane structures, and metabolic signaling messengers[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. Lipids were highly acquired in cancer cells, showing highly abnormal proliferation through lipid uptake enhancement, lipolysis, and fatty acid synthesis[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. In tumor cells, abnormal lipid metabolism promotes malignant biological processes like migration, invasion, and angiogenesis[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eRecently, various inhibitors that intervene in lipid synthesis have been investigated in preclinical studies and early clinical trials and showed anti-cancer effects [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. However, mechanisms regulating lipid in cancer cells and the influence of lipid metabolism on the tumor microenvironment (TME) are incompletely understood. Thus, the development of cancer treatment by targeting altered lipid metabolism was minimal. In malignant gliomas, tumor tissues' lipid levels have been reported to be higher compared with normal tissues[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Revealing the molecular mechanism of altered lipid components involved in gliomas has become research hotspot. For example, extracellular lipid loading augments hypoxic paracrine signaling and stimulates the glioma malignancy process and immune cell infiltration[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Triglyceride- rich lipoproteins (TRLs) processing was facilitated, and a source of lipid nutrients for glioma cells was provided by GPIHBP1[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. However, how lipid metabolism influence prognosis and the immune process in glioma progression need further systematic research.\u003c/p\u003e \u003cp\u003eIn this study, systematic and comprehensive research was conducted on the characteristics of the intersecting lipid metabolism- and immune-related genes in glioma. Firstly, the characteristics of clinical and molecular subtypes of gliomas samples could be well stratified by the intersecting genes. Furtherly, through the lasso regression algorithm and Cox regression analysis, a 16-gene risk signature was developed, and the association with the infiltration level and abundance of tumor-infiltrating immune cells (TIICs) was investigated. Finally, the signal pathways positively related to risk score were mainly involved in the immunosuppressive process identified by GSEA. Taken together, all the results might show a novel insight into the research of glioma malignancy and immunotherapy.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eData and samples collection\u003c/h2\u003e \u003cp\u003eWe collected 698 glioma samples with whole-genome RNA-seq data and clinical information in the TCGA database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://cancergenome.nih.gov/\u003c/span\u003e\u003cspan address=\"http://cancergenome.nih.gov/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), used as a training set. In addition, four hundred thirteen samples with complete follow-up information screened out of 693 samples were obtained from CGGA (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.cgga.org.cn\u003c/span\u003e\u003cspan address=\"http://www.cgga.org.cn\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) part B, which was used as a validation set.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003ePatient samples\u003c/h2\u003e \u003cp\u003e The Institutional Ethics Committee approved this study of the Faculty of Medicine at Renmin Hospital of Wuhan University. Informed consent was obtained from all patients whose tissues were used. In total, 6 control samples from patients with cerebral hemorrhage and 24 lower-grade glioma samples (WHO grade 2\u0026ndash;3), and 40 GBM samples were collected during May 2019 and April 2021. All patients were not treated with chemotherapy or radiotherapy before surgery.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eScreening of DEGs regarding the level of TIICs\u003c/h2\u003e \u003cp\u003eR package \"estimate\" was used to estimate the infiltrating level of TIICs in each sample. Kaplan\u0026ndash;Meier curve was plotted by R package \"survival\" [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e] (log-rank p\u0026thinsp;\u0026lt;\u0026thinsp;0.05 was considered significant). All glioma samples were labeled with a high score or low score group by the median value of Immune Score, Stromal Score, and ESTIMATE Score. Differential expressed genes (DEGs) were identified regarding the high- vs. low-Immune and Stromal score groups by R Package \"limma\" [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e] separately (False discovery rate (FDR)\u0026thinsp;\u0026lt;\u0026thinsp;0.05 and the abs of fold change after transformation of log2 were larger than one considered significant.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eAcquisition of lipid metabolism-related genes\u003c/h2\u003e \u003cp\u003eLipids metabolism-related gene set (REACTOME_METABOLISM_OF_LIPIDS) was obtained in Molecular Signatures Database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.broad.mit.edu/gsea/msigdb/\u003c/span\u003e\u003cspan address=\"http://www.broad.mit.edu/gsea/msigdb/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003ePPI network construction and Functional annotation of target genes\u003c/h2\u003e \u003cp\u003eNodes with confidence of interactive relationship larger than 0.40 were obtained from the STRING database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.string-db.org/\u003c/span\u003e\u003cspan address=\"http://www.string-db.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), then, the protein-protein interaction (PPI) network was reconstructed by Cytoscape software[\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. Finally, by R package \"clusterProfiler\" [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e], then Kyoto Encyclopedia of Genes and Genomes(KEGG)pathways and Gene Ontology༈GO༉analysis were performed. (pathways with both p and q values were below 0.05 were considered significantly enriched).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eConsensus Clustering of Glioma\u003c/h2\u003e \u003cp\u003eThrough the R package \"Consensus Cluster Plus\", k-means consensus clustering was carried out in698 glioma samples based on the expression profile of these 67 genes [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. In addition, quantitative stability evidence determined the optimal number of the glioma clusters.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003eDifferent Clinical Characteristics analysis among clusters\u003c/h2\u003e \u003cp\u003eThe clinical and pathological characteristics data and Tumor immune cell infiltration levels corresponding to the glioma in TCGA were selected for correlation analysis with different molecular subtypes by Wilcoxon rank-sum or Kruskal\u0026ndash;Wallis rank-sum test depending on the number of comparisons. The results were visualized as a heatmap by R package \"pheatmap\".\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eConstruction of risk signature\u003c/h2\u003e \u003cp\u003eUnivariate COX regression analysis was performed by the \"survival\" package to select genes with prognostic value in glioma from 67 genes (p value\u0026thinsp;\u0026lt;\u0026thinsp;0.05). The Least Absolute Shrinkage and Selection Operator (LASSO) regression algorithm[\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e] was used to formulate a risk signature (The penalty parameter λ was chosen based on a 10-fold cross-validation). Genes and their regression coefficients were obtained according to the minimum λ value. The formula was:\u003c/p\u003e \u003cp\u003eRisk score\u0026thinsp;=\u0026thinsp;exprgene(1)\u0026times;coefficientgene(1)\u0026thinsp;+\u0026thinsp;exprgene(2)\u0026times;coefficientgene(2)+⋯+exprgene(n) \u0026times;coefficientgene(n).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003ePrognosis analysis of risk signature\u003c/h2\u003e \u003cp\u003ePatients were separated into high- and low-risk groups in TCGA-glioma and CGGA cohorts by the median risk score. Then, evaluate the prognostic significance of the risk signature by Kaplan-Meier survival curves (performed through R package \u0026ldquo;survival\u0026rdquo;) in TCGA-glioma cohort and verified in CGGA cohort (log-rank test p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 was considered significant). Next, univariate and multivariate Cox regression analyses (performed through R packages \u0026ldquo;'survival\u0026rdquo; and \u0026ldquo;forestplot\u0026rdquo;) investigated independent prognostic factors, including the risk score in glioma. Subsequently, investigate the specificity and sensitivity of risk score in the prediction of 1-, 3- and 5-year overall survival (OS) by analyzing the receiver operating characteristic (ROC) curve [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eStudent's t or chi-squared tests tested differences in clinicopathological features and the level of TIICs of risk score.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eGene Set Enrichment Analysis\u003c/h2\u003e \u003cp\u003eIn the Molecular Signatures Database, Hallmark and C7 gene sets were downloaded, which were used as the target gene sets to investigate the gene sets associated with risk score in the whole transcriptome of all glioma samples in TCGA performed by the software GSEA-3.0. (NOM p\u0026thinsp;\u0026lt;\u0026thinsp;0.05 and FDR q\u0026thinsp;\u0026lt;\u0026thinsp;0.05 were considered significant).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eImmune infiltration Profile\u003c/h2\u003e \u003cp\u003eThe abundance of TIICs was estimated using CIBERSORT[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e] computational method in low and high-risk groups separately. Pearson correlation analysis and the Wilcoxon test investigate the correlation between TIICs and risk scores in glioma. In addition, for predicting the efficacy of immunotherapy, the correlation between risk score and the expression level of immune checkpoints, including PD-1 and PD-L1, was investigated by the Wilcoxon test.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cp\u003e \u003cb\u003eThe research flow is shown in\u003c/b\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eIdentification of 67 immune- and lipid metabolism-related genes\u003c/h2\u003e \u003cp\u003eExcluding the cases with incomplete clinical information and overall survival time less than 3 months, 598 Cases were in TCGA and 657 cases in CGGA. Samples with higher immune infiltrating levels, including Stromal, Immune, and ESTIMATE score, had worse overall survival (P-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05, Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eA-C). The results showed that it is reasonable for the immune components in TME of glioma to indicate glioma patients' prognosis. A total of 4033 DEGs were obtained among high and low Immune Score groups, and 4629 DEGs were identified between different Stromal Score groups. The top 50 DEGs in immune score and stromal score goups were respectively shown in figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eD, E. Then, we obtained the 67 intersecting genes among lipid metabolism related genes, DEGs of immune score and DEGs of stromal score (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). GO and KEGG analysis results confirmed that these genes are involved in lipid metabolism-related biological processes and pathways. PPI networks also showed that the 67 genes were closely related (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB-D).\u003c/p\u003e\u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eConsensus clustering of glioma samples\u003c/h2\u003e \u003cp\u003eTwo robust clusters with clustering stability increasing between k\u0026thinsp;=\u0026thinsp;2\u0026ndash;9 were determined by k-means consensus clustering of the samples in TCGA-glioma (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e3\u003c/span\u003eA-C). We have also explored the stability of subtyping by heirarchical clustering and partitioning around medoids (figure \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e), and found that K-means can achieve better grouping performance. PCA's (principal component analysis) results showed that different clusters have significantly different components when k\u0026thinsp;=\u0026thinsp;2 (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e3\u003c/span\u003eD). Overall survival analysis showed that the Cluster1 subgroup predicted a better prognosis in glioma patients than the Cluster2 subgroup (P-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05, Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e3\u003c/span\u003eE). Furthermore, we observed that striking differences were determined by consensus clustering in the clinical and molecular features of the two glioma subgroups (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e3\u003c/span\u003eF). In the TCGA database, Cluster2 was significantly correlated with higher Stromal Score, higher Immune Score, 1p19q non-codeletion, IDH wild type, older age group (\u0026gt;\u0026thinsp;45 years old), and higher WHO grade. Moreover, we have validated these results in CGGA cohorts (figure \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003e). These results showed that the clustering based on 67 could distinguish the clinical, pathological, and immune characteristics of glioma samples and predict the prognosis of patients with glioma.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eConstruction of a 16-gene risk signature\u003c/h2\u003e \u003cp\u003eFifty-two genes with prognostic significance were selected through univariate cox regression analysis (P-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05, Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). Then, 16 genes were generated by LASSO regression analysis, including PIK3R6, PLIN2, CYP2E1, PLA2G5, PLBD1, FABP5, PON1, HEXB, CYP17A1, MTMR7, ARSJ, HILPDA, PPARG, FABP7, ANGPTL4, and ACSL6 (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e4\u003c/span\u003eB-D). The risk score of each sample was then calculated using the coefficients and expression value of these 16 genes through the formula listed above. Then by the median value of risk score, all glioma samples are divided into two groups (low and high risk group). There was a significant difference between the two groups about the correlation between risk score and clinicopathological factors of glioma patients in the TCGA and CGGA cohorts, the clinicopathological characteristics and corresponding number of patients were shown in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e\u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e \u003cp\u003eValidation set TCGA RNA-seq cohort\u003c/p\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;587)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c7\" namest=\"c5\"\u003e \u003cp\u003eTraining set CGGA RNA-seq cohort\u003c/p\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;309)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFeatures\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLow-risk Score(n\u0026thinsp;=\u0026thinsp;297)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eHigh-risk Score(n\u0026thinsp;=\u0026thinsp;290)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eP-value\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eLow-risk Score(n\u0026thinsp;=\u0026thinsp;206)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eHigh-risk Score(n\u0026thinsp;=\u0026thinsp;207)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003eP-value\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u0026lt;=45\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e191\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e92\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e132\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e115\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u0026gt;\u0026thinsp;45\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e108\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e198\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e74\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e92\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGender\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.62\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.44\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFemale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e131\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e115\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e96\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e86\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMale\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e166\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e175\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e110\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e121\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWHO Grade\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eII\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e169\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e42\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e72\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e126\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIII\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e127\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e107\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e104\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e57\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIV\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e141\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIDH status\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWT\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e207\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e168\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e147\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMut\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e285\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e83\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e38\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e60\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1p/19q status\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCodel\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e140\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e72\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNoncodel\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e157\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e281\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e134\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e192\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eDifference analysis of risk score among clinicopathological subtypes\u003c/h2\u003e \u003cp\u003eGlioblastoma had a higher risk score than lower grade gliomas in the heatmap. Risk score was positively associated with immune and stromal score, 1p/19q non-codeletion, IDH wildtype, and \u0026gt;\u0026thinsp;45 age group in the training set (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e4\u003c/span\u003eE).\u003c/p\u003e \u003cp\u003eFurthermore, we classified the patients into different groups according to the different clinical and pathological features of samples. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e5\u003c/span\u003e and Figure \u003cspan refid=\"MOESM4\" class=\"InternalRef\"\u003eS4\u003c/span\u003e, in \u0026gt;\u0026thinsp;45 age, higher grade, wildtype IDH, and non-codeletion 1p19q groups, glioma samples had higher risk scores (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). However, there was no significant difference of risk score among different gender in both training and validation cohorts. In addition, in the TCGA cohort, patients in cluster 2 had higher risk scores (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). In the CGGA-B cohort, samples with unmethylated MGMT promoter and accepting radiotherapy had a higher risk score (p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). While accepting Temozolomide chemotherapy or not had no significant difference on risk score (p\u0026thinsp;\u0026gt;\u0026thinsp;0.05).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003ePrognostic value of risk signature constructed\u003c/h2\u003e \u003cp\u003eIn TCGA, patients with higher risk scores had a significantly poor prognosis as shown in the Kaplan-Meier curve performed among all low and high grade gliomas (P-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05, Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e6\u003c/span\u003eA-C), and the area under ROC curves for predicting overall survival were 0.838 (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e6\u003c/span\u003eD). Consistently, patients with higher risk scores were also correlated to worse clinical outcomes in gliomas in CGGA (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e6\u003c/span\u003eE-G). The AUC for predicting survival time in the CGGA dataset is 0.848 (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e6\u003c/span\u003eH). In addition, the risk score was demonstrated as an independent prognostic factor of prognosis in both TCGA and CGGA cohorts by Cox regression analysis (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). These results suggested that the immune- and lipid metabolism-related risk signature could be a prognostic predictor for gliomas with relatively high sensitivity and specificity.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e\u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e \u003cp\u003eUnivariate analysis\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003eMultivariate analysis\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariables\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003eHR\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e95%CI\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003eP-value\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cb\u003eHR\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u003cb\u003e95%CI\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e\u003cb\u003eP-value\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGrade\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4.986\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.873\u0026ndash;6.421\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eP\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.993\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e1.456\u0026ndash;2.729\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eP\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGender\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.011\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.747\u0026ndash;1.368\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.944\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.934\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.682\u0026ndash;1.278\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.669\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4.863\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e3.391\u0026ndash;6.975\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eP\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e2.178\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e1.380\u0026ndash;3.439\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eP\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIDH mutation status\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.090\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.063\u0026ndash;0.129\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eP\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.332\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.194\u0026ndash;0.569\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eP\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e1p19q codeletion status\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.217\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.128\u0026ndash;0.370\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eP\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.572\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.306\u0026ndash;1.071\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0.081\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRisk Score\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.269\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1.225\u0026ndash;1.314\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eP\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.109\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e1.051\u0026ndash;1.171\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003eP\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003eCorrelation of Risk score with the Proportion of TIICs\u003c/h2\u003e \u003cp\u003eTwenty-one kinds of immune cell profiles in low (n\u0026thinsp;=\u0026thinsp;349) and high (n\u0026thinsp;=\u0026thinsp;349) risk glioma samples were constructed(Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e7\u003c/span\u003eA-D). Macrophages M2 (the median value of the component proportion was 33.89%), Monocytes (16.33%), and T cells CD4 memory resting (13.08%) were the prominent TIICs in the high-risk group. T cells regulatory (Tregs) were positively correlated with Macrophages M0 (rho\u0026thinsp;=\u0026thinsp;0.47, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). Meanwhile, Monocytes were significantly negatively correlated with macrophages M0 (r=-0.79, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). In the low-risk score group, the Macrophages M2 (26.65%), Monocytes (26.45%), and T cells CD4 memory resting (11.13%) were the prominent TIICs. As the Fig.\u0026nbsp;\u003cspan refid=\"Fig11\" class=\"InternalRef\"\u003e7\u003c/span\u003eE-I shown, the abundance of T cells regulatory (Tregs) (r\u0026thinsp;=\u0026thinsp;0.25, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and Macrophages M2 (r\u0026thinsp;=\u0026thinsp;0.32, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) were significantly positively correlated with risk score. In contrast, the abundance of NK cells activated (r=-0.37, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and T cells CD4 memory resting (r=-0.16, p\u0026thinsp;\u0026lt;\u0026thinsp;0.05) were significantly negatively correlated with risk score. We verified these results in the CGGA cohort (figure \u003cspan refid=\"MOESM5\" class=\"InternalRef\"\u003eS5\u003c/span\u003e). These results further indicated that the immune activity in TME of glioma might be associated with the level of risk score.\u003c/p\u003e\u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003eRisk score could predict the effect of immunotherapy\u003c/h2\u003e \u003cp\u003eAs Fig.\u0026nbsp;\u003cspan refid=\"Fig13\" class=\"InternalRef\"\u003e8\u003c/span\u003eA-F shown, GSEA analyses were carried out for investigating the immune signature gene sets related to the samples in TCGA-glioma, indicating that the gene sets \"Genes up-regulated in myeloid-derived suppressor cells from colon tumors \", \"Genes down-regulated in activated T lymphocytes \" and \"Genes up-regulated in CD4 T cells \" were positively associated with the high-risk group, while the gene sets \"Genes up-regulated in comparison of naive CD4 T cells versus CD4 and effector memory T cells \", \"Genes up-regulated in comparison of unstimulated NK cells versus those stimulated with IL2\" and \"Genes down-regulated in comparison of wild type CD8 effector T cells \" were negatively related to the high-risk group. According to the ranking of NES values from high to low, the top five genes sets positively enriched in high-risk score group and top five terms positively enriched in low-risk group were shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig13\" class=\"InternalRef\"\u003e8\u003c/span\u003eG.\u003c/p\u003e\u003cp\u003eThen, the expression level of immune checkpoints, including PD-1 and PD-L1, was significantly higher in the high risk group (P-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05, Fig.\u0026nbsp;\u003cspan refid=\"Fig13\" class=\"InternalRef\"\u003e8\u003c/span\u003eH, I), the expression of immune checkpoints is essential for immune escape and treatment with therapies. These results indicated that anti-PD-1 or PD-L1 immunotherapy might have a better therapeutic effect in glioma patients with a higher risk score.\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eMetabolomic associated studies have provided new molecular mechanisms into the malignant process of brain tumors[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. The lipid compositions of tissues are altered in malignant gliomas compared with normal brain tissues, particularly with the exclusive presence of cholesterol esters and significant elevation of unsaturated fatty acids and phosphatidylcholine in tumor tissues[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. Previous studies have identified glucose-related and amino acid-related risk signatures for the biological process in glioma and the survival of glioma patients through bioinformatic profiling. The characteristics of novel lipid metabolism-related biomarkers need a systematical understanding of glioma. Our study first identified a lipid metabolism-related and immune-associated risk signature, which had instructional significance in the differentiation of malignancy degree of tumor and prognosis judgment. A risk score is used for meaningful signature construction as a standard approach [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. Then, we constructed a 16-gene risk signature to connect lipid metabolism, immune microenvironment, and prognosis of glioma. The 16-gene signature could predict glioma's prognosis in different subgroups and distinguish the status of IDH mutation type and 1p/19q co-deletion among tumors. Furthermore, the signature was furtherly confirmed as an independent risk factor for poor prognosis of patients with glioma by performing a cox regression analysis. As a novel biomarker in glioma, risk score may be a better prognostic indictor compared to single vital clinical factor like age, WHO grade, which were correlated with the prognosis of patients[\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. Previous studies have shown that the seven factors including age, histopathology, IDH status, 1p/19q status, radiotherapy, chemotherapy and recurrence were identified as independent prognostic factors for high grade glioma patients[\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. In this study, we confirmed that the lipid metabolism related risk score could also be used as an independent risk factor in gliomas. Moreover, the area under the receiver operating characteristic curve values for overall survival probabilities was 0.838, which is close to the prediction efficiency of previous autophagy related prognosis models[\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eAccumulating shreds of evidence suggest that lipid metabolism regulates the tumor-associated immune cells. The activation program of M1 macrophages tends to utilize glycolysis. However, the M2 macrophages mainly resort to fat oxidation (FAO)[\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. T cells were also influenced by lipid metabolism, including differentiation, activation, and migration. Thus, lipid metabolism could also play an essential role in modulating T cell-mediated immunity. In this study, the risk score was correlated to the abundance of M2 macrophages and T cells (including CD4 and CD8), which indicated that the signature, an immune-associated biomarker, could distinguish between the anti-tumor promoting-tumor escape of characteristics in TIME.\u003c/p\u003e \u003cp\u003eMoreover, the correlation of lipid-metabolism-related risk signature and distribution of TIICs furtherly provides the potential relationship between lipid metabolism and the immune microenvironments in gliomas. Samples with higher risk scores are often associated with higher levels of immune infiltration, which contains mainly inhibitory immune components, such as M2 macrophages. It may be one of the reasons for the poor prognosis of patients in the high-risk group.\u003c/p\u003e \u003cp\u003eImmune checkpoint inhibitor therapy is emerging as a valid treatment for diverse solid tumors, including anti-PD-1/PD-L1. However, the therapeutic efficacy in glioma largely depends on the activity of tumor-killing immune cells and the expression level of an immune checkpoint, and optimal outcomes have been demonstrated by various preclinical studies[\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. Different genomic subtypes or molecular profiles are the main challenges in response to PD-1/PD-L1 checkpoint blockades[\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. Therefore, we detected the expression of immune checkpoints in both low-risk and high-risk groups, respectively. Compared with the high-risk group, the expression level of PD-L1 and PD-1 were significantly decreased in the low-risk group. These results indicated a better efficacy and more sensitivity of immune checkpoint inhibitors (ICIs) therapy for patients with higher risk scores in glioma. However, resistance to ICIs therapy was attributed to macrophage polarization[\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. Due to the high proportion of macrophages in the high-risk group, inducing macrophages to differentiate into M1 tumor-killing types may provide us with new ideas. In addition, with the help of mediators and cell adhesion molecules, it is necessary to promote the homing of peripheral CD4 and CD8 T cells across the blood-brain barrier to improve the level of tumor immune cells infiltration, ultimately improve the therapeutic effect of ICIs. Consequently, we have identified and validated a novel and reliable risk signature, which could accurately predict the effect of ICIs treatment and provide appropriate treatment strategies according to the risk score of different populations with glioma, considering the level of two vital immune checkpoints and the characteristics of immune microenvironment.\u003c/p\u003e \u003cp\u003eHowever, certain limitations should be noted in the present work. Firstly, the present study conducted independent external validation, yet all the changes across glioma cases derived from diverse geographical regions might not be covered, since all information and tissues were retrospectively obtained from public databases. Secondly, there are different microenvironment features among various tumor sites, like the invasion margin or the tumor core. Thirdly, the expression of lipid metabolism-related markers could be different among immune cells and tumor cells, making it difficult to assess the lipid-metabolism and immune status among diverse cells in gliomas. As a result, our results should be further validated in the well-designed, multicenter, prospective studies (including conducting next-generation sequencing for glioma tissues), and the expression pattern of lipid metabolism-related genes may also need to be investigated in the single-cell RNA sequence data.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn conclusion, we identified that the lipid metabolism-related gene set could distinguish gliomas' clinical and molecular features. A 16-genes risk signature developed related to immunity and lipid metabolism was confirmed as a novel biomarker that could predict the prognosis and the effect of immunotherapy for patients with gliomas.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cdiv class=\"DefinitionList\"\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eLGG\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003elower-grade glioma\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eWHO\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eWorld Health Organization\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eTIME\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003etumor immune microenvironment\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eTME\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003etumor microenvironment\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eTCGA\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eThe Cancer Genome Atlas\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eCGGA\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eChinese Glioma Genome Atlas\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eOS\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eoverall survival\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eTIICs\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003etumor immune infiltrating cells\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eICIs\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eImmune checkpoints inhibitors\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eGO\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eGene Ontology\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eKEGG\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eKyoto Encyclopedia of Genes and Genomes\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003ePPI\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003eprotein-protein interaction\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eFig\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003efigure\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv class=\"DefinitionListEntry\"\u003e \u003cdiv class=\"Term\"\u003eTab\u003c/div\u003e \u003cdiv class=\"Description\"\u003e \u003cp\u003etable.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePublicly available datasets were analyzed in this study. This data can be found below:\u003c/p\u003e\n\u003col\u003e\n \u003cli\u003eTCGA, https://www.cancer.gov/;\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eCGGA, http://www.cgga.org.cn/;\u0026nbsp;\u003c/li\u003e\n \u003cli\u003eSTRING, https://string-db.org/cgi/input.pl\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no conflict of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the National Natural Science Foundation of China (No.82072764).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026apos; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eQX Chen and LG Ye contributed to the conception and design of this study. SW Liang and XY Zhang contributed to the analysis and interpretation of data. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe gratefully acknowledge The Cancer Genome Atlas pilot project, which made the genomic data and clinical data of glioma available.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eOstrom QT, Patil N, Cioffi G, Waite K, Kruchko C, Barnholtz-Sloan JS. CBTRUS Statistical Report: Primary Brain and Other Central Nervous System Tumors Diagnosed in the United States in 2013\u0026ndash;2017. Neuro Oncol. 2020;22:iv1\u0026ndash;96.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAldape K, Brindle KM, Chesler L, Chopra R, Gajjar A, Gilbert MR, et al. Challenges to curing primary brain tumours. Nat Rev Clin Oncol. 2019;16:509\u0026ndash;20.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStupp R, Taillibert S, Kanner A, Read W, Steinberg D, Lhermitte B, et al. Effect of Tumor-Treating Fields Plus Maintenance Temozolomide vs Maintenance Temozolomide Alone on Survival in Patients With Glioblastoma: A Randomized Clinical Trial. JAMA. 2017;318:2306\u0026ndash;16.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSampson JH, Gunn MD, Fecci PE, Ashley DM. Brain immunology and immunotherapy in brain tumours. Nat Rev Cancer. 2020;20:12\u0026ndash;25.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAhmed KA, Chinnaiyan P. Applying metabolomics to understand the aggressive phenotype and identify novel therapeutic targets in glioblastoma. Metabolites. 2014;4:740\u0026ndash;50.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGriffin JL, Kauppinen RA. A metabolomics perspective of human brain tumours. FEBS J. 2007;274:1132\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011;144:646\u0026ndash;74.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePavlova NN, Thompson CB. The Emerging Hallmarks of Cancer Metabolism. Cell Metab. 2016;23:27\u0026ndash;47.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan Meer G, Voelker DR, Feigenson GW. Membrane lipids: where they are and how they behave. Nat Rev Mol Cell Biol. 2008;9:112\u0026ndash;24.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBeloribi-Djefaflia S, Vasseur S, Guillaumond F. Lipid metabolic reprogramming in cancer cells. Oncogenesis. 2016;5:e189.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFisher KE, Pop A, Koh W, Anthis NJ, Saunders WB, Davis GE. Tumor cell invasion of collagen matrices requires coordinate lipid agonist-induced G-protein and membrane-type matrix metalloproteinase-1-dependent signaling. Mol Cancer. 2006;5:69.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLuo X, Cheng C, Tan Z, Li N, Tang M, Yang L, et al. Emerging roles of lipid metabolism in cancer metastasis. Mol Cancer. 2017;16:76.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGeng F, Guo D. Lipid droplets, potential biomarker and metabolic target in glioblastoma. Intern Med Rev (Wash D C). 2017;3.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCheng C, Ru P, Geng F, Liu J, Yoo JY, Wu X, et al. Glucose-Mediated N-glycosylation of SCAP Is Essential for SREBP-1 Activation and Tumor Growth. Cancer Cell. 2015;28:569\u0026ndash;81.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWard PS, Thompson CB. Metabolic reprogramming: a cancer hallmark even warburg did not anticipate. Cancer Cell. 2012;21:297\u0026ndash;308.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOffer S, Menard JA, P\u0026eacute;rez JE, de Oliveira KG, Chandran VI, Johansson MC, et al. Extracellular lipid loading augments hypoxic paracrine signaling and promotes glioma angiogenesis and macrophage infiltration. J Exp Clin Cancer Res. 2019;38:241.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHu X, Matsumoto K, Jung RS, Weston TA, Heizer PJ, He C et al. GPIHBP1 expression in gliomas promotes utilization of lipoprotein-derived nutrients. Elife. 2019;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eUpton A, Arvanitis TN. Using evolutional properties of gene networks in understanding survival prognosis of glioblastoma. IEEE J Biomed Health Inf. 2014;18:810\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRitchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015;43:e47.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, et al. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 2003;13:2498\u0026ndash;504.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYu G, Wang L-G, Han Y, He Q-Y. clusterProfiler: an R package for comparing biological themes among gene clusters. OMICS. 2012;16:284\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu Y-Q, Chai R-C, Wang Y-Z, Wang Z, Liu X, Wu F, et al. Amino acid metabolism-related gene expression-based risk signature can better predict overall survival for glioma. Cancer Sci. 2019;110:321\u0026ndash;33.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGao J, Kwan PW, Shi D. Sparse kernel learning with LASSO and Bayesian inference algorithm. Neural Netw. 2010;23:257\u0026ndash;64.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHeagerty PJ, Lumley T, Pepe MS. Time-dependent ROC curves for censored survival data and a diagnostic marker. Biometrics. 2000;56:337\u0026ndash;44.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSrivastava NK, Pradhan S, Gowda GAN, Kumar R. In vitro, high-resolution 1H and 31P NMR based analysis of the lipid components in the tissue, serum, and CSF of the patients with primary brain tumors: one possible diagnostic view. NMR Biomed. 2010;23:113\u0026ndash;22.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTosi MR, Tugnoli V. Cholesteryl esters in malignancy. Clin Chim Acta. 2005;359:27\u0026ndash;45.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHughey JJ, Butte AJ. Robust meta-analysis of gene expression using the elastic net. Nucleic Acids Res. 2015;43:e79.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQu S, Qiu O, Hu Z. The prognostic factors and nomogram for patients with high-grade gliomas. Fundamental Research [Internet]. 2021;1:824\u0026ndash;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.sciencedirect.com/science/article/pii/S2667325821001345\u003c/span\u003e\u003cspan address=\"https://www.sciencedirect.com/science/article/pii/S2667325821001345\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQu S, Liu S, Qiu W, Liu J, Wang H. Screening of autophagy genes as prognostic indicators for glioma patients. Am J Transl Res [Internet]. e-Century Publishing Corporation; 2020;12:5320\u0026ndash;31. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://pubmed.ncbi.nlm.nih.gov/33042422\u003c/span\u003e\u003cspan address=\"https://pubmed.ncbi.nlm.nih.gov/33042422\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuang SC-C, Everts B, Ivanova Y, O\u0026rsquo;Sullivan D, Nascimento M, Smith AM, et al. Cell-intrinsic lysosomal lipolysis is essential for alternative activation of macrophages. Nat Immunol. 2014;15:846\u0026ndash;55.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePearce EL, Pearce EJ. Metabolic pathways in immune cell activation and quiescence. Immunity. 2013;38:633\u0026ndash;43.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCloughesy TF, Mochizuki AY, Orpilla JR, Hugo W, Lee AH, Davidson TB, et al. Neoadjuvant anti-PD-1 immunotherapy promotes a survival benefit with intratumoral and systemic immune responses in recurrent glioblastoma. Nat Med. 2019;25:477\u0026ndash;86.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBerghoff AS, Kiesel B, Widhalm G, Rajky O, Ricken G, W\u0026ouml;hrer A, et al. Programmed death ligand 1 expression and tumor-infiltrating lymphocytes in glioblastoma. Neuro Oncol. 2015;17:1064\u0026ndash;75.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSoto-Pantoja DR, Wilson AS, Clear KY, Westwood B, Triozzi PL, Cook KL. Unfolded protein response signaling impacts macrophage polarity to modulate breast cancer cell clearance and melanoma immune checkpoint therapy responsiveness. Oncotarget [Internet]. Impact Journals LLC; 2017;8:80545\u0026ndash;59. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://pubmed.ncbi.nlm.nih.gov/29113324\u003c/span\u003e\u003cspan address=\"https://pubmed.ncbi.nlm.nih.gov/29113324\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Lipid metabolism, immune profiling, risk signature, immunotherapy, bioinformatics","lastPublishedDoi":"10.21203/rs.3.rs-4408953/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4408953/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eLipid metabolism reprogramming is critical in various biological processes and is considered a hallmark in cancers. The expression data of mRNA and corresponding follow-up information were obtained from TCGA used as a training set and the CGGA used as a validating set. Based on the expression of genes involved in lipid metabolism, 550 glioma samples of the training set were clustered by unsupervised classification method. Then, we construct a lipid metabolism-related risk signature based on the Lasso regression algorithm. The biological mechanism related to risk score was investigated by gene sets enrichment analysis (GSEA). 67 lipid metabolism- and immune-related genes were identified. Two robust groups were yielded by consensus clustering of the 550 samples. Subgroup2 correlated with a significantly better clinical outcome compared with Subgroup1. A 16-genes risk signature was constructed, and the overall survival of patients is dramatically better in the low-risk than the high-risk group. Consistently, the 16-gene signature showed pretty prognostically predicting ability by the receiver operating characteristic curve with areas under curve more than 0.8 in both TCGA and CGGA. Furthermore, the risk score was identified as an independent prognostic factor for glioma. Moreover, samples with a high-risk score were correlated with a higher level of immune infiltration and associated with a higher expression of immune checkpoints, which indicated an inhibitory tumor immune microenvironment. Our study demonstrated a new sight of lipid metabolism-related and immune-associated genes and constructed a 16-gene risk signature to predict prognosis and immunotherapy for glioma patients.\u003c/p\u003e","manuscriptTitle":"A lipid metabolism and immune-related gene signature for predicting the prognosis of human gliomas","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-05-31 20:25:52","doi":"10.21203/rs.3.rs-4408953/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"007f2b56-9f4c-44c7-92bd-fb90f705bb1d","owner":[],"postedDate":"May 31st, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-07-18T06:31:51+00:00","versionOfRecord":[],"versionCreatedAt":"2024-05-31 20:25:52","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4408953","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4408953","identity":"rs-4408953","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}